article
stringlengths 0
745k
| abstract
stringlengths 0
92.7k
|
|---|---|
ly@xmath1 blobs ( labs ) are radio - quiet , giant ( @xmath13@xmath14 kpc ) ly@xmath1 nebulae often discovered in overdense regions of star - forming galaxies at high redshifts .
this new population of labs may be related to important physical processes in galaxy formation . however , the physical origin of the extended ly@xmath1 nebulae is mysterious .
although similar giant ly@xmath1 nebulae are often seen around high redshift radio galaxies ( hzrgs ) , they are thought to be related to their radio jets and are presumably not the same population as the radio - quiet labs .
there are at least three possibilities for the origin of labs : ( 1 ) ly@xmath1 cooling radiation resulting from gravitational heating , ( 2 ) photoionization by metal free massive ( popiii ) stars , or obscured starburst and agn @xcite , and ( 3 ) shock heating by starburst - driven galactic superwind @xcite .
recent observations have revealed that labs show large ( @xmath15 500 km s@xmath7 ) ly@xmath1 velocity widths @xcite and that a significant fraction of labs display high luminosity infrared dust continuum emission @xcite .
these properties suggest ( 2 ) and/or ( 3 ) as the possible origin of the labs , and a close relationship to the formation of massive galaxies .
ssa 22 lab1 at @xmath2 is the best suitable target to reveal the origin of labs .
its spatial extent of @xmath16 kpc and a ly@xmath1 luminosity of @xmath17 ergs s@xmath7 places this as one of the largest and brightest objects among the known labs @xcite . in a very deep ly@xmath1 image taken with the subaru prime - focus camera ( suprime - cam ) , bubble like structures are seen in lab1 , which may be evidence for galactic superwind @xcite .
@xcite demonstrated in their numerical simulations that such bubbles can be produced by galactic superwind from a proto - galaxy . in their model , at first , multiple gas clumps are distributed throughout a dark matter halo .
subsequently , starbursts occur coincidently in these clumps through interactions or mergers , and drive large scale gas outflows .
the system evolves into an elliptical galaxy at the present epoch .
ssa 22 lab1 has been observed @xcite with the submillimeter common - user bolometer array ( scuba ) on the james clerk maxwell telescope .
an unresolved submillimeter source was detected towards lab1 with a @xmath10 flux density of @xmath18 mjy , which makes this one of the brightest submillimeter sources at high redshifts detected to date @xcite . however , the angular resolution of @xmath9 of the scuba is insufficient to precisely identify the optical counterpart and to investigate the spatial extent of the star - formation activity in lab1 .
although several faint optical components were detected near the center of lab1 with hubble space telescope ( hst ) observations @xcite , it has not been confirmed that they are optical counterparts to the submillimeter source yet .
a radio source was detected with a 1.4 ghz flux density of @xmath19jy ( @xmath20 ) at one of the optical components , j1 , which is possibly the counterpart to the submillimeter source @xcite .
however , a very red @xmath21 band source ( @xmath22 ) was detected at another optical component , j2 @xcite , which is @xmath0 from j1 .
very recently , j2 was also identified as a mid - infrared source with spitzer / irac observations and it is also possibly the counterpart to the submillimeter source @xcite .
it is also possible that the submillimeter emission of lab1 may not be associated with compact source(s ) , but has a large spatial extent due to galactic superwind similar to the nearby starburst galaxy , m82 , in which the distribution of dust is spread out beyond the central ( optical ) starburst region . in order to examine the spatial extent of star - formation activity and to seek the relation between the production of labs and galactic superwinds , higher angular resolution and more sensitive submillimeter observations
are required . the submillimeter array ( sma , * ? ? ?
* ) provides an angular resolution of @xmath0 , which is about eight times higher than that with scuba .
it is also sensitive enough to detect bright submillimeter sources at high redshifts in a full night of observation . for these reasons ,
the sma is a very powerful instrument to examine the morphology of submillimeter sources at high redshifts .
for example , @xcite have successfully detected submillimeter emission at the @xmath23 level for two 20 mjy submillimeter galaxies ( smgs ) , smm j123711 + 6222212 and mips j142824.0 + 352619 , using the sma , and they obtained firm upper limits of @xmath24 to the source sizes . in this letter we report a result of @xmath0 resolution interferometric submillimeter observation of ssa22 lab1 with the sma and discuss the nature of the object .
we assume a flat universe of @xmath25 , @xmath26 , and @xmath27 km s@xmath7 mpc@xmath7 ( @xmath28 corresponds to a physical length of 7.6 kpc at @xmath2 ) .
observations were carried out on 2006 september 24 ( 1 track , 4.5 hours on source ) using eight antennas in the compact configuration of the sma .
the sis receivers were tuned to a center frequency of 345 ghz ( 869@xmath29 ) in the upper sideband ( usb ) , yielding 335 ghz ( 895@xmath29 ) in the lower sideband ( lsb ) .
the sma correlator had 2 ghz total bandwidth in each sideband .
the adopted phase reference center of the target source was @xmath1 ( j2000 ) @xmath30 , @xmath31 ( j2000 ) @xmath32 @xcite .
the track was taken under good atmospheric opacity ( i.e. @xmath33 ) .
the range of ( unprojected ) baseline lengths was 16 69 meters .
the sma data were calibrated using the caltech software package mir , modified for the sma .
antenna based passband calibration was done using two bright qsos , j2225@xmath340457 ( 3c446 , 1.4 jy ) and j2148 + 0657 ( 1.0 jy ) , as well as uranus ( 69.0 jy ) and callisto ( 5.0 jy ) .
antenna based time - dependent complex gain calibration was carried out using two qsos , j2225@xmath340457 ( 1.4 jy ; 6@xmath35 away from the target ) and j2148 + 0657 ( 1.0 jy ; 10@xmath35 ) .
the astrometry was checked by applying the phase calibration to a nearby qso j2218@xmath340335 ( 0.2 jy ; 4@xmath35 ) and we assess the accuracy of the astrometry to be @xmath36 .
the astrometric accuracy of all of these calibrators is better than 10 milliarcsec , and the positional reference is the international celestial reference frame ( icrf , * ? ? ?
finally , absolute flux calibration was performed using callisto .
data calibrations were carried out by two independent groups , which allowed us to confirm the repeatability and the robustness of the results .
imaging was carried out in miriad @xcite .
the natural weighted beam gave a synthesized beam size of @xmath4 ( fwhm , p.a .
= @xmath37 ) .
the rms noise after combining the two sidebands ( effectively at @xmath38 ) is 1.4 mjy .
the fwhm of the primary beam is @xmath8 .
figure [ map ] shows the 880@xmath29 continuum map of ssa 22 lab1 .
no significant emission is detected at the 3@xmath6 limit of 4.2 mjy beam@xmath7 in the field of view of @xmath8 .
if we adopt a dust temperature of @xmath39 k and a power - law emissivity spectral index of @xmath40 @xcite , the extrapolated @xmath3 flux density of lab1 is 16.2 mjy from the scuba @xmath10 flux density of 17.4 mjy . if there were a single unresolved source , it would be detected at @xmath41 level in the @xmath3 map . thus the non - detection suggests the existence of a spatially extended submillimeter emission associated with lab1 assuming that the scuba detection is real .
alternatively , if there are multiple compact sources , the flux density of each source is lower than the 3@xmath6 limit of 4.2 mjy beam@xmath7 and therefore at least four sources are needed to produce the total flux density of 16.2 mjy .
widely spaced distribution of the sources is also required to satisfy the @xmath42 limit ; the separation of the sources should be larger than the beam size of @xmath0 ( 15 kpc ) .
we note that what appears like artifacts in the map actually arises from high side - lobe level that extends n - s of the synthesized beam .
the synthesized beam has up to 70% side - lobes because the source is at zero declination .
the noise peaks in the resultant maps also carry this high side - lobe , resulting in structure that appears like diagonal ridges .
the theoretical noise level derived using the observed sma parameters are consistent with the observed rms of 1.4 mjy . in order to check the appearance of spatially extended sources in the sma 880@xmath29 map
, we carried out simulations of the sma observations using @xmath43 in miriad ( figure [ sim ] ) . in this simulation , we consider three model sources , each of which has a single gaussian profile with the fwhm of 1 , 4 , and 8 .
we assume that the sources have a total flux density of 16.2 mjy intrinsically and are placed at the phase center of the maps .
we use the same observational parameters ( i.e. the configuration of the sma antennas , the coordinates of the source , and the hour angle range ) for this simulation . the rms noise of 1.4 mjy beam@xmath7 in the simulated maps is the same as that of the observed map .
the simulation shows that we could detect the source with fwhm of 1 at 10@xmath6 level , and with fwhm of 4 at 3@xmath6 level in the observed map .
however , the source is resolved out when the diameter of the source is larger than fwhm of 8 .
note that these simulations do not include possible pointing and phase calibration errors .
thus , a conservative lower limit of the fwhm of the spatial extent of the submillimeter emission is @xmath11 .
since the source is unresolved in the previous @xmath9 resolution scuba map @xcite , a conservative upper limit of the fwhm is @xmath9 .
the fwhm of the submillimeter emission associated to lab1 should range from @xmath11 to @xmath9 , which corresponds to @xmath44 kpc at @xmath2 , if it is indeed a single source .
figure [ uv ] shows the visibility amplitude versus projected baseline with @xmath45 error bars . assuming the source position is at the phase center
, we calculate the vector averaged amplitudes from the visibility data in circular bins of @xmath46 , @xmath47 , @xmath48 , and @xmath49 k@xmath50 .
each bin contains about 3000 8000 raw visibility points .
the visibility amplitude shows a marginal ( 2.7@xmath6 ) excess at the bin that represents data for @xmath46 k@xmath50 .
we also plot the total @xmath3 flux density extrapolated from the scuba measurement ( @xmath51 mjy ) at the projected baseline of zero . in order to constrain the spatial extent of the submillimeter emission
, we fit a gaussian profile to the visibility data and the total flux density data .
the best fit gaussian source has a flux density of @xmath52 mjy and a fwhm of @xmath53 , which corresponds to @xmath54 kpc at @xmath2 .
the suggested source size of @xmath55 from the visibility data is consistent with the range of the source size of @xmath56 expected from the simulations . to increase the detectability of the extended submillimeter emission
, we tapered the visibility data with a gaussian that has a fwhm of @xmath57 ( figure [ taper ] ) .
the synthesized beam size of the map is @xmath58 and the rms noise is 2.8 mjy beam@xmath7 .
however , we could not detect the extended submillimeter emission in the map at the 3 @xmath6 limit of 8.4 mjy beam@xmath7 .
the peak flux density of 6.0 mjy beam@xmath7 ( 2.1 @xmath6 ) is expected for the source with the fwhm of @xmath55 and the total flux density of 16.2mjy .
thus the non - detection in the lower resolution map is consistent with the expected peak flux density of the extended submillimeter emission . in summary ,
the data suggest that the spatial extent of the submillimeter emission of lab1 should be larger than @xmath11 ( @xmath12 kpc ) , if the source has a gaussian profile .
however , the s / n of the data is too low to distinguish whether the submillimeter emission comes from a single extended source , widely spaced multiple compact sources , or a combination of both .
we discuss possible interpretations of the spatially extended submillimeter emission of lab1 .
one possible ( and probably most likely ) interpretation is spatially extended star - formation in lab1 .
there is evidence to support this interpretation .
it seems that the uv continuum emission is also spatially extended in lab1 .
figure [ spcam ] ( left ) shows @xmath59-band ( uv continuum at the rest frame of lab1 ) image contours superposed on the 880@xmath29 map .
the resolution of the @xmath59-band image is @xmath28 .
the astrometric system of the image is defined by the reference frame of the 2mass all - sky catalog of point sources @xcite , and the typical rms error is less than @xmath60 . in figure [ spcam ] ( left ) , we label uv continuum sources that are possibly associated with lab1 , c11 @xcite , j1 4 sources @xcite , and a1 6 .
the uv continuum sources are widely spaced and have the median separation of @xmath61 .
the spectroscopic redshift of c11 was measured to be the same as lab1 @xcite .
there is a marginal ( @xmath62 ) excess of submillimeter emission at the position of c11 , suggesting that c11 is one of the counterparts of the submillimeter emission .
a2 shows a marginal narrow - band deficit ( @xmath63 mag ) , which suggests it has ly@xmath1 absorption at @xmath2 and is associated with lab1 .
we derived the photometric redshifts of the uv continuum sources inside the ly@xmath1 nebula using palomar @xmath64-band image @xcite and suprime - cam @xmath65 , @xmath66 , @xmath59 , @xmath67 , and , @xmath68 bands images @xcite with hyperz .
we found that the sources brighter than @xmath69 magnitude ( i.e. c11 , j1 3 , and a1 3 ) show photometric redshifts between @xmath70 and @xmath71 .
these redshifts are consistent with @xmath2 , if we consider the estimated uncertainty of the photometric redshifts of @xmath72
. moreover diffuse uv continuum emission seems to connect these sources with each other .
the diffuse uv continuum emission may be tidal tails and bridges due to interactions of these sources .
the photometric redshifts of the uv continuum sources and the connecting diffuse uv continuum emission suggest that most of the uv continuum sources are associated with lab1 .
it is possible that the distribution of the submillimeter emission is similar to that of the uv continuum emission .
we derive a star - formation rate ( sfr ) in lab1 of 1400 @xmath73 yr@xmath7 from the far - infrared ( fir ) luminosity of @xmath74 @xcite and the relation between far - infrared ( fir ) luminosity and sfr , @xmath75 .
the large sfr of lab1 suggests that intense starbursts occur in lab1 .
we also derive an average sfr surface density of @xmath76 @xmath73 yr@xmath7 kpc@xmath77 from dividing the sfr by the area of the submillimeter emission , @xmath78 , where @xmath79 is a half size of the fwhm of the submillimeter emission of 40 kpc .
the average sfr surface density of lab1 is comparable to those of local starburst galaxies estimated from fir luminosity and h@xmath1 size @xcite .
thus the large spatial extent of the submillimeter emission and the high average sfr surface density suggest that starbursts occur throughout the large area in lab1 .
we also derive a sfr in lab1 of 220 @xmath73 yr@xmath7 from the uv luminosity density at @xmath80 of @xmath81 for the uv continuum emission inside the ly@xmath1 nebula and the relation between uv luminosity density and sfr , @xmath82 .
we do not correct for dust extinction in this calculation .
the sfr derived from the uv continuum emission is about an order of magnitude lower than that derived from the submillimeter emission .
the difference suggests that most of the uv continuum emission produced from star - formation activity in lab1 is attenuated by dust .
there may be buried , unidentified uv continuum sources in lab1 .
it is worth noting here that recent spitzer / mips @xmath83 observations have also given us a hint that labs have multiple dusty sources .
@xcite detected two or three @xmath83 sources at the positions of two labs at @xmath84 , which are possibly associated with the labs .
the separations of these @xmath83 sources are @xmath85 kpc , similar to or somewhat larger than the separations of the possible optical counterparts ( i.e. uv continuum sources ) of lab1 .
the widely spaced dusty sources may be a common feature of labs .
some part of the submillimeter emission can be interpreted as thermal emission from spatially extended dust expelled from starburst regions by galactic superwind .
since lab1 shows possible evidence for galactic superwind ( i.e bubble like structures seen in its ly@xmath1 nebula ) , it may have a large amount of dust entrained in gas outflows .
figure [ spcam ] ( right ) shows ly@xmath1 image contours superposed on the 880@xmath29 map . while the s / n of the submillimeter emission is low and may not be adequate for conclusive determinations , the submillimeter features seem to correlate with the high surface brightness regions of ly@xmath1 emission .
the possible correlation in distributions between submillimeter and ly@xmath1 emissions is consistent with this interpretation .
@xcite found a ly@xmath1 velocity gradient around c11 and argued that a bipolar outflow seen in m82 is a local analog for c11 .
in fact , m82 shows not only an extended h@xmath1 emission - line nebula associated with the bipolar gas outflow @xcite but also extended dust emission .
@xcite found that @xmath86 dust emission extends to 6 kpc from the central starburst region of m82 , and the extended emission accounts for about one - third of the total @xmath86 flux with irac observations .
thus , it is possible that the thermal emission from the spatially extended dust expelled from starburst regions also contributes the spatially extended submillimeter emission of lab1 . in this picture ,
some of the extended ly@xmath1 and uv continuum emission of lab1 may be light scattered by spatially extended dust . the sunyaev - zeldovich ( s - z ) increment is another possible interpretation for the spatially extended submillimeter emission . there is evidence that lab1 lies in a deep dark matter potential well made by the proto - cluster at @xmath2 @xcite .
the dynamical mass of lab1 is estimated to be @xmath87 m@xmath88 from the size and the velocity width of ly@xmath1 emission @xcite .
however , a 450@xmath29 emission with a flux density of @xmath89 mjy was detected toward lab1 and the measured 450/850 @xmath29 ratio is an order of magnitude higher than the expected one from the s - z increment @xcite .
thus , the s - z increment is not a likely interpretation .
the _ non - detection _ of submillimeter emission in the sma map may be due to time - variability of dust emission around a buried agn , but past observational results provide evidence that this is unlikely .
the spectral energy distribution of lab1 suggests that the contribution of non - thermal emission to submillimeter wavelength must be negligible .
the sma observations of lab1 are about five years after the scuba observations . if the source shows time - variability , the submillimeter flux density had to decrease by a factor of four or more in five years ( one year in rest frame ) . in order to produce the time - variability in such time - scale , the required size of dust distribution around a nucleus is less than about one light year ( @xmath90 pc ) . taking the fir luminosity of @xmath74 and the dust temperature of @xmath39 k , we estimate the minimum source size of @xmath91 kpc . for the optically thin case ,
the source size is estimated to be much larger .
thus , the minimum source size is inconsistent with the required source size for time - variability . in order to investigate the relation of lab1 to other luminous submillimeter populations at high redshifts such as smgs and hzrgs
, we compare the spatial extents of the submillimeter emission of these objects .
the spatial extents of the submillimeter emission of smgs are 5 20 times smaller than that of lab1 .
@xcite estimated the median submillimeter source size to be @xmath92 ( 4 kpc ) from @xmath93 resolution millimeter continuum and co imaging of eight smgs at @xmath94 with the iram plateau de bure interferometer ( pdbi ) .
@xcite directly constrain the upper limits of the submillimeter source size of @xmath24 from @xmath0 resolution submillimeter imaging of two luminous smgs with the sma .
although some of the smgs are resolved into two compact sources @xcite , most smgs are single compact sources .
thus the majority of the smgs appear to be much more compact in submillimeter than lab1 .
the spatial extent of the submillimeter emission of hzrgs is similar to or somewhat larger than that of lab1 .
high resolution millimeter observations with the iram pdbi revealed the spatially extended dust and gas in hzrgs .
@xcite found that the millimeter and co emissions in a radio galaxy 4c 60.07 at @xmath95 extends over @xmath96 kpc .
also found that evidence for spatially resolved dust emission at scales of @xmath97 to @xmath57 in b3 j2330 + 3927 at @xmath2 .
@xcite found that five of seven hzrgs are resolved even with the @xmath9 scuba beam and the fwhm of the submillimeter emission ranges from 50 to 250 kpc . since hzrgs are known to lie in overdense regions as well as lab1 , both of hzrgs and lab1 may have spatially extended starbursts induced by interactions and mergers of gas clumps in overdense environments at high redshifts .
the @xmath0 resolution submillimeter observations of ssa22 lab1 provide evidence that the associated submillimeter emission extends larger than @xmath11 ( @xmath13 kpc ) . the most likely interpretation for the extended submillimeter emission
is spatially extended star - formation in lab1 .
the derived sfr surface density suggests that starbursts occur throughout the large area in lab1 .
the hint of a possible correlation in distributions between submillimeter and ly@xmath1 emissions suggests that the thermal emission from the spatially extended dust expelled from starburst regions by galactic superwind may contribute the spatially extended submillimeter emission of lab1 .
the spatial extent of submillimeter emission of lab1 is similar to those of hzrgs rather than smgs .
further data will be required to determine whether spatially extended submillimeter emission is a common feature of labs and to investigate the relationship between radio - quiet labs and hzrgs .
we thank takashi hattori , tomoki saito , masao mori , keiichi umetsu , koichiro nakanishi , masakazu kobayashi , thomas greve , masayuki umemura , and hiroyuki hirashita for insightful discussion .
we also thank ian smail for encouragement .
this work is supported by the grant - in - aid for the 21st century coe `` center for diversity and universality in physics '' from the ministry of education , culture , sports , science and technology ( mext ) of japan .
osterbrock , d. e. 1989 , research supported by the university of california , john simon guggenheim memorial foundation , university of minnesota , et al .
mill valley , ca , university science books , 1989 , 422 p. , | we present @xmath0 resolution submillimeter observations of the submillimeter luminous giant ly@xmath1 blob ( lab1 ) in the ssa 22 protocluster at redshift @xmath2 with the submillimeter array ( sma ) .
although the expected submillimeter flux density is 16 mjy at @xmath3 , no emission is detected with the @xmath4 ( @xmath5 kpc ) beam at the 3@xmath6 level of 4.2 mjy beam@xmath7 in the sma field of view of @xmath8 .
this is in contrast to the previous lower angular resolution ( @xmath9 ) observations where a bright ( 17 mjy ) unresolved submillimeter source was detected at @xmath10 toward the lab1 using the submillimeter common - user bolometer array on the james clerk maxwell telescope .
the sma non - detection suggests that the spatial extent of the submillimeter emission of lab1 should be larger than @xmath11 ( @xmath12 kpc ) .
the most likely interpretation of the spatially extended submillimeter emission is that starbursts occur throughout the large area in lab1 .
some part of the submillimeter emission may come from spatially extended dust expelled from starburst regions by galactic superwind .
the spatial extent of the submillimeter emission of lab1 is similar to those of high redshift radio galaxies rather than submillimeter galaxies . |
we report results of _ ab initio _ calculations of excitation energies , oscillator strengths , and transition rates in yb - like ions with nuclear charges @xmath2 ranging from 72 to 100 .
the ions considered here , starting from doubly ionized hf iii , all have @xmath4 ground states .
we do not consider yb i and lu ii both of which have a @xmath5 ground state configuration . in recent publications @xcite , the spectra of re vi , os vii , and ir viii were studied and energies levels of the @xmath4 , @xmath6 , @xmath7 , @xmath8 and @xmath9 configurations were determined .
the cowan hartree - fock code @xcite with relativistic and correlation options was used in refs .
@xcite to calculate energy levels and to carry out least - squares adjustments of energies .
although we do not consider yb i and lu ii here , it should be noted that @xcite recently carried out elaborate calculations of electric - dipole amplitudes in atomic ytterbium .
moreover , @xcite listed energies for 249 levels of yb i and 40 levels of lu ii . in the present paper ,
we use relativistic many - body perturbation theory ( rmbpt ) to determine energies of the 14 even - parity @xmath4 , @xmath6 , and @xmath5 states and the 36 odd - parity @xmath7 , @xmath8 , and @xmath9 states for yb - like ions .
we illustrate our calculation with a detailed study of re vi , @xmath2= 75 .
our first - order rmbpt calculations include both the coulomb and retarded breit interactions but our second - order calculations are limited to the coulomb interaction only .
reduced matrix elements , line strengths , oscillator strengths , and transition rates are determined for all allowed and forbidden electric - dipole transitions between even - parity ( @xmath10 ) and odd - parity ( @xmath11 ) states .
retarded e1 matrix elements are evaluated in both length and velocity forms .
rmbpt calculations that start from a local potential are gauge independent order - by - order , provided `` derivative terms '' are included in second- and higher - order matrix elements and careful attention is paid to negative - energy states .
the present calculations start from a nonlocal [ xe]@xmath3 dirac - fock ( df ) potential and consequently give gauge - dependent transition matrix elements .
second - order correlation corrections compensate almost exactly for the gauge dependence of the first - order matrix elements and lead to corrected matrix elements that differ by less than 5% in length and velocity forms for all of the ions considered here .
energies from the present calculation agree well with results given in refs .
@xcite for low - lying levels , but disagree substantially for various highly - excited levels , as discussed later .
.[tab0 ] possible two - particle states in the yb - like ions [ cols= " < , < , < , < " , ] [ tab - com ] in table [ tab - com ] ,
transition energies are compared with recent measurements @xcite .
we obtain good agreement with the experiment for low - lying @xmath4 , @xmath6 , and @xmath7 levels .
we also we obtain reasonable agreement ( 500 -1000 @xmath12 ) for highly - excited @xmath8 levels in re
vi , os vii , and ir viii ions .
however , we disagree substantially ( 2000 -10000 @xmath12 ) for @xmath8 levels in pt ix , au x , and hg xi ions . as mentioned in refs .
@xcite , mixing between levels of the @xmath8 configuration and the core excited configurations with a @xmath13 hole could be very important for pt ix , au x , and hg xi ions .
indeed , substantial mixing of the @xmath14 and @xmath15 configurations was found in ref .
@xcite . from this , we conclude that the present values for @xmath8 configurations in pt ix , au x , and hg xi are less reliable ( @xmath16 1% ) than data for re vi , os vii , and ir viii .
the fine - structure intervals for the @xmath17 , @xmath18 , and @xmath19 terms of @xmath7 configuration divided by @xmath20 are shown in fig .
[ spl - pd ] .
the fine structures of these levels do not follow the land rules even for small @xmath2 ; the @xmath17 levels are partially inverted , while the @xmath18 and @xmath19 levels show regular ordering of the fine - structure splittings for both low and high @xmath2 .
the unusual splittings are caused by changes from @xmath21 to @xmath22 coupling and by mixing from other triplet and singlet states .
comparisons are made with experimental data from @xcite in fig .
[ spl - pd ] .
excellent agreement ( 0.3% 3% ) is found for the four intervals @xmath23@xmath24 , @xmath25@xmath26 , @xmath27@xmath28 , and @xmath28@xmath29 .
the agreement decreases for the @xmath30@xmath23 and @xmath26@xmath31 intervals , from 10% for re@xmath32 to 2% for hg@xmath33 .
experimental energies for other ions would be very helpful in confirming the @xmath2 dependence shown in fig .
[ spl - pd ] .
trends of the @xmath2 dependence of line strengths and oscillator strengths , are shown in figs .
[ fig - line ] and [ fig - osc ] . in fig .
[ fig - line ] , we illustrate the @xmath2 dependence of line strengths for the four @xmath4@xmath7 transitions calculated ( a ) by rmbpt and ( b ) in first - order to illustrate once again the importance of including second - order matrix elements . comparing the curves in fig .
[ fig - line]a and fig .
[ fig - line]b , we can see that the @xmath2 dependence is the same for both cases except for the small region @xmath2 = 8088 .
the singularities in this region are already present for uncoupled dipole matrix elements .
the second - order contribution for the @xmath34@xmath35 uncoupled matrix element is shown in fig .
[ dip - lv ] .
similar singularities appeared for other uncoupled matrix elements between even - parity @xmath36=2 states and odd - parity @xmath36=3 states .
consequently , those singularities also appear in the coupled matrix elements and the line strengths shown in fig .
[ fig - line ] .
the @xmath2 dependence is more complicated for transitions between even - parity states with @xmath36=1 and odd - parity states with @xmath36=0 , as seen in fig .
[ fig - osc ] , where we give oscillator strengths for four transitions calculated ( a ) by rmbpt and ( b ) in first - order .
the oscillator strength curves are not smooth functions of @xmath2 , even in first - order .
the singularities shown in the four curves of fig .
[ fig - osc]b arise from the strong mixing between three states of the odd - parity complex with @xmath36=0 .
the mixing of @xmath37 , @xmath38 , and @xmath39 states was discussed earlier in fig .
[ en - odd]a where it was shown that avoided crossings occur near @xmath2=76 and 84 .
in summary , a systematic second - order rmbpt study of the energies of the @xmath4 , @xmath6 , @xmath5 , @xmath7 , @xmath8 , and @xmath9 states of yb - like ions has been presented .
these calculations agree with existing experimental energy data for intermediate @xmath2=75 to 80 at the level of 500 - 1000 @xmath12 for low - lying @xmath4 , @xmath6 , and @xmath7 levels .
they provide a smooth theoretical reference database for the identification of lines
. also presented is a systematic second - order relativistic rmbpt study of reduced matrix elements , line strengths , oscillator strengths , and transition rates for allowed and forbidden electric - dipole transitions in yb - like ions with nuclear charges ranging from @xmath2 = 72 to 100 .
the dipole matrix elements include retardation and correlation corrections from coulomb and breit interactions .
both length and velocity forms of the matrix elements are evaluated , and @xmath40 % differences , caused by the non - locality of the starting df potential , are found between the two forms .
we believe that our results will be useful in analyzing existing experimental data and planning new experiments .
there remains a paucity of experimental data for many of the higher ionized members of this sequence , both for term energies and for transition probabilities and lifetimes .
additionally , matrix elements from the present calculations will provide basic theoretical input for calculations of reduced matrix elements , oscillator strengths , and transition rates in three - electron lu - like ions .
the work of w.r.j . and m.s.s .
was supported in part by national science foundation grant no .
phy-01 - 39928 .
u.i.s . acknowledges partial support by grant no .
b503968 from lawrence livermore national laboratory .
the work of j.r.a was performed under the auspices of the u. s. department of energy by the university of california , lawrence livermore national laboratory under contract no .
we also acknowledge helpful discussions with professor a.n . ryabtsev .
here , we consider the second - order double - excitation contribution to explain singularities in the @xmath2 dependence of energy levels given in figs .
[ en - eve ] and [ en - odd ] . a typical contribution from one of the double - excitation diagram for the second - order interaction energy
has the form @xcite @xmath41 where @xmath42 are products of angular coupling coefficients and slater integrals @xcite , @xmath43 is the single - particle energy for state @xmath44 , and @xmath45 designates a single - particle valence state ( @xmath46 ) . for the case of a [ xe]@xmath3 core ,
sums over @xmath44 and @xmath47 include @xmath48 , @xmath49 , @xmath50 , @xmath51 , @xmath52 , @xmath53 , and all states with principal quantum number @xmath545 .
we must , however , exclude terms with pairs ( @xmath55 ) that are included in model space .
therefore , we remove the pairs ( @xmath56 ) , ( @xmath6 ) , and ( @xmath57 ) since they are in the even - parity model space and pairs ( @xmath7 ) , ( @xmath8 ) , and ( @xmath9 ) in the odd - parity model space . as an example , let us consider the singularity near @xmath2=85 in the curve describing the energy of the @xmath5 level in fig .
[ en - eve]b .
the denominator in eq .
( [ eq1 ] ) is @xmath58 for the matrix element @xmath59 . for an ion with @xmath2=85
, @xmath60 becomes very small when ( @xmath55)=(@xmath61 ) since @xmath62=-9.488404 a.u .
, @xmath63=-12.069226 a.u . , and @xmath64=-6.932272 a.u . , giving @xmath60=0.02469 .
this one term dominates the matrix element @xmath59 and increases its size by a factor of six in comparison with values for neighboring @xmath2 , explaining the singulatity at @xmath2=85 in fig .
[ en - eve]b .
the explanation of the sharp features in the curves describing energies of the @xmath39 and @xmath65 levels in fig .
[ en - odd ] is similar .
a typical contribution from one of the second - order correlation corrections to the dipole matrix element ( @xmath66 has the form @xcite
@xmath67\propto \sum_{i}% \frac{z_{iv}x_{k}(v^{\prime } w^{\prime } wi)}{\epsilon _ { i}+\epsilon _ { w}-\epsilon _ { v^{\prime } } -\epsilon _ { w^{\prime } } } \ , , \ ] ] where @xmath68 is a single - electron dipole matrix element . in the sum over @xmath69 , only terms with vanishing denominators are excluded . for the @xmath70 transition , we obtain @xmath71}\hspace{4em } \nonumber \\ & & \propto \sum_{i}\frac{z(i,5d_{5/2})x_{k}(5d_{5/2}6p_{1/2}\ 5d_{3/2}i)}{% \epsilon _
{ i}+\epsilon ( 5d_{3/2})-\epsilon ( 5d_{5/2})-\epsilon ( 6p_{1/2})}\ , .\end{aligned}\ ] ] for the case @xmath69=@xmath51 and @xmath2=84 , the denominator becomes @xmath72 this one term dominates the entire matrix element . for the @xmath73 transition , the denominator becomes very small at @xmath2=88 when @xmath69=@xmath51 .
@xmath74 again , this single term dominates the matrix element . a typical contributions from the second - order rpa correction for dipole matrix element ( @xmath66 has the form @xmath75\propto \sum_{i}% \frac{z_{nb}x_{k}(wnv'b)}{\epsilon _ { n}+\epsilon _ { w}-\epsilon _ { v'}-\epsilon _ { b}}\ , , \ ] ] where the index @xmath76 designates core states and @xmath47 designates an excited state .
for the special case of the @xmath77 - 6p_{1/2}5d_{5/2}[1]$ ] transition , we obtain @xmath78 } \hspace{2em } \\ & & \propto \sum_{n}\sum_{b}\frac{z(b , n)x_{k}(5d_{3/2}b\ 6p_{1/2}n)}{\epsilon _ { n}+\epsilon ( 5d_{3/2})-\epsilon ( 6p_{1/2})-\epsilon _ { b}}\ , .\end{aligned}\ ] ] in the case of @xmath76=@xmath79 and @xmath47=@xmath48 for nuclear charge @xmath2=88 , the denominator becomes @xmath80 as before , the small value of the denominator leads to an anomalous increase in the size of the rpa matrix element . | excitation energies , oscillator strengths , and transition rates are calculated for @xmath0@xmath1 electric dipole transitions in yb - like ions with nuclear charges @xmath2 ranging from 72 to 100 .
relativistic many - body perturbation theory ( rmbpt ) , including the retarded breit interaction , is used to evaluate retarded e1 matrix elements in length and velocity forms .
the calculations start from a [ xe]@xmath3 core dirac - fock potential .
first - order rmbpt is used to obtain intermediate coupling coefficients , and second - order rmbpt is used to determine matrix elements .
a detailed discussion of the various contributions to energy levels and dipole matrix elements is given for ytterbium like rhenium , @xmath2=75 .
the resulting transition energies are compared with experimental values and with results from other recent calculations .
trends of excitation energies , line strengths , oscillator strengths , and transition rates as functions of nuclear charge @xmath2 are shown graphically for selected states and transitions .
these calculations are presented as a theoretical benchmark for comparison with experiment and theory . |
considerable effort has been made in the past few years to implement qubits in nanoscale solid state structures .
one of the most promising candidates are spin qubits confined in electrostatically defined quantum dots ( qds ) embedded in gaas structures @xcite .
a universal set of quantum gates has been demonstrated in gaas double quantum dots ( dqd ) through the achievements of single - spin rotations and the two - spin exchange interaction that generates the @xmath4 gate @xcite . despite these advances , coherence times are limited by the hyperfine interaction , which couples the trapped electron spin in the quantum dot to the spin-@xmath5 nuclei of the gaas substrate .
the resulting nuclear fields cause rapid electron spin dephasing , leading to an inhomogenous dephasing time @xmath6 . as each electron spin
is coupled to approximately one million nuclei , the resulting behavior of the coupled electron - nuclear spin system is complicated and leads to rich dynamics that are sensitive to experimental parameters @xcite . the hyperfine interaction has traditionally been viewed as a nuisance . however , a recent experiment demonstrates that generation of a controlled nuclear field gradient can be used to drive fast spin rotations @xcite .
the development of quantum control methods in semiconductor quantum dots that are based on nuclear spin interactions could lead to new paradigms for single spin control . .
the spin states for the implementation of the qubit are the hybridized singlet @xmath7 and the triplet @xmath1 .
( b ) a coherent superposition of the qubit states is generated by lzs interferometry .
( c ) the initialized @xmath8 is swept through the avoided crossing by means of an applied linear gate voltage pulse @xmath9 .
the final state is a coherent superposition of @xmath7 and @xmath1 generated by lzs tunneling . at @xmath10 ,
the system evolves in the external magnetic field @xmath11 for a time @xmath12 before a reverse gate voltage pulse brings the system back to @xmath13 where a qpc measurement is performed to determine the singlet state return probability , @xmath14.,scaledwidth=48.0% ] we theoretically show that hyperfine interactions can be harnessed for quantum control in gaas semiconductor quantum dots . in the presence of an external magnetic field @xmath11 , which splits the triplet states , the hyperfine interaction results in an avoided crossing between the spin singlet @xmath7 and spin triplet @xmath15 , which form the basis of a new type of spin qubit . coherent quantum control of this qubit has already been experimentally achieved through landau - zener - stckelberg ( lzs ) interferometry @xcite , a technique previously used to coherently manipulate superconducting qubits @xcite and which is based on the interference due to repeated lzs tunneling events @xcite .
the original lzs problem studies a two level system which exhibits an avoided crossing when an external control parameter is changed .
if the control parameter is time dependent , the system can be brought from an initial state through the avoided crossing .
this passage may result in a change of populations and relative phase of the states . in the lzs version ,
the system is driven from @xmath16 and the difference in energy between the two states is a linear function of time , @xmath17 , which leads to the well - known result for the non - adiabatic transition probability @xmath18 . for the dqd system , the usual infinite - time asymptotic theory describing lzs interferometry can not be used .
the avoided crossing originates from the hyperfine interaction between the electronic spins and the nuclear spins whose fluctuations result in a poorly defined crossing position .
the phenomena observed in the experiments can only be properly described using finite - time lzs theory @xcite . to prove that our formalism describes correctly the coherent manipulation of the @xmath7-@xmath15 based qubit
, we will compare it to the experimentally measurable quantity @xmath14 , the singlet return probability .
we then show how single qubit operations can be engineered either by the euler angle method for rotations or by only using lzs interferometry .
finally , we demonstrate that a two qubit gate can be achieved by capacitively coupling two @xmath7-@xmath15 qubits .
in contrast to the @xmath7-@xmath19 qubit , where the rotation rate is set by a charge - noise - susceptible exchange energy , the rotation rate in the @xmath7-@xmath15 qubit is set by the zeeman energy and approaches @xmath20 for modest magnetic fields of @xmath21 .
the spin preserving hamiltonian , @xmath22 with @xmath23 and @xmath24 describes the coupling between two electrons in a dqd in a magnetic field @xmath11 .
@xmath25 denotes the effective land g - factor ( @xmath26 for gaas ) , @xmath27 the bohr magneton , the @xmath28 and @xmath29 label the dot number and spin .
the first term is the single - particle energy of the confined electrons , the second accounts for the coulomb energy @xmath30 of two electrons on the same qd , and the last for tunneling with strength @xmath31 between the dots .
the diagonalization of the first two terms of eq
. leads to the relevant charge states of a dqd : the singlets @xmath32 , @xmath33 , @xmath34 and triplets @xmath35 where @xmath36 denotes the charge configuration of the dots ( see fig .
[ energy_diagram](a ) ) .
the other states can be neglected as they have energies much higher than those considered here .
the degeneracy of the singlets @xmath33 and @xmath34 at @xmath37 is lifted by the inter - dot tunneling , resulting in a splitting of @xmath38 .
the hyperfine interaction @xmath39 between the electron spins @xmath40 and the nuclear spins @xmath41 opens a splitting @xmath42 at the degeneracy point @xmath43 of the singlet @xmath7 and the triplet @xmath15 . here , @xmath44 is the overhauser ( effective nuclear ) field operator .
the sum runs over the @xmath45 nuclear spins in dot @xmath46 and @xmath47 is the hyperfine coupling constant with the @xmath48-th nucleus in dot @xmath46 , with @xmath49 the electron wave function , @xmath50 the volume of the unit cell and @xmath51 the hyperfine coupling strength . from now on , since we assume symmetric dots , we have @xmath52 . introducing @xmath53 and @xmath54 , we write @xmath55 to determine which spin states are relevant for our theory , we consider @xmath56 from the asymptotic lzs model and the result of ref . for multiple level crossings to estimate the order of magnitude of the transition probabilities .
the initialization of the system is done by preparing a singlet @xmath57 @xmath58 , then @xmath59 is swept to achieve @xmath60 . during this operation , the system goes through three avoided crossings ( cf fig .
[ energy_diagram ] ) .
to estimate the matrix element @xmath61 , we use the experimentally found value of @xmath62 from ref . .
the matrix element entering @xmath56 at the avoided crossing between @xmath57 and the excited singlet state @xmath63 is given by @xmath64 , with @xmath65 .
the order of magnitude of @xmath66 is taken between @xmath67 .
we find @xmath68 and @xmath69 .
these results show that population of the excited singlet and @xmath70 level are negligible .
@xmath71 can also be neglected because it does not cross with any other level , it splits from @xmath72 due to the exchange coupling @xcite .
near the @xmath7-@xmath1 crossing , the dynamics can be restricted to the hilbert space spanned by @xmath73 , @xmath74 , and @xmath75 and described by @xmath76 where we can neglect an additive term @xmath77 , with @xmath78 the detuning of the dots . according to our previous estimate
, we can reduce eq . to a @xmath79 hamiltonian which only takes into account the lowest hybridized singlet @xmath80 and the triplet @xmath81 , where @xmath82 , with @xmath83 .
the energy associated with the lowest hybridized singlet is @xmath84 and the energy of the triplet is @xmath85 .
the hamiltonian describing the dynamics of the lowest energy states in the vicinity of @xmath7-@xmath1 can therefore be written as @xmath86 another relevant quantity derived from eq .
is the degeneracy position @xmath43 of @xmath7 and @xmath15 , given by the funnel - shaped function @xmath87
we model the overhauser field classically , such that it acts on the electron spin as a magnetic field @xmath88 with @xmath89 and its physical properties are given by a statistical distribution that reflects the quantum fluctuations of the nuclear ensemble . at typical operating temperatures and external magnetic fields @xmath90 , where @xmath91 and @xmath92 are the nuclear @xmath93 factor and magneton . in this limit
we can assume the nuclei to be completely unpolarized , resulting in a gaussian distribution of nuclear fields@xcite @xmath94 , with @xmath95 and @xmath96 .
the effective hamiltonian describing the qubit dynamics around the @xmath7-@xmath1 avoided crossing is given by @xmath97 we include the nuclear zeeman spliting in the @xmath1 energy , @xmath98 , where @xmath99 and @xmath100 .
the classical approximation of the overhauser fields is possible because the nuclear state changes only slightly after a single sweep @xcite . to obtain an analytical expression of the lzs propagator
( see appendix [ sec : lzsprop ] ) , we linearize the difference in energy @xmath101 around @xmath102 @xcite by assuming @xmath103 , we find @xmath104 here @xmath105 is the rate at which the external voltage gates are ramped .
control over @xmath66 can therefore be achieved by modifying @xmath105 . to test our model we first compute the singlet return probability @xmath14 , an experimentally observable quantity , as a function of the final detuning @xmath10 and waiting time
@xmath12 , @xmath106 with @xmath107 where @xmath108 $ ] are the backward and forward lzs propagators and @xmath109 $ ] describes the evolution of the system during the waiting time @xmath12 at the final detuning position @xmath10 with @xmath110 . from the semi - classical model as a function of the waiting time @xmath12 and final detuning @xmath10 .
we find nanosecond oscillation periods and the dephasing time @xmath111 , in good agreement with experiment .
we used @xmath112 , @xmath113 , @xmath114 and @xmath65 .
( b ) @xmath14 ( blue ) as a function of @xmath12 for @xmath115 ( horizontal line in ( a ) ) . we plot ( orange ) @xmath116 where @xmath117 , @xmath118 and @xmath119 is extracted from a best fit.,scaledwidth=48.0% ] we evaluate @xmath14 by numerical sampling . instead of estimating @xmath120 from @xmath121 and @xmath122 of the qds , here we use the experimentally determined @xcite @xmath62 to derive @xmath123 .
we have @xmath124 , with @xmath125 . in fig . [ eff_res ]
we show @xmath14 as a function of @xmath10 and @xmath12 for @xmath113 , @xmath126 , @xmath127 , and @xmath65 .
we use a square pulse with a ramping time @xmath128 fixed to @xmath129 and the initial detuning @xmath13 is varied to reach different values of @xmath10 .
we identify coherent oscillations as a function of @xmath12 and @xmath10 . from a best fit ,
we obtain the decoherence time @xmath130 , which agrees well with experiment .
the decoherence is mainly due to the fluctuations of @xmath131 .
the period of the temporal oscillations is @xmath132 for @xmath115 ( see fig . [ eff_res](b ) ) . for a fixed @xmath11
, a shorter period can be obtained for smaller @xmath10 , the fastest oscillations being defined by the zeeman energy . to further decrease
the period the external @xmath11 field can be increased and hence the qubit manipulation could be done in a time scale of @xmath133 for @xmath134 , which would allow @xmath135 coherent operations within @xmath136 . in the exchange gate
demonstrated in ref . , @xmath137 increases with @xmath59 , which results in faster dephasing for faster rotations .
in contrast , here the rotation rate is set by the zeeman energy , which is independent of @xmath59 far from the @xmath7-@xmath1 avoided crossing . as a result
, the coherent oscillation frequency can be increased without making the qubit more susceptible to gate voltage fluctuations by simply increasing @xmath11 .
far from the avoided crossing the level spacing is independent of detuning , similar to the `` sweet spot '' in superconducting qubits @xcite . for @xmath138 .
theoretical values are obtained by finding the detuning @xmath139 for which the theoretical and experimental oscillation periods match .
a suitable @xmath140 is then chosen such that @xmath141 .
the experiments used passive filtering of square pulses to reduce @xmath56 and thereby increase the visibility of the oscillations .
as a result , oscillations in the experimental data are delayed to longer times due to finite rise time effects .
the theory points are obtained with a perfect square pulse.,scaledwidth=48.0% ] the model predicts coherent oscillations in @xmath14 for @xmath142 , i.e. in the case where the qubit has not passed the avoided crossing .
it can be explained within finite - time lzs theory , but not with the conventional asymptotic landau - zener formula . in other words
, even if @xmath143 , we have @xmath144 , which illustrates the non - adiabatic character of the problem . for the pulse conditions used in ref .
, oscillations are not observed for @xmath145 , most likely due to charge dephasing .
the coherence time of an admixture of @xmath146 and @xmath147 charge state has been measured to be @xmath148 for gaas qds@xcite and sets the time scale at which the system must be driven to observe oscillations for @xmath142 .
a finite - time effect in agreement with the experimental data is the dependence of the oscillation amplitude @xmath14 on the pulse length .
finally , we show in fig .
[ exp_theo_fit ] a comparison between experiment and theory for @xmath138 . the experimental data were obtained from the setup used in ref . .
the experiments used passive filtering of square pulses to reduce @xmath56 and thereby increase the visibility of the oscillations . as a result
, oscillations in the experimental data are delayed to longer times due to finite rise time effects .
the passage through the avoided crossing can be interpreted as a rotation ( see appendix [ sec : lzsrot ] ) , @xmath149 by an angle @xmath150 around the axis @xmath151 , see fig .
[ single_gate ] . here
@xmath152 and @xmath153 is a unit vector where @xmath154 is the dimensionless coupling strength . since @xmath155 and @xmath150 are functions of the same experimental parameters @xmath156 , it is not straight forward to find them simultaneously in order to build a given single qubit rotation .
for instance , fixing two parameters and tuning the third one will simultaneously change @xmath155 and @xmath150 limiting the achievable rotations angles .
nevertheless , the situation is not hopeless and several composite methods can be engineered to achieve any rotation .
we present here three methods , each of them having their own advantages . since rotations by an angle @xmath157 around the @xmath158-axis are available by letting the qubit evolve in an external @xmath11 field , we would like @xmath151 to be in the @xmath159-plane in order to build any rotation by the euler angle method . below
we show that this is possible if for example ( see appendix [ sec : lzsrot ] ) the propagation times are equal , @xmath160 .
however , a @xmath161-rotation from @xmath162 to @xmath163 would take an exponentially long time with a single lzs transition , since it corresponds to a fully adiabatic transition .
however , this problem can be circumvented by sequentially applying several lzs transitions .
a @xmath161-rotation from @xmath162 to @xmath163 can be achieved in @xmath164 for two consecutive and identical lzs transitions .
the single qubit gates can also be implemented with a pure lzs interferometry technique , similar to the one used to control superconducting qubits .
this method requires sequential driving of the qubit through the avoided crossing .
the different passages result in a series of lzs transitions each of them corresponding to a rotation of the qubit . by tuning @xmath105 and choosing @xmath165 with different ratios for @xmath166 ,
any qubit rotations can be achieved within a nanosecond .
since finite - time effects are present in the system , we can think about a control method where the qubit is operated on the @xmath146 charge configuration side ( `` sweet region '' ) .
this requires the preparation of a @xmath167 state @xcite and pulses with rise times shorter than @xmath20 which do not drive the system through the avoided crossing unless a measurement is required .
the qubit manipulation is achieved through finite - time lzs interferometry , where tuning @xmath105 and the propagation time @xmath168 allows to achieve any desirable angle .
, the rotation axis lays in the @xmath159-plane .
in addition with the rotations generated around the @xmath158-axis by letting the qubit evolve in the external magnetic field , any rotation can be achieved by the euler angle method .
( b ) if the propagation times are different the rotations axis may not lay ( see appedix [ sec : lzsrot ] ) in the @xmath159 plane . in this case
, a pure lzs interferometry technique can be used to generate any rotation.,scaledwidth=48.0% ] for all those methods , an arbitrary single qubit rotation can be expressed as a series of a forward sweep -wait -backward sweep operators , @xmath169 which reduces to eq . for @xmath170 .
the proposed methods require a maximum of @xmath171 .
it is important to notice that the rotation axis and the final measurable angles will not be @xmath151 , @xmath172 , and @xmath173 , but rather @xmath174 , @xmath175 , and @xmath176 , where the brackets denote the averaging over the nuclear spin bath . a similar scheme with @xmath170 has been proposed for the @xmath7 - @xmath71 qubit @xcite .
to complete the set of quantum gates , a two - qubit operation such as @xmath177 is required .
we consider the hamiltonian @xmath178 where @xmath179 is the single qubit hamiltonian and @xmath180 describes the capacitive coupling @xcite between two adjacent qds belonging to different @xmath181 qubits
. has different values whenever the charge state of the control qubit is @xmath182 ( a ) or @xmath146 ( b ) .
the later case results in @xmath183 which suppresses any lzs transition.,scaledwidth=48.0% ] tunneling between the dots of the different qubits can be suppressed by an appropriate gate voltage .
if the control qubit , qubit-1 , is in a @xmath57 state , @xmath184 and the dynamics of the target qubit , qubit-2 , is reduced to the case of a single qubit , see fig .
[ double_gate ] .
when the control qubit is in a @xmath146 charge configuration , the target qubit is influenced by the interdot coulomb interaction . in this case , the dynamics of the target qubit can be described by eq . by replacing @xmath185 .
in particular , this affects the position of the avoided crossing . for a system with two dqd separated by a distance @xmath186 , where @xmath187 is the approximate radius of one qd , the intradot coulomb interaction @xmath188 is comparable to the interdot coulomb interaction @xmath189 resulting in @xmath190 . from the previous discussion , we know that a @xmath161-rotation is possible within @xmath148 if @xmath191 . in the case where the avoided crossing is at @xmath192 the same lzs sequence
will leave the target qubit unchanged , even within a finite - time theory since the separation between the two avoided crossings is @xmath193 . therefore , we estimate the @xmath177 gate time to be @xmath194 .
let us consider the case where the control qubit is in the @xmath57 state , which is the logical @xmath195 of the qubit , such that the target and control qubit are not capacitively coupled , @xmath127 . for @xmath196 , @xmath112 , @xmath197 , and @xmath198
we find for @xmath199 and @xmath200 in eq .
@xmath201 this example shows the almost perfect realization of a conditional @xmath202 operation which corresponds to a cnot gate up to single - qubit gates . to show that this method produces a cnot gate , we consider the case where the control qubit is in the @xmath203 state , which is the logical @xmath204 of the qubit .
we estimate a lower bound for the strength of the capacitive coupling between the qubits to be @xmath205 ( see above ) . in this case
, the target qubit evolution takes the form @xmath206 which is close to @xmath207 and demonstrates the possibility of generating a cnot gate with the proposed method . notice that our choice for @xmath208 corresponds to a propagation time @xmath209 such that the total gate time to achieve controlled-@xmath202 is @xmath210 .
the fidelity is @xmath211\right|}^2/16 \simeq 0.918 $ ] .
a more accurate cnot gate can be engineered by fine tuning the parameters entering the lzs propagator .
we have demonstrated that coherent control of the @xmath7-@xmath1 qubit can be achieved using lzs interferometry .
hyperfine interactions lead to an avoided crossing between @xmath7 and @xmath1 states , which allows for efficient quantum control .
moreover , we predict that in the limit of fast rise - time pulses coherent oscillations in @xmath14 should be observed even without going through the avoided crossing .
this phenomenon is a finite time effect which we have theoretically described using the general finite - time lzs theory and it can be used to operate the qubit in the @xmath146 charge configuration side ( `` sweet region '' ) . our scheme can be extended to dqd in materials with few nuclear spins ( graphene , cnt , si ) .
in such cases , the avoided crossing between the qubit states can be achieved by engineering a dqd in the presence of micro - magnets which provide the in - plane gradient magnetic field for the realization of the lzs based gates @xcite .
the qubit will moreover benefit from the lack of the inhomogeneous broadening due to the overhauser fields and exhibit an extended @xmath212 .
in gaas dqds the method proposed in ref .
could be used to extend @xmath212 without cancelling the gradient field .
other schemes to polarize the nuclear spins @xcite or reduce their fluctuations @xcite also exist .
we acknowledge funding from the dfg within spp 1285 , for 912 and sfb 767 .
research at princeton was supported by the sloan and packard foundations , darpa award n66001 - 09 - 1 - 2020 , and the nsf through dmr-0819860 and dmr-0846341 .
in this appendix we follow the work of vitanov and garraway @xcite and consider , without loss of generality , a two - level system whose eigenenergies @xmath213 and @xmath214 are time dependent , @xmath215 and their difference is a linear function of time @xmath216 .
furthermore , we assume the levels to be coupled with strength @xmath217 .
the matrix representation of the system s hamiltonian is given by @xmath218 the time evolution of such a system is described by the time - dependent schrdinger equation @xmath219 with @xmath220 .
after substitution of into eq . , a coupled system of first order ordinary differential equations is obtained align _ 1 ( t ) & = -(t ) c_1 ( t ) + c_2 ( t ) , [ lzs_diffeq_syst:1 ] + _ 2 ( t ) & = c_1 ( t ) + ( t ) c_2 ( t ) .
[ lzs_diffeq_syst:2 ] by deriving eq . with respect to time and substituting eqs .
and into the newly obtained ordinary second order differential equation , we obtain @xmath221 it is convenient to introduce dimensionless parameters before solving eq . , here we introduce the dimensionless time @xmath222 which we substitute in eq . to obtain @xmath223 where @xmath224 is the dimensionless coupling strength . the solution of eq .
is @xmath225 where @xmath226 are parabolic cylinder functions , which solve the weber equation@xcite @xmath227 and can be obtained from eq . by writing the expression in brackets as @xmath228 and substituting @xmath229 .
@xmath230 is obtained by inserting eq . into eq . and
using the property @xmath231 one finds @xmath232 .
\label{c2_weber_dimless } \end{split}\ ] ] to find the constants @xmath233 and @xmath234 , we consider initial conditions given by @xmath235 and @xmath236 and the wronskian relation @xmath237 we solve the system of equation given by eqs . and for @xmath233 and @xmath234 using the wronskian property , we find align _ 1 & = , [ const_diff_eq:1 ] + _ 2 & = .
[ const_diff_eq:2 ] substituting eqs . and into eqs . and and having in mind that we are looking for the evolution operator @xmath238 giving the final state knowing the initial one @xmath239 we finally find the lzs propagator @xmath240 with @xmath241 \end{split } \label{u11}\ ] ] and @xmath242 .
\end{split } \label{u12}\ ] ] in the original lzs problem @xmath102 is defined at the energy levels crossing .
a situation where @xmath243 and @xmath244 corresponds to drive the system through the avoided crossing .
the case @xmath243 and @xmath143 corresponds to stop the system before it goes through the avoided crossing . finally , @xmath245 and @xmath244 corresponds to a system which is initially prepared after the avoided crossing .
the expression of the lzs propagator can be expressed with simpler functions when the argument @xmath246 and the parameter @xmath247 , in this case the parabolic cylinder functions can be expanded asymptotically@xcite . the necessary asymptotic forms to expand eqs . and are @xmath248 @xmath249 @xmath250 @xmath251 where we have defined @xmath252 and @xmath253 using the above expressions and writing @xmath254 to fulfill the condition for the expansion we find , @xmath255 and @xmath256 where @xmath257 , and @xmath258 are respectively given by eq .
and eq . for @xmath259 .
we noticed that the asymptotic expansions , , , and are valid for the weaker condition @xmath260 , as already reported in ref . for the expansion of @xmath261 and @xmath262 .
in quantum mechanics the rotation operator @xmath263 by an angle @xmath264 around an axis @xmath155 of a two - level system has the representation @xmath265 @xmath268\\ & \phantom{=}+ { \mathrm{e}}^{-\frac{\pi\eta^2}{2 } } \cos\varsigma\left[\cos\theta_{{\mathrm{f}}}\cos\theta_{{\mathrm{i } } } - \sin\theta_{{\mathrm{f}}}\sin\theta_{{\mathrm{i}}}\right ] , \end{aligned } \label{costheta}\ ] ] @xmath273\right.\\ & \phantom{=\frac{1}{\sin\frac{\vartheta}{2 } } ( } \left.+ { \mathrm{e}}^{-\frac{\pi \eta^2}{2}}\sin\varrho\left[\cos\theta_{{\mathrm{i}}}\sin\theta_{{\mathrm{f } } } - \cos\theta_{{\mathrm{f}}}\sin\theta_{{\mathrm{i}}}\right]\right ) , \end{aligned } \label{nx}\ ] ] @xmath274\right.\\ & \phantom{\frac{1}{\sin\frac{\vartheta}{2 } } ( } \left.+ { \mathrm{e}}^{-\frac{\pi \eta^2}{2}}\cos\varrho\left[-\cos\theta_{{\mathrm{i}}}\sin\theta_{{\mathrm{f } } } - \cos\theta_{{\mathrm{f}}}\sin\theta_{{\mathrm{i}}}\right]\right ) , \end{aligned } \label{ny}\ ] ] @xmath275\right.\\ & \phantom{\frac{1}{\sin\frac{\vartheta}{2 } } ( } \left.+ { \mathrm{e}}^{-\frac{\pi \eta^2}{2}}\sin\varrho\left[-\cos\theta_{{\mathrm{f}}}\cos\theta_{{\mathrm{i } } } + \sin\theta_{{\mathrm{f}}}\sin\theta_{{\mathrm{i}}}\right]\right ) . \end{aligned } \label{nz}\ ] ] | we theoretically demonstrate that nuclear spins can be harnessed to coherently control two - electron spin states in a double quantum dot .
hyperfine interactions lead to an avoided crossing between the spin singlet state and the @xmath0 triplet state , @xmath1 .
we show that a coherent superposition of singlet and triplet states can be achieved using finite - time landau - zener - stckelberg interferometry . in this system
the coherent rotation rate is set by the zeeman energy , resulting in @xmath2 nanosecond single spin rotations .
we analyze the coherence of this spin qubit by considering the coupling to the nuclear spin bath and show that @xmath3 , in good agreement with experimental data .
our analysis further demonstrates that efficient single qubit and two qubit control can be achieved using landau - zener - stckelberg interferometry . |
clusters of galaxies are the largest gravitationally - collapsed structures in the universe .
studying properties such as the number density of clusters and details of their growth from the highest density perturbations in the early universe , offers insight into the underlying cosmology ( e.g. * ? ? ?
* ; * ? ? ?
the study of galaxy clusters has been transformed with the launch of powerful x - ray telescopes such as _ chandra _ and _ xmm _ , which have allowed the study of the x - ray emitting intracluster medium ( icm ) with unprecedented detail and accuracy .
cluster properties have been used widely in the determination of cosmological parameters .
cosmological studies utilising clusters include investigating the cluster temperature function ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , scaling relations used as a mass proxy such as the luminosity - mass ( e.g * ? ? ? * ) and the temperature - mass ( e.g. * ? ? ?
* ) relations , using the gas mass fraction , f@xmath6 @xcite , and the cluster mass function ( e.g. * ? ? ?
* ) , to place constraints on various cosmological parameters . since one of the most important ingredients of these cosmological studies is the cluster mass , large efforts have been undertaken to accurately determine this quantity .
one such method involves the construction of radial temperature and gas density profiles of the icm , and under the assumption of hydrostatic equilibrium , the cluster mass can be determined .
observations using x - rays have become a well established method of estimating cluster masses , and , at the present time , arguably the most reliable .
however , constructing temperature profiles for individual clusters for use in a hydrostatic mass analysis generally require long telescope exposure times , and not is not feasible for large samples of clusters . therefore , deriving well calibrated scaling relations between simple cluster observables and mass is of crucial importance for using clusters as cosmological probes ( e.g. * ? ? ?
the x - ray luminosity ( @xmath7 ) is one of the easiest cluster properties to obtain , and has had a rich history in its scaling with mass ( @xmath8 ) . under the assumption of self - similarity @xcite , the @xmath2 relation ( throughout this work we use the notation @xmath2 when generally discussing the luminosity - mass relation ) is expected to follow a relationship of @xmath9 .
however , observational studies of the @xmath2 relation have found a slope steeper than the self - similar expectation , @xmath32 .
the most widely accepted theory for the steep slope of the @xmath2 relation , is due to heating from sources such as supernovae ( sne ) and active galactic nuclei ( agn ) .
this causes gas to be expelled from the inner region , hence suppressing the luminosity .
this effect should be larger in lower mass systems due to the shallower potential well , which therefore causes the observed steepening of the @xmath2 relation .
a complication of measuring the @xmath2 relation is that the cluster samples used are traditionally x - ray selected , with x - ray flux limited samples suffering from two forms of selection bias , malmquist bias , where higher luminosity clusters are detectable out to higher redshifts and so occupy a larger survey volume , and eddington bias , where in the presence of intrinsic or statistical scatter in luminosity for a given mass , objects above a flux limit will have above - average luminosities for their mass . due to the steep slope of the cluster mass function ,
the eddington bias is amplified , resulting in a net movement of lower mass objects into a flux limited sample . the consequence of biases on the observed @xmath2 relation is to bias the normalisation high and the slope low .
therefore , taking these biases into account is paramount when modeling cluster scaling relations , in order to uncover the true nature of any non - gravitational heating which drives departures from self - similar behavior with mass or redshift .
although scaling relation studies have had a rich history , at the present time only a small number of published relations attempt to account for selection biases , while @xcite provides the most robust handling of selection effects to date .
deriving cluster masses through x - ray observations with the assumption that the icm is in hydrostatic equilibrium , is not always valid , as some clusters have complex temperature structures due to processes such as merger events .
this assumption is thought to be closely satisfied by clusters that are dynamically relaxed .
clusters that appear to host a cool core ( cc ) are frequently used for mass derivations as they are believed to be dynamically relaxed .
many methods are used to infer the presence of a cc , including measuring the central temperature drop ( e.g. * ? ? ?
* ) , the central cooling time , the core entropy ( e.g. * ? ? ? * ) and the cuspiness of the gas density profile @xcite .
frequently , scaling relations involving the cluster mass are used as a mass proxy to calculate the masses of large cluster samples .
this first requires constructing scaling relations for dynamically relaxed clusters , and then inferring the cluster masses from these relations .
it has been found however that using cc clusters sometimes results in larger scatter of the @xmath2 scaling relation compared to non - cool core ( ncc ) clusters ( e.g. * ? ? ? * ) .
@xcite found that by defining a cluster sub - sample using clusters appearing both dynamically relaxed and hosting a cool core , the luminosity - temperature ( @xmath10 ) relation appears self - similar , compared to unrelaxed and non - cool core clusters .
while this method may be the preferred choice for defining sub - samples of clusters for mass derivations , this limits the cluster sample size used for mass calculations .
this paper aims to measure the masses for a complete sample of 35 clusters to measure the x - ray observable mass scaling relations .
we aim to define a sub - sample of clusters that are both dynamically relaxed and host a cc for which the assumption of hydrostatic equilibrium is valid and accurate x - ray masses can be obtained . by constructing scaling relations between x - ray observables and mass for this sub - sample of clusters , these will be used as a mass proxy for the remaining clusters .
the outline of the paper is as follows . in sect .
[ sec : data ] we discuss the sample selection and data analysis .
[ sec : analysis ] details the cluster analysis and determines the cool - core and morphological status of each cluster is the sample .
notes on individual clusters are given in sect .
[ sec : notes ] .
our results are presented in sect .
[ sec : xscaling ] .
the discussion and conclusions are presented in sect .
[ sec : disc ] and sect .
[ sec : conc ] respectively . throughout this paper
we assume a _
wmap_9 cosmology of h@xmath11=69.7 km s@xmath5 mpc@xmath5 , @xmath12=0.282 and @xmath13=0.718 @xcite .
the sample of clusters used in our analysis was defined by the conditions given in @xcite .
the sample of clusters represents a volume - limited sample of x - ray luminous clusters taken from the rass - based , x - ray flux limited rosat brightest cluster sample ( bcs ) of @xcite and its low - flux extension ( ebcs , @xcite ) .
@xcite imposed a lower cutoff in x - ray luminosity of l@xmath14 = 6 @xmath15 10@xmath16 erg s@xmath5 ( the limit based upon a cosmology assuming @xmath12=0.3 , @xmath13=0.7 and h=0.7 ) , corresponding to a sample of 36 clusters within the redshift range 0.15 @xmath17 z @xmath17 0.30 .
figure [ fig : lz ] plots the luminosity - redshift distribution of the bcs and ebcs , with the yellow shaded region highlighting the region enclosed by the luminosity and redshift cuts defining our sample selection .
note that two of the clusters were dropped from the cluster sample , as detailed below , leading to a final sample of 34 clusters .
the cluster a689 satisfies these selection criteria , but was noted in the original detection as having a significant portion of its flux coming from embedded point sources , and was therefore excluded from the cluster sample .
the source of the embedded point source was found to be a central bl - lac object @xcite , with the re - analysis determining a cluster x - ray luminosity @xmath1810 times lower than quoted in the bcs , well below the sample cutoff x - ray luminosity .
furthermore , we found that the quoted redshift for the cluster zw5768 in @xcite was incorrect .
after further investigation using an updated redshift ( see @xmath19 [ sec : notes ] ) , we determined that the cluster would not satisfy the sample selection , and was subsequently dropped from the cluster sample . for the x - ray analysis we obtained _ chandra _ observations to complete the sample and downloaded archived observations of the remaining clusters from the _ chandra _ data archive . , width=332 ] [ cols="<,^,^,^,^,^,^,^ " , ]
in this paper we have presented the @xmath2 relation for our sample , for the properties derived within an @xmath20 calculated from the rcc @xmath1 relation ( see @xmath19 [ sec : mt ] ) . here
we present the @xmath2 relations using hydrostatic masses for all clusters , and the luminosities derived within the hydrostatic @xmath20 ( r@xmath21 ) .
table [ tab : hydroresults ] lists the properties of the clusters within r@xmath21 .
the @xmath22-@xmath23 relation using hydrostatic masses is plotted in figure [ f.hydrolm ] , and the best fit paramteres are given in table [ t.lmhydro ] .
lccccc + & r@xmath21 & l@xmath24 & l@xmath25 & t@xmath26 & m@xmath27 + cluster & ( mpc ) & ( 10@xmath16 ergs s@xmath5 ) & ( 10@xmath16 ergs s@xmath5 ) & ( kev ) & ( 10@xmath28m@xmath29 ) + a2204@xmath30 & 1.36@xmath31 & [email protected] & [email protected] & 9.23@xmath33 & 8.25@xmath34 + rxj1720.1 + 2638@xmath30 & 1.34@xmath35 & [email protected] & [email protected] & 8.01@xmath36 & 7.91@xmath37 + a586 & 1.05@xmath38 & [email protected] & [email protected] & 6.02@xmath39 & 3.80@xmath40 + a1914 & 1.42@xmath41 & [email protected] & [email protected] & 9.37@xmath42 & 9.41@xmath43 + a665 & 1.56@xmath44 & [email protected] & [email protected] & 8.81@xmath45 & 12.78@xmath46 + a115 & 1.07@xmath47 & [email protected] & [email protected] & 7.21@xmath48 & 4.15@xmath49 + a520 & 1.27@xmath50 & [email protected] & [email protected] & 7.91@xmath51 & 7.04@xmath52 + a963 & 1.08@xmath53 & [email protected] & [email protected] & 5.77@xmath54 & 4.36@xmath55 + a1423@xmath30 & 1.09@xmath56 & [email protected] & [email protected] & 6.26@xmath36 & 4.44@xmath57 + a773 & 1.35@xmath58 & [email protected] & [email protected] & 8.15@xmath59 & 8.55@xmath60 + a1763 & 1.40@xmath61 & [email protected] & [email protected] & 8.82@xmath62 & 9.57@xmath63 + a2261 & 1.18@xmath64 & [email protected] & [email protected] & 8.07@xmath65 & 5.79@xmath66 + a1682 & 1.01@xmath64 & [email protected] & [email protected] & 6.70@xmath67 & 3.60@xmath68 + a2111 & 1.14@xmath69 & [email protected] & [email protected] & 7.62@xmath70 & 5.25@xmath71 + z5247 & 0.89@xmath44 & [email protected] & [email protected] & 5.09@xmath72 & 2.45@xmath73 + a267 & 1.00@xmath74 & [email protected] & [email protected] & 4.99@xmath75 & 3.48@xmath76 + a2219 & 1.49@xmath77 & [email protected] & [email protected] & 11.53@xmath78 & 11.57@xmath79 + a2390 & 1.58@xmath80 & [email protected] & [email protected] & 11.38@xmath81 & 13.89@xmath82 + z2089@xmath30 & 0.91@xmath83 & [email protected] & [email protected] & 4.41@xmath84 & 2.67@xmath85 + rxj2129.6 + 0005@xmath30 & 1.13@xmath69 & [email protected] & [email protected] & 6.79@xmath86 & 5.14@xmath87 + a1835@xmath30 & 1.40@xmath88 & [email protected] & [email protected] & 9.85@xmath33 & 9.91@xmath89 + a68 & 1.32@xmath90 & [email protected] & [email protected] & 8.31@xmath91 & 8.28@xmath92 + ms1455.0 + 2232@xmath30 & 1.02@xmath77 & [email protected] & [email protected] & 5.38@xmath93 & 3.81@xmath94 + a2631 & 1.24@xmath95 & [email protected] & [email protected] & 7.02@xmath96 & 7.03@xmath97 + a1758 & 1.62@xmath98 & [email protected] & [email protected] & 8.82@xmath99 & 15.75@xmath100 + a1576 & 1.17@xmath38 & [email protected] & [email protected] & 7.81@xmath101 & 5.86@xmath102 + a697 & 1.59@xmath103 & [email protected] & [email protected] & 12.58@xmath104 & 15.01@xmath105 + rxj0439.0 + 0715@xmath30 & 1.09@xmath106 & [email protected] & [email protected] & 6.42@xmath59 & 4.59@xmath107 + rxj0437.1 + 0043@xmath30 & 1.23@xmath108 & [email protected] & [email protected] & 7.44@xmath109 & 6.87@xmath110 + a611 & 1.13@xmath111 & [email protected] & [email protected] & 8.47@xmath112 & 5.36@xmath113 + z7215 & 1.13@xmath114 & [email protected] & [email protected] & 6.98@xmath115 & 5.38@xmath116 + z3146@xmath30 & 1.25@xmath69 & [email protected] & [email protected] & 8.27@xmath117 & 7.30@xmath118 + a781 & 1.11@xmath64 & [email protected] & [email protected] & 6.57@xmath119 & 5.10@xmath120 + a2552 & 1.24@xmath121 & [email protected] & [email protected] & 8.32@xmath122 & 7.27@xmath123 + l|ccccc + relation & method & @xmath124 & @xmath125 & @xmath126 & @xmath127 + @xmath22-@xmath128 & bces & @xmath129 & @xmath130 & 7/3 & @xmath131 + @xmath132-@xmath128 & m10 & @xmath133 & @xmath134 & 7/3 & @xmath135 + @xmath22-@xmath128 & m10 & @xmath136 & @xmath134 & 7/3 & @xmath135 + @xmath137-@xmath128 & m10 & @xmath138 & @xmath139 & 2.69 & @xmath135 +
here we show images , gas density and temperature profiles for our cluster sample .
figure [ fig : a2204 ] shows an adaptively smoothed image of the cluster measuring 3@xmath153 mpc on a side ( left ) , the emmisivity profile with the best fitting gas density profile ( middle ) and the temperature profile ( right ) with the best fitting three dimentional model ( red ) and the corresponding projected profile ( blue ) for the cluster a2204 .
figures [ fig : a586]-[fig : a2552 ] show the rest of the cluster sample . | we present the results of work involving a statistically complete sample of 34 galaxy clusters , in the redshift range 0.15@[email protected] observed with _ chandra_. we present the calibration of the mass - temperature ( mt ) relation using hydrostatic mass estimates for the most dynamically relaxed clusters , and use this relation as a mass proxy for the full cluster sample .
we find that the slope of the @xmath1 relation follows the self - similar expectation , and is consistent with previously published relations .
we investigate the luminosity - mass ( lm ) relation for the cluster sample , utilising a method to fully account for selection biases .
we find that the difference in normalisation of the @xmath2 relation with and without accounting for selection effects is @xmath32 . for a cluster of luminosity 10@xmath4 erg s@xmath5
, we find that the mass estimated from the @xmath2 relation when we account for selection effects is @xmath340% higher compared to the sample @xmath2 relation ( not accounting for selection effects ) .
[ firstpage ] galaxies : clusters : general - x - rays : galaxies : clusters |
sagittarius a * ( sgr a * ) , located at the galactic center , marks a dark mass of just over @xmath0 @xcite . at present there is no credible alternative to a supermassive black hole @xcite .
its proximity makes it the best studied astronomical black hole candidate , one for which there is strong evidence that an event horizon exists @xcite .
a variety of observations and theoretical models imply that the millimeter emission region lies within several schwarzschild radii of the black hole ( @xmath1as ) . directly resolving the region
provides a powerful probe of the structure and dynamics near the horizon .
general relativity predicts that sgr a * will have a photon ring and associated shadow approximately @xmath2as in diameter @xcite .
spatially resolved observations thus hold great promise to assess the nature of the emission region ( e.g. , whether the millimeter - wavelength emission arises from a thick accretion disk or weak jet ) as well as to test general relativity in the strong gravity regime ( e.g. , via the shape and size of the shadow ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
? * ; * ? ? ?
. for this purpose the event horizon telescope ( eht ) is being assembled .
comprised of new and existing telescopes at 1.3 mm and 0.87 mm , the eht is a global array for very long baseline interferometry ( vlbi ) observations of nearby supermassive black holes , including sgr a * @xcite . uniquely among the many telescopes that observe sgr a * , the eht resolves structures on the scale of a few schwarzschild radii in the inner accretion and outflow region .
this resolution is well matched to the scales of the predicted physical and astrophysical features .
previously published eht observations have used either the correlated flux density @xcite or polarization @xcite on long baselines to infer the structure of sgr a*. in this work , we focus on a third eht data product , closure phases . in a radio interferometric array ,
each baseline produces a complex observable known as the visibility , which is effectively a fourier component of the source image .
the visibility can be decomposed into two quantities : an amplitude and a phase .
both parts of the visibility contain information about the structure of the observed source .
the amplitude alone can be sufficient to characterize the approximate size of a source ( * ? ? ? * and others ) and
even permit modelling of the source structure ( e.g. , * ? ? ?
* ; * ? ? ?
* ) , but most of the detailed structural information is contained in the phase @xcite .
for instance , @xcite demonstrated that the inclusion of phase data from just a few telescopes would nail down the spin vector of the black hole in an accretion flow model of sgr a*. at the high frequencies at which the eht observes , visibility phases are easily corrupted by rapidly varying tropospheric delays , primarily due to water vapor .
a more robust phase observable is the closure phase , or sum of visibility phases along a closed loop of three baselines @xcite .
closure phases are immune to atmospheric phase fluctuations and to most other phase errors that are station - based rather than baseline - based in origin , such as phase variations in the receiver and local oscillator system at each station @xcite .
closure phases that are neither zero nor 180 indicate that the source structure is not point - symmetric at the resolution of the observing array @xcite .
nonzero closure phases have been detected on bright quasar sources with the eht and used to model the structure of these sources @xcite , but the relative weakness of sgr a * has heretofore only allowed a weak upper limit to be placed on the absolute value of its closure phase on the california - hawaii - arizona triangle @xcite . in this paper
we report on detections of nonzero closure phases in sgr a * , providing the first direct indication of asymmetric emission near the black hole .
multiple measurements of the closure phase of sgr a * were obtained .
we summarize the observing setup and methods of analysis in sections [ observations ] and [ analysis ] , describe the results of the dataset in section [ results ] , examine implications for the quiescent and variable structure of sgr a * in section [ discussion ] , and comment on future prospects for improved data in section [ conclusions ] .
the eht obtained detections of sgr a * on closed triangles of baselines among stations in arizona , california , and hawaii in 2009 , 2011 , 2012 , and 2013 . in all cases , two 480-mhz bands , centered at 229.089 and 229.601
ghz ( hereafter called low and high bands , respectively ) , were observed .
a hydrogen maser was used as the timing and frequency standard at all sites ( but see section [ 2011 ] ) .
the two bands were correlated and post - processed independently .
digital backends and phased - array processors channelized each 480 mhz band into 15 channels of 32 mhz each .
data were recorded on the disk - based mark 5b+ and mark 5c systems @xcite and then correlated on the haystack mark 4 vlbi correlator @xcite with a spectral resolution of 1 mhz and an accumulation period of either 0.5 s or 1 s. left - circular polarization ( lcp ) was always observed , and right - circular polarization ( rcp ) was observed in later experiments as well .
we report only on closure quantities that do not mix polarizations .
one or more telescopes from each of three sites in arizona , california , and hawaii participated in each set of observations .
the arizona radio observatory ( aro ) submillimeter telescope ( smt ) on mt .
graham , arizona was used in all cases . at the california and hawaii sites , the capabilities of the instruments evolved through the years , transitioning to recording coherently phased sums of connected dishes . over the years of data analyzed here , the configuration of vlbi recording at these sites evolved as described below . the combined array for research in millimeter - wave astronomy ( carma ) in eastern california participated .
observations always consisted of two vlbi stations , one of which was a single 10.4-m antenna . a second 10.4-m antenna participated in 2009 and part of 2011 . from 2011
day 091 onward , the second antenna was replaced by a more sensitive phased array of up to eight telescopes ( including both the 10.4-m and 6.1-m antennas ) .
three observatories on mauna kea , hawaii participated in observations : the submillimeter array ( sma ) , the james clerk maxwell telescope ( jcmt ) , and the caltech submillimeter observatory ( cso ) .
the sma consisted of a phased array of up to eight telescopes @xcite .
table [ obs - table ] summarizes the telescopes participating in each set of observations along with one - letter station codes , used hereafter .
typical fringe spacings are 60 @xmath3as on hawaii - arizona baselines , 70 @xmath3as on hawaii - california baselines , and 300 @xmath3as on california - arizona baselines
. two stations of the same polarization were used at the carma site in all experiments and on mauna kea in 2011 .
on arcsecond scales , extended thermal structures near sgr a * contribute to the millimeter - wavelength interferometer response ( e.g. , * ? ? ? * ) .
examination of the correlated flux density as a function of baseline length indicates that this emission is resolved out on baselines longer than @xmath4 k@xmath5 , or a projected baseline length of 26 m at @xmath6 mm .
the intrasite vlbi baselines ( cd , df , eg , jp , and op ) were longer than @xmath7 k@xmath5 except in 2009 .
data quality at 1.3 mm is highly dependent on weather conditions , which are different from day to day and often variable on any given day as well .
the sensitivity of the eht was generally better in later years due to the inclusion of phased arrays on mauna kea and at the carma site .
lllll c & carma ( single ) & lcp & 2009 - 2011 ( 088 - 090 ) & both + & & & 2011 ( 091 - 094 ) & low + d & carma ( single ) & lcp & 2009 - 2011 & both + & & & 2012 - 2013 & low + e & carma ( single ) & rcp & 2013 & low + f & carma ( phased ) & lcp & 2011 ( 091 - 094 ) & high + & & & 2012 - 2013 & both + g & carma ( phased ) & rcp & 2012 - 2013 & both + j & jcmt & lcp & 2009 , 2011 ( 088 ) & both + & & rcp & 2012 - 2013 & both + o & cso & lcp & 2011 ( 090 - 094 ) & both + p & sma ( phased ) & lcp & 2011 - 2013 & both + s & smt & lcp & 2009 - 2013 & both + t & smt & rcp & 2012 - 2013 & both sgr a * was observed on days 093 , 095 , 096 , and 097 , although there were no detections on the carma - hawaii baselines on day 095 . the observing array consisted of the smt , the jcmt , and two individual carma antennas each operating as co - located vlbi sites but using the same hydrogen maser as a time and frequency standard .
calibrated amplitudes from days 095 , 096 , and 097 have been published in @xcite .
there was significant power aliased into the observing band at the carma stations because the 90 phase - switching normally used to separate the sidebands from the double - sideband mixers was disabled during vlbi scans .
this was not an issue for vlbi baselines between sites , for which natural fringe rotation was rapid enough to wash out the contribution from the opposite sideband .
however , the other sideband was clearly visible in the fringe - rate spectrum on the intrasite cd baseline , introducing a nonclosing phase error on only the cd baseline .
additionally , stations c and d were not separated by a projected length of 20 k@xmath5 . as a result of these two effects , measured
closure phases on the cdj and cds triangles are nonzero ( see section [ consistency ] ) and are therefore excluded from our analysis .
+ sgr a * was observed on days 088 , 090 , 091 , 092 , and 094 , although day 092 suffered from uncharacteristically high atmospheric turbulence at the carma site . the observing array consisted of the smt , two stations at carma , and two stations at hawaii .
one station consisted of a single antenna ( d ) .
a second single antenna ( c ) was used in the low band on all days and the high band on days 088 and 090 .
station c was replaced with the phased - array processor ( f ) in the high band starting with day 091 . at hawaii ,
the jcmt was used as a standalone antenna on day 088 , and the cso was used on the other days .
the second station at hawaii was a phased array that summed signals from sma antennas plus either the cso ( day 088 ) or the jcmt ( other days ) . while hydrogen masers were used at all sites , the digital backend sampler clocks , which are the final mix in the signal chain ,
were erroneously driven off of the local rubidium clock at carma on days 088 - 092 .
an analysis of calibrator sources indicated that this setup did not affect phase closure .
further details can be found in @xcite .
sgr a * was observed on day 075 , 080 , and 081 , although only day 081 provided usable data on all three baselines .
each site provided dual - circular polarization observations , with the two polarizations coming from different telescopes at mauna kea .
disk failures caused the loss of lcp data from station s in the high band .
+ sgr a * was observed on days 080 , 081 , 082 , 085 , and 086 .
the zenith opacity at the carma site was unusually low , dipping to @xmath8 at one point , resulting in high sensitivity on the carma baselines on some nights .
the failure of a mark 5b+ recording system caused the loss of one polarization in one band at phased carma on most nights .
calibrated visibility amplitudes have been published in @xcite . in this work
we adopt the sign conventions of @xcite and @xcite .
the delay on baseline @xmath9 is positive if the signal arrives at station @xmath10 after station @xmath11 .
a positive delay produces a positive visibility phase modulo @xmath12 ambiguities .
the closure phase on a triangle of three baselines is defined to be the directed sum of the visibility phases in order : @xmath13 .
obtaining a closure phase requires detecting the source on a closed triangle of three baselines . in practice ,
source detection is accomplished by finding a peak in the scan amplitude in a multidimensional space defined by delays and the delay - rate ( residual to the correlator model values ) .
the haystack observatory postprocessing system ( hops ) provides tools to search delay / rate space and determine the signal - to - noise ratio ( s / n ) and probability of false detection associated with the peak .
the rapidly variable troposphere at 1.3 mm introduces large phase fluctuations on each baseline , introducing challenges for fringe finding .
when fringes are strong , coherent vector integration along the entire length of the scan is sufficient to detect the fringe despite the substantial coherence losses due to the rapidly varying phase .
once the delays are well determined , the atmospheric phase variations are mitigated by segmenting the data at a cadence shorter than the timescale over which the tropospheric phase changes appreciably . since the tropospheric contribution to the measured visibility phases close , the visibility phases can be closed on a per - segment basis and then averaged over the length of the scan to produce a measurement of the closure phase of the source . the bispectral s / n of the resulting averaged closure phase can vary depending on the choice of segmentation time @xcite . however , evaluated closure phases at different segmentation times are self - consistent provided that the segmentation time does not greatly exceed the coherence timescale .
when fringes are weak , as is often the case on baselines between hawaii and the mainland , hops supports additional strategies to aid in fringe detection .
delay closure can be used to set tight search windows , aiding in the detection of marginal fringes .
phase self - calibration on two strong baselines to a sensitive station can be used to mitigate atmospheric fluctuations on the third baseline of a triangle .
weak fringes can sometimes be detected using incoherent averaging , in which data are segmented at a cadence comparable to the coherence time of the atmosphere , and then those segments are scalar - averaged @xcite . since the optimal strategy for fringe detection and closure phase evaluation varies depending on the sensitivity of each station and atmospheric conditions , multiple strategies were employed , tailored to the particular characteristics of each dataset . when multiple measurements of the same closure phase were obtained through different processings , only the data point with the highest bispectral s / n was retained .
the discarded duplicate points are consistent with the retained data points ( top panels of fig .
[ fig - consistency ] ) , indicating that the particular methods chosen for fringe detection and closure phase evaluation do not significantly bias the data .
there are two classes of triangles on which we obtain closure phases .
triangles that include two vlbi stations from the same site ( e.g. , dfs ) should produce closure phases that are trivially zero to within measurement error . on the intrasite baseline of these `` trivial '' triangles , the large - scale emission in the galactic center ( on scales @xmath14 )
is resolved out .
sgr a * is then pointlike , causing the intrinsic source phase to be zero .
the two long baselines effectively sample the same @xmath15 point , adding a source phase on one baseline and subtracting it on the other .
there is no evidence of nonzero closure phases on sgr a * or other sources in our data on the trivial triangles .
`` nontrivial '' closure phases on triangles involve one carma station , one station in hawaii , and the smt ; these may be nonzero due to source structure . as another consistency check
, we examined measurements of closure phases that should be identical to within their errors .
it is possible that variations in sgr a * may cause fluctuations in the closure phase from scan to scan .
however , during any particular scan , it is possible to obtain more than one estimate of trivial and nontrivial closure phases due to duplications among the stations .
the closure phases obtained in the low and high bands should be identical , since the fractional frequency difference between the observing bands is very small .
closure phases on , e.g. , the fps ( lcp ) and gjt ( rcp ) triangles should be identical , since sgr a * exhibits almost no circular polarization at these frequencies @xcite .
similarly , simultaneous closure phases on pairs of triangles that share the same sites but with different stations ( e.g. , dps and fps ) should provide measurements of the same value . as expected , the pairwise differences of substantially identical closure phases are consistent with a unit gaussian distribution centered on zero when the differences are normalized by the quadrature sum of the errors of the closure phases ( bottom panels of fig .
[ fig - consistency ] ) .
this also provides evidence that the closure phase errorbars are correctly estimated .
in total we detect 181 unique nontrivial closure phases for sgr a * on the california - hawaii - arizona triangle .
we additionally detect 233 trivial closure phases . detected
scan - averaged closure phases are listed in table [ data - table ] .
the data are shown in figure [ fig - data ] .
there are more data points in later epochs due to the increased sensitivity provided by phased sma and , later , phased carma .
medians of the nontrivial closure phases are presented in table [ table - medians ] along with bootstrap estimates of the 95% confidence interval of the median , derived from random resampling with replacement .
rrrrrrrrrrrrr 2009 & 93 & 11.5417 & h & cjs & -21.4 & 3.78 & -2548.3 & -1691.2 & 3044.2 & 1591.4 & -495.8 & 99.8 + 2009 & 93 & 11.9583 & l & cjs & 17.1 & 4.37 & -2658.9 & -1553.0 & 3191.9 & 1425.8 & -533.0 & 127.2 + 2009 & 93 & 11.9583 & l & djs & 13.3 & 4.57 & -2658.9 & -1553.0 & 3191.9 & 1425.8 & -533.0 & 127.1 + 2009 & 93 & 12.2917 & l & djs & -14.0 & 4.09 & -2724.4 & -1438.7 & 3282.6 & 1288.4 & -558.2 & 150.3 + 2009 & 93 & 12.6250 & l & cjs & -11.1 & 4.25 & -2769.2 & -1322.1 & 3348.2 & 1147.6 & -579.1 & 174.5 + 2009 & 93 & 13.1250 & h & cjs & 9.6 & 8.74 & -2796.5 & -1144.7 & 3398.5 & 932.6 & -602.0 & 212.1 + 2009 & 93 & 13.4583 & h & cjs & 4.3 & 7.03 & -2788.0 & -1026.1 & 3399.6 & 788.2 & -611.6 & 237.9 + 2009 & 93 & 13.4583 & l & cjs & -3.0 & 7.30 & -2788.0 & -1026.1 & 3399.6 & 788.2 & -611.6 & 237.9 + 2009 & 93 & 13.8750 & h & cjs & 18.3 & 6.97 & -2747.3 & -879.2 & 3364.4 & 608.7 & -617.0 & 270.5 + 2009 & 93 & 13.8750 &
l & cjs & 33.7 & 8.53 & -2747.3 & -879.2 & 3364.4 & 608.7 & -617.0 & 270.5 the median closure phase ( + 6.3 ) on the california - hawaii - arizona triangle is positive at high statistical significance . in a larger run of @xmath16 bootstrap - resampled datasets ,
every median was positive .
this result is also robust against the exclusion of data from the day with the largest closure phase ( 2013 day 080 ) ; the resulting dataset has a median closure phase of 5.0 with a 95% lower bound of 3.1 . for comparison , the median _ trivial _ closure phase is 0.4 , consistent with zero ( 95% range : @xmath17 to @xmath18 ) as expected .
the median nontrivial closure phase of + 6.3 is too large to be attributable to instrumental effects ( appendix ) and is nearly identical to the + 6.7 measured in an independent analysis of 2013 data by r .- s .
lu et al .
( in preparation ) using both the mark 4 and difx correlators .
ccrrrr 2009 & 093 & 11 & 9.6 & @xmath1911.1 & 17.7 + 2009 & 096 & 3 & 7.1 & & + 2009 & 097 & 10 & 8.4 & 0.7 & 13.5 + 2009 & all & 24 & 8.4 & 0.7 & 13.5 + 2011 & 088 & 7 & 13.6 & @xmath190.4 & 29.0 + 2011 & 090 & 2 & 10.0 & & + 2011 & 091 & 5 & 5.7 & @xmath195.9 & 11.7 + 2011 & 094 & 17 & @xmath190.3 & @xmath197.2 & 2.7 + 2011 & all & 31 & 2.6 & @xmath193.5 & 5.7 + 2012 & 081 & 25 & 3.1 & @xmath191.8 & 6.5 + 2013 & 080 & 28 & 16.0 & 6.7 & 20.2 + 2013 & 081 & 10 & 7.2 & @xmath197.7 & 12.7 + 2013 & 082 & 15 & 10.3 & @xmath190.5 & 14.1 + 2013 & 085 & 32 & 6.5 & 0.5 & 7.1 + 2013 & 086 & 16 & 3.0 & @xmath191.6 & 6.3 + 2013 & all & 101 & 6.9 & 5.6 & 9.4 + all & all & 181 & 6.3 & 4.3 & 7.0 care must be taken not to overinterpret differences between subsamples of the dataset .
when individual days of data are processed in multiple ways , the median closure phase can differ by a few degrees .
this is particularly true for data taken in 2009 and 2011 ( before the sensitivity of the observing array was increased ) or for subsamples consisting of only a few measurements , where the inclusion or exclusion of one or two marginal detections can have a greater impact on the derived median value .
for instance , the median closure phase on 2011 day 094 increases from @xmath20 to @xmath21 when only self - calibrated data points are considered .
the dataset taken as a whole is large enough to be robust against the details of the processing to within a few tenths of a degree . since each baseline samples different @xmath15 points at different times due to earth rotation , it would be possible for the closure phase to vary as a function of time even if sgr a * did not exhibit variability
indeed , there is a trend for the measured california - hawaii - arizona closure phase to be larger later in the observing track ( fig .
[ cphase - vs - gst ] ) .
the @xmath22 per degree of freedom for the best - fit line is 1.43 , indicating that there is additional variability and/or that the increase with time is not linear .
this increase with gst is significant at a level of @xmath23 using a kendall tau test .
the expected functional form as a function of gst is model - dependent , although the general trend for the california - hawaii - arizona closure phase to increase with time over an observing night provides an important constraint for physically motivated models of the 1.3 mm emission region in sgr a*.
our data clearly demonstrate that the closure phase on the california - hawaii - arizona triangle is nonzero , with a trend for the magnitude of the closure phase to increase over the course of a night .
the sign and approximate value of the closure phase are consistent among multiple observing epochs over four years . in this section
we consider the implications of these results .
the detection of nonzero closure phase is an unambiguous indication that these eht data are resolving structure in the image of sgr a*. two robust conclusions are that the morphology of the emission from sgr a * at 1.3 mm can not exhibit point symmetry and that the millimeter emission is asymmetric on scales of a few schwarzschild radii .
the sign of the closure phase resolves 180 rotational ambiguities in models . as an example
, the best - fit parameters for the @xcite model find that the rotation axis of the accretion disk points toward either @xmath1952 or @xmath24128 ( @xmath25 error @xmath26 ) east of north .
this pair of directions was obtained from visibility amplitude information alone , which can not discriminate between the two directions because the fourier transforms of an image and the same image rotated 180 are identical modulo a sign flip in phase .
therefore , visibility phase or closure phase information is required to break the 180 degeneracy .
the @xmath24128 direction is consistent in sign with our measured closure phases .
adding the new closure phase data is likely to result in better estimates of model parameters and to provide stronger constraints on models , including those that allow for deviations from general relativity ( e.g. , * ? ? ?
the detection of nonzero but small closure phases in the image of sgr a * demonstrates the power of imaging observations in placing strong constraints on its accretion flow geometry .
in particular , the data favor emission morphologies that are connected rather than those composed of disjoint regions at horizon scales . at this point , it is instructive to look at the images generated in different general relativistic magnetohydrodynamic ( grmhd ) simulations in order to explore in more detail how our observations can be used to constrain various configurations that are physically plausible and are consistent with all other currently available data .
the parameters of grmhd simulations are typically calibrated in order to reproduce the broadband spectrum of sgr a * as well as the overall size of its emitting region at 1.3 mm . even with these constraints imposed ,
however , the images they generate can be quite different from each other , depending on the prescription for the plasma thermodynamics that was employed as well as on the initial magnetic field configuration and tilt of the torus that was used to feed the black hole . in a set of simulations referred to by @xcite as standard and normal evolution ( sane ) , the magnetic flux remains modest ( e.g. , * ? ? ?
* ; * ? ? ?
if the electron temperature is assumed to be at a constant ratio with the ion temperature everywhere in the flow , the generated images typically show continuous crescent - like brightness distributions @xcite . on the other hand ,
if the electrons in the jet are allowed to be heated much more strongly than within the disk , the images are characterized by bright regions from the inner walls of the jets , dissected by the cooler accretion disks @xcite .
the relative brightness of the two regions ( and their exact shape ) depends on the inclination of the observer .
images with even more disjoint bright regions arise in magnetically arrested disk simulations ( mad in the terminology of @xcite ; see also @xcite and @xcite for similar magnetically dominated simulations and their applications to eht observations of m87 ) and are dominated by emission from the footpoints of the jets @xcite .
finally , images with separated bright regions arise naturally in grmhd simulations in which the accreting material is fed to the black hole from a plane that has a tilt with respect to the black - hole spin @xcite because of the presence of standing shocks in these flows .
comparing the detailed predictions of these simulations to our data is beyond the scope of the current paper .
however , motivated by the rather general properties of the disjoint geometries in the images exhibited by some of the grmhd simulations , we discuss below how the observations reported here can be used to constrain such configurations .
the gross characteristics of the jet - dominated images with disjoint bright regions described above ( see * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) can be captured by a simple geometric model composed of two separated regions that are symmetric and identical except for a difference in the brightness of each component .
the closure phases of such a configuration will be equivalent to an even more reduced model in which the two regions are replaced with point sources . in this reduced model ,
the visibilities are analytically calculable as @xmath27 , where @xmath28 represents the amplitude ratio of the two components , and @xmath29 and @xmath30 refer to their separation east and north , respectively .
figure [ two - point - sources ] shows the separations between two point sources that are consistent with our data .
these separations would produce california - hawaii - arizona closure phases that are between @xmath31 and @xmath32 , the range of values implied by the best - fit line in figure [ cphase - vs - gst ] , at all gst times for which we have measured closure phases .
both @xmath33 ( one point source ) and @xmath34 ( two equal point sources ) are inconsistent with our data , since each would produce closure phases that are identically zero or @xmath35 . for a separation comparable to the shadow diameter ,
the closure phase data imply that a disconnected two - component model would be oriented roughly east - west with the brighter component located to the west ( figure [ two - point - sources ] ) .
if the emission from sgr a * is coming from the footpoints of a mad - type jet , this is inconsistent with the orientations of the @xcite jet or @xcite clockwise disk ( see extended discussion in @xcite ) but aligned with the preferred axis of the intrinsic emission at 7 mm @xcite .
closure phase measurements at 86 ghz already rule out asymmetric jet structures on larger angular scales from a few hundred microarcseconds to a few milliarcseconds @xcite .
the closure phase on the california - hawaii - arizona triangle is consistent in sign and magnitude , to within measurement error , from day to day .
these observations span a 4-year timescale that is much longer than the orbital period at the innermost stable circular orbit , which ranges from a few minutes to about half an hour depending on the spin of the black hole .
it is also larger than the lense - thirring precession timescale for a tilted accretion disk unless the effective outer accretion flow radius is very large or the black hole spin is very small @xcite .
misalignment of the spin axes of the accretion disk and black hole could produce two different observational consequences .
first , it is possible that the inner disk could have a stable but slowly precessing structure .
examination of each epoch of the data in the context of disk ( e.g. , * ? ? ?
* ) or geometric crescent models @xcite may be able to place limits on the amount of precession .
second , misalignment could result in multiple bright regions in the accretion flow due to standing shocks .
grmhd simulations of the accretion flow find that the separation between these regions is comparable to the diameter of the photon ring , although the emission pattern can be quite complicated in general @xcite . since the standing shocks can travel faster than the lense - thirring precession speed , the accretion flow would be expected to have substantially different structure in different years and very likely on different days within each year . the predicted closure phases over the course of a day on the california - hawaii - arizona triangle for the 515h model of @xcite mimic the rough range of closure phases observed , including a general trend of increasing value with gst .
further study is required to determine whether such a model predicts excess variability in closure phases and long - baseline amplitudes beyond what is observed .
the increased sensitivity of the eht in upcoming years will be helpful for examining variability on intraday timescales .
the discovery of the g2 gas cloud on an orbit with a close approach to sgr a * @xcite sparked interest in the possibility that the accretion rate of sgr a * would increase due to the introduction of additional material into the accretion flow .
most of the material in g2 did not pass through pericenter until after the final epoch of observations reported herein @xcite , and mounting evidence suggests that g2 contains a star and is therefore not a pure gas cloud @xcite .
in any case , the infall timescale is on the order of years @xcite , so it would not be expected for there to be observational signatures of the g2 event in these data .
however , there is evidence that g2 is a knot in a larger gas streamer that also includes the g1 gas cloud , which reached pericenter in the middle of 2001 @xcite .
if so , the accretion flow of sgr a * could be supplemented with material from g1 or other gas in the streamer , with the caveat that some of the material deposited in the outer accretion flow may be carried away by outflows rather than making it to the inner region traced by the 1.3 mm emission @xcite .
the consistency of closure phases across multiple epochs from 2009 through 2013 provides evidence against large changes in the accretion rate over this period , consistent with the results of radio and millimeter - wavelength monitoring during the g2 encounter @xcite .
the grmhd simulations of @xcite are instructive .
when the accretion rate is decreased , the effective size of the accretion region decreases , with the result that the hawaii - arizona and hawaii - california baselines do not adequately resolve the 1.3 mm emission region , causing the predicted closure phases to drop very close to zero . as the accretion rate
is increased , the effective size of the emission region becomes larger , producing larger closure phases as the fringe spacing of the hawaii - arizona and hawaii - california baselines becomes better matched to the asymmetric emission region .
these larger closure phases persist even at the largest accretion rates modelled by @xcite , where the shadow of the emission region is obscured by the high optical depth of the accretion flow .
this behavior is also seen when the accretion rate of the @xcite radiatively inefficient accretion flow models is varied .
the average california - hawaii - arizona closure phase may therefore provide information complementary to visibility amplitudes in determining the overall accretion rate of the sgr a * system .
scattering in the tenuous plasma of the interstellar medium affects the image of sgr a * at radio wavelengths .
the biggest effect of this scattering is to blur the image of sgr a * , causing its apparent size to vary approximately as the square of the observing wavelength ( * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* and many others ) .
a secondary effect of scattering is to introduce variable substructure within the scattered image @xcite .
both of these effects can modify vlbi observables .
the formalism of @xcite and @xcite distinguishes between three different regimes of scattering . in the snapshot regime ,
diffractive scattering from small - scale inhomogeneities dominates . in the average regime , diffractive scattering
is quenched , but refractive scintillation from large - scale inhomogeneities persists . in the ensemble - average regime , both diffractive and refractive scintillation are suppressed , and the scattering produces a deterministic blurring of the image . due to the intrinsic size of sgr a * as well as the integration time and bandwidth used in vlbi observations , the average regime is applicable to eht observations of sgr a * over the course of a single night .
the ensemble of many nights of observations will tend statistically toward the ensemble - average regime , in which the observed visibilities are the intrinsic visibilities downweighted by a real gaussian whose width in baseline space is inversely related to the size of the scattering ellipse . in the ensemble - average regime ,
the effects of scattering blurring of the image are invertible and do not affect closure phases @xcite
. however , in the average regime , image distortions and refractive substructure can introduce nonclosing phases .
the magnitude of these effects may be up to 50 mjy in the visibility domain , with peak effect at baselines near the length at which the ensemble - average visibility of a point source falls to @xmath36 @xcite
. this could be as large as a 10% effect on top of source visibilities of approximately 500 mjy on the baselines between hawaii and the mainland us @xcite , corresponding to a phase noise of @xmath37 .
however , this refractive phase noise will be partially correlated when the antennas are located within a few thousand kilometers of each other , so the net effect on closure phases on the california - hawaii - arizona triangle will be smaller .
because refractive phase noise will fluctuate randomly about zero on timescales of about 1 day @xcite , it can not account for the nonzero closure phases that we measure , which show a strong tendency to be positive . nor can refractive phase noise account for the dependence of closure phase on gst .
thus , our measurements are a secure indication of intrinsic source asymmetry and do not merely reflect scattering - induced asymmetrical substructure .
however , refractive variations may contribute to smaller interday variations in closure phase .
further observations at a range of wavelengths will be required to characterize the properties of the turbulent scattering screen and better understand its contribution to the apparent variability of sgr a*.
we have obtained 181 measurements of the closure phase at 1.3 mm on the california - hawaii - arizona triangle from 2009 to 2013 .
the median closure phase is nonzero at high statistical significance .
this provides the first direct evidence that the structure of the 1.3 mm emission region is asymmetric on spatial scales comparable to the diameter of the shadow around the black hole that is predicted by general relativity .
if the 1.3 mm emission arises from a mad - like jet whose emission is concentrated in two disjoint bright regions separated by the shadow diameter , our data place a very strong constraint on its orientation .
the data also provide important constraints for parameters of other outflow and accretion models of sgr a*. the constancy of the sign of the closure phase argues for the persistence of asymmetric quiescent structure in sgr a * likely coupled with some structural variability , consistent with simulations of a dynamic , spin - aligned accretion disk . while it is not currently possible to entirely disentangle the effects of variability in the structure of the emitting material around sgr a * from apparent substructure introduced by variations in the scattering screen , the california - hawaii - arizona closure phases indicate that refractive phase noise is not dominant on baselines between hawaii and the western us .
these results are encouraging for producing an image of the quiescent emission by averaging several nights of data to mitigate intrinsic source variability and refractive substructure @xcite .
closure phase measurements will soon provide much stronger constraints on source structure . in the near term , eht observations in 2015 and beyond will incorporate additional observatories , including the large millimeter telescope in mexico , providing closure phase data on new triangles with higher angular resolution .
increasing data rates , starting with dual - polarization 2 ghz observations in 2015 , will provide increased sensitivity that will result in larger detection rates and smaller random errors on each closure phase measurement , allowing interday and intraday variability to be tracked more accurately .
completion of the 1.3 mm vlbi array including phased alma @xcite , the south pole telescope , the iram 30-m telescope at pico veleta , and the northern extended millimeter array at plateau de bure will produce very sensitive data with good baseline coverage , culminating in the ability to reconstruct model - independent images of sgr a * , m87 , and other sources @xcite .
the eht is supported by multiple grants from the national science foundation ( nsf ) and a grant from the gordon & betty moore foundation ( gbmf-3561 ) to s. s. d. the sma is a joint project between the smithsonian astrophysical observatory and the academia sinica institute of astronomy and astrophysics . the aro is partially supported through the nsf university radio observatories program .
the jcmt was operated by the joint astronomy centre on behalf of the science and technology facilities council of the uk , the netherlands organisation for scientific research , and the national research council of canada .
funding for ongoing carma development and operations was supported by the nsf and carma partner universities .
a. e. b. receives financial support from the perimeter institute for theoretical physics and the natural sciences and engineering research council of canada through a discovery grant .
research at perimeter institute is supported by the government of canada through industry canada and by the province of ontario through the ministry of research and innovation .
m. h. acknowledges support from mext / jsps kakenhi .
l. l. and g. n. o .- l .
acknowledge the financial support of dgapa , unam , and conacyt , mexico .
the eht gratefully acknowledges equipment donations from xilinx inc .
_ facilities : _
there are many places where potential errors may be introduced into vlbi data , including real - world imperfections in the signal chain and inaccuracies in the input model for correlation .
many potential sources of error close . among nonclosing errors ,
some introduce additional random ( zero - mean ) error into each closure phase estimate , while others may introduce biases . in order to characterize the significance of our results , we examine which errors
might potentially bias closure phase measurements . in this appendix
we consider potential sources of error from both theoretical and empirical perspectives .
for the latter , we appeal to the data themselves to characterize potential biases .
in addition to the sgr a * data reported in this manuscript , two other sources from the 2013 observations provide high - s / n data to test whether potential sources of error introduce measurable biases into closure phase measurements . the source with
the largest correlated flux density on long baselines in 2013 was bl lac , for which a long series of consecutive scans on day 086 provides a large sample with high bispectral s / n on all triangles , often exceeding 100 on the fps and gjt triangles .
another bright source that was observed over five nights in 2013 , 3c 279 , shows evidence of high polarization and complicated polarimetric structure on long baselines .
we use data from these sources to estimate an upper limit of the magnitude of potential biases by considering matched pairs of simultaneously measured closure phases from two different data subsets ( rcp vs. lcp , high band vs. low band , etc . ) on both nontrivial and trivial triangles .
each telescope uses a hydrogen maser as its timing and frequency standard .
the maser signal is very stable over timescales of minutes but usually exhibits a slow drift over longer timescales .
the difference between a one pulse - per - second ( pps ) signal from the maser and a pps signal from the global positioning system ( gps ) is logged over time , from which a time offset and drift rate are calculated .
these parameters serve as inputs to the model used for correlation .
the correlator model also takes as inputs the locations of the telescopes and the celestial coordinates of the source .
the location of each telescope , including the phase center of the phased arrays , has been measured with gps .
errors on the order of centimeters to tens of centimeters may be possible due to a combination of gps measurement errors and continental drift .
source coordinates are obtained from longer - wavelength vlbi catalogues and automatically corrected for precession .
however , these coordinates may contain errors on the order of milliarcseconds , and in any case the centroid of emission at 1.3 mm may be different from that measured at a longer wavelength due both to frequency - dependent source structure and to intrinsic source variability . because of these effects , as well as a rapidly varying atmosphere , the a priori correlator parameters are close , though not perfectly correct , for modelling the delay and rate of a fringe .
fringe finding is required post - correlation in order to find residual delays and rates to compensate for errors in the model . however , the total quantities ( delay and rate ) , the sum of the model and residual quantities , are independent of the input model provided that the a priori model was within the effective correlator search windows ( set by the accumulation period and spectral resolution ) .
thus , small station - based clock and position errors would not be expected to introduce biases into the closure phases calculated later in postprocessing .
the receivers on the eht telescopes were set up to receive left and right circularly polarized emission from the sky . since no feed or polarizer ( such as a quarter - wave plate ) is perfect , a system set up to receive lcp will nevertheless detect a small portion of rcp emission , and vice versa .
the full equations for the correlated quantities , including polarization leakage , are @xmath38 \and \nonumber \\ l_1l_2^ * = g_{1l}g_{2l}^ * & [ & ( i_{12}-v_{12 } ) e^{i(+\varphi_1-\varphi_2 ) } \nonumber \\ & & + d_{1l } d_{2l}^ * ( i_{12}+v_{12 } ) e^{i(-\varphi_1+\varphi_2 ) } \nonumber \\ & & + d_{1l } p_{12 } e^{i(-\varphi_1-\varphi_2 ) } \nonumber \\ & & + d_{2l}^ * p_{21}^ * e^{i(+\varphi_1+\varphi_2 ) } ] , \label{pol - equations}\end{aligned}\ ] ] where numeric subscripts indicate antennas , letter subscripts indicate the sense of the circularly polarized feed , asterisks indicate complex conjugation , @xmath39 indicates complex gain terms , @xmath40 indicates polarization leakage terms , @xmath41 and @xmath42 indicate stokes parameters representing the source total intensity and circular polarization , @xmath43 indicates the combination of stokes parameters representing the source linear polarization , and @xmath44 indicates the field rotation angle @xcite .
the field rotation angle depends on the mount of the telescope and the location of the receiver ; for some eht telescopes it is equal to the parallactic angle , and for others it is the parallactic angle plus or minus the elevation angle .
sgr a * exhibits very little ( @xmath45 ) circular polarization at 1.3 mm @xcite , so circular polarization would not be expected to introduce an error of more than a few tenths of a degree in the closure phase from the terms without leakage ( @xmath40 ) in equations ( [ pol - equations ] ) . as the field rotation angles rotate , the stokes i terms vary as the _ difference _ of the field rotation angle between the two stations . in the absence of polarization leakage ,
the rotations introduced by the field rotation angles on three closing baselines cancel , producing no net change in the measured closure phase .
in contrast , sgr a * exhibits high linear polarization at 1.3 mm .
the median effective linear polarization fraction ( @xmath46 ) ranges from about 5% on intrasite and smt - carma baselines to 35% on the jcmt - carma baseline , with substantial additional variability @xcite .
instrumental polarization leakage ( @xmath40 ) terms for the eht range from 1% at the sma to 11% at the smt , with a median value of about 5% .
polarization leakage can contaminate the rr and ll visibilities via the product @xmath47 , which for sgr a * amounts to a few percent .
this is not large enough for our measured median closure phase of @xmath48 to be attributable to polarization leakage .
leakages vary as the _ sum _ of the field rotation angles of the stations on each baseline , with effects on rcp and lcp closure phases that are approximately symmetric but opposite in sign . assuming the median baseline - dependent polarization fractions and average polarization angles measured for sgr a * in 2013 as well as the @xmath40-terms derived during instrumental polarization calibration , the typical biases expected to be introduced by polarization leakage are @xmath49 over most of an observing track , with a maximum effect of about @xmath50 .
the closure phase data on the most sensitive nontrivial lcp and rcp triangles ( fps and gjt , respectively ) do not have sufficient s / n to be able to detect this difference .
the trivial dfs and egt closure phases are consistent with each other to less than @xmath51 , and the slope of their difference with time is consistent with zero .
neither bl lac nor 3c 279 shows a statistically significant bias between the lcp and rcp closure phases on any triangle ( including the nontrivial california - hawaii - arizona triangle ) or a slope with time . as can be seen in equation ( [ pol - equations ] ) , the complex gain ( @xmath39 ) terms apply to the source term of each visibility as well as all of the leakage ( @xmath40 ) terms .
the closure phase is the argument of the product of three visibilities .
denoting gain - corrected source quantities ( bracketed terms in equation ( [ pol - equations ] ) ) by @xmath52 $ ] , the measured closure phase is @xmath53[r_2r_3^*][r_3r_1^*]\right ) \nonumber \\ & = & \arg\left([r_1r_2^*][r_2r_3^*][r_3r_1^*]\right).\end{aligned}\ ] ] thus , station - based , frequency - independent complex gain errors do not introduce additional closure phase errors beyond those already present due to polarization leakage .
hops provides partial mitigation of bandpass effects .
the amplitude of each station is automatically normalized to the autocorrelation response on a per - channel level .
this can theoretically still result in small errors when the autocorrelation response does not match the desired crosscorrelation response for instance , if the if response reduces sensitivity to sky signals near the edge of the band , or if subchannel structure in the bandpass response is significant .
eht telescopes do not have a pulse calibration system to align the phases of each channel .
instead , the eht data reduction uses manual instrumental phase calibration on very bright sources to derive phase offsets to apply to each channel on a per - station basis .
the signal for all frequencies within each 480 mhz band passes through the same chain of electronics before being sampled by a single backend . as a result , the manual phases required to flatten the phase structure of the bandpass vary smoothly across the band .
before they are removed , channel - to - channel phase differences are typically a few degrees to @xmath54 , with larger differences sometimes seen at the edges of the band .
data from bl lac indicate that any biases introduced by bandpass effects do not exceed a few tenths of a degree .
high - band and low - band closure phases on nontrivial triangles agree to less than a degree .
an additional test in which the low band was subdivided into two pieces ( the first 8 and the last 7 channels ) found consistent closure phases on all triangles at the level of @xmath55 or better .
whitney , a. , ruszczyk , c. , romney , j. , & owens , k. 2010 , sixth international vlbi service for geodesy and astronomy , proceedings from the 2010 general meeting ,
vlbi2010 : from vision to reality , ed .
d. behrend & k. d. baver , 373 | the galactic center black hole sagittarius a * ( sgr a * ) is a prime observing target for the event horizon telescope ( eht ) , which can resolve the 1.3 mm emission from this source on angular scales comparable to that of the general relativistic shadow .
previous eht observations have used visibility amplitudes to infer the morphology of the millimeter - wavelength emission .
potentially much richer source information is contained in the phases .
we report on 1.3 mm phase information on sgr a * obtained with the eht on a total of 13 observing nights over 4 years .
closure phases , the sum of visibility phases along a closed triangle of interferometer baselines , are used because they are robust against phase corruptions introduced by instrumentation and the rapidly variable atmosphere .
the median closure phase on a triangle including telescopes in california , hawaii , and arizona is nonzero .
this result conclusively demonstrates that the millimeter emission is asymmetric on scales of a few schwarzschild radii and can be used to break 180 rotational ambiguities inherent from amplitude data alone .
the stability of the sign of the closure phase over most observing nights indicates persistent asymmetry in the image of sgr a * that is not obscured by refraction due to interstellar electrons along the line of sight . |
the charmonium spectroscopy has been recently pushed forward with the unexpected discovery of many new charmonium - like resonances in the @xmath16 factories at slac , kek and cesr .
the @xmath16-factories , originally constructed to test matter - antimatter asymmetries or cp - violation , within or beyond the standard model , discovered a number of interesting charm and hidden charm mesons which do not seem to have a simple @xmath17 structure .
the first of these xyz states is the @xmath18 that was observed by the belle collaboration as a narrow peak near @xmath19 mev in the @xmath20 invariant mass distributions in @xmath21 decays @xcite .
belle also observed that the rate of the decay @xmath22 was comparable to @xmath23 @xcite , with the @xmath24 or @xmath25 coming from @xmath26 or @xmath27 which would imply the @xmath18 is a mixture of both @xmath6 and @xmath28 @xcite ( see however @xcite for alternative explanations of that ratio ) .
the close value of the @xmath18 mass to the sum of the masses @xmath29 mev led to consider the @xmath18 as a molecule - like bound state of a @xmath30 and a @xmath31 meson , and much work on the properties of these possible systems has been done @xcite . the next xyz states , that we will consider in this paper , are the @xmath32 , the @xmath3 , the @xmath5 and the @xmath33 .
the @xmath32 was observed in the double - charmonium production reaction @xmath34 with mass @xmath35 mev and width @xmath36 mev @xcite .
after that belle also observed a @xmath9 mass peak in the @xmath37 reaction @xcite . whereas the @xmath38 has been observed ,
there is no signal for the @xmath39 or the @xmath40 decays . because of that and the fact that the @xmath41 and @xmath42 were also produced in double - charm production , it was believed that the @xmath32 could have @xmath43 , being a @xmath44 charmonium state ( @xmath45 ) , but there are problems with this assignment since in this case the @xmath32 should have a mass @xmath46 @xmath47 mev or even higher @xcite .
thus , it seems very unlikely that the @xmath32 is a @xmath17 state .
belle has observed a state with a decay mode @xmath48 in @xmath49 decays @xcite and babar has confirmed it @xcite , although the values for the mass and width reported by babar are smaller than belle s values ( @xmath50 mev and @xmath51 mev is reported by belle and @xmath52 mev and @xmath53 mev by babar ) . in principle , the mass and width of the @xmath3 suggest a radially excited p - wave charmonium state but then the @xmath54 would be the dominant decay mode and has not been observed .
hence , it seems very unlikely that the reported @xmath3 and @xmath32 represent different decay modes of the same state @xcite .
the @xmath55 assignment in the case that the @xmath3 were a charmonium state is not very clear .
indeed , for a charmonium state with @xmath43 ( @xmath56 ) the mass is a little low and for @xmath57 the mass is too high @xcite , also if it had a simple @xmath17 structure one would expect that the open charm decay modes would be dominant , and the other , @xmath40 , negligible .
also , the large partial width @xmath48 , estimated above @xmath58 mev @xcite , is quite larger than the measured partial widths for any of the observed hadronic transitions between charmonium states .
belle has recently observed a mass peak in the @xmath9 system in the @xmath59 reaction @xcite .
this state with a mass of @xmath60 mev and a width of @xmath61 mev has been called @xmath5 .
the production mechanism ensures that it has @xmath62 .
the known charmonium states seen from a @xmath59 reaction have @xmath63 , thus , although this state could be identified with a @xmath64 ( @xmath45 ) or @xmath65 ( @xmath66 ) charmonium state , the mass predicted in the first case would be smaller ( @xmath47 mev ) and higher ( @xmath67 mev ) in the second case @xcite .
recently , the cdf collaboration at fermilab has announced a narrow peak near the @xmath68 threshold , which is designated as @xmath69 , observed in the @xmath70 decay , it has a mass @xmath71 mev and a width @xmath72 mev @xcite .
the width observed is quite different from the one reported by belle for the @xmath5 @xcite , suggesting that one is taking about a different state .
another point is the structure and quantum numbers .
there are some predictions : in @xcite the authors solved the schrdinger equation from the potential obtained using effective lagrangians and they found molecular solutions for @xmath73 and @xmath69 with @xmath74 , concluding that the @xmath73 and the @xmath69 are molecular partners . in @xcite ,
the authors assume that the @xmath3 and the @xmath69 are hadronic molecules with quantum numbers @xmath7 or @xmath8 whose constituents are the charm vectors @xmath9 for the @xmath3 and @xmath75 for the @xmath69 and they calculate the decay rates of the observed modes @xmath76 and @xmath77 for the case @xmath7 .
the coupling constants are determined by means of the compositeness condition @xcite and the results for these decay modes support the molecular interpretation of the @xmath73 and the @xmath69 . in @xcite
the authors use qcd sum rules to evaluate the mass of a possible mesonic state that couples to a molecular @xmath75 current . whereas in @xcite
a mass @xmath78 gev is found , concluding that it is possible to describe the @xmath69 as a @xmath79 molecular state , in @xcite they find a larger value of the mass @xmath80 gev . in @xcite
the authors suggest possible assignments for the charmonium - like `` x , y , z '' states , for example the @xmath5 could be assigned to a @xmath81 state .
also in @xcite the authors find that the p - charmonium @xmath82 associated to the @xmath69 is problematic because of the upper limit of the branching ratio of the decay @xmath83 computed of the order of @xmath84 . in @xcite a tetraquark structure is suggested for this state , while in @xcite it is argued than the peak at this energy is just a @xmath85 threshold effect .
the @xmath33 is also reported by belle as a peak in the spectrum of @xmath39 mesons produced in @xmath86 collisions , with mass and width @xmath87 mev and @xmath88 mev . in the production process
the two photons can only produce @xmath39 in a @xmath89 or @xmath8 states .
the belle measurements favors the @xmath8 hypothesis , making the assignment of the @xmath33 to the @xmath90 charmonium state possible @xcite , following the arguments given in @xcite . in this paper
we will propose a theoretical explanation on the nature of some of these xyz states , providing structure and quantum numbers for them .
our work is based on the hidden gauge symmetry ( hgs ) formalism for the interaction of vector mesons , which was introduced by bando - kugo - yamawaki @xcite .
this hidden gauge formalism has been applied in @xcite for the @xmath91 interaction , giving rise to two bound @xmath92 states that could be identified with the @xmath93 and @xmath94 .
after that , the radiative decay of these states into @xmath86 was studied in @xcite , obtaining results in good agreement with the pdg @xcite .
the work of @xcite was later extended to @xmath95 @xcite , where several states were found that could also be identified with some of the pdg .
recently , the @xmath96 , @xmath97 interaction was studied within the same formalism @xcite , generalizing to @xmath98 the basic hidden gauge lagrangians , but breaking the @xmath98 symmetry in the vector exchange diagrams .
some states were found which could be identified with the @xmath99 and the @xmath100 , giving a prediction of the quantum numbers for the last one @xmath101 in the case the correspondence was correct , and providing a reasonable explanation for the small width of this state .
also , a new @xmath102 state was predicted with @xmath103 , a mass close to @xmath104 mev and width of about @xmath105 mev .
the present paper follows the steps of @xcite , investigating possible vector - vector states of hidden charm , using a unitary approach in coupled channels including all the channels involved for charm @xmath0 and strangeness @xmath1 .
our starting point is the lagrangian , which involves the interaction of vector mesons amongst themselves , coming from the formalism of the hidden gauge symmetry ( hgs ) for vector mesons @xcite @xmath107 where the symbol @xmath108 stands for the trace in the @xmath98 space and @xmath109 is given by @xmath110\ , \label{vmunu}\ ] ] with @xmath111 given by @xmath112 and @xmath113 mev the pion decay constant . using the value of @xmath111 in eq .
( [ g ] ) is one of the ways to account for the ksfr relation @xcite which is tied to vector meson dominance @xcite .
the vector field @xmath114 is represented by the @xmath98 matrix which is parametrized by 16 vector mesons including the 15-plet and singlet of @xmath98 , @xmath115 where the ideal mixing has been taken for @xmath26 , @xmath116 and @xmath117 .
the interaction of @xmath118 gives rise to a contact term coming from @xmath119[v_\mu , v_\nu]$ ] @xmath120 depicted in fig .
[ fig : fig1 ] a ) , and on the other hand it gives rise to a three vector vertex @xmath121 depicted in fig .
[ fig : fig1 ] b ) .
this latter lagrangian gives rise to a @xmath122 interaction by means of the exchange of one of the vectors , as shown in figs .
[ fig : fig1 ] c ) , d ) .
lagrangian : a ) four vector contact term , eq .
( [ lcont ] ) ; b ) three - vector interaction , eq .
( [ l3v ] ) ; c ) @xmath123 and @xmath124 channels from vector exchange ; d ) @xmath125 channel for vector exchange.,width=604 ] the @xmath98 structure of the lagrangian allows us to take into account all the channels within @xmath98 which couple to certain quantum numbers . in the present work we shall present results for the case of all the channels which couple to @xmath9 or @xmath126
the formalism is the same used in @xcite and @xcite .
some approximations were made there which make the formalism handy and reliable , by neglecting the three - momentum of the vector mesons with respect to their masses .
it is interesting to see that with this approximation one obtains @xcite from the hidden gauge approach the chiral local lagrangians which are used to study the interaction of pseudoscalar mesons among themselves and the pseudoscalar mesons with vector mesons and with baryons @xcite .
the channels that we are interested in are those with charm @xmath0 and strangeness @xmath1 , they are , in the case of @xmath6 : @xmath127 , @xmath128 , @xmath129 , @xmath130 , @xmath131 + @xmath132 , @xmath133 , @xmath134 , @xmath135 , @xmath136 where the magnitude between parenthesis is the sum of the masses of the two meson involved , and for @xmath28 : @xmath127 , @xmath129 , @xmath130 , @xmath137 , @xmath138 , @xmath139 .
we are not interested in the case of @xmath140 which was considered in @xcite and where no bound states or resonances were found .
consider now the @xmath141 reaction , see fig .
[ fig : fig2 ] . in order to get the amplitude we use the lagrangian of eq .
( [ lcont ] ) obtaining interaction . ]
@xmath142 where the indices @xmath143 and @xmath144 correspond to the particles with momenta @xmath145 and @xmath146 in fig .
[ fig : fig2 ] .
it is straightforward to write down all amplitudes for the other channels . in the approximation of neglecting the three - momenta of the vector mesons , only the spatial components of the polarization vectors
are nonvanishing , the orbital angular momentum is @xmath147 , and then one can obtain easily spin projection operators @xcite into spin 0 , 1 , 2 states , which are given below : @xmath148 where the order @xmath149 in the polarization vectors is understood .
we can then write the combination of polarization vectors appearing in eq .
( [ eq : rhodvertex ] ) in terms of the spin combinations and thus we obtain the kernel of the interaction which will be later on used to solve the bethe - salpeter equation
. however , it is practical to construct the isospin combinations before the spin projection is done . recalling that we have isospin doublets @xmath150 and @xmath151 , the @xmath6 and @xmath58 combinations are written as @xmath152 we then find the amplitudes in the isospin base by forming linear combinations of the amplitudes in the particle base weighted by the clebsh - gordan coefficients as given in eq .
( [ eq : isospincomb ] ) , @xmath153 these amplitudes , after projection into the spin channels , give rise to the following kernels ( potential ) for @xmath6 , @xmath154 and also for the case of @xmath28 , @xmath155 in the same way , we proceed to calculate the contact term projected in isospin and also spin between two of anyone of the involved channels .
continuing with the @xmath156 reaction , now we want to calculate the amplitude of the first diagram in fig .
[ fig : fig6 ] . in order to do that , we follow the same procedure as in @xcite , starting off the three - vector vertex which is given by the lagrangian of eq .
( [ l3v ] ) and is depicted in fig . [
fig : fig5 ] in the case of @xmath157 .
the amplitude of fig .
[ fig : fig5 ] can be automatically calculated rewriting the lagrangian of eq .
( [ l3v ] ) as , @xmath158 as it was explained in @xcite , within the approximation of neglecting the three - momenta of external vectors , the @xmath159 corresponds to the exchanged vector , simplifying the calculation .
as already mentioned , this corresponds to the consideration of only the @xmath125-wave . the vertex function corresponding to the diagram of fig .
[ fig : fig5 ] is given by @xmath160 with this basic structure , and considering all the particles involved in the exchange , we can readily evaluate the amplitude of the first diagram of fig .
[ fig : fig6 ] to obtain . ]
@xmath161 in the vector exchange diagrams the exchange of one heavy - vector meson becomes supressed with respect to the exchange of a light vector by the ratio @xmath162 with @xmath163 and @xmath164 . to project in s - wave one must do the following replacements : @xmath165 where @xmath166 means the projection over s - wave , and @xmath167 , @xmath168 , @xmath169 , @xmath170 and @xmath171 , with @xmath172 , is the mass of each particle , and so on for the other products of momenta .
thus , eq . ( [ exchange ] ) can be written as @xmath173 the isospin projections taking into account the three diagrams of fig .
[ fig : fig6 ] and eq .
( [ eq : isospincomb ] ) give us @xmath174 now using the equations for the spin projections we can split the terms into their spin parts and we obtain @xmath175 for the vector exchange diagrams we obtain the same expression for all the spin states .
one must do the same calculations for every channel , but , one can see that all the terms corresponding to the t - channel have the structure @xmath176 and those that corresponds to the u - channel @xmath177 in tables [ tab : am1 ] , [ tab : am2 ] , [ tab : am3 ] , [ tab : am4 ] , [ tab : am5 ] and [ tab : am6 ] of the appendix we give the @xmath178 and @xmath179 amplitudes projected into isospin and spin .
the amplitudes , @xmath180 , @xmath40 , @xmath181 , @xmath92 , @xmath182 , @xmath183 , @xmath180 , @xmath40 , @xmath184 , @xmath185 for @xmath6 and @xmath186 , @xmath92 , @xmath187 , @xmath188 , @xmath189 for @xmath28 , are not in the tables because the interaction is exactly zero in our model . also , the @xmath95 amplitudes that involve the @xmath190 , @xmath92 , @xmath182 , @xmath183 and @xmath185 channels are in the tables v - x ( contact term ) and xviii - xxi ( exchange term ) in @xcite . here
we have used the constant @xmath191 with @xmath192 mev for light mesons and @xmath193 , @xmath194 , @xmath195 with @xmath196 mev @xcite , @xmath197 mev @xcite , @xmath198 mev @xcite for the @xmath199 , @xmath200 and @xmath117 mesons in the channels that involve one light meson and the @xmath117 meson . ] . as we see in these tables
the exchange term is equal for @xmath63 and @xmath201 and the exchange of one heavy vector meson is supressed by the factor @xmath202 which is equal to @xmath203 in the case of @xmath204 and @xmath205 .
we can observe that the interaction at the @xmath9 threshold becomes attractive and very strong for @xmath206 , @xmath207 and @xmath208 amplitudes , which will lead to bound states of two vector mesons , basically @xmath9 and/or @xmath209 , in the region of interest around @xmath2 mev .
the formalism that we are using is also allowed for @xmath125-channel vector exchange and we can have the diagram of fig .
[ fig : fig8 ] .
but we found in @xcite that this leads to a @xmath210-wave interaction for equal masses of the vectors , and only to a minor component of @xmath125-wave in the case of different masses @xcite , hence , we do not consider these terms .
-channel @xmath27 exchange diagram . ]
the results of the amplitudes discussed in the section 2.2 and 2.3 provide the kernel or potential @xmath211 to be used in the bethe - salpeter equation in its on - shell factorized form , @xmath212 the potencial @xmath211 here is a @xmath213 matrix in @xmath6 with the amplitudes obtained from the channels @xmath127 , @xmath128 , @xmath129 , @xmath130 , @xmath131 + @xmath132 , @xmath133 , @xmath134 , @xmath135 , @xmath136 in its elements for each spin @xmath214 independently . also , in @xmath28
, @xmath211 is a @xmath215 matrix whose elements come from the transition potentials between the channels @xmath127 , @xmath129 , @xmath130 , @xmath137 , @xmath138 , @xmath139 . in eq .
( [ bethe ] ) @xmath216 is a diagonal matrix where its elements are the two meson loop function @xmath217 for each channel @xmath218 : @xmath219 which upon using dimensional regularization can be recast as @xmath220 where @xmath221 is the total four - momentum of the two mesons and @xmath210 is the three - momentum of the mesons in the center - of - mass frame : @xmath222 analogously , using a cut off one obtains @xmath223 } \ , \label{loopcut}\ ] ] where @xmath224 stands for the cut off in the three - momentum , @xmath225 and the square of center - of - mass energy @xmath226 . in the complex plane , for a general @xmath227 , the loop function in the second riemann sheet can be written as @xcite : @xmath228 where @xmath229 refer to the loop function in the second riemann sheet and @xmath230 is the loop function in the first riemann sheet given by eqs .
( [ dimreg ] ) and ( [ loopcut ] ) for each channel @xmath218 .
once we introduce the potentials obtained from the tables [ tab : am1 ] , [ tab : am2 ] , [ tab : am3 ] , [ tab : am4 ] , [ tab : am5 ] and [ tab : am6 ] ( in the appendix ) as a kernel @xmath211 in eq .
( [ bethe ] ) , we evaluate the transition matrix @xmath231 between channels and look at @xmath232 in the real axes and look for poles in the second riemann sheet of the complex plane .
if these poles are not very far from the real axis they occur in @xmath233 .
since the only meaningful physical quantity is the value of the amplitudes for real @xmath227 , only poles not too far from the real axes would be easily identified experimentally as a resonance .
the amplitude in eq .
( [ bethe ] ) close to a pole looks like @xmath234 the constants @xmath235 ( @xmath236 ) , which provide the coupling of the resonance to one particular channel are calculated by means of the residues of the amplitudes .
we have set the parameters of eq .
( [ dimreg ] ) as follow : we have fixed the value of @xmath237 as @xmath238 mev for all the channels , the subtraction constant for the loops containing @xmath95 mesons , that we call as @xmath239 , is set to @xmath240 in order to find the position of the pole @xmath241 as in @xcite , for the channels where the two mesons are heavy we put @xmath242 , value that is chosen to get the position of the pole found in @xmath1 around @xmath243 mev , and for the rest of the channels , @xmath40 , @xmath244 , @xmath245 , we also put @xmath246 .
one could do a study of the theoretical uncertainties for the positions and widths to account for underlying su(4 ) symmetry breaking of the model .
this study was done in @xcite and one finds similar results here . in summary , using the freedom of the model to use @xmath247 or @xmath248 and @xmath249 , but readjusting the substraction constants to obtain one of the resonances with the same mass as the experiment , one finds changes in the masses smaller than @xmath250 mev and changes in the couplings smaller than @xmath251 .
we find four poles in @xmath6 and one pole in @xmath28 .
the pole positions and the coupling constans @xmath235 are given in tables [ tab : res1 ] , [ tab : res2 ] , [ tab : res3 ] , [ tab : res4 ] and [ tab : res5 ] .
in @xmath6 we have found one pole with a mass @xmath252 mev for each spin @xmath214 and from the couplings we see that they couple stronger to @xmath9 . for @xmath63 and @xmath253 ,
the width found is very small , @xmath254 and @xmath255 mev respectively , but for @xmath201 the width is comparatively larger @xmath256 mev .
another pole is also found for @xmath6 and @xmath201 with mass @xmath257 mev and width @xmath258 mev which couples now stronger to @xmath209 . in the @xmath28 sector
, we find a pole only for @xmath201 with mass @xmath259 and width @xmath260 that couples mostly to @xmath9 . in @xcite ,
the xyz mesons observed up to now are listed in table 1 . to make the appropiate correspondence ,
let us quote from @xcite the properties of the xyz states discussed , which we show in table [ tab : new ] .
c|c|c|c|c|c state & m ( mev ) & @xmath261 ( mev ) & @xmath55 & decay modes & production modes + @xmath4 & @xmath262 & @xmath263 ' '' '' ' '' '' & @xmath8 & @xmath39 & @xmath264 + @xmath32 & @xmath265 & @xmath266 ' '' '' ' '' '' & @xmath267 & @xmath268 & @xmath269 + @xmath3 & @xmath270 & @xmath271 ' '' '' ' '' '' & @xmath267 & @xmath40 & @xmath272 + & @xmath273 & @xmath274 ' '' '' ' '' '' & & & + @xmath5 & @xmath275 & @xmath276 ' '' '' ' '' '' & @xmath267 & @xmath277&@xmath278 + looking at this table , we see that there are three states with mass around @xmath243 mev .
the identification is subtle .
at first one could think that the three states around @xmath243 mev could correspond to our states found around this mass .
however , all the three experimental states have c - parity positive , while our state at @xmath279 mev has c - parity negative . hence
, this state is a prediction and does not correspond to any of the experimental ones .
we call it @xmath280 where @xmath210 stands for prediction . the other two states that appear in table [ tab : new ] with masses around @xmath243 mev are the @xmath4 and the @xmath3 .
the @xmath4 has been seen in the spectrum of @xmath39 produced in @xmath264 collisions with mass @xmath281 mev and @xmath282 mev and the two - photon production process can only produce @xmath39 in a @xmath89 or @xmath8 state . from the angular distributions the belle measurement strongly favors the @xmath8 hypothesis , and for this reason we associate this state to our @xmath8 state , which is found with mass and width @xmath283 mev and @xmath284 mev .
for the @xmath3 there are different measurements of the mass and width , whereas babar reports @xmath285 mev and @xmath286 mev , the belle s values are @xmath287 mev and @xmath288 mev .
even though the difference in the values of the mass is not large , there is a considerable uncertainty in the width .
this is the state that we associate to our @xmath89 state , which we find with mass @xmath289 mev and width @xmath290 mev .
the point in favor of this assignment is the result of the calculation of @xmath291)\to \omega j/\psi)$ ] , that can be done straightforward by means the formula : @xmath292)\to \omega j/\psi)=\frac{p\,|g_{y\omega j/\psi}|^2}{8 \pi\,m_y^2 } \label{decay}\ ] ] with @xmath210 the momentum of @xmath26 in the resonance rest frame .
taking the coupling @xmath293 mev from table [ tab : res1 ] , we obtain @xmath291)\to \omega j/\psi)=1.52 $ ] mev , which is compatible with @xmath294 mev , obtained from the measured product of branching fractions @xmath295 ( @xmath296 , reported by belle , and @xmath297 according to babar ) together with the assumption that @xmath298 is less than or equal to @xmath299 , the typical value for allowed @xmath300 charmonium decays .
thus , we find a natural explanation on why this rate is much larger than it would be should it correspond to hadronic transitions between charmonium states @xcite .
the state @xmath32 of table [ tab : new ] does not decay into @xmath40 @xcite .
this is a reason not to associate it to our @xmath301)$ ] .
this state does not fit into our vector - vector scheme , suggesting it has a different nature .
the state found with @xmath302 , @xmath303 mev and @xmath258 mev can be clearly identified with the @xmath5 by the proximity of mass , width and quantum numbers . in the next section
, we are going to include the @xmath39 channel by means a box diagram as it was done in @xcite .
@xmath304 , @xmath305=0^+[0^{++}]$ ] .couplings @xmath306 in units of mev for @xmath6 , @xmath63 . [ cols="^,^,^,^,^",options="header " , ] for @xmath28 and @xmath201 in the main channel @xmath9.,title="fig:",width=264 ] + for @xmath6 and @xmath63 before and after the inclusion of the @xmath39 box diagrams.,title="fig:",width=340 ] +
we have made a full study of the vector - vector interaction in the @xmath0 and @xmath1 sector using the hidden gauge formalism .
the interaction comes from contact terms plus vector meson exchange in the @xmath123-channel .
we have found a strong attraction in the @xmath6 , @xmath214 and @xmath28 , @xmath201 sectors , enough to bind the vector - vector system . by looking for poles in the second riemann sheet ,
we have found five resonances , three of which can be associated with the experimental data : the state @xmath307)$ ] to the @xmath3 , the @xmath308)$ ] to the @xmath33 and the @xmath309)$ ] to the @xmath5 .
there is no experimental counterpart for our state @xmath310)$ ] , which is thus a prediction of our model .
these three states with mass around @xmath243 mev are basically composed by @xmath9 , and decay into pairs of light vectors like @xmath190 , or light vector - heavy vector as @xmath40 .
our model predicts another state around @xmath311 mev , @xmath309)$ ] , which we identify with the @xmath5 state in base to the proximity of mass and widths and c - parity .
this resonance has @xmath302 and is mostly @xmath209 . in the @xmath28 sector , the attraction is weak and we find only one resonance in the case of @xmath201 , @xmath312)$ ] , the possible association of this state to the @xmath32 is unlikely since our state can decay to @xmath39 , though with small intensity , but this decay is not seen for the @xmath32 .
the width that we obtain is also considerably larger than that of the @xmath32 , @xmath313 mev .
as we have shown in this work , the region around @xmath243 mev is very interesting and there could be more resonances not yet seen in this region .
the findings of this work should motivate the experimentalist to look into this region in the channels that involve light vector - light vector or light vector - heavy vector like @xmath190 and @xmath188 .
this work is partly supported by dgicyt contract number fis2006 - 03438 .
we acknowledge the support of the european community - research infrastructure integrating activity study of strongly interacting matter ( acronym hadronphysics2 , grant agreement n. 227431 ) under the seventh framework programme of eu
. 99 s. k. choi _ et al .
_ ( belle collaboration ) , phys . rev .
lett . * 91 * , 262001 ( 2003 ) d. e. acosta _ et al . _
[ cdf ii collaboration ] , phys .
lett . * 93 * , 072001 ( 2004 ) v. m. abazov _ et al . _
[ d0 collaboration ] , phys .
lett . * 93 * , 162002 ( 2004 ) b. aubert _ et al . _ [ babar collaboration ] , phys .
d * 71 * , 071103 ( 2005 ) k. abe , et al .
( belle collab . )
belle - conf-0540,arxiv : hep - ex/0505037 .
pap . lp-2005 - 175 .
lepton - photon interact
. high energy , xxii , uppsala , swed .
( 2005 ) f. e. close and p. r. page , phys .
b * 578 * , 119 ( 2004 ) e. s. swanson , phys .
b * 588 * , 189 ( 2004 ) e. braaten and m. kusunoki , phys . rev .
d * 72 * , 054022 ( 2005 ) d. gamermann and e. oset , arxiv:0905.0402 [ hep - ph ] .
m. b. voloshin and l. b. okun , jetp lett .
* 23 * , 333 ( 1976 ) [ pisma zh .
* 23 * , 369 ( 1976 ) ] .
a. de rujula , h. georgi and s. l. glashow , phys .
* 38 * , 317 ( 1977 ) .
n. a. tornqvist , z. phys .
c * 61 * , 525 ( 1994 ) m. b. voloshin , phys .
b * 579 * , 316 ( 2004 ) d. gamermann and e. oset , eur .
j. a * 33 * , 119 ( 2007 ) k. abe _ et al . _ , phys .
* 98 * , 082001 ( 2007 ) p. pakhlov _ et al . _
[ belle collaboration ] , phys .
lett . * 100 * , 202001 ( 2008 ) t. barnes , s. godfrey and e. s. swanson , phys .
d * 72 * , 054026 ( 2005 ) k. abe _ et al . _ [ belle collaboration ] , phys . rev .
lett . * 94 * , 182002 ( 2005 ) b. aubert _ et al . _ [ babar collaboration ] , phys .
* 101 * , 082001 ( 2008 ) s. l. olsen , arxiv:0901.2371 [ hep - ex ] .
s. godfrey and s. l. olsen , ann .
nucl . part .
sci . * 58 * , 51 ( 2008 ) t. aaltonen et al . ( the cdf collaboration ) [ arxiv:0903.2229 [ hep - ph ] ] .
x. liu and shi - lin zhu .
[ arxiv:0903.2529v2 [ hep - ph ] ] t. branz , t. gutsche and v. e. lyubovitskij [ arxiv:0903.5424v1 [ hep - ph ] ] zhi - gang wang [ arxiv:0903.5200v1 [ hep - ph ] ] r. m. albuquerque , m. e. bracco , m. nielsen [ arxiv:0903.5540v1 [ hep - ph ] ] bai - qing li and kuang - ta chao [ arxiv:0903.5506v1 [ hep - ph ] ] x. liu [ arxiv:0904.0136v2 [ hep - ph ] ] fl .
stancu [ arxiv:0906.2485v1 [ hep - ph ] ] e. van beveren , g. rupp [ arxiv:0906.2278v1 [ hep - ph ] ] m. bando , t. kugo , s. uehara , k. yamawaki and t. yanagida , phys .
* 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 ) r. molina , d. nicmorus and e. oset , phys .
d * 78 * , 114018 ( 2008 ) h. nagahiro , j. yamagata - sekihara , e. oset and s. hirenzaki , [ arxiv:0809.3717 [ hep - ph ] ] . c. amsler
_ et al . _
[ particle data group ] , phys .
b * 667 * , 1 ( 2008 ) .
l. s. geng and e. oset , l. s. geng and e. oset , phys .
d79 ( 2009 ) 074009 .
r. molina , h. nagahiro , a. hosaka , e. oset [ arxiv:0903.3823v1 [ hep - ph ] ] .
h. nagahiro , l. roca , a. hosaka and e. oset , phys .
d * 79 * 014015 ( 2009 ) riazuddin and fayyazuddin , phys .
rev . * 147 * , 1071 ( 1966 ) .
sakurai , currents and mesons ( university of chicago press , chicago il 1969 ) j. gasser and h. leutwyler , nucl .
b * 250 * , 465 ( 1985 ) .
u. g. meissner , rept .
phys . * 56 * , 903 ( 1993 ) .
m. c. birse , z. phys .
a * 355 * , 231 ( 1996 ) g. ecker , j. gasser , h. leutwyler , a. pich and e. de rafael , phys .
b * 223 * , 425 ( 1989 ) .
j. f. sun , d. s. du and y. l. yang , eur .
j. c * 60 * , 107 ( 2009 ) l. roca , e. oset and j. singh , phys .
d * 72 * , 014002 ( 2005 ) f. s. navarra , m. nielsen and m. e. bracco , phys .
d * 65 * 037502 ( 2002 )
c|c|c|c & contact & v - exchange & @xmath314 $ ] + | we study the vector - vector interaction within the framework of the hidden gauge formalism for the channels with quantum numbers charm @xmath0 and strangeness @xmath1 in the energy region around @xmath2 mev . by looking for poles in the complex plane
we find three resonances that could be identified by the mass , width and quantum numbers with the @xmath3 , @xmath4 and @xmath5 , these poles appear with isospin @xmath6 and @xmath7 , @xmath8 and @xmath8 respectively .
whereas the @xmath3 and @xmath4 are coupled more strongly to @xmath9 , the @xmath5 is basically a @xmath10 molecular state .
another two extra resonances appear in our approach with @xmath11 and @xmath12 which are not found in the pdg with masses @xmath13 mev and widths @xmath14 mev respectively .
@xmath15 departamento de fsica terica and ific , centro mixto universidad de valencia - csic , institutos de investigacin de paterna , aptdo .
22085 , 46071 valencia , spain + |
@xmath0 bond - alternating heisenberg chains have attracted a great amount of attention both theoretically and experimentally .
the system shows a quantum phase transition between the haldane - gap phase and the dimer phase depending on the bond - alternating ratio @xcite .
thermodynamic properties at the gapless point @xcite and in the dimer phase @xcite were studied experimentally .
the results were compared with theoretical results and good agreement was obtained @xcite .
the elementary excitation was investigated by quantum monte carlo @xcite and exact diagonalization @xcite methods . in the haldane - gap phase close to the uniform chain
, the low - lying excitations are expected to be scattering states of the domain walls in the hidden antiferromagnetic ordering @xcite . in the vicinity of the isolated dimer system ,
the low - lying excitations are well described by the @xmath0 magnon @xcite .
dynamical structure factors in the haldane - gap and dimer phases were calculated by the exact diagonalization method @xcite . in both phases , the distributions of the scattering intensity of the one - magnon mode and the excitation continuum were investigated .
it is known that a feature of the haldane - gap phase appears in transverse magnetic fields .
in fact , the energy spectrum in transverse fields was measured for the anisotropic haldane - gap compound @xmath3 ( abbreviated to nenp ) by inelastic neutron - scattering experiments @xcite and the satisfying agreement between the theoretical @xcite and experimental results was achieved .
many @xmath0 bond - alternating heisenberg - chains compounds possess noticeable single - ion anisotropy .
characteristic spin dynamics in the haldane - gap and dimer phases may thus emerge in transverse magnetic fields .
however , little knowledge had been obtained about dynamical properties of @xmath0 bond - alternating heisenberg chains in transverse magnetic fields . quite recently
, inelastic neutron - scattering experiments on the dimer - phase compound @xmath1 ( abbreviated to ntenp ) in transverse magnetic fields ( @xmath4 ) revealed fascinating aspects of dynamical properties @xcite . in @xmath5
, the triplet magnon excitation is lifted into the lower @xmath6 branch and the higher @xmath7 branch owing to a single - ion anisotropy .
the observed scattering intensity of the higher @xmath7 branch was anomalously weak at @xmath8 as compared with that in the anisotropic haldane - gap compound @xmath2 ( abbreviated to ndmap ) , where the wave number is represented in the extended zone scheme . in @xmath9 ,
the degenerate lower branch is lifted . as @xmath4 increases , the excitation energy of the highest ( lowest ) branch increased ( decreased ) , while that of the middle branch was almost independent of @xmath4 . in ntenp ,
the intensity of the highest branch decreased rapidly with increasing @xmath4 and disappeared at @xmath10 , whereas the lower two branches survived up to @xmath11 .
note that @xmath12 is the critical field where the excitation gap closes . in @xmath13
, there appeared only one gapped branch at @xmath8 which continued from the middle branch in @xmath14 .
these findings are in contrast with those observed in nenp @xcite and ndmap @xcite under transverse magnetic fields : in @xmath15 three branches showed conspicuous intensity up to @xmath11 and in @xmath13 three gapped branches were observed .
it is desirable to clarify the origin of the different spin dynamics between these systems . in this paper , we calculate the dynamical structure factor ( dsf ) of the @xmath0 bond - alternating heisenberg chain with a single - ion anisotropy in transverse magnetic fields , using a continued - fraction method based on the lanczos algorithm @xcite . by shifting the bond - alternating ratio systematically ,
dynamical properties in the haldane - gap phase and the dimer phase are investigated . in sec .
ii , we briefly summarize the method for the numerical calculation . in sec .
iii , we show the results for the dsf , turning our attention to the behavior at @xmath8 . in sec .
iv , the characteristic field dependence in the haldane - gap phase and dimer phase is discussed in viewpoint of the field dependence of the excitation continuum .
the origin of different spin dynamics observed in ntenp and ndmap is also discussed .
v is devoted to the summary .
let us consider the @xmath0 bond - alternating heisenberg chain with a single - ion anisotropy in transverse magnetic fields described by the following hamiltonian , @xmath16 where @xmath17 and @xmath18 is a bond - alternating ratio .
the periodic boundary condition is applied .
the dfs can be expressed as @xcite @xmath19 where @xmath20 is the eigenstate of @xmath21 with the lowest eigenvalue @xmath22 , @xmath23 with @xmath24 being the total number of spins , and @xmath25 .
the lattice constant between neighboring two spins is set to unity .
therefore , @xmath26 corresponds to the boundary of the brillouin zone in case of @xmath27 .
we set @xmath28 and @xmath29 .
the energy is measured in units of @xmath30 . in the expression ( [ ham2 ] ) , @xmath31 is the static structure factor and @xmath32 is represented in the form of the continued fraction , which can be calculated numerically by lanczos algorithm .
the total contribution of @xmath32 for a fixed @xmath33 is normalized to unity , because the sum rule , @xmath34 , has to be satisfied . instead of taking @xmath35 , we set @xmath36 . in finite - size systems , therefore , @xmath32 consists of a finite number of lorentzians .
the integrated value of each lorentzian with respect to @xmath37 in given @xmath33 is called as the residue @xcite . to discuss whether an excited state forms the isolated mode or the excitation continuum in the thermodynamic limit , the criterion proposed by takahashi @xcite may be effective . according to this criterion
, the residue of an excited state in the continuum tends to decrease with increasing the system size @xmath24 , while the residue for an isolated mode hardly depends on @xmath24 .
this method was successfully applied @xcite to explain spin dynamics of nenp observed by inelastic neutron - scattering measurements @xcite . adopting the parameter sets @xmath38 , and @xmath39
, we calculate the dfs up to @xmath40 spin systems in @xmath5 and @xmath41 spin systems in @xmath42 .
the parameter sets @xmath43 , and @xmath39 describe ndmap @xcite , ntenp @xcite , and the dimer - phase compound nmaop @xcite , respectively . according to the phase diagram @xcite
, the parameter set @xmath44 belongs to the haldane - gap phase and is located nearer the gapless line than ndmap @xcite . in the dimer phase , ntenp is situated nearer the gapless line than nmaop . in the next section ,
we show the results for @xmath45 and @xmath46 .
the dsf s in @xmath5 are shown in fig .
[ fig1 ] . except for ndmap ,
the results are shown in the extended zone scheme .
the area of the circle is proportional to the scattering intensity .
we turn our attention to the behavior at @xmath8 . in @xmath44 and ntenp , the next - lowest - excited state in @xmath47 with relatively large intensity
is located close to the lowest excited state , while in @xmath48 the next - lowest - excited state lies well above the lowest excited state . in ndmap and nmaop
, there appear appreciable gaps between them in both @xmath47 and @xmath48 .
and @xmath49 at @xmath5 for @xmath40 .
the area of the circle is proportional to the scattering intensity . ] .
insets : the @xmath24 dependence of the residue at @xmath8 . ] in fig .
[ fig2 ] , we show the finite - size effects of the residues of the lowest excited states in @xmath45 and @xmath49 . as shown in the insets , the residues of @xmath47 and @xmath48 in ndmap and nmaop show little size dependence , indicating that their lowest excited states form the isolated modes . in @xmath44 and ntenp , on the contrary , the residues in @xmath47 decrease with increasing @xmath24 , while the residues in @xmath48 scarcely depend on @xmath24 . in spite of such appreciable size dependence
, we have to be careful to decide the characteristic of the lowest excited state of @xmath47 in @xmath44 and ntenp . since their next - lowest - excited states are located close to the lowest excited states as shown in fig .
[ fig1 ] , the residues of the lowest excited states may be suffering from the higher energy states . in @xmath5 . ]
the excitation energy and the intensity of the watched lower excited states are extrapolated to @xmath50 .
their @xmath24 dependence is shown in fig .
we first evaluate the ratio of the intensity of the lowest excited state in @xmath47 to that in @xmath48 at @xmath5 .
the lowest excited states in @xmath47 and @xmath48 at @xmath5 are the @xmath7 and @xmath6 branches , respectively .
the resultant ratios are @xmath51 for ndmap , @xmath52 for @xmath44 , @xmath53 for ntenp , and @xmath54 for nmaop .
our results for ndmap and ntenp are consistent with the experimental results @xcite .
is enlarged by a factor of @xmath55 as compared with the intensity in the other figures .
the solid and broken lines denote the excitation energies for the isolated mode and lower edge of the excitation continuum , respectively .
the critical fields are @xmath56 for ndmap , @xmath57 for @xmath44 , @xmath57 for ntenp , and @xmath58 for nmaop . ]
the field dependence of the excitation energy and the intensity at @xmath8 is investigated in the same way .
the results are summarized in fig .
in magnetic fields , the lowest excited state in @xmath48 separates into two branches .
as @xmath4 increases , the lower branch shifts to the lower energy region and is softened at @xmath12 , which results in the quantum phase transition .
the critical fields are evaluated as @xmath56 for ndmap , @xmath57 for @xmath44 , @xmath57 for ntenp , and @xmath58 for nmaop .
note that the gapped excitation energy is fitted well with @xmath59 , while the softened mode is proportional to @xmath60 at @xmath61 @xcite .
we first discuss the behavior in @xmath47 . as @xmath4 increases , the lowest excited state in @xmath47 shifts towards the higher energy region , while the next - lowest - excited state shifts to the lower energy region . in ndmap , the isolated mode never meets the excitation continuum even around @xmath11 .
therefore , the distinct intensity of the isolated mode appears in @xmath15 . in @xmath44 ,
the isolated mode merges into the excitation continuum around @xmath62 , where the intensity of the isolated mode becomes so weak .
note that there are many excited states above the lower edge of the excitation continuum .
however , their intensity is so small that they are invisible in the present scale . in ntenp and nmaop
, the intensity of the lowest excited states in @xmath47 diminishes with increasing @xmath4 and disappears around @xmath63 and @xmath64 , respectively , where the lowest excited states cross the excitation continua . after the crossing , the replaced lowest excited states reach @xmath65 at @xmath61 in ntenp .
judging from the results for @xmath44 and ntenp , at @xmath5 their lowest excited states in @xmath47 form the isolated modes . in @xmath48 ,
on the other hand , the excitation continua of the four systems are located in the higher energy region up to @xmath61 .
accordingly , two isolated modes appear in @xmath15 in @xmath48 .
dependence of the residue at @xmath8 in @xmath13 ( left figures ) .
@xmath66 for ndmap , @xmath67 for @xmath44 , @xmath66 for ntenp , and @xmath68 for nmaop . the extrapolated intensity and the excitation energy in @xmath47 and @xmath48
are presented in the same figure ( right figures ) .
insets : extrapolation of the intensity ( left figures ) and the excitation energy ( right figures ) to @xmath50 . ]
we next investigate dynamical properties in @xmath13 using the same method . in @xmath13 ,
the lowest - lying excitation becomes gapless , yielding the peak with the largest intensity in @xmath69 irrespective of @xmath4 .
we disregard the behavior of this peak . in fig .
[ fig5 ] , we show the results for ndmap at @xmath66 , @xmath44 at @xmath67 , ntenp at @xmath66 , and nmaop at @xmath68 .
the residues of the peaks are shown in the left figures , and the extrapolated intensity and energy are presented in the right figures .
the extrapolation of the intensity and the excitation energy are shown in the insets . in ndmap
, two appreciable peaks appear in @xmath47 , while one peak appears in @xmath48 .
their residues are almost independent of @xmath24 and the extrapolated intensity takes nonzero values . therefore ,
these three peaks are from the isolated modes .
note that at @xmath11 the lower isolated mode in @xmath47 emerges at @xmath70 , after the lower branch in @xmath48 becomes softened . in @xmath44
, one appreciable peak appears in @xmath47 and @xmath48 , respectively .
their residues hardly depend on @xmath24 and the extrapolated intensity takes nonzero values .
thus , these two peaks are from the isolated modes . since the excitation continuum at @xmath66 is located around @xmath71 , only the lower branch of @xmath47
emerges . in ntenp
, only one noticeable peak appears in @xmath48 .
its residue scarcely depends on @xmath24 , while the residue of the lowest excited state in @xmath47 decreases with increasing @xmath24 .
the results indicate that the lowest excited state in @xmath48 forms the isolated mode , while that in @xmath47 forms the lower edge of the excitation continuum .
in fact , the intensity of the latter state extrapolated to @xmath50 is much smaller than the extrapolated former one as shown in the inset .
the lower edge of the excitation continuum of @xmath47 is evaluated as @xmath72 at @xmath66 , which makes the isolated mode in @xmath47 unstable in @xmath73 . on the contrary ,
the excitation continuum of @xmath48 lies in @xmath74 .
accordingly , in ntenp only one peak emerges in @xmath48 . in nmaop
, one peak appears in @xmath47 and @xmath48 , respectively .
their residues hardly depend on @xmath24 , indicating that they are from the isolated modes . in @xmath13 , the excitation continuum in @xmath47 shifts to the low energy region . at @xmath68 its lower edge
is located at @xmath72 , which is close to the isolated mode .
therefore , as @xmath4 increases in @xmath13 , the isolated mode in @xmath47 probably disappears .
by contrast , the excitation continuum of @xmath48 lies in @xmath75 , yielding the stable isolated mode in @xmath48 even in @xmath13 .
@xmath45 is active on the excitation process that conserves the @xmath76 component of the ground state , while @xmath49 is active on the excitation process that changes the @xmath76 component of the ground state by @xmath77 . in @xmath78 , the ground state with @xmath79 and @xmath80 takes place instead of the singlet ground state in @xmath81 .
in fact , we have confirmed that @xmath82 and @xmath83 in @xmath84 for our four parameter sets . in @xmath78 , therefore , the excitations to the singlet state and to the triplet state with @xmath85 component make main contributions to the lower and higher isolated modes of @xmath47 , respectively .
we now discuss the observable dsf in the reduced zone scheme . in @xmath44 ,
the residues of @xmath45 and @xmath49 in @xmath86 show a typical pattern for the excitation continuum @xcite as shown in fig .
the excitation energies of the isolated modes at @xmath8 are smaller than that around @xmath87 even in transverse fields .
when @xmath88 , therefore , in the reduced zone scheme one isolated mode appears in @xmath89 and two isolated modes appear in @xmath90 . in @xmath13 ,
one isolated mode emerges in @xmath89 and @xmath90 , respectively . in ntenp and nmaop ,
the excitation energies take almost the same values at the symmetric wave numbers about @xmath26 even in transverse fields .
the same feature was already known in the dimer phase for @xmath91 and @xmath5 @xcite . as shown in fig .
[ fig2 ] , the residues of @xmath45 and @xmath49 in @xmath86 show typical behavior for the isolated mode @xcite
. therefore , in the reduced zone scheme the lowest excited states of @xmath89 and @xmath90 form the isolated modes in @xmath5 .
as @xmath4 increases , in ntenp and nmaop the intensity of the highest branch in @xmath89 decreases and disappears at @xmath63 and @xmath92 , respectively . in @xmath13 , only one isolated mode of @xmath90 emerges in ntenp , while the isolated modes of respective @xmath89 and @xmath90 emerge in nmaop . .
full triangles denote @xmath47 and gray circles denote @xmath48 . on the solid and gray lines , appreciable scattering intensity appears .
the broken lines for @xmath12 are guides to eyes .
, title="fig : " ] .
full triangles denote @xmath47 and gray circles denote @xmath48 . on the solid and gray lines , appreciable scattering intensity appears .
the broken lines for @xmath12 are guides to eyes .
, title="fig : " ] .
full triangles denote @xmath47 and gray circles denote @xmath48 . on the solid and gray lines , appreciable scattering intensity appears .
the broken lines for @xmath12 are guides to eyes .
, title="fig : " ] .
full triangles denote @xmath47 and gray circles denote @xmath48 . on the solid and gray lines
, appreciable scattering intensity appears .
the broken lines for @xmath12 are guides to eyes .
, title="fig : " ] in inelastic neutron - scattering experiments on ntenp and ndmap @xcite , the dsf s were observed as a superposition of @xmath45 and @xmath49 . using our numerical results
, we summarize the corresponding results in fig .
[ fig6 ] . on the solid and gray lines ,
the isolated modes possess appreciable intensity . in ntenp
, the highest branch disappears at @xmath93 , which is larger than the value @xmath94 estimated experimentally @xcite . above @xmath12 , only one gapped branch emerges in ntenp , whereas three gapped branches appear in ndmap .
the field dependence of the dsf s shown in fig .
[ fig6 ] reproduce qualitatively the experimental results for ntenp and ndmap @xcite . in ntenp @xcite and ndmap @xcite , the phase transitions to the three - dimensional ordered states
were observed by specific heat measurements in transverse fields @xmath95 at low temperatures .
this phase transition is caused by the interchain interaction @xmath96 . in ntenp and ndmap ,
their values are estimated as @xmath97 @xcite , and @xmath98 and @xmath99 depending on the direction @xcite , respectively . judging from our numerical results , the interchain interactions in ntenp and ndmap are so weak that they do not have serious effects on dynamical properties in @xmath13 .
the characteristic dynamical properties are attributed to the different field dependence of the excitation continuum in @xmath47 .
as the system approaches the gapless line in both phases , the excitation continuum of @xmath47 easily shifts to the low energy region in weak fields . in the haldane - gap phase this feature makes the highest isolated mode of @xmath44 unstable and invisible in @xmath78 , whereas in ndmap three isolated modes appear even in @xmath78 .
when @xmath78 , in the dimer phase , only one gapped mode of @xmath48 appears in ntenp , while the lower isolated mode of @xmath47 emerges close to @xmath100 in addition to the isolated @xmath48 mode . as shown so far , the excitation continuum of @xmath47 in the dimer phase shifts to the lower energy region close to @xmath70 as compared with that in the haldane - gap phase
. such different field dependence of the excitation continuum may be intrinsic in the dimer phase and the haldane - gap phase .
we have investigated the dsf of the @xmath0 bond - alternating heisenberg chain with a single - ion anisotropy in transverse magnetic fields , using a continued fraction method based on the lanczos algorithm .
we have shown that the excitation continuum in @xmath45 causes characteristic field dependence of the dsf in the haldane - gap phase and the dimer phase .
our results well reproduce the different field dependence of experimental findings for ntenp and ndmap .
we would like to thank professor m. hagiwara , dr .
l. p. regnault , and dr .
zheludev for useful comments and valuable discussions .
part of our computational programs are based on titpack version 2 by h. nishimori .
numerical computations were carried out at the yukawa institute computer facility , kyoto university , and the supercomputer center , the institute for solid state physics , university of tokyo .
this work was supported by a grant - in - aid for scientific research from the ministry of education , culture , sports , science , and technology , japan .
_ note added._-during the completion of this manuscript , we were informed that in ntenp at @xmath5 the ratio of the lowest - lying energy in @xmath47 to that in @xmath48 is described by the sine - gordon quantum field thory @xcite .
the result agrees well with our numerical result .
i. affleck and f. d. m. haldane , phys .
b * 36 * , 5291 ( 1987 ) .
m. hagiwara , y. narumi , k. kindo , m. kohno , h. nakano , r. sato , and m. takahashi , phys .
lett . * 80 * , 1312 ( 1998 ) .
m. kohno , m. takahashi , and m. hagiwara , phys .
b * 57 * , 1046 ( 1998 ) .
h. nakano , m. hagiwara , t. chihara , and m. takahashi , j. phys .
66 * , 2997 ( 1997 ) .
y. narumi , m. hagiwara , m. kohno , and k. kindo , phys .
* 86 * , 324 ( 2001 ) .
n. tateiwa , m. hagiwara , h. aruga - katori , and t. c. kobayashi , physica b * 329 - 333 * , 1209 ( 2003 ) .
y. narumi , k. kindo , m. hagiwara , h. nakano , a. kawaguchi , k. okunishi , and m. kohno , phys .
b * 69 * , 174405 ( 2004 ) . s. yamamoto , phys .
b * 51 * , 16128 ( 1995 ) .
k.totsuka , y. nishiyama , n. hatano , and , m. suzuki , j. phys . : condens .
matter * 7 * , 4895 ( 1995 ) .
t. suzuki and s. suga , phys .
b * 68 * , 134429 ( 2003 ) . l. p. regnault , i. zaliznyak , j. p. renard , and c. vettier , phys . rev .
b * 50 * , 9174 ( 1994 ) .
h. tasaki , j. phys . : condens .
matter * 3 * , 5875 ( 1991 ) .
m. hagiwara , l. p. regnault , a. zheludev , a. stunault , n. metoki , t. suzuki , s. suga , k. kakurai , y. koike , p. vorderwisch , and j. h. chung , to appear in phys .
. lett . * 94 * , 177202 ( 2005 ) ; cond - mat/0501207 .
a. zheludev , z. honda , c. l. broholm , k. katsumata , s. m. shapiro , a. kolezhuk , s. park and y. qiu , phys .
b * 68 * , 134438 ( 2003 ) .
e. r. gagliano and c. a. balseiro , phys .
lett . * 59 * , 2999 ( 1987 ) .
m. takahashi , phys .
b * 50 * , 3045 ( 1994 ) .
s. ma , c. broholm , d. h. reich , b. j. sternlieb , and r. w. erwin , phys .
lett . * 69 * , 3571 ( 1992 ) .
a. zheludev , y. chen , c. l. broholm , z. honda , and k. katsumata , phys .
b * 63 * , 104410 ( 2001 ) . y. narumi , m. hagiwara , r. sato , k. kindo , h. nakano , and m. takahashi , physica b * 246 - 247 * , 509 ( 1998 ) . t. tonegawa , t. nakano , and m. kaburagi , j. phys .
* 65 * , 3317 ( 1996 ) ; w. chen , k. hida and c. sanctuary , j. phys .
. jpn . * 69 * , 237 ( 2000 ) ; a. koga , phys .
a * 296 * , 243 ( 2002 ) . in ref .
15 , the phase diagram is represented on the basis of the model hamiltonian ; @xmath101\mathbf{s}_{i } \cdot \mathbf{s}_{i+1 } + \tilde{d } \sum_{i } ( s^{z}_{i})^{2}. $ ] to see the position on this phase diagram , therefore , the parameters have to be described by @xmath102 .
the parameter sets @xmath38 , and @xmath39 used in this paper are converted into @xmath103 , and @xmath104 , respectively .
the first two parameter sets are in the haldane - gap phase and the latter two ones are in the dimer phase .
j. l. cardy , nucl .
b * 270 * , 186 ( 1986 ) ; h. w. j. blte , j. l. cardy , and m. p. nightingale , phys .
lett . * 56 * , 742 ( 1986 ) ; i. affleck , phys . rev .
56 * , 746 ( 1986 ) .
z. honda , h. asakawa , and k. katsumata , phys .
lett . * 81 * , 2566 ( 1998 ) .
a. zheludev , t. masuda , b. sales , d. mandrus , t. papenbrock , t. barnes , and s. park , phys .
b * 69 * , 144417 ( 2004 ) . j. tamaki and m. oshikawa , the physical society of japan 2005 spring meeting . | we calculate dynamical structure factors of the @xmath0 bond - alternating heisenberg chain with a single - ion anisotropy in transverse magnetic fields , using a continued fraction method based on the lanczos algorithm . in the haldane - gap phase and the dimer phase ,
dynamical structure factors show characteristic field dependence .
possible interpretations are discussed in viewpoint of the field dependence of the excitation continuum .
the numerical results are in qualitative agreement with recent results for inelastic neutron - scattering experiments on the @xmath0 bond - alternating heisenberg - chain compound @xmath1 and the @xmath0 haldane - gap compound @xmath2 in transverse magnetic fields . |
since its introduction in 1993,@xcite the density matrix renormalisation group method ( dmrg ) has seen tremendous use in the study of one - dimensional systems.@xcite various improvements such as real - space parallelisation,@xcite the use of abelian and non - abelian symmetries@xcite and multi - grid methods@xcite have been proposed .
most markedly , the introduction@xcite of density matrix perturbation steps allowed the switch from two - site dmrg to single - site dmrg in 2005 , which provided a major speed - up and improved convergence in particular for systems with long - range interactions .
nevertheless , despite some progress,@xcite ( nearly ) two - dimensional systems , such as long cylinders , are still a hard problem for dmrg .
the main reason for this is the different scaling of entanglement due to the area law:@xcite in one dimension , entanglement and hence matrix dimensions in dmrg are essentially size - independent for ground states of gapped systems , whereas in two dimensions , entanglement grows linearly and matrix dimensions roughly exponentially with system width . as a result ,
the part of the hilbert space considered by dmrg during its ground state search increases dramatically , resulting mainly in three problems : firstly , the dmrg algorithm becomes numerically more challenging as the sizes of matrices involved grow ( we will assume matrix - matrix multiplications to scale as @xmath4 throughout the paper ) .
secondly , the increased search space size makes it more likely to get stuck in local minima .
thirdly , while sequential updates work well in 1-d chains with short - range interactions , nearest - neighbour sites in the 2-d lattice can be separated much farther in the dmrg chain .
therefore , improvements to the core dmrg algorithm are still highly worthwhile . in this paper
, we will adopt parts of the amen method@xcite developed in the tensor train / numerical linear algebra community to construct a strictly single - site dmrg algorithm that works without accessing the ( full ) reduced density matrix .
compared to the existing _ centermatrix wavefunction formalism _
( cwf),@xcite we achieve a speed - up of @xmath5 during each application of @xmath0 to @xmath1 in the eigensolver during the central optimisation routine , where @xmath3 is the dimension of the physical state space on each site .
the layout of this paper is as follows : section [ sec : notation ] will establish the notation .
section [ sec : cwf ] will recapitulate the density matrix perturbation method and the cwf .
section [ sec : subexp ] will introduce the subspace expansion method and the heuristic expansion term with a simple two - spin example .
the trictly ingle - ite dmrg algorithm ( dmrg3s ) will be presented in section [ sec : dmrg3s ] alongside a comparison with the existing cwf . as both the original perturbation method and the heuristic subspace expansion require a _ mixing factor _
@xmath6,@xcite section [ sec : alpha ] describes how to adaptively choose @xmath6 for fastest convergence .
numerical comparisons and examples will be given in section [ sec : numexps ] .
the notation established here closely follows the review article ref .. consider a state @xmath1 of a system of @xmath7 sites .
each site has a physical state dimension @xmath8 , e.g. @xmath9 , @xmath10 for a system of @xmath11 @xmath12 spins : @xmath13 in practice , the dimension of the physical basis is usually constant , @xmath14 , but we will keep the subscript to refer to one specific basis on site @xmath15 where necessary .
it is then possible to decompose the coefficients @xmath16 as a series of rank-3 tensors @xmath17 of size @xmath18 respectively , with @xmath19 .
the coefficient @xmath20 can then be written as the matrix product of the corresponding matrices in @xmath21 : @xmath22 the maximal dimension @xmath23 is called the _ mps bond dimension_. in typical one - dimensional calculations , @xmath24 , but for e.g. @xmath25 cylinders , @xmath26 is often necessary .
it is in these numerically demanding cases that our improvements are of particular relevance .
similarly , a hamiltonian operator can be written as a _ matrix product operator _ ( mpo ) , where each tensor @xmath27 is now of rank 4 , namely @xmath28 : @xmath29 @xmath30 is called the _ mpo bond dimension_. we will usually assume that for most @xmath15 , @xmath31 and @xmath32 . in practice , this holds nearly everywhere except at the ends of the chain , where the @xmath33 grow exponentially from @xmath34 to @xmath35 .
the basis of @xmath36 ( @xmath27 ) of dimension @xmath37 ( @xmath38 ) is called the left - hand side ( lhs ) basis , whereas the basis of dimension @xmath33 ( @xmath39 ) is the right - hand side ( rhs ) basis of this tensor . for simplicity , @xmath33 , @xmath8 and @xmath39 can also refer to the specific basis ( and not only its dimension ) when unambiguous . instead of @xmath36 , we will also write @xmath40 ( @xmath41 for a left ( right ) normalised mps tensor : @xmath42
if we then define the contractions @xmath43 we can rewrite @xmath1 from as @xmath44 that is , when only considering one specific bond @xmath45 , the left and right mps bases at this bond are built up from the states generated by the mps tensor chains to the left and right of the bond .
individual elements of an mps basis are therefore called `` state '' .
furthermore , define @xmath46 and @xmath47 with summation over all possible indices .
similarly , @xmath48 and @xmath49 . with these contractions , it is possible to write @xmath50 for any @xmath51 $ ] .
dmrg then works by _ sweeping _ over the system multiple times . during each sweep ,
each site tensor @xmath36 is sequentially _ updated _ once with each update consisting of one optimisation step via e.g. a sparse eigensolver and possibly one _ enrichment _ step during which the left or right mps basis of @xmath36 is changed in some way . depending on the exact implementation , updates may work on one ( single - site dmrg ) or two sites ( two - site dmrg ) at a time .
the enrichment step may be missing or implemented via density matrix perturbation or subspace expansion .
during single - site dmrg , only a single mps tensor @xmath36 on site @xmath15 is optimised at once .
compared to two - site dmrg , the search space is reduced by a factor of @xmath52 , leading to a speed - up of at least @xmath53 per iteration.@xcite however , since the left and right bases of the tensors @xmath36 are fixed and defined by the environment ( @xmath54 and @xmath55 ) , this approach is likely to get stuck . while also occurring
if there are no symmetries implemented on the level of the mps , this issue is most easily visible if one considers @xmath56 symmetries:@xcite assume that all basis states to the right of the rhs bond of @xmath36 transform as some quantum number @xmath57 .
if we now target a specific sector , e.g. @xmath58 overall , then on the lhs of this bond ( i.e. from the left edge up to and including @xmath36 ) , all states must transform as @xmath59 . in this configuration
, it is impossible for a local change of @xmath36 to add a new state that transforms as , say , @xmath60 , to its right basis states , as there would be no corresponding state @xmath61 to the right of that bond , rendering the addition of the state moot from the perspective of the local optimiser , as its norm will be zero identically .
a concrete example of this issue is given in section [ sec : numexps : stuckspins ] .
dmrg is a variational approach on the state space available to mps of a given bond dimension . as such
, the algorithm must converge into either the global or a local minimum of the energy in this state space .
hence , we will call all cases where dmrg converges on an energy substantially higher than the minimal energy achievable with the allowed mps bond dimension cases where dmrg is stuck in _ local minima_.
this convergence problem has been solved by white ( 2005).@xcite in the following , we will assume a left - to - right sweep , sweeping in the other direction works similarly , but on the left rather than right bonds .
after the local optimisation of the tensor @xmath36 , the reduced density matrix @xmath62 is built on the next bond .
this is the reduced density matrix resulting from tracing out the part of the system to the left of bond @xmath45 .
@xmath63 is then perturbed as @xmath64 the new @xmath65 is then used to decide on a new set of basis states on the rhs of @xmath36 , with the inverse mapping from the new to the old basis being multiplied into each component of @xmath66 .
the mixing factor @xmath6 is a small scalar used to control the perturbation .
a new scheme to find the optimal choice of @xmath6 is discussed in section [ sec : alpha ] . in a standard single - site dmrg calculation ,
the reduced density matrix @xmath63 is never used .
more importantly , even building @xmath63 on a given bond @xmath45 will not yield a density matrix that can be used in , as it only contains the @xmath33 states existing on that bond already without knowledge of the @xmath37 states on the bond one step to the left .
in other words , it is not possible to choose the optimal set @xmath67 based only on @xmath33 , rather , one requires also @xmath8 and @xmath37 .
the centermatrix wavefunction formalism@xcite was developed to cope with this problem .
given a site tensor @xmath68 on a left - to - right sweep , it introduces a `` centermatrix '' @xmath69 and replaces the original site tensor as @xmath70 @xmath40 is constructed to be left - orthogonal and is essentially an identity matrix mapping the left basis @xmath37 and the physical basis @xmath8 onto a complete basis containing all @xmath71 states on its right .
the new basis is `` complete '' in the sense that all states reachable from the left bond basis @xmath37 and the local physical basis @xmath8 are contained within it . the contents of @xmath36 are placed in @xmath72 accordingly and the original state remains unchanged .
the reduced density matrix is then @xmath73 and has access to all @xmath71 states , as required above .
a perturbation of @xmath63 according to hence allows the introduction of new states .
the dmrg optimisation step can work on @xmath72 alone , with @xmath74 built prior to optimisation of @xmath72 from the expanded @xmath40 . during each eigensolver step ,
the effective hamiltonian on site @xmath15 has to be applied onto @xmath72 .
the application is done by contraction of @xmath75 , @xmath76 and @xmath69 at cost @xmath77 per step .
after optimisation , the perturbation is added .
its computational cost is dominated by the calculation of @xmath78 at @xmath79 .
the bond between @xmath40 and @xmath72 can then be truncated down to @xmath35 using @xmath80 and the remaining parts of @xmath72 are multiplied into @xmath66 to the right .
the resulting algorithm converges quickly for one - dimensional problems and performs reasonably well for small cylinders .
however , both the cost of the applications of @xmath0 to @xmath1 as @xmath81 as well as the large density matrix @xmath82 cause problems if @xmath35 and @xmath83 become large .
the idea of using _ subspace expansion _ instead of density matrix perturbation originates @xcite in the tensor train / numerical linear algebra community .
there , a stringent proof was given regarding the convergence properties of this method when the local tensor @xmath84 of the residual @xmath85 is used as the expansion term . here
, we will only use the method of subspace expansion and substitute a numerically much more cheaply available expansion term .
the following section is divided into three parts : firstly , we will explain the concept of subspace expansion acting on two neighbouring mps tensors @xmath36 , @xmath86 .
secondly , the expansion term employed in dmrg3s is introduced and motivated .
thirdly , a simple example is described . in the following
, we will describe subspace expansion of the rhs basis of the current working tensor , as it would occur during a left - to - right sweep .
assume a state @xmath1 described by a set of tensors @xmath87 .
at the bond @xmath45 , we can then decompose the state as a sum over left and right basis states as in eq . .
now we _ expand _ the tensor @xmath88 by some expansion term @xmath89 for each individual physical index component : @xmath90 \quad.\ ] ] this effectively expands the rhs mps basis of @xmath36 from @xmath33 to @xmath91 .
similarly , expand the components of @xmath92 with zeros : @xmath93 \quad.\ ] ] the appropriately - sized block of zeros only multiplies with the expansion term @xmath94 . in terms of a decomposition as in
, this is equivalent to @xmath95 \left [ \begin{matrix } r_{i+1 } \\ 0 \end{matrix } \right ] |\sigma_1 \ldots \sigma_i\rangle \otimes |\sigma_{i+1},\ldots,\sigma_l\rangle \label{eq : subexp - post}\ ] ] where @xmath96 is the result of multiplying @xmath54 and @xmath97 , with the @xmath98 in the second expression similarly resulting from the @xmath98 in @xmath66 .
while the state @xmath1 remains unchanged , the local optimiser on the new site @xmath66 can now choose the initially - zero components differently if so required : the necessary flexibility in the left-/right basis states to escape local minima has been achieved without referring to the density matrix .
note that while orthonormality of @xmath66 is lost , we do not need it between the enrichment step on site @xmath15 and the optimisation step on site @xmath99 .
the orthonormality of @xmath36 can be restored via singular value decomposition as usual .
furthermore , it is usually necessary to truncate the rhs basis of @xmath100 down from @xmath101 to @xmath35 immediately following the expansion : this preserves the most relevant states of the expansion term while avoiding an exponential explosion of bond dimensions .
when sweeping from right to left , the left rather than right mps basis of the current working tensor is expanded , with the left tensor @xmath102 being zero - padded as opposed to the right tensor @xmath66 : @xmath103 \\
a_{i-1}^{\sigma_{i-1 } } & \to \tilde a_{i-1}^{\sigma_{i-1 } } = \left [ \begin{matrix } a_{i-1}^{\sigma_{i-1 } } & 0 \end{matrix}\right ] \quad.\end{aligned}\ ] ] using the exact residual as the expansion term is computationally expensive : the term @xmath104 can be updated locally and is mostly unproblematic , but the subtraction of @xmath105 and subsequent re - orthonormalisation is costly and has to be done after each local optimisation , as the current value of @xmath106 changes .
this exact calculation is hence only possible for @xmath107 , which is far too small to tackle difficult two - dimensional problems .
instead , we propose the very cheaply available terms @xmath108 to be used during left - to - right sweeps and @xmath109 for use during right - to - left sweeps with some scalar mixing factor @xmath6 . in the regime
where the exact residual can be computed , these terms work essentially equally well . this expression for @xmath97 can be heuristically motivated as follows : is equivalent to the partial projection of latexmath:[$h current bond . hence , in the ground state and ignoring numerical errors , the rhs basis of this @xmath97 is identical to that of @xmath36 .
truncation from @xmath111 to @xmath33 is then possible without inducing errors .
numerically , it seems possible to choose @xmath6 arbitrarily large without hindering convergence or perturbing the state too much in simple ( one - dimensional ) problems .
however , if the chosen maximal bond dimension @xmath35 is insufficient to faithfully capture the ground state of the given system , @xmath6 has to be taken to zero eventually to allow convergence .
otherwise , @xmath97 will continuously add new states and disturb the result of the eigensolver , which is optimal at this specific value of @xmath35 but not an eigenstate of @xmath0 yet .
the cost of a single subspace expansion is @xmath112 for the calculation of @xmath97 , potentially @xmath113 for the addition to @xmath36 and @xmath66 respectively and @xmath114 for the svd of an @xmath115 matrix formed from @xmath100 .
if we restrict the svd to @xmath35 singular values , then the resulting matrices will be of dimension @xmath116 , @xmath117 and @xmath118 respectively .
the first can be reformed into @xmath119 at cost @xmath120 and the second and third multiplied into @xmath66 at cost @xmath121 .
the total cost of this step is dominated by the cost of the svd at @xmath122 , which is still cheaper than the calculation of the perturbation term in , not considering the other costs associated to using the density matrix for truncation . in the following , we will demonstrate and illustrate the method of subspace expansion at the simple example of a system of two spins with @xmath124 from @xmath125 to @xmath126 as it would occur during a left - to - right sweep .
assume the hamiltonian @xmath127 with mpo - components @xmath128 \\
w_2 & = \left [ \frac{1}{\sqrt{2 } } s_- \quad \frac{1}{\sqrt{2 } } s_+ \quad s_z \right]^t \quad.\end{aligned}\ ] ] let the initial state be an @xmath129 mps , described by components @xmath130 \quad & a_1^\downarrow = \left [ \sqrt{1-a^2 } \right ] \\
b_2^\uparrow = [ b ] \quad & b_2^\downarrow = \left [ \sqrt{1-b^2 } \right]\end{aligned}\ ] ] where square brackets denote matrices in the mps bond indices . due to the standard normalisation constraints
, there are only two free scalar variables here , @xmath131 and @xmath132 .
subspace expansion of @xmath133 is straightforward ( keep in mind that @xmath134 for convenience ) : @xmath135 \\
p_1^\downarrow & = w_1^{\downarrow\uparrow } a_1^\uparrow + w_1^{\downarrow\downarrow } a_1^\downarrow \\ & = \left [ \begin{matrix } 0 & \frac{a}{\sqrt{2 } } & - \sqrt{1-a^2 } \end{matrix } \right]\end{aligned}\ ] ] resulting in @xmath136 and @xmath137 directly after the expansion : @xmath138 \\
a_1^{\prime\downarrow } & = \left [ \begin{matrix } \sqrt{1-a^2 } & 0 & \frac{a}{\sqrt{2 } } & - \sqrt{1-a^2 } \end{matrix } \right ] \\
b_2^{\prime\uparrow } & = \left [ \begin{matrix } b \\ 0 \\ 0 \\ 0 \end{matrix } \right ] \quad b_2^{\prime\downarrow } = \left [ \begin{matrix } \sqrt{1-b^2 } \\ 0 \\ 0 \\ 0 \end{matrix } \right ] \quad.\end{aligned}\ ] ] normalising @xmath136 via a singular value decomposition as @xmath139 and multiplying @xmath140 gives : @xmath141 \\
a_1^{\prime\prime\downarrow } & = \left [ \begin{matrix } 0 & 1 \end{matrix } \right ] \\
sv^\dagger & = \left [ \begin{matrix } a & \frac{\sqrt{1-a^2}}{\sqrt{2 } } & 0 & a \\
\sqrt{1-a^2 } & 0 & \frac{a}{\sqrt{2 } } & - \sqrt{1-a^2 } \end{matrix}\right ] \\
b_2^{\prime\prime\uparrow } & = \left [ \begin{matrix } a b \\ \sqrt{1-a^2 } b \end{matrix } \right ] \\
b_2^{\prime\prime\downarrow } & = \left [ \begin{matrix } a \sqrt{1-b^2 } \\
\sqrt{1-a^2 } \sqrt{1-b^2 } \end{matrix}\right]\quad.\end{aligned}\ ] ] as expected , the final state @xmath142 is still entirely unchanged , but there is now a one - to - one correspondence between the four entries of @xmath143 and the coefficients @xmath144 in the computational basis , making the optimisation towards @xmath145 trivial .
we can now combine standard single - site dmrg ( e.g.ref .
, p. 67 ) with the subspace expansion method as a way to enrich the local state space , leading to a trictly ingle - ite dmrg implementation ( dmrg3s ) that works without referring to the density matrix at any point . with the notation from section [ sec : notation ] ,
the steps follow mostly standard single - site dmrg . in an outermost loop ,
the algorithm sweeps over the system from left - to - right and right - to - left until convergence is reached .
criteria for convergence are e.g. diminishing changes in energy or an overlap close to @xmath34 between the states at the ends of subsequent sweeps . the inner loop sweeps over the system , iterating over and updating the tensors on each site sequentially .
each local update during a left - to - right sweep ( right - to - left sweeps work analogously ) consists of the following steps : 1 .
optimise the tensor @xmath36 : use an eigensolver targeting the smallest eigenvalue to find a solution @xmath146 to the eigenvalue problem @xmath147 @xmath148 is the new current energy estimate .
this first step dominates the computational cost .
2 . build @xmath149 according to using @xmath150 .
build an appropriately - sized zero block @xmath151 after the dimensions of @xmath97 are known .
3 . subspace - expand @xmath152 with @xmath153 and @xmath66 with @xmath151 .
4 . apply a svd to @xmath154 and truncate its right basis to @xmath33 again , resulting in @xmath155 .
5 . multiply the remainder of the svd ( @xmath156 ) into @xmath157 .
build @xmath74 from @xmath155 , @xmath158 and @xmath27 .
calculate a new energy value after truncation based on @xmath159 , @xmath160 , @xmath161 and @xmath162 . use this energy value and @xmath148 to adapt the current value of @xmath6 ( cf
. section [ sec : alpha ] ) .
continue on site @xmath99 . of these ,
step 2 and 3 implement the actual subspace expansion , whereas all others are identical to standard single - site dmrg .
it is important to note that the only change from standard single - site dmrg is the addition of an enrichment step via subspace expansion .
therefore , this method does not interfere with e.g. real - space parallelised dmrg,@xcite the use of nonabelian symmetries@xcite or multi - grid methods.@xcite to analyse the computational cost , we have to take special care to ensure optimal ordering of the multiplications during each eigensolver iteration in .
the problem is to contract @xmath163 , with @xmath158 and @xmath164 , @xmath165 and @xmath166 .
the optimal ordering is then @xmath167 : 1 .
contract @xmath158 and @xmath168 over the left mps bond at cost @xmath169 .
multiply in @xmath27 over the physical bond of @xmath36 and the left mpo bond at cost @xmath170 .
3 . finally contract with @xmath162 over the right mpo and mps bonds at cost @xmath171 .
the total cost of this procedure to apply @xmath0 to @xmath1 is @xmath172 . assuming large @xmath173 is small , this gives a speed - up in the eigensolver multiplications of @xmath174 over the cwf approach , which takes @xmath175 .
in addition to this speed - up , the subspace expansion is considerably cheaper than the density matrix perturbation .
since the perturbation / truncation step can often take up to 30% of total computational time , improvements there also have a high impact . at the same time
, the number of sweeps at large @xmath35 needed to converge does not seem to increase compared to the cwf approach ( cf .
section [ sec : numexps ] ) and sometimes even decreases .
both density matrix perturbation and subspace expansion generally require some small mixing factor @xmath6 to moderate the contributions of the perturbation terms .
the optimal choice of this @xmath6 depends on the number of states available and those required to represent the ground state , as well as the current speed of convergence .
too large values for @xmath6 hinder convergence by destroying the improvements made by the local optimiser , whereas too small values lead to the calculation being stuck in local minima with vital states not added for the reasons given in section [ sec : cwf : rhopert ] .
the correct choice of @xmath6 hence affects calculations to a large degree , but is also difficult to estimate before the start of the calculation .
( colour online ) energies of the state at different points during a single update : before optimisation , the state has some initial energy @xmath176 .
local optimisation via the eigensolver takes this energy down by @xmath177 to @xmath178 .
subsequent truncation causes a rise in energy by @xmath179 with the final value at the end of this update being @xmath180 . ]
[ fig : energy - levels - alpha ] displays the individual steps within a single update from the energy perspective : let @xmath181 denote the gain in energy during the optimisation step and let @xmath182 denote the subsequent rise in energy during the truncation following the enrichment step .
@xmath183 only occurs if some enrichment ( either via density matrix perturbation or subspace expansion ) has occurred , otherwise there would be no need for any sort of truncation .
we can hence control the approximate value of @xmath182 via @xmath6 , which leads to a simple adaptive and computationally cheap algorithm : if @xmath179 was very small or even negative ( after changing the optimised state by expansion of its right basis ) during the current update , we can increase @xmath6 during the next update step on the next site .
if , on the other hand , @xmath184 , that is , if the error incurred during truncation nullified the gain in energy during the optimisation step , we should reduce the value of @xmath6 at the next iteration to avoid making this mistake again . in practice , it seems that keeping @xmath185 gives the fastest convergence .
given the order - of - magnitude nature of @xmath6 , it is furthermore best to increase / decrease it via multiplication with some factor greater / smaller than @xmath34 as opposed to adding or subtracting fixed values .
some special cases for very small @xmath181 ( stuck in a local minimum or converged to the ground state ? ) and @xmath186 or @xmath187 have to be considered , mostly depending on the exact implementation .
it is unclear whether there is a causal relation between the optimal choice of @xmath6 and the ratio of @xmath188 or whether both simply correlate with a proceeding dmrg calculation : at the beginning , gains in energy are large and @xmath6 is optimally chosen large , whereas later on , energy decreases more slowly and smaller values of @xmath6 are more appropriate .
it is important to note that this is a tool to reach convergence more quickly .
if one is primarily interested in a wavefunction representing the ground state , the calculation of a new @xmath6 at each iteration comes at essentially zero cost .
if , however , the aim is to extrapolate in the truncation error during the calculation , then a fixed value for @xmath6 is of course absolutely necessary .
in this sub - section , we will give a short example of how dmrg can get stuck in a local minimum even on a very small system .
consider @xmath189 @xmath190 spins with isotropic antiferromagnetic interactions and open boundary conditions .
the @xmath56 symmetry of the system is exploited on the mps basis , with the overall @xmath191 forced to be zero .
the initial state is constructed from 20 linearly independent states , all with @xmath192 sites on the very right at @xmath193 and @xmath194 in total .
the quantum number distribution at each bond is plotted in fig .
[ fig : qnums - dist ] as black circles .
dmrg3s is run with subspace expansion disabled , i.e. @xmath195 throughout the calculation .
the algorithm `` converges '' to some high - energy state at @xmath196 . the resulting quantum number distribution ( red squares in fig . [
fig : qnums - dist ] ) shows clear asymmetry both between the left and right parts of the system and the @xmath197 and @xmath198 sectors at any given bond .
it is also visible that while some states are removed by dmrg3s without enrichment , it can not add new states : the red squares only occur together with the black filled circles from the input state .
if we enable enrichment via subspace expansion , i.e. take @xmath199 , dmrg3s quickly converges to a much better ground state at @xmath200 .
the quantum numbers are now evenly distributed between the left- and right parts of the system and @xmath201 symmetry is also restored .
( colour online ) the quantum number distribution as counted from the right at each bond of a @xmath202 system with @xmath190 and @xmath203 .
the artificial input state is shown with black circles .
two dmrg calculations have then been done on this input state , once with no enrichment term ( @xmath195 , red squares ) and once with subspace expansion enabled ( @xmath204 , blue diamonds ) .
it is clearly visible that without enrichment , dmrg3s can reduce some weights to zero , but can not add new states red only occurs together with black .
as soon as enrichment is enabled , dmrg3s restores @xmath205 symmetry and reflective symmetry over the 10@xmath206 bond and finds a much better ground state . ] in the following subsections , we will compare the two single - site dmrg algorithms cwf and dmrg3s when applied to four different physical systems : a @xmath12 heisenberg spin chain with periodic boundary conditions , a bosonic system with an optical lattice potential , a fermi - hubbard model at @xmath207 and quarter - filling and a system of free fermions at half - filling .
each algorithm is run at three different values of @xmath208 from the same initial state and run to convergence .
this way , it is possible to both observe the behaviour of the methods at low and high accuracies . the usual setup in dmrg calculations of starting at small @xmath35 and increasing @xmath35 slowly while the calculation progresses makes it unfortunately very difficult to compare between the three methods .
this is because different methods require different configurations to converge optimally .
we therefore restrict ourselves to fixed @xmath35 throughout an entire calculation , even though all methods could be sped up further by increasing @xmath35 slowly during the calculation .
errors in energy compared to a numerically exact reference value @xmath209 are plotted as a function of sweeps and cpu time .
it should be stressed that this error in energy is not directly comparable to the truncation error traditionally used in two - site dmrg or the variance @xmath210 sometimes considered in single - site dmrg .
even small differences in energy can lead to vastly different physical states and reaching maximal accuracy in energy is crucial to ensure that the true ground state has been reached .
furthermore , a traditional two - site dmrg ( 2dmrg ) calculation without perturbations is done and its error in energy and runtime to convergence is compared to the two single - site algorithms . here , _
convergence _ is defined as a normalised change in energy less than @xmath211 ( for @xmath212 ) resp .
@xmath213 ( for @xmath214 ) .
the _ runtime to convergence _ is the cpu time used until that energy was output by the eigensolver for the first time .
all calculations were performed on a single core of a xeon e5 - 2650 .
firstly , we consider a @xmath12 heisenberg spin chain with @xmath215 sites and periodic boundary conditions implemented on the level of the hamiltonian as a simple link between the first and last site : @xmath216 @xmath56 symmetries are exploited and the calculations are forced in the @xmath58 sector .
( colour online ) spin chain eq . : normalised error in energy as a function of sweeps ( left ) and cpu time used ( right ) of the two single - site algorithms at different @xmath217 .
dmrg3s shows both a speed - up and an improved convergence per sweep compared to cwf , with a long tail of slow convergence very visible for cwf at high accuracies . ]
.[tab : numexps : spinchain]spin chain eq . :
normalised error in energy at convergence and runtime to convergence of all three methods .
dmrg3s is consistently faster than cwf , whereas the energies provided by 2dmrg are not comparable in accuracy .
[ cols=">,^,^,^",options="header " , ] finally , we consider a model of free fermions on a chain of 100 sites with hamiltonian @xmath218\;. \label{eq : numexps : fermions}\ ] ] the maximally delocalised wavefunction found in the ground - state of this system is notoriously difficult for mps formats in general to reproduce faithfully . at the same time
, most other parameters are identical ( @xmath3 , @xmath7 , @xmath35 ) or very close ( @xmath83 ) to those in the fermi - hubbard model from section [ sec : numexps : fermi - hubbard ] .
the calculation is done using @xmath219 and @xmath220 symmetries at half - filling with @xmath221 fermions and @xmath203 .
the choice of @xmath35 is the same as for the fermi - hubbard system , namely @xmath222 .
we used @xmath223 as the reference value , since all methods converged to this ground - state energy at @xmath224 .
the results in tab .
[ tab : numexps : fermions ] and fig .
[ fig : numexps : fermions ] mostly follow the previous results for locally interacting systems : accuracies of all methods are essentially identical , whereas time to convergence varies between the methods . at small @xmath35 , there are some speed - ups of dmrg3s over cwf , largely due to better convergence behaviour per sweep , whereas a significant advantage of dmrg3s becomes visible at larger @xmath35 , when numerical operations become cheaper compared to the cwf method .
correspondingly , the speed - up from cwf to dmrg3s increases from @xmath225 at @xmath226 to @xmath227 at @xmath224 .
similarly , the larger numerical cost of two - site dmrg becomes more noticeable at larger @xmath35 , with the speed - up between 2dmrg and dmrg3s increasing from @xmath228 at @xmath226 to more than @xmath229 at @xmath224 .
compared to the non - critical fermi - hubbard system from section [ sec : numexps : fermi - hubbard ] , we observe larger errors in energy at fixed @xmath35 , as expected .
correspondingly , as more eigenvalues contribute significantly , convergence of both the eigenvalue solver and the singular value decompositions becomes slower , leading to a slow - down of all three methods .
the new strictly single - site dmrg ( dmrg3s ) algorithm results in a theoretical speed - up of @xmath230 during the optimisation steps compared to the centermatrix wavefunction formalism ( cwf ) , provided that @xmath231 is small .
further , convergence rates per sweep are improved in the important and computationally most expensive high - accuracy / large-@xmath35 phase of the calculation .
in addition , auxiliary calculations ( enrichment , normalisation , etc . ) are sped up and memory requirements are relaxed .
numerical experiments confirm a speed - up within the theoretical expectations compared to the cwf method .
the efficiency of single - site dmrg in general compared to the traditional two - site dmrg was substantiated further by a large speed - up at comparable accuracies in energy .
we would like to thank s. dolgov , d. savostyanov and i. kuprov for very helpful discussions . c. hubig acknowledges funding through the exqm graduate school and the nanosystems initiate munich .
f. a. wolf acknowledges support by the research unit for 1807 of the dfg .
17ifxundefined [ 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 link:\doibase 10.1103/physrevlett.69.2863 [ * * , ( ) ] link:\doibase 10.1103/physrevb.48.10345 [ * * , ( ) ] link:\doibase 10.1103/revmodphys.77.259 [ * * , ( ) ] link:\doibase 10.1016/j.aop.2010.09.012 [ * * , ( ) ] link:\doibase 10.1103/physrevb.87.155137 [ * * , ( ) ] link:\doibase 10.1209/epl / i2002 - 00393 - 0 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.109.020604 [ * * , ( ) ] link:\doibase 10.1103/physrevb.72.180403 [ * * , ( ) ] link:\doibase 10.1126/science.1201080 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.109.067201 [ * * , ( ) ] link:\doibase 10.1146/annurev - conmatphys-020911 - 125018 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.90.227902 [ * * , ( ) ] link:\doibase 10.1103/revmodphys.82.277 [ * * , ( ) ] link:\doibase 10.1088/1742 - 5468/2007/10/p10014 [ * * , ( ) ] link:\doibase 10.1137/140953289 [ * * , ( ) ] @noop ( ) , http://nbn-resolving.de/urn:nbn:de:bvb:19-159631 [ `` , '' ] ( ) , | we introduce a strictly single - site dmrg algorithm based on the subspace expansion of the alternating minimal energy ( amen ) method .
the proposed new mps basis enrichment method is sufficient to avoid local minima during the optimisation , similarly to the density matrix perturbation method , but computationally cheaper .
each application of @xmath0 to @xmath1 in the central eigensolver is reduced in cost for a speed - up of @xmath2 , with @xmath3 the physical site dimension .
further speed - ups result from cheaper auxiliary calculations and an often greatly improved convergence behaviour .
runtime to convergence improves by up to a factor of 2.5 on the fermi - hubbard model compared to the previous single - site method and by up to a factor of 3.9 compared to two - site dmrg .
the method is compatible with real - space parallelisation and non - abelian symmetries . |
it has long been realized that a high - altitude observing platform located in the stratosphere and thus above a significant fraction of the earth s atmosphere could offer image quality competitive with space - based platforms .
this was the motivation behind the series of stratoscope i and ii balloon flights that ran in the late 1950s , 1960s , and early 1970s flying 0.3 to 0.9 m telescopes to an altitude of 24 km ( 80 kft ) and obtaining 0.2 arcseconds resolution images of the sun , planets , and selected stars and galaxies @xcite .
stratoscope images along with recent atmospheric turbulence studies @xcite have shown that near diffraction - limited image quality can be achieved at altitudes at or above 20 km ( 65 kft ) where the telescope is above @xmath0 95% or more of the atmosphere .
since the end of the stratoscope missions , few high - altitude balloon flights have carried optical and near - infrared astronomical telescopes and detectors .
nasa s highly successful multi - million cubic foot , high - altitude balloons flown at altitudes of 30 to 40 km ( 100 130 kft ) have largely been limited to the arctic and antarctic summers and have typically involved heliophysics , x - ray , gamma - ray , particle astrophysics , and ir / sub - mm programs that are unaffected by daylight observing conditions .
only a few high altitude balloon flights , like the recent heliophysics sunrise telescope @xcite , have been conducted outside of the polar regions
. however , such high altitude , daylight balloon missions are generally not suitable for a broad spectrum of general astronomical observing programs requiring dark sky observing conditions .
the few nighttime high - altitude astronomical balloon flights that have occurred have been limited to relatively short duration times of a week or less @xcite .
despite an ever increasing number of space missions , there has been renewed interest in recent years for exploring the use of high - altitude balloon flights for nighttime astronomical research .
this has resulted in a number of papers discussing possible lighter - than - air ( lta ) vehicles and telescope arrangements for optical and infrared observations from non - polar locations @xcite .
a self - propelled , high - altitude , long endurance ( hale ) stratospheric airship capable of keeping station over a desired geographic location would be a highly attractive platform for a variety of astronomical and other science missions @xcite .
a solar - powered airship operating at altitudes near 20 km , where the stratospheric winds are lightest could , in principle , remain aloft for days , weeks , or even months thus serving as a general purpose astronomical observatory for night observations covering a broad set of targets having a wide range of declinations .
besides avoiding so - called `` no - fly zones '' over some countries that restrict free - floating balloon flights over their territory , a station - keeping airship could provide simple and continuous line - of - sight telemetry allowing for high - bandwidth data communication to a single ground station . in basic terms
, a stratospheric airship differs from a conventional airship or blimp in terms of cruising altitude , balloon fabric , and propulsion .
blimps have thick and robust gas envelopes , are flown at relatively low altitudes ( @xmath1 m ) , at low speeds ( @xmath2 m / s ) , and are powered by conventional piston engines . their advantage over airplanes
is their ability to stay aloft and hover for long durations without refueling and to do so at a relatively low cost of energy consumption ( see the recent historical review of airships by liao and pasternak @xcite ) .
the possibility of relatively low construction and operations costs have made airships attractive for a host of potential uses .
for example , the us department of defense ( dod ) has funded several high - altitude airship designs and test programs over the last decade with the goal of developing a reliable low cost stratospheric , long duration platform which could provide wide area surveillance and communications capabilities with good air defense .
recent dod projects include southwest research institute s ( swri ) sounder and hisentinel vehicles @xcite and lockheed - martin s high altitude airship ( haa ) and high altitude endurance - demonstrator ( hale - d ) airships . unfortunately , despite considerable effort and expense , no self - propelled airship built by any manufacturer has flown at stratospheric altitudes for more than one day .
the current record for a high altitude airship flight duration may still be the high platform ii vehicle built by raven industries and flown in the late 1960s at 20.4 km ( 67 kft ) for a few hours @xcite . a 2007 nasa study of a variety of lta and heavier - than - air ( hta ) unmanned hale vehicles found lta vehicle concepts attractive in terms of performance but were viewed as carrying a high technical risk @xcite .
this assessment was arrived at , in part , due to the fact that the design and construction of a high - altitude airship poses several major obstacles including large envelope size , extremely lightweight and fragile balloon fabric for lifting gas containment , energy storage and power systems , launch and recovery operations , diurnal thermal management , and high - altitude propulsion motors and propellers @xcite . a more recent 2012 assessment of us military airship efforts ( gao report 13 - 81 ) also gave an unfavorable outlook for the future development and deployment of high altitude airships . in reviewing various recent hale airship efforts ,
the report noted that many have been either terminated or have suffered `` significant technical challenges , such as overweight components , and difficulties with integration of software development , which , in turn , have driven up costs and delayed schedules . '' despite such setbacks , strong interest in the development of a high - altitude , long endurance airship persists .
several commercial telecommunication companies continue to pursue hale airship development because such platforms could provide communication and data services to consumers in rural or remote areas @xcite and would combine some of the best features of satellite and fixed wireless services such as short transmission delay times , low propagation loss , and relatively large service areas @xcite .
airship programs such as the recently completed european hapcos project ( http://www.hapcos.org ) , the japanese stratospheric airship platform study @xcite , the google internet balloon project ( `` project loon '' ) , and thales alenia space consortium s `` stratosbus '' are among some of the more recent efforts to use balloons for telecommunications purposes .
one of the most difficult problems in airship design is propulsion power .
while stratospheric wind speeds are lowest ( 5 15 m / s ) at altitudes around 20 km ( 65 kft ) , wind can vary significantly both daily and throughout the year , exceeding 25 m / s at times and even higher in gusts . at these speeds ,
wind force on a conventional natural shape balloon is considerable , driving airship designers toward aerodynamic balloon shapes with low form drag values and propulsion systems involving large solar arrays or hydrogen fuel cells . the form or shape drag force @xmath3 acting on a vehicle moving through a fluid of density @xmath4 at speed @xmath5 is @xmath6 where @xmath7 is the drag area ( equal to the projected frontal area ) of the vehicle and @xmath8 is the coefficient of form drag corresponding to the particular shape of the vehicle .
similarly , the frictional drag force @xmath9 is @xmath10 where the area @xmath11 is the wetted surface " and the coefficient @xmath12 is the skin frictional drag coefficient ( which depends on the viscosity of the fluid ) . to illustrate the wind induced drag forces on an airship
, we will consider the hisentinel50 airship built by swri .
this vehicle was cylindrical in shape with length @xmath13 m and diameter @xmath14 m. its frontal area was @xmath15 m@xmath16 , its wetted area was @xmath17 m@xmath16 with drag coefficients estimated at @xmath18 and @xmath19 . at an altitude of 65 kft
the air density is @xmath20 kg / m@xmath21 meaning that for a wind speed of 10 m / s , its total drag force is @xmath22 this force is the thrust needed to oppose its wind - induced drag .
the power the airship needs to match this wind force and thereby enable it to keep station is @xmath23 this amount of power is relatively small and practical for an airship using photovoltaic ( pv ) panels and lightweight electric motors .
but this example represents a fairly favorable scenario in terms of mild stratospheric winds of just 10 m / s at the sweet spot " altitude around 20 km plus a very low drag airship design . since drag is proportional to the square of velocity and power
is drag times velocity , propulsion power is really proportional to @xmath24 .
thus airship power requirements increase rapidly with wind speed . for instance , using the same airship numbers above but now for a wind speed of 30 m / s , the airship s total drag force increases to nearly 370 n requiring 11 kw of power to keep station .
this is a considerable amount of power to generate in order to maintain the airship floating above its desired position point , apart from any power that might be required by the airship s payload .
however , even at lower wind speeds , having an airship keep station could be challenging . if the airship s overall drag forces were twice as large due perhaps to a larger form drag coefficient for the airship or caused by a large and highly non - streamlined mission payload shape , the power required would increase by a factor of two .
this would mean some 20 kw of power would then be needed for station - keeping , an amount difficult to generate using pv panels alone on this relatively modest sized airship . since steady wind speeds around 30 m / s
are not exceptionally rare at 20 km , this means that strict year - round station - keeping for such an airship might simply not be possible .
a radically different approach for establishing a lighter - than - air stratospheric station - keeping platform involves tethering the vehicle to a ground station .
this scheme again would keep the platform s altitude to 20 km or so as to take advantage of the lightest stratospheric winds and hence the lowest drag forces on the airship .
however , no tethered high altitude stratospheric aerostat has been successfully flown for even one full diurnal cycle , although several attempts were made by french atmospheric scientists in the late 1970s @xcite .
the main obstacles include aviation restrictions , tether strength and weight , the tether winch , and tether wind drag .
storms and wind gusts in the troposphere can generate large transient wind loads on the tether , the winch , and the vehicle itself especially during initial deployment and recovery . despite this , a tethered stratospheric aerostat offers some distinct advantages over powered airships .
these include no propulsion motors or propellers allowing for higher mass payloads , no large solar panel arrays to power the propulsion motors , and no large batteries for nighttime propulsion .
in addition , the advent of technically advanced , high tensile strength materials such as ultra high molecular weight polyethylene ( uhmwpe ) such as spectra and dyneema ) , polybenzobisoxazole ( pbo ) such as zylon , and liquid crystal polymers such as vectran , kevlar , and technora has made the concept of a tethered stratospheric aerostat more practical than in the past .
several papers concerning the feasibility and flight properties of a tethered aerostat at altitudes around 20 km have appeared recently .
these include a study of a sea - anchored stratospheric , long duration balloon @xcite , the construction , launch and operation of tethered stratospheric balloons as alternatives for satellites @xcite , and investigations of the dynamic response for a high altitude tethered balloon aerostat and tether line to winds and their effects on payload pointing stability @xcite .
the chief advantage of the tethered lta platform scheme is simplicity . in principle , a land or sea deployed tether to a stratospheric balloon from a launch site with favorable tropospheric winds , few aviation hazards or flight restrictions , and seasonal periods of low stratospheric wind speeds , might allow flight durations exceeding a few days .
however , weather conditions throughout the tropospheric column ( e.g. , surface and low altitude winds and gusts , storms and downdrafts ) along with tether mass and tether wind loading may severely restrict its applicability and flight duration .
as is done for low altitude aerostats , most high - altitude tethered airship models have the tether attached to a ground - based winch which must be operated so as to limit the tension on the tether below its minimum breaking strength . despite
a number of articles discussing this approach @xcite , the only partially successful series of flights seems to have been done by atmospheric researchers in the 1970s @xcite and , to our knowledge , no high - altitude tethered aerostats have been attempted since .
here we describe an alternative means of establishing a stratospheric station - keeping lta platform that makes use of a tether . during certain times of the year at mid- and low latitudes , winds in the upper stratosphere
move in nearly the opposite direction than the wind in the lower stratosphere . a balloon or airship at high altitude
could be tethered to a heavier - than - air glider tug " at a lower altitude where the wind blows essentially in the opposite direction . by adjusting the aerodynamic configuration of the tug
, wind forces acting on it can be made to counteract those acting on the airship .
n in winter to nearly 40@xmath25n in summer .
the seasons and latitudes are reversed in the southern hemisphere . adapted from plots taken from the european centre for medium - range weather forecasts ( ecmwf ) era-40 website . ]
an example configuration exploiting this naturally occurring wind shear is shown in figure [ fig:1 ] .
the airship and its payload float at an altitude around 24 km ( 80 kft ) while the tug flies some 7 km lower at around 17 km ( 55 kft ) .
the tether connecting them is shorter and hence lighter than it would need to be if it were to extend to the ground and it does not penetrate the turbulent weather of the troposphere .
the tug s relatively high altitude places it well above the maximum operating ceilings of all commercial aircraft ( 43 kft ) and private or corporate jets ( 51 kft ) thereby greatly reducing aviation restrictions and hazards .
wind at the tug s altitude is generally stronger and the air denser than higher up meaning the tug can be relatively small and still develop the necessary forces to balance that experienced by the upper airship .
this approach to a station - keeping capability depends upon stratospheric wind shear that is , the difference in wind speed and direction between the altitude of the airship and that of the tug .
figure [ fig:2 ] shows plots of wind speed and direction as a function of altitude and latitude where altitude is indicated by the associated atmospheric pressure .
although these plots are multi - year averages for each season , they illustrate the basic stratospheric wind shear phenomenon .
each plot is annotated with the example altitudes discussed above and with a range of latitudes for which favorable conditions prevail .
although the plots of figure [ fig:2 ] and the results of other stratospheric wind studies @xcite indicate the existence of a usable stratospheric wind shear , such multi - year average plots do not reflect the variable day - to - day wind conditions that would actually govern the behavior of the proposed system .
such day - by - day wind direction differences at altitudes of 16.7 and 24.4 km ( 55 and 80 kft ) are shown in figure [ fig:3 ] .
each of the four plots is for a 60-day interval in the spring of the years 2000 , 2005 , 2010 , and 2013 for the atmosphere above hilo , hawaii ( latitude + 19.8@xmath25 ) and assembled from radiosonde data available from the university of wyoming s upper air sounding website ( http://weather.uwyo.edu ) .
typically , two radiosonde flights are made each day and both measurements are plotted when available .
the plot for 2013 shows the most recently available data .
more details may be found in @xcite .
km above hilo , hawaii for four years based on radiosonde data . a difference of 180@xmath26 ( solid line ) indicates directly opposing winds .
, title="fig : " ] km above hilo , hawaii for four years based on radiosonde data .
a difference of 180@xmath26 ( solid line ) indicates directly opposing winds . ,
title="fig : " ] + km above hilo , hawaii for four years based on radiosonde data .
a difference of 180@xmath26 ( solid line ) indicates directly opposing winds .
, title="fig : " ] km above hilo , hawaii for four years based on radiosonde data .
a difference of 180@xmath26 ( solid line ) indicates directly opposing winds . , title="fig : " ] data for a tug altitude of 16.7 km ( 55 kft ) were extracted from the radiosonde database within a relatively small altitude range ( @xmath27 0.2 km ) , while the airship s altitude was allowed to vary by @xmath28 km so as to reflect the likelihood of altitude variations due to diurnal heating effects . in cases of missing radiosonde data within these altitude ranges ,
we interpolated between the two closest values .
although the data shown in figure [ fig:3 ] cover an upper altitude range centered at 24.4 km , nearly 75% of the measurements plotted correspond to values taken at altitudes between 23.5 and 24.3 km .
it is important to note that not all sounding data covering these time intervals are shown in these plots . besides some missing radiosonde data ( typically just a few days during a month )
, we do not show wind direction differences that exceed 70@xmath26 .
large variations in upper air flows can occasionally lead to unfavorable wind conditions for several days each month .
this is the reason that during the year 2000 we show wind direction differences for march 16 may 15 rather than april 1 may 30 . during that year , the wind direction reversal formed over hawaii about two weeks earlier than typically seen . during the four periods shown , the number of 12-hour periods during which easterly and westerly wind direction were greater than 70@xmath26 apart were 23 in 2000 , 18 in 2005 , 15 in 2010 , and 20 in 2013 .
however , on many of these occasions , wind speeds were relatively low at one or both altitudes . keeping in mind these limitations , the plots of figure [ fig:3 ] illustrate that between 16.7 and 24.4 km ( 55 kft and 80 kft ) the stratospheric wind directions are within 45@xmath26 of being 180@xmath26 apart for the majority of the days shown .
the best of these two - month periods is april and may 2005 when over 85% of the time the upper and lower altitude winds were within 30@xmath26 of being 180@xmath26 apart .
the worst two month period shown occurred in 2013 . marked differences year to year is not surprising .
this is , after all , weather , and weather patterns can change significantly from one year to the next .
however , the regular appearance of such opposing wind flows between stratospheric layers only some 7 km apart can be exploited to maintain the geographical location of a high - altitude platform without the need of propulsion power . because of seasonal wind variations above a particular geographic location , stratospheric wind shear will not permit year - round station - keeping .
suitable opposing winds are found around @xmath29 latitude in hemispheric summers , but in spring and fall they are found at lower latitudes around 15 to 25@xmath26 ( see figure [ fig:2 ] .
this is shown in figure 4 where we plot wind direction differences at 15.2 km and 24.4 km ( 50 kft and 80 kft ) for the months of june and july in the years 2000 and 2010 over denver , colorado ( latitude + 39.8@xmath25 ) .
although there is considerable scatter , the lower to upper stratospheric wind shear is still within 45 degrees of being directly opposite over 75% of the time .
these plots also show that the wind shear can be experienced by a tug at lower altitudes , here at 15.2 km ( 50 kft ) .
the seasonal shift in latitude of the stratospheric wind shear means that in order to operate year - round the airship and tug will need to move north or south some 20 30 degrees in latitude during the course of a year .
a shift in latitude of the wind shear may be partially responsible for some of the unfavorable wind shear days seen in the spring months over hawaii ( see fig.3 ) .
km above denver , colorado for the years 2000 and 2010 based on radiosonde data .
a difference of 180@xmath26 ( solid line ) indicates directly opposing winds .
, title="fig : " ] km above denver , colorado for the years 2000 and 2010 based on radiosonde data .
a difference of 180@xmath26 ( solid line ) indicates directly opposing winds .
, title="fig : " ] +
the operation of the proposed station - keeping system depends on balancing the aerodynamic forces on the airship with those acting on the tug .
the air density at the tug s altitude of 17 km is roughly three times that at the airship s altitude of 24 km .
in addition , wind speeds are generally two to four times greater at the lower altitude than at the higher altitude
. the tug will therefore experience drag forces some 10 to 50 times higher than the airship even if the two vehicles are of similar size and shape .
if the airship is streamlined so as to minimize drag , the tug could be made quite compact and lightweight while still developing the counter force necessary for hold the airship steady in the wind .
table [ shipforce ] shows sample wind induced drag force calculations for three airship sizes and shapes .
these computations assumed an airship altitude of 23.7 km ( 78 kft ) and a variety of ambient wind speeds .
the listed drag force values were calculated assuming only form and surface drag .
case 1 is an airship similar in size to swri s streamlined hisentinel80 airship , case 2 is a `` super - sized '' hisentinel80 , and case 3 is for a spherical balloon having a displaced volume similar to that of hisentinel80 in case 1 . comparing cases 1 and 3 ,
it is clear that having a streamlined airship versus a spherical balloon lowers the total wind drag force by about a factor of 20 .
also , going from a small to a larger streamlined airship ( cases 1 and 2 ) the system gains a factor of nearly 10 in potential lift while the total drag force increases by only a factor around 2.5 .
* case 1 : * hisentinel80 : d = 15 m , l = 60 m ; volume : 10600 m@xmath30 + balloon : 320 kg ( @ 0.1
kg / m@xmath31 ) ; helium : 70 kg ; displaced air : 510 kg + ccccc drag & & & wind speed & + parameters & 5 m / s & 10 m / s & 20 m / s & 30 m / s + c@xmath32 = 0.03 ; a@xmath33 = 177 m@xmath31 & 3 n & 13 n & 51 n & 114 n + c@xmath34 = 0.003 ; a@xmath35 = 3500 m@xmath31 & 6 n & 25 n & 100 n & 227 n + total drag force & 9 n & 38 n & 151 n & 341 n + + * case 2 : * super - hisentinel : d = 25 m , l = 100 m ; volume : 49100 m@xmath30 + balloon : 885 kg ( @ 0.1
kg / m@xmath31 ) ; helium : 325 kg ; displaced air : 2350 kg + ccccc drag & & & wind speed & + parameters & 5 m / s & 10 m / s & 20 m / s & 30 m / s + c@xmath32 = 0.03 ; a@xmath33 = 490 m@xmath31 & 9 n & 35 n & 141 n & 317 n + c@xmath34 = 0.003 ; a@xmath35 = 9000 m@xmath31 & 16 n & 65 n & 260 n & 583 n + total drag force & 25 n & 100 n & 401 n & 900 n + + * case 3 : * spherical balloon : d = 28 m ; volume : 11500 m@xmath30 + balloon : 250 kg ( @ 0.1 kg / m@xmath31 ) ; helium : 75 kg ; displaced air : 550 kg + ccccc drag & & & wind speed & + parameters & 5 m / s & 10 m / s & 20 m / s & 30 m / s + c@xmath32 = 0.5 ; a@xmath33 = 615 m@xmath31 & 185 n & 740 n & 3000 n & 6700 n + c@xmath34 = 0.003 ; a@xmath35 = 2460 m@xmath31 & 4 n & 18 n & 70 n & 160 n + total drag force & 190 n & 760 n & 3100 n & 6900
n + the drag values listed in table [ shipforce ] must be comparable to the wind drag numbers for the lower altitude tug vehicle which are listed in table [ tugforce ] for a range of wind speeds likely to be encountered at the tug s altitude of around 17 km .
as an example , we have adopted a tug design in the form of a conventional glider consisting of a narrow fuselage and thin , high - aspect wings with high lift - to - drag ratios .
we have assumed some sort of variable drag device as part of the tug with a form drag force proportional to an adjustable area of the device .
the table shows drag force values resulting from both open and closed configurations .
as table [ tugforce ] shows , it appears feasible for a tug to generate drag forces covering the complete wind speed range calculated for the two streamlined airship cases in table [ shipforce ] ( cases 1 and 2 ) but not for a spherically shaped airship ( case 3 ) .
this again disfavors a spherical airship shape .
tug : d = 0.75 m , l = 4 m fuselage + 4 m drag device , altitude = 16.7 km ( 55 kft ) + ccccc vehicle & drag & & wind speed & + component & parameters & 10 m / s & 20 m / s & 30 m / s + fuselage + wings & c@xmath32 = 0.12 ; a@xmath33 = 1.8 m@xmath31 & 2.0 n & 9.0 n & 20 n + & c@xmath34 = 0.03 ; a@xmath34 = 7.1 m@xmath31 & 2.0 n & 8.0 n & 19 n + drag device closed & c@xmath34 = 0.03 ; a@xmath34 = 7.1 m@xmath31 & 2.0 n & 8.0 n & 19
n + `` '' opened & c@xmath32 = 1.0 ; a@xmath33 = 28 m@xmath31 & 270 n & 1090 n & 2450 n + total force range & & 6 - 275 n & 26 - 1115 n & 60 - 2490 n + single tether : altitude : 0 to 20 km ; total tether length : 20 km ; c@xmath32 = 1.0 + crrrrrr & & & altitude & & + dyneema & 0 - 5 km & 5 - 10 km & 10 - 15 km & 15 - 20 km & + ( sk78 ) & 0.96 kg / m@xmath30 & 0.56 kg / m@xmath30 & 0.30 kg / m@xmath30 & 0.13 kg / m@xmath30 & totals + 5 mm ; mbs 3300 kg & 480 kg & 280 kg & & & 760 kg + mass : 15 kg / km & 75 kg & 75 kg & & & 150 kg + 3 mm : mbs 1400 kg & & & 90 kg & 40 kg & 130 kg + mass : 5 kg / km & & & 25 kg & 25 kg & + & & & & & 1090 kg + tug - airship : altitudes : 17 and 24 km ; total tether length : 12 km ; c@xmath32 = 1.0 + crrrrrr & & & altitude & & + dyneema & ... & ... & 17 - 21 km & 21 - 24 km & + ( sk78 ) & ... & ... & 0.10
kg / m@xmath30 & 0.06 kg / m@xmath30 & totals + 2 mm : mbs 450 kg & & & 16 kg & 7 kg & 23 kg + mass : 2.4 kg / km & & & 17 kg & 12 kg & + & & & & & 52 kg +
lastly , we show in table [ tetherforce ] estimated wind loading values for both a ground - tethered high altitude aerostat and our proposed high altitude airship - tug scheme . here
we have assumed a constant wind speed of 20 m / s at all altitudes .
although there are a variety of tether materials that could be used in either scheme , for these sample calculations we chose dyneema sk78 as the tether material .
there are stronger tether material options which have higher breaking strengths but these numbers serve to give a sense of mass and wind loads at various altitudes and hence required tether strength . for the single ground - tethered scheme , we chose a 5 mm tether for altitudes 0 10 km ( 0 - 33 kft ) and a 3 mm tether for altitudes 10 20 km ( 43 - 65 kft )
. a thicker tether might be required at lower altitudes since it transverses the whole troposphere where more severe transient wind loads are likely to be experienced .
in contrast , a thinner 2 mm cord was chosen for the the airship - tug tether since wind loading conditions are far more benign above the jet stream and most storms at an altitudes 17 km ( 55 kft ) and higher .
comparison of the two high - altitude tethered airship approaches in table [ tetherforce ] shows that a single tether will experience just under one metric ton of horizontal wind loading plus tension due to 200 kg of tether mass .
although this estimate assumes a constant wind speed of 20 m / s along the entire 20 km length , these wind loads and tether mass could actually be an underestimate .
it is unlikely that the tether would be as short as 20 km given wind loading and varying wind directions and speeds from the ground winch up to the altitude of 20 km , and thus a tether length as much as 30 km is probably more realistic . in that case , again dividing the tether into 3 mm and 5 mm thicknesses but now each 15 km long an even greater wind loading might exist while the total tether mass increases to around 300 kg . in real life , the situation might be even less favorable since tropospheric wind speeds often exceed 20 m / s and can even be over 50 m / s in the jet stream . at a wind speed of 40 m
/ s , just a 1 km long section of a 5 mm thick tether at an altitude around 10 km ( 30 kft ; @xmath36 kg / m@xmath30 ) would have a wind load of 150 kg for this short section . in any case
, a minimum break strength ( mbs ) safety factor for a single tether with the chosen thicknesses is low and much less than the usually desired factor of 5 or more .
thus , such tether weight and wind loading estimates would seem to pose serious operational challenges for maintaining a stratospheric airship with a grounded tether complicating the ground tether approach further .
while , as in the single tether case , a considerably longer tether will be needed in reality than just the 7 km altitude separation of airship and tug , a shorter and thinner tether in an airship - tug scheme offers both a lower tether mass and wind loading . a tether of sk78 dyneema 12 km
long will have a combined mass load and wind load well below the tether s mbs of 450 kg .
for example , even a relatively high 30 m / s wind speed over the entire 12 km long 2 mm tether at altitudes between 17 and 24 km will only generate a total wind load of less than 100 kg .
the airship - tug station - keeping arrangement discussed above uses the naturally occurring seasonal stratospheric wind shear to provide the needed energy to keep the system on station .
the payload carrying platform s altitude around 80 kft is also much higher than that of a self - propelled airship at 65 kft thereby providing wider horizon to horizon coverage of the earth and better upward viewing image quality .
this tether scheme also avoids several problems associated with a ground - based tethered platform ; namely , little if any aviation hazard , no winch , no stormy weather to fly through , and a shorter tether meaning less tether weight and wind loading .
in addition , the tether is expected to be always under some tension so slack issues that can arise in a ground - based winch tether arrangement are reduced .
wind loading at altitudes above 15 to 17 km ( 50 to 55 kft ) should also be relatively low even in high wind conditions , making a thin and lightweight tether practical .
there are several key components of the concept that will determine its reliability and effectiveness .
the higher - altitude lta platform must be constructed so as to have no appreciable fabric or seam leaks of lifting gas ( i.e. , hydrogen or helium ) thus permitting long float durations of weeks to months .
both it , the tug and the tether should be as lightweight as possible enabling the greatest payload mass in relation to the balloon s lift capability .
ideally , the upper lta platform would also have a streamlined aerodynamic shape so as to lessen wind drag forces as much as possible .
it should also have some directional lift capability such as through a rear vertical stabilizer so as to help steer it into or against the prevailing winds and be designed for flexibility in payload mounting configuration . for example , astronomers may want a top - mounted telescope that has unobstructed access to targets near the zenith , while earth scientists may prefer down pointing instruments . however , the most critical component of the proposed concept is perhaps the tug vehicle .
we conceive the tug as taking the form of a ultra - lightweight glider with intrinsically low drag .
it could develop the forces needed to counter drift of the airship in two ways : deployment of a variable drag device such as a parachute or umbrella like device or variable pitch propeller(s ) , or it could generate appropriate aerodynamic forces with its wings .
drag is necessarily in the direction of airflow , so it may seem that the high - drag configuration would only work if the winds at the two altitudes exactly oppose .
but if the airship were a dirigible " design , it could develop aerodynamic forces that are not precisely parallel to wind direction .
similarly , the tug could be controlled to fly in a direction that produced the necessary tether force over a wide range of angles relative to the wind direction .
the combination of a semi - steerable lta airship and a maneuverable drone - like tug with variable lift capability could allow the system to keep station in a variety of wind combinations .
it could even maneuver to find better wind conditions , and climb and descend to some degree as needed .
the tug will need to be able to generate it own power to serve its operating flight systems and possibly to be self - propelled to some extent .
in addition to solar pv power stored in batteries , the tug could be equipped with a propeller to serve as a variable drag device and power from the propeller could be used to generate electricity both day and night .
a cruder wind force balancing scheme was proposed in 1969 by r. bourke @xcite in a raytheon company report .
he described a concept in which a conventional balloon floating in the stratospheric easterlies could deploy a parachute into the lower stratospheric westerlies to provide a drag force to overcome the balloon s drift . using available wind data available at the time
, bourke concluded that this arrangement could work for certain months of the year , mainly during summer months at mid - latitudes .
but he also found that the altitude and latitude of the lowest stratospheric winds varied seasonally leading to difficulties in maintaining accurate station keeping .
nonetheless , he viewed the concept as `` provocative in its intrinsic simplicity '' .
however , to our knowledge no high - altitude balloon plus drag chute system was ever deployed and tested by raytheon or anyone else .
our scheme differs substantially from that proposed by bourke .
he suggested that the upper altitude balloon have self - propulsion capabilities and proposed a simple drag chute lowered from the balloon using a winch only as a supplemental element to aid the airship s station - keeping ability .
in contrast , our concept consists of a passive and ideally aerodynamically - shaped , stratospheric balloon or airship tethered to a lower altitude robotic tug vehicle that can precisely control its aerodynamic wind forces .
our stratospheric airship would have no self - propulsion element but could have some directional steering capabilities much like that demonstrated in a high altitude wing guidance system @xcite .
bourke s use of a winch - lowered drag chute may have been an attempt to simplify the balloon launch .
our scheme could also include some sort of tether storage system possibly attached to the tug vehicle in an effort to better control deployment and recovery of both upper and lower vehicles .
one application for a stratospheric platform would be wide - field , high resolution optical and near - infrared imaging of astronomical targets . the value of high angular resolution imaging for astronomy can not be overstated .
the chief reason for the enormous impact of the hubble space telescope ( hst ) across a wide spectrum of research topics despite its modest size mirror ( 2.4 m ) has been its ability to obtain diffraction - limited imaging due to its location above earth s atmosphere .
however , with no repair or refurbishment missions currently planned , hubble s expected useful lifetime will probably end before the year 2020 due instrument failures or degradation of its batteries , solar panels , pointing gyros , and associated equipment @xcite . with no present follow - up optical / uv space mission to hubble
, its loss may mean astronomical high - resolution imaging might be confined for the near future to small space telescopes or ground - based adaptive optics ( ao ) instruments which employ one or more natural or laser guide stars to correct for atmospheric turbulence .
unfortunately , ao instruments work best in the infrared and under good seeing conditions and provide limited field - of - view ( @xmath37 1 arcmin ) with strehl ratios less than 60% @xcite . a reliable lta platform situated at an altitude of 20 km or higher
should , if properly equipped , provide image quality competitive with space - based telescopes .
such an observatory could provide sub - arcsecond imaging with short response times at a much lower cost than a comparable space - based telescope .
for example , at an altitude of 24 km ( 80 kft ) an astronomical telescope would be above the weather and all but @xmath0 2.5% of the atmosphere .
it would experience virtually perfectly clear skies every night with image quality at or approaching the diffraction limit of the main aperture .
thus , an optical telescope located at such stratospheric altitudes with a mirror just 0.5 m in diameter ( 20-inch ) with sufficient pointing stability and large ccd arrays could provide wide - field images with fwhm = 0.25 arcsecond at 500 nm , making it virtually superior to any ground - based imaging system .
being above the weather , it could provide such data quality night after night for as long as the platform remained at this altitude . the lack of appreciable water vapor , dust and other particulates in the remaining atmosphere above these altitudes such a platform would also enjoy excellent atmospheric transmission .
light scattering from moonlight would be expected to be minimal and not a major factor in scheduling faint target observations , making most observing time effectively astronomical `` dark time . ''
this feature would greatly enhance the platform s ability to respond rapidly to opportunities for observations of faint transient targets such as supernovae and gamma - ray bursters .
also , unlike low earth orbit ( leo ) satellites such as hst , data transfer to and from a high - altitude station - keeping observatory could involve simple line - of - sight communications running continuously to a single ground station .
finally , a stratospheric astronomical observatory could also provide reliable science support for a host of space - based missions at an estimated cost of a few percent of a conventional leo satellite .
we have described a new method for establishing a near station keeping , stratospheric lta vehicle at low and mid - latitudes .
this concept uses the naturally occurring seasonal wind shear between upper and lower layers of the stratosphere to provide forces that counter platform wind drift and allow it to keep station over a specified geographical location .
we have necessarily left out many details about the architecture .
these include platform migration issues in order to follow seasonal variations in latitude where optimal stratospheric wind shears are found , launch and recovery problems and solutions , specific airship and tug design constraints , and science payload arrangements to permit unobscured horizon - to - horizon observations .
if this method is shown to be practical , then the quest for the long - sought method of station keeping a scientific hale platform may finally be realized , within season and latitude restrictions .
this concept could provide the means for obtaining high quality data rivaling space - based platforms but at a small fraction of the cost .
the development of an affordable stratospheric platform that could keep station for weeks or months would be a powerful new tool for a variety of users and could be a game - changer for astronomical , atmospheric , and earth - science research , as well as for a host of other applications including military surveillance and civil telecommunications services .
the authors gratefully acknowledge valuable advice and conversations about high - altitude lta science platforms from participants in the w. m. keck institute for space studies ( kiss ) workshop entitled `` airships : a new horizon for science , '' especially jeff hall , steve lord , steve smith , mike smith , and workshop co - leads sarah miller , lynne hillibrand , and jason rhodes .
djuknic , g .m . ,
freidenfelds , j. , & okunev , y. : establishing wireless communications services via high - altitude aeronautical platforms : a concept whose time has come ? ieee communications magazine , pp 128135 , september 1997 equchi , k. et al . :
feasibility study program on stratospheric platform airship technology in japan .
aiaa s 13th lighter - than - air systems technology conference , norfolk va ( aiaa 99 - 3912 ) ( 1998 ) norfolk va ( aiaa 99 - 3912 ) .
grace , d. , thornton , j. , chen , g. , white , g.p . , & tozer , t.c .
: improving the system capacity of broadband services using multiple high altitude platforms .
ieee transactions on wireless communications , * 4 * , no .
2 , p 700 ( 2005 ) habib , a. , vernin , j. , benkhaldoun , z. , & lanteri , h. : single star scidar : atmospheric parameters profiling using the simulated annealing algorithm .
monthly notices of the royal astronomical society , volume 368 , issue 3 , pp .
1456 - 1462 ( 2006 ) hibbitts , c. a. , young , e. , kremic , t. , & landis , r. : science measurements and instruments for a planetary science stratospheric balloon platform .
proceedings of the aerospace conference , ieee . 2 - 9
march 2013 .
big sky , mt . isbn : 978 - 1 - 4673 - 1812 - 9 , id.178 ( 2013 ) hoegemann , c. k. , chueca , s. , delgado , j. m. , et al . : cute scidar : presentation of the new canarian instrument and first observational results .
spie 5490 , advancements in adaptive optics , 774 ( october 25 , 2004 ) ; doi:10.1117/12.551795 lee , m. , smith , s. , & androulakakis , s. : the high altitude lighter than air airship efforts at the us army space and missile defense command / army forces strategic command .
18th aiaa lighter - than - air systems technology conference , 4 - 7 may 2009 , seattle , washington aiaa 2009 - 2852 perotti , f. , della - ventura , a. , sechi , g. , et al . : balloon - borne observations of ngc 4151 using the miso telescope . in : non - solar gamma - rays ; proceedings of the symposium , bangalore , india , may 29-june 9 , 1979 .
oxford , pergamon press , ltd .
, p. 67 - 70
( 1980 ) .
rigaut , f. , neichel , b. , boccas , m. , et al . :
gemini multiconjugate adaptive optics system review - i. design , trade - offs and integration .
monthly notices of the royal astronomical society , * 437 * , p.2361 - 2375 , 437 ( 2014 ) smith , m.s . , and rainwater , e.l .
: applications of scientific ballooning technology to high altitude airships .
aiaa s 3rd annual aviation technology , integration , and operations , denver co , 17 - 19 november 2003 smith , i.s . ,
fortenberry , m.l . ,
lee , m. & judy , r. : hisentinel80 : flight of a high altitude airship .
presented at the 11th aiaa aviation aviation technology , integration , and operations conference , 19th aiaa lighter - than - air , virginia beach , va . ,
september 2011 .
von appen - schnur , g. f. , kueke , r. , schaefer , i. , & stenvers , k .- h . : aerostatic platforms , past , present , and future : a prototype for astronomy ?
spie * 4014 * , p. 226 - 236 ,
airborne telescope systems , ramsey k. melugin ; hans - peter roeser ; eds ( 2000 ) welsh , b. y. , boksenberg , a. , anderson , b. , & towlson , w. a. : high resolution ultra - violet observations of alpha lyrae using the university college london balloon - borne telescope system . astronomy and astrophysics , * 126 * , 335 - 340 ( 1983 ) wilson , r. w. , wooder , n. j. , rigal , f. , & dainty , j. c. : estimation of anisoplanatism in adaptive optics by generalized scidar profiling .
monthly notice of the royal astronomical society , volume 339 , issue 2 , pp .
491 - 494 ( 2003 ) | during certain times of the year at middle and low latitudes , winds in the upper stratosphere move in nearly the opposite direction than the wind in the lower stratosphere . here
we present a method for maintaining a high - altitude balloon platform in near station - keeping mode that utilizes this stratospheric wind shear .
the proposed method places a balloon - borne science platform high in the stratosphere connected by a lightweight , high - strength tether to a `` tug '' vehicle located in the lower or middle stratosphere . using aerodynamic control surfaces , wind - induced aerodynamic forces on the tug can be manipulated to counter the wind drag acting on the higher altitude science vehicle , thus controlling the upper vehicle s geographic location .
we describe the general framework of this station - keeping method , some important properties required for the upper stratospheric science payload and lower tug platforms , and compare this station - keeping approach with the capabilities of a high altitude airship and conventional tethered aerostat approaches .
we conclude by discussing the advantages of such a platform for a variety of missions with emphasis on astrophysical research . |
starting with early considerations of bronstein , mead , wheeler , pauli , heisenberg and others ( see @xcite for a nice historical view on the subject ) , there exists a vast literature dealing with the reasons why putting together general relativity with quantum mechanics asks for a revision of the classical concept of spacetime .
a common feature is the emergence of noncommutativity in the description of spacetime itself , as opposed to quantum mechanics alone where noncommutativity lives on the phase space .
the appearance of spacetime noncommutativity has a wide range of motivations , from pure mathematics to phenomenological arguments .
we shall not try to be exhaustive here , and simply point out some directions of research that have been through interesting developments in the last years .
we shall make the distinction between _ noncommutative spacetimes _ intended as spaces whose coordinates no longer commute , and _ spectral geometries _ intended as a space whose algebra of functions is non - necessarily commutative .
_ noncommutative spacetimes _ , as recalled in section 2 , are obtained as a deformation of a usual space by trading the ( commutative ) coordinate functions @xmath0 of a manifold with coordinate operators @xmath1 satisfying non - trivial commutation relations . besides the seminal quantum spacetime model of doplicher , fredenhagen , roberts @xcite treated in l. tomassini talk @xcite , such spaces are present in many - if not all - approaches to quantum gravity , including loop quantum gravity , string theory ( see p. aschieri and r. szabo text @xcite ) , as well as in more phenomenology - oriented models like doubly special relativity , as illustrated by s. bianco in his presentation @xcite .
noncommutative spacetimes also emerged very early as a possible solution to ultraviolet divergencies in quantum field theory , especially in the work of snyder @xcite . quantum field and gauge theories on noncommutative spacetimes
have thus developed as a theory on their own , independently of any consideration on quantum gravity .
recent advances are presented in the texts of t. juric , s. meljanac , a , samasarov one the one side @xcite , and j .- c .
wallet & a. gr for noncommutative gauge theories @xcite on the other @xcite . nevertheless , revisiting the work of snyder in the light of nowadays quantum - gravity problematic offers an intriguing point of view , which is presented in v. astuti paper @xcite
. _ spectral geometries _
@xcite is the subject of section 3 .
it consists in a generalization of gelfand duality between locally compact spaces and @xmath2-algebras , so that to encompass all the aspects of riemannian geometry @xcite beyond topology .
it furnishes a geometrical interpretation of the lagrangian of the standard model of elementary particles @xcite , as well as some possibilities to go beyond @xcite .
recent progress on that matter are reported in the contribution of a. devastato @xcite .
finally f. besnard @xcite discusses the extension of this approach to the pseudo - riemannian case .
spacetime as a pseudo - riemannian manifold loses sense at planck scale @xmath3 this is because an arbitrary accurate localization process requires to concentrate an arbitrary amount of energy in a small volume , yielding the creation of a black hole . to maintain an operational meaning to the measurement process
, one should impose some non - zero minimal uncertainties in the simultaneous measurement of spacetime coordinates .
this can be realized by promoting the coordinates functions @xmath4 to operators @xmath5 satisfying the non - commutative relation @xmath6 = q_{\mu\nu}.\label{eq:1}\ ] ] the behavior of under a poincar transformation marks the difference between two classes of models , that both have given birth to an extended literature and many discussions , sometimes quite vivid . for simplicity ,
let us assume that the commutator of two coordinates is a central element ( although some models of non - central commutators have been investigated , see in tomassini s paper ) , that is @xmath7 = i\lambda_p^2 \theta_{\mu\nu}\mathbb i\ ] ] where @xmath8 is the identity operator in the hilbert space on which the @xmath5 are represented and @xmath9 is an antisymmetric matrix . obviously is not invariant under the action of the poincar group @xmath10 since @xmath11 & = [ a_\mu , a_\nu]{{\mathbb i}}+ q_\alpha([a_\mu{{\mathbb i } } , \lambda_\nu^\alpha{{\mathbb i } } ] - [ a_\nu{{\mathbb i}},\lambda_\mu^\alpha{{\mathbb i } } ] ) + \lambda_\mu^\alpha[q_\alpha , q_\beta]\lambda_\nu^\beta \\&= i\lambda_p^2\,\lambda_\mu^\alpha \lambda_\nu^\beta \theta_{\alpha\beta}{{\mathbb i}}\neq [ q'_\mu , q'_\nu].\end{aligned}\ ] ] one may ask that @xmath12 transforms under the conjugate action of the poincar group , yielding the _ poincar covariant _ model of doplicher , fredenhagen , roberts . the planck length , viewed as the norm of the tensor @xmath13 is poincar invariant , and there is no modification of the dispersion relation @xmath14
. applications of this model to cosmology are presented in * luca tomassini * s _ noncommutative friedmann spacetimes from penrose - like inequalities _ @xcite .
alternatively one may require the _ poincar invariance _ of the relation by imposing @xmath15 = i\lambda_p^2 \theta_{\mu\nu}.\label{eq:2}\ ] ] this means that the symmetry group of the quantum space is no longer the poincar group , but a quantum group deformation of it ( the so called @xmath16-poincar quantum group ) , characterized by a non - trivial commutation relation for translations @xmath17
= i\theta_{\mu\nu}- i\theta^{\alpha\beta}\lambda^\mu_\alpha \lambda^\nu_\beta.\ ] ] under this symmetry , @xmath18 is again lorentz invariant but there is now a modification of the dispersion relation @xmath19 the possible experimental signature of have been intensively explored , also for the lie algebra- like noncommutativity ( @xmath20-poincar deformation ) @xmath6
= \kappa \,\epsilon_{\mu\nu}^\rho q_\rho . \label{eq:3}\ ] ] such quantum - group deformations of poincar symmetries provide a useful tool to implement the idea of _ doubly special relativity _ , that is the implementation of the planck length @xmath18 as an invariant scale , in a similar manner as the speed of light @xmath21 @xcite .
this has been interpreted later as a geometry where the space of momenta is curved ( _ relative locality _ ) .
recent developments on that matter are treated in * stefano bianco * _ phenomenology from the dsr - deformed relativistic symmetries of 3d quantum gravity via the relative - locality framework _ @xcite .
quantum field theories on noncommutative spacetime were put at the front of the scene by ( open ) string theory , when it was observed that @xmath22-brane world volume acquire a noncommutative deformation in the background of a non - zero b - field .
this , together with new developments on closed string and nonassociative algebras , is recalled in * paolo aschieri & richard j. szabo * _ triproducts , nonassociative strar products and geometry of r - flux string compactifications _
@xcite .
however one should not forget that quantum field and gauge theories on noncommutative spacetimes have been originally introduced independently of quantum gravity , as a tool to avoid ultraviolet divergencies .
this was the original idea of snyder , that has been somehow subsumed by renormalization .
nevertheless the idea that noncommutative spaces offer a beautiful ground to understand better quantum field and gauge theories , especially their renormalization properties , has been intensively investigated in the last decade . a scalar field model on a generalized @xmath20-minkowski space
is presented in * tajron juric , stjepan meljanac , and andjelo samsarov * _ light - like @xmath20-deformation and scalar field theory via drinfeld twist _ @xcite . the recent advances in gauge theory on noncommutative spacetimes @xcite
are reported in * jean - christophe wallet * _ spectral theorem in noncommutative field theories : jacobi dynamics _
@xcite .
finally , to go back to quantum gravity , let us also mention that snyder s ideas re - thought from a quantum gravity perspective yields intriguing results .
recently , it has been used to question how much of the noncommutativity of the coordinates actually survive the description of physical systems .
this is the object of * valerio astuti * _ covariant quantum mechanics applied to noncommutative geometry _ @xcite .
on may question the physical meaning of the noncommutative coordinate @xmath5 in . as elements of the ( abstract ) polynomial algebra they generate , the spectrum of the @xmath5 s is not real , which makes their interpretation as physical observable problematic .
a solution is to represent the @xmath5 s on some hilbert space , as this is done in quantum mechanics . to do so , it is convenient to view the @xmath23 s as affiliated to the algebra of compact operators , as pointed out in the doplicher - fredenhagen - roberts .
but this indicates that the algebra generated by the coordinates may not be the most accurate tools to describe a quantum space , an algebra suitably associated to the @xmath5 s can do a better job . with this idea in minds ,
one has no reason to restrict one s attention to noncommutative deformations of commutative coordinates : there are much more noncommutative algebras to play with ! this idea is enforced by gelfand duality between commutative @xmath2-algebras and locally compact spaces , which suggests that a natural definition of a _ noncommutative geometry _ is an object such that its `` algebra of functions '' ( and not only its coordinates ) is noncommutative .
a spectral triple consists in a involutive algebra @xmath25 , a faithful representation on @xmath26 , a densely defined operator @xmath22 on @xmath26 such that @xmath27 $ ] is bounded and @xmath28^{-1}$ ] is compact for any @xmath29 and @xmath30 . imposing a set of further conditions
, one defines _
real _ spectral triples , whose paradigmatic example is @xmath31 that is the algebra of smooth functions on a closed spin manifold @xmath32 , acting on the hilbert space of square integrable spinors , with @xmath33 the dirac operator .
conversely , one has the following reconstruction theorem @xcite : given a real spectral triple @xmath24 with @xmath25 unital commutative , then there exists a compact oriented riemannian spin manifold @xmath32 such that @xmath34 .
these conditions still makes sense for non - commutative @xmath25 , so that one defines a _ noncommutative geometry _ as a spectral triple @xmath35 where @xmath25 is non necessarily commutative .
to summarize : @xmath36 spectral triple turn out to be a powerful tool to describe the standard model of elementary particles from a purely geometric point of view .
the starting point is to view spacetime no more as a manifold , as in general relativity , but as the product of a manifold by a matrix geometry .
this allows to incorporate in the geometry the degrees of freedom of the standard model .
more precisely , one considers the _ almost - commutative _ algebra @xmath37 where @xmath38 is a finite dimensional algebra that carries the gauge group of the standard model ( which is retrieved as the group of unitaries of @xmath38 ) .
it acts on the space of fermions , that is @xmath39 ( two colored quarks , one neutrino and one electron make @xmath40 , that multiplies @xmath41 for the chirality , another @xmath41 for antiparticles and @xmath42 for the number of generations ) .
there is a finite dimensional dirac operator @xmath43 , that is a @xmath44 matrix whose entries are the yukawa coupling of the fermions and the mixing angles of quarks and neutrinos .
a general formula for product of spectral triples yields the generalized dirac operator @xmath45 .
bosonic fields are generated by the so - called fluctuations of the metric , that is the substitution of @xmath22 with the covariant dirac operator @xmath46 where @xmath47 is a selfadjoint element of the set of generalized @xmath48-forms @xmath49\right\}$ ] .
the asymptotic expansion @xmath50 of the _ spectral action _
@xmath51 @xcite where @xmath52 is a cutoff parameter and @xmath53 a smooth approximation of the characteristic function of the interval @xmath54 $ ] yields the bosonic lagrangian of the standard model ( including the higgs ) minimally coupled to einstein - hilbert action ( in euclidean signature ) . in other terms
the standard model appears as a purely gravitational theory , but on a ( slightly ) noncommutative space . as a bonus
, the higgs field comes out as the component of a connection in the noncommutative part of the geometry . practically , the spectral action provides relations between the parameters of the theory at a putative energy of unification . in particular the mass term of
the higgs appears as a function of the input of the models , namely the yukawa couplings of fermions . assuming the big - desert hypothesis
, the running of this mass under the flow of the renormalization group yields a prediction for the mass of the higgs of @xmath55 gev , a value ruled out by tevatron in @xmath56 . since then the higgs - brout - englert boson has been discovered with a mass around @xmath57 gev .
this mass is problematic , or at least intriguing , because it lies just below the threshold of stability , meaning that electroweak vacuum is a metastable state rather than a stable one .
one solution to stabilize the electroweak vacuum is to postulate there exists another scalar field , called @xmath58 , suitably coupled to the higgs .
chamseddine and connes have noticed in @xcite that taking into account this new scalar field in the spectral action , by promoting the yukawa coupling of the right neutrino ( which is one of the constant component of the matrix @xmath43 ) to a field , @xmath59 then one obtains the correct coupling to the higgs as well as a way to pull back the mass of the higgs from @xmath55 to @xmath60 . in @xcite , as reported in * agostino devastato * _ noncommutative geometry , grand symmetry and twisted spectral triple _ @xcite , its is shown how the substitution can be obtained as a fluctuation of the dirac operator , but in a slightly modified version inspired by the notion of _ twisted spectral triple _ introduced previously by connes and moscovici .
the field @xmath58 thus appears as a higgs - like field associated to a spontaneous symmetry breaking to the standard model of a `` grand symmetry '' model where the spin degrees of freedom ( @xmath61 acting on the space of spinors ) are mixed with the internal degrees of freedom ( @xmath38 acting on the space of particles ) .
spectral triple provides a generalization of riemannian geometry to the noncommutative setting , but there is no reconstruction theorem for pseudo - riemannnian manifolds . once computed the spectral action , one makes a wick rotation @xmath62 , as done for instance in the path integral approach to quantum gravity .
however one might like to make sense of minkovskian signature from the beginning .
in _ two roads to noncommutative causality _
@xcite , * fabien besnard * presents a state of the art , including his own recent results , on various attempts to incorporate a causal structure in spectral triples .
let us end this discussion by our own contribution , stressing the algebraic approach on how to put time into the game .
this is the idea , sometimes advertised by connes as `` the heart of noncommutative geometry '' , that time involution is intrinsically contained within the noncommutativity of the algebra .
namely , given a von neumann algebra @xmath25 acting on an hilbert space @xmath63 together with a vector @xmath64 in @xmath63 cyclic and separating for the action of @xmath25 , one defines by tomita - takesaki modular theory a @xmath48-parameter group of automorphism @xmath65 .
connes has shown that the group obtained from another state @xmath66 differs from the former only by unitaries , that is @xmath67 where the unitary intertwining is given by connes cocycle @xmath68 .
hence there is a canonical group of outer automorphism @xmath69 canonically associated to the von neumann algebra @xmath25 , where @xmath70 the physical interpretation of this group as a time evolution is enforced by the fact that @xmath71 satisfies with respect to @xmath64 the same properties as does in statistical physics the time evolution @xmath72 with respect to a gibbs equilibrium state , namely the kms condition .
an interesting framework to test the interpretation of the modular group as a real `` physical '' time is algebraic quantum field theory , where the von neumann algebra is the algebra @xmath73 of local observables associated to an open region @xmath74 of minkowski spacetime . in particular , for @xmath75 the rindler wedge
, it is well known that the modular group of the vacuum vector is generated by the boosts .
hence it has a geometrical action whose orbits are the trajectories of uniformly accelerated observers .
the kms condition is interpreted as the vacuum being a thermal equilibrium state for this observer , with a temperature proportional to the acceleration ( see @xcite for a recent critical view on this interpretation , though ) .
a similar analysis holds for double - cone regions of minkowski spacetime and yields a correction to the unruh temperature for an observer with a finite life - time @xcite , see fig .
[ fig1 ] .
interestingly , a similar analysis for a double - cone in a bidimensional boundary conformal field theory permits to compute explicitly the action of connes cocycle .
the field in the double - cone is determined by its components @xmath76 on each interval on the boundary defining the double - cone ( fig .
[ fig2 ] ) .
there is a state whose modular group has pure geometrical action @xmath77 ( but the orbit is not the trajectory of an observer with constant acceleration ) while the modular group for the vacuum mixes this geometric action with a mixing of the components @xmath78 where @xmath79 is a suitable parametrization of the orbit and @xmath80 is the mixing matrix . in this sense ,
the action of the unitary cocyle is non - geometric , and amounts to mix the components of the conformal field on the two intervals . for a further interpretation of this ,
see @xcite and @xcite . | noncommutative geometry in its many incarnations appears at the crossroad of various researches in theoretical and mathematical physics : from models of quantum space - time ( with or without breaking of lorentz symmetry ) to loop gravity and string theory , from early considerations on uv - divergencies in quantum field theory to recent models of gauge theories on noncommutative spacetime , from connes description of the standard model of elementary particles to recent pati - salam like extensions .
we list several of these applications , emphasizing also the original point of view brought by noncommutative geometry on the nature of time .
this text serves as an introduction to the volume of proceedings of the parallel session `` noncommutative geometry and quantum gravity '' , as a part of the conference `` conceptual and technical challenges in quantum gravity '' organized at the university of rome _ la sapienza _ in september 2014 . |
we have extended the standard staggered quark qcd action by the addition of a 4-fermion term with a @xmath0 chiral flavour symmetry , generated by @xmath1 @xcite .
( for earlier , related work see @xcite . ) introducing the auxiliary fields @xmath2 to render the action quadratic we have @xmath3 + { n_f \over 8}\sum_s \dot{\psi}^{\dag } a^{\dag } a\dot{\psi } \nonumber\\ \!\ ! & & \!\!\ !
-\sum_{\tilde{s}}\frac{1}{8}n_f\gamma(\sigma^2+\pi^2 ) + \frac{1}{2}\sum_{\tilde{s}}(\dot{\sigma}^2+\dot{\pi}^2 ) \nonumber\\ \!\ ! & & \!\!\ ! + \frac{1}{2}\sum_l(\dot{\theta}_7 ^ 2+\dot{\theta}_8 ^ 2 + \dot{\theta}_1^{\ast}\dot{\theta}_1 + \dot{\theta}_2^{\ast}\dot{\theta}_2 + \dot{\theta}_3^{\ast}\dot{\theta}_3)\end{aligned}\ ] ] for our molecular dynamics lagrangian , where @xmath4 the advantage of this action is that the dirac operator remains non - singular when @xmath5 so that we can work in the chiral limit .
note that we have an exact chiral symmetry generated by @xmath6 which means we will have a massless pion at @xmath5 when this symmetry is spontaneously broken .
the 4-fermion term is an irrelevant operator , so this theory should have the same continuum limit continuum qcd as the standard action .
@xmath7 qcd is believed to have a second order finite temperature chiral phase transition for @xmath5 which weakens to a rapid crossover , with no actual phase transition , for @xmath8 .
the critical exponents that describe the neighbourhood of this critical point are expected to be those of an @xmath9 spin model in 3 dimensions .
( see @xcite for a recent review of lattice qcd thermodynamics and critical exponents . ) for the staggered lattice formulation the critical exponents are expected to be @xmath10 or @xmath9 .
( for recent work on staggered critical exponents see @xcite . ) here the advantage of being able to work at zero quark mass is clear . we can hope to see clear evidence for the second order nature of the phase transition . in addition
one can measure certain critical exponents directly .
in particular one can measure the exponent @xmath11 from @xmath12 where @xmath13 for mean field theory , @xmath14 for @xmath9 , and @xmath15 for @xmath10 . in the high temperature phase , the common @xmath16 and @xmath17 screening masses increase from zero as @xmath18 where @xmath19 for mean field theory , @xmath20 for @xmath9 and @xmath21 for @xmath10 .
the same critical exponent governs the vanishing of the @xmath17 mass below the transition .
if we now fix @xmath22 and decrease the mass @xmath23 from a finite value to zero , @xmath24 vanishes as @xmath25 where @xmath26 for mean field theory , @xmath27 for @xmath9 , and @xmath27 for @xmath10 . of the fluctuation quantities ( susceptibilities ) , the most accessible is the `` magnetic '' susceptibility @xmath28\ ] ] which scales as @xmath29
our simulations are being carried out using hybrid molecular dynamics with noisy fermions to accommodate @xmath7 . because of the exact flavour @xmath30 symmetry at @xmath5 , the direction of symmetry breaking in the ( @xmath24,@xmath31 ) space ( or ( @xmath17,@xmath16 ) space is not determined . on a finite lattice ,
this direction rotates during the run , forcing us to use @xmath32 or @xmath33 as our order parameter on each configuration .
( here @xmath34 should be taken to mean a lattice average , not an ensemble average . )
this estimate differs from the true value by @xmath35 , which can only be removed by working at more than 1 spatial volume for each @xmath36 .
we are currently running on @xmath37 , @xmath38 , @xmath39 , and @xmath40 lattices at @xmath41 , and on a @xmath39 at @xmath42 . whereas our earlier simulations at @xmath43 and @xmath44 showed a first order transition ,
all @xmath45 combinations above show evidence for the expected second order transition . @xmath24 and
the wilson line show a sharp , but not discontinuous , transition with no sign of metastability , unlike the previous case .
figure [ fig : pbp ] shows the @xmath46 dependence of the chiral condensate for @xmath45 and @xmath47 .
in addition , close to the transition , observables show clear signs of critical slowing down with large fluctuations over many thousands of time units .
( we have simulated for as many as 39,000 time units at a single @xmath46 value . ) such a time evolution for the chiral condensate close to the transition is shown in figure [ fig : pbp_time ] .
the @xmath43 , @xmath47 runs show signs of critical fluctuations , but the transition occurs over a very narrow range and it will require more statistics to tell if it has become second order or remains first order .
it is clear that we will need to extend our runs close to the phase transition .
this is a consequence of the critical slowing down close to the second order transition .
this currently limits our ability to extract accurate critical exponents due in part to the fact that our errors are poorly determined and partly because it prevents accurate extrapolation to infinite volume .
in addition , the scaling window could be very narrow requiring us to consider more @xmath46 values close to the transition . despite this the @xmath24 data looks promising , and suggests a critical exponent @xmath48 , however , this is on the basis of statistically poor fits .
the @xmath45 , @xmath47 data shows relatively modest finite volume effects . for @xmath43 , @xmath47
, we find @xmath49 .
for @xmath45 , @xmath44 , @xmath50 , while for @xmath45 , @xmath47 , @xmath51 .
figure [ fig : masses ] shows the @xmath17 and @xmath16 screening masses for the @xmath40 , @xmath47 simulations .
the @xmath17 screening mass shows the correct behaviour , decreasing to zero as @xmath46 is increased through the transition , and increasing from zero , degenerate with the @xmath16 above the transition . below the transition ,
the pion mass is consistent with zero .
it appears doubtful that these masses can be determined accurately enough to obtain a good estimate for @xmath52 .
the @xmath53 mass can be calculated , but even assuming it remains distinct from the @xmath54 mass above the transition , it will be difficult to tell how much of this is due to the explicit @xmath30 symmetry breaking provided by the 4-fermion interaction .
@xmath7 lattice qcd with massless quarks and a weak 4-fermion interaction appears to have the expected second order transition , at least for @xmath55 .
more work is needed to clarify the @xmath43 case . with more statistics the @xmath45 simulations should produce an accurate determination of the critical exponent @xmath11 .
moving to finite mass at @xmath22 should allow an accurate determination of @xmath53 .
hadronic screening masses need further analysis .
other order parameters remain to be analysed .
unfortunately , there is no obvious way to include 4-fermion interactions with full @xmath56 chiral flavour symmetry .
supported by doe contract w-31 - 109-eng-38 and nsf grant nsf - phy-96 - 05199 .
computing was provided by nersc .
999 j. b. kogut , j .- f .
laga and d. k. sinclair , phys .
d58 , 034504 ( 1998 ) , and references contained therein .
k. i. kondo , h. mino , and k. yamawaki , phys .
d39 , 2430 ( 1989 ) ; r. c. brower , y. shen and c .- i .
tan , boston university preprint buhep-94 - 3 ( 1994 ) ; r. c. brower , k. orginos and c .- i .
tan , nucl .
b(proc . suppl . ) 42 , 42 ( 1995 ) .
e. laermann , nucl .
63a - c , 114 ( 1998 ) . c. w. bernard , et al .
d55 , 6861 ( 1997 ) . | we have simulated lattice qcd with an irrelevant 4-fermion interaction and 2 zero mass quarks .
the chiral phase transition is observed to be second order and we discuss extraction of critical exponents . |
downscaling magnetoelectronic devices to the nanometer regime implies that electron - electron interaction effects become prominent , as has been amply demonstrated by many experimental studies on the coulomb blockade in double tunnel junctions with ferromagnetic elements.@xcite much of the theoretical work focusses on f@xmath0n@xmath0f spin valves , in which the island is a normal metal ( n ) and the contacts are ferromagnets ( f ) with variable magnetization directions .
initially , the interest was mainly focussed on the giant magnetoresistance , _
i.e_. the difference in the transport properties for parallel or antiparallel magnetizations.@xcite more recently , the interplay between spin and interaction effects for non - collinear magnetization configurations has attracted quite some interest.@xcite a single - electron spin - valve transistor ( sv - set ) is an f@xmath0n@xmath0f spin - valve with a sufficiently small normal - metal ( n ) island that is coupled to the ferromagnetic leads by tunnel barriers .
when the electrostatic charging energy of the island is larger than the thermal energy , charge transport can be controlled at the level of single electron charges by varying bias and gate voltage , as is well known for non - magnetic sets.@xcite with spin - dependent electron tunneling rates and sufficiently long spin - decay lifetimes , a spin accumulation ( or non - equilibrium magnetization ) that strongly affects electron transport may build up in the nonmagnetic island . in this article , we discuss the transport characteristics of metallic sv - sets in the coulomb blockade regime , allowing for arbitrary , noncollinear magnetization directions .
in particular , we examine the influence of exchange effects through f@xmath0n tunnel contacts on the spin accumulation in the center island , presenting a more complete discussion compared to that in ref . .
we argue that two separate exchange effects have to be taken into account . on one hand
, there is the non - local interface exchange , let us call it x1 in the following . in scattering theory for non - interacting systems
it is described by the imaginary part of the spin - mixing conductance,@xcite while in the context of current - induced magnetization dynamics x1 acts as an effective field.@xcite such an effective field has been found experimentally to strongly affect the transport dynamics in spin valves with mgo tunnel junctions.@xcite this effect has recently also been involved to explain magnetoresistance effects in carbon nanotube spin valves@xcite and called spin - dependent interface phase shifts.@xcite the second exchange term ( x2 ) is an interaction - dependent exchange effect due to virtual tunneling processes that is absent in non - interacting systems .
it has been considered for islands in the electric quantum limit , in which transport is carried by a single quantized level only.@xcite the x2 effect is potentially attractive for quantum information processing , since it allows to switch on and off effective magnetic fields in arbitrary directions just by a gate electric potential .
we compute here x2 for a metallic island in which size quantization is not important .
we find that both exchange effects are of comparable magnitude and affect the transport properties in a characteristic way , but can be separated in principle by employing the gate dependence of x2 .
the paper is organized as follows . in sec .
[ model ] we introduce the model system for the sv - set . in sec .
[ exchange ] the two relevant types of exchange processes are discussed .
charge and spin transfer rates are determined in sec .
[ transport ] . finally , we present results for the transport characteristics as a function of magnetic configuration , gate voltage and applied magnetic field in sec . [ results ] .
an sv - set ( see fig . [ schematic]a ) is composed of a small metallic cluster in contact with two large ferromagnetic electron reservoirs in thermal equilibrium characterized by magnetization directions @xmath1 and @xmath2 with @xmath3 ( for @xmath4 ) , where @xmath5 and @xmath6 . .
( b ) the magnetization directions in the leads define an angle @xmath7.,width=324 ] the f@xmath8n contacts are tunneling barriers with conductances that depend on the electron spin , @xmath9 for the majority and @xmath10 for the minority spin in the ferromagnet .
the total conductance for contact @xmath11 is then given by @xmath12 and the contact polarization is defined as @xmath13 .
the resistances @xmath14 are taken to be much larger than the resistance quantum @xmath15 , which , at low enough temperatures and bias voltages , allows us to study the blockade of transport by the coulomb interaction .
the electron tunneling rates are governed by the change of electrostatic energy of the whole circuit upon transfer of an electron .
the capacitances of the junctions @xmath16 determine the charging energy of the island .
we limit our considerations to islands formed by metallic clusters for which the thermal energy ( @xmath17 ) is much larger than the average single - particle energy separation ( reciprocal density of states ) @xmath18 , but much smaller than the single - electron charging energy .
therefore , many levels on the island participate in the transport and we may treat the electronic spectrum as continuous . for a gold cluster with a diameter of 10 @xmath19 , @xmath20 approximately corresponds to a temperature of 2 k. the kondo physics of quantum dots connected to ferromagnetic leads@xcite is suppressed in this regime .
since the currents flowing into and out of the cluster are spin - polarized , the island may become magnetized .
the number of unpaired spins on the island is limited by spin - flip scattering , which we parametrize by a spin - flip relaxation time @xmath21 .
there is evidence from several experiments that the spin - flip times in metallic nanoparticles can be much longer than in bulk systems , which implies that the effects of a spin accumulation on the island should be taken into account.@xcite for later convenience we introduce the spin - flip conductance parameter @xmath22 .
we assume that the energy relaxation on the island is fast compared to the electron dwell time .
the total hamiltonian for the sv - set is @xmath23 where @xmath24 is the hamiltonian for the normal metal cluster in the orthodox model for coulomb blockade @xmath25 here @xmath26 is a creation operator for an electron state with orbital index @xmath27 and spin @xmath28 where the @xmath29-axis is chosen as spin quantization axis .
the hamiltonian includes an electrostatic interaction energy which depends on the junction capacitances @xmath16 , the gate voltage @xmath30 and the excess number of electrons on the island @xmath31 .
the gate voltage shifts the potential and induces a charge @xmath32 .
we assume that the gate capacitance @xmath33 , and in the following we use @xmath34 .
the energy levels in the two ferromagnetic leads ( denoted by @xmath4 ) are spin - dependent : @xmath35 the operators @xmath36 create electrons with spin @xmath37 in the spin - quantization axis along @xmath38 .
it is convenient to introduce annihilation operators @xmath39 for electrons in the normal metal defined for a quantization axis in the direction of @xmath38 .
the relation between operators in the two bases is then @xmath40 , expressed in terms of the spin @xmath41 rotation matrix @xmath42{cc}\cos \theta_{\alpha}/2 & \sin \theta_{\alpha}/2\\ -\sin \theta_{\alpha}/2 & \cos \theta_{\alpha}/2 \end{array } \right ) .\ ] ] then , for each contact , a tunneling hamiltonian @xmath43 describes the coupling to the island .
the tunneling coefficients are assumed to not significantly depend on energy on the scale of the charging energy .
we discuss the exchange contribution , represented by the hamiltonian @xmath44 , in the next section .
here we discuss two different exchange effects that affect the electrons in the normal metal island attached to magnetic contacts .
these two flavors arise when the transport properties for an sv - set are determined to lowest order in the tunnelling probabilities .
the non - local exchange coupling between ferromagnetic films through a normal metal spacer is an important effect that determines the ground state of magnetic multilayers ( see ref . for a recent review ) .
electrons in a normal metal that are reflected at a contact to a ferromagnet , pick up a phase depending on the electron spin relative to the magnetization direction . in sufficiently clean and narrow f@xmath0n@xmath0f structures ,
quantum well states are formed in n whose energy depend on the magnetic configuration through the spin - dependent phase . by a rotation of the magnetization directions
the energy spectrum and fermi energy varies , causing the ground state energy to depend on the relative angle @xmath7 . in metallic multilayers with a suitable spacer thickness , this can lead to an antiparallel ground state , which displays the celebrated giant magnetoresistance when the magnetizations are forced into a parallel direction by a magnetic field . even when the ground state energies are not significantly affected by the exchange coupling , configuration - dependent quantized states can still be observed in transport .
this has been shown for high - quality planar tunnel junctions@xcite as well as spin valves in which the node is formed by single carbon nanotubes with a quantized energy spectrum.@xcite in ref .
the effect of interfacial phase shifts on the magnetoresistance of ballistic quantum wires between ferromagnetic leads was calculated .
the spin - dependent phase shifts give rise to a slightly different quantization condition , which can spin - split the energy levels .
since we are here interested in classical islands with a continuous electron spectrum , we calculate energy shifts for a semiclassical island using the bohr - sommerfeld quantization rule in appendix [ appshifts ] . here
we consider the limit of tunnel junctions between a normal metallic island and ferromagnetic electrodes .
the torques on the ferromagnets are then very small .
the exchange coupling does not significantly disturb the ferromagnets in this limit , but persists to affect transport .
the present study focusses on the charge transport properties in the limit of small tunneling matrix elements , thus from the outset excluding resonant tunneling , co - tunneling or kondo type physics .
the states on the island may be size quantized , _
i.e_. the energy level spacing exceeds the thermal energy ( quantum dot ) , or , in the opposite limit , better described by a semicontinuous density of states ( classical dot ) .
here we concentrate on the latter , _
i.e_. semiclassical , diffuse , or chaotic islands , for which it can be shown quite generally that equilibrium spin currents are suppressed.@xcite the state of the island is then characterized by a semiclassical charge and spin distribution function that has to be determined self - consistently as a function of the junction parameters and the applied voltages . for non - interacting systems ,
the spin and charge currents through an f@xmath0n interface are determined not only by the conventional conductances @xmath9 and @xmath10 introduced above , but also by the complex spin - mixing conductances @xmath45,@xcite which are discussed in sec .
[ transport ] .
the real part @xmath46 is the material parameter that is proportional to the spin - transfer torque.@xcite the imaginary part @xmath47 reflects the spin - dependent interface phase shifts and affects the magnetization and spin accumulation dynamics as an effective exchange magnetic field parallel to the magnetization direction.@xcite @xmath47 is relatively small for intermetallic interfaces,@xcite but is in general comparable in magnitude to the other conductance parameters.@xcite the non - local interface exchange has been discussed in similar terms for spin valves consisting of luttinger liquids with ferromagnetic contacts.@xcite the blocking of transport by the coulomb charging is usually described by fermi s golden rule ( see below ) , which employs a probability ( squared matrix elements ) and energy conservation . as long as the charging energy is much smaller than atomic energy scales ( like the fermi energy )
, the junction parameters such as the interface transparency and spin - mixing conductance are unaffected and the coulomb blockade is governed by the energy conservation criterion only .
this implies that the exchange effect can be described by the @xmath47 of the bare junction .
it remains to parametrize the exchange in the limit of the tunneling hamiltonian , _
i.e_. to lowest order in the interface transmission .
we show below that this is achieved by adding the following exchange term @xmath44 to the hamiltonian for the two leads : @xmath48 the energy shifts @xmath49 , see eq .
( [ shifts ] ) , are proportional to the inverse density of states , but they remain relevant for small level splitting because the dwell time is inversely proportional to the average energy level separation or inverse density of states @xmath50 .
this hamiltonian is an effective zeeman splitting caused by an exchange magnetic field in the direction of the magnetization , see sec .
[ spinacc ] .
the interface exchange term x1 is a property of the separate interfaces and they contribute independently .
the second type of exchange ( x2 ) felt by the spins on the island is a property of the entire device .
it originates from virtual tunneling processes , corresponding to single - electron transfer from and to the cluster . in the tunneling regime , this process can be treated and understood in terms of perturbation theory . in the absence of tunneling , the number of electrons on the island is a good quantum number .
the perturbation by the contact to the electrodes allows mixing in of states in which the number of electrons on the island is changed by unity , at the cost of the charging energy . in second order perturbation theory
this results in an energy gain represented by a sum over ( virtually ) excited states in which the coulomb energy appears in the denominator and the tunneling probability in the numerator . when the leads are non - magnetic , these virtual processes correspond to a quantum correction to the average charge on the central electrode.@xcite this effect depends strongly on the applied gate voltage .
when the unperturbed @xmath51 @xmath52 particle ground state is tuned in energy just above the @xmath53 particle state , the quantum correction will be large and positive ( negative ) . at the degeneracy point ,
perturbation theory breaks down , but the ensuing divergence can be controlled by taking into account finite temperatures .
when the tunneling probabilities to the ferromagnetic contacts are spin dependent , the deviations from the exact quantized charge on the island become spin - dependent , and therefore lead to a net excess of spins in the ground state that depends on the configuration of the contact magnetizations . for a symmetric spin valve it is easy to see that the island ground state magnetization due to these virtual processes x2 is maximal for parallel magnetizations and vanishes for antiparallel ones .
the additional exchange affects non - equilibrium electron transport , in contrast to higher - order so - called co - tunneling processes , to the same order as the in and out - tunneling processes . for a quantum - dot island with a single quantized level , knig and
martinek@xcite showed that in the case of _ non - collinear _ magnetizations the non - equilibrium spins on an island injected by a finite source - drain voltage are dephased by precessing around the effective exchange field .
this effect was also discussed for few - level quantum dots.@xcite since x1 discussed in sec .
[ nonlocal ] is a material constant , the gate voltage dependence of x2 provides a handle for an experimental discrimination of the two effects .
we derive an expression for the effective x2 exchange field for a classical sv - set in sec .
[ spinacc ] .
we compute the transport characteristics of the sv - set in lowest - order perturbation theory@xcite for a diffusive or chaotic island in the sequential tunneling regime .
the rate equations lead to a probability distribution for the excess number of charges @xmath31 .
the excess spin accumulation @xmath54 contains a large number of spins and we are interested in its average value in the steady state that is found from the condition @xmath55 .
the operators for the excess number of electrons on the island and on the two leads are @xmath56 and @xmath57 respectively .
the unpaired spin angular momentum on the cluster is written as @xmath58 , where @xmath59 is the vector of pauli spin matrices .
it is convenient to introduce a vector chemical potential @xmath60 in the island , with size @xmath61 ( see also ref . ) , where @xmath62 is the density of states at the fermi energy .
we can take into account stoner enhancement intra - island exchange effects in terms of the static susceptibility @xmath63 and we may also write @xmath64 .
we denote the unit vector in the direction of the spin accumulation by @xmath65 .
the charge current is equal to the expectation values for the rate of change of @xmath31 . in terms of the tunneling hamiltonian @xmath66
the time - evolution is given by @xmath67 \nonumber \\ & = \frac{i}{\hbar}\sum_{\alpha kqs^{\prime}}t_{kqs^{\prime}}^{\alpha } a_{\alpha ks^{\prime}}^{\dagger}c_{\alpha qs^{\prime}}+h.c.,\end{aligned}\ ] ] we use the interaction representation , and write the total hamiltonian as @xmath68 to second order in @xmath69 we have @xmath70 \right \rangle _ { \circ},\ ] ] where @xmath71 denotes an expectation value with respect to hamiltonian @xmath72 .
the electrochemical potentials of the two reservoirs are @xmath73 and @xmath74 .
it is convenient to introduce grand canonical hamiltonians including the chemical potentials as@xcite @xmath75 the time dependence @xmath76 can be formulated in terms of the projection operators @xmath77 where @xmath78 is the unit matrix , by making use of the equality @xmath79 _ { s^{\prime } s^{\prime \prime}}c_{ps^{\prime \prime}}.\end{aligned}\ ] ] the leads and the island are supposed to be in thermal equilibrium , so that @xmath80 , with fermi - dirac distribution @xmath81 , where @xmath82 is the inverse temperature . using the expression for the matrix elements @xmath83 _ { s^{\prime } s^{\prime } } & = \frac{1}{2}\left ( 1+s^{\prime}s^{\prime \prime}\hat{s}\cdot \overrightarrow{m}_{\alpha}\right ) , \\
\text{with } s^{\prime},s^{\prime \prime } & \in \ { \uparrow,\downarrow \}=\{+,-\},\nonumber\end{aligned}\ ] ] the rate of change of the number of electrons on the island reads @xmath84,\nonumber\end{aligned}\ ] ] where @xmath85 and @xmath86 .
the relation between the up and down spin conductances ( @xmath87 and @xmath10 ) and the tunneling coefficients is @xmath88 , where @xmath89 is the value of @xmath90 at the fermi energy averaged over all the modes .
@xmath91 is the spin - dependent density of states in ferromagnet @xmath11 . in the low - bias regime
considered here we can linearize eq .
( [ dndt ] ) in @xmath92 and @xmath93 .
the resulting expression for the rate for electron tunneling through contact @xmath11 , increasing the excess number of electrons @xmath31 from @xmath94 to @xmath95 , is denoted by @xmath96 . the analogous rate for removing one electron when @xmath31 is @xmath95 is @xmath97 .
explicitly , we find @xmath98 now that we have determined the tunneling rates we can write down the master equation for electron transport in the orthodox model .
we consider a regime in which @xmath99 , and restrict ourselves to a gate voltage range for which the excess number of electrons @xmath31 alternates between @xmath94 and @xmath100 knowing that the results will periodically repeat with this period .
the center of the coulomb oscillation for transitions between @xmath101 and @xmath95 electrons is at @xmath102 .
the steady state on the island is characterized by a constant spin accumulation ( to be determined below ) and the probabilities @xmath103 and @xmath104 that there are @xmath94 or @xmath95 excess electrons .
we have @xmath105 .
the rate equation for the probabilities is @xmath106 from the condition of detailed balance , @xmath107 , we find @xmath108 the expression for the conductance of the sv - set as a function of the spin accumulation can now be calculated and reads @xmath109 .
\label{cond}\ ] ] the steady - state spin accumulation is found by setting the total rate of change of @xmath54 to zero .
there are several contributions to the dynamics of the spin accumulation : @xmath110 the first two terms are due to the tunneling processes , the remaining ones to exchange , external magnetic fields and spin flip .
we start from @xmath111 , \ ] ] with an expectation value that to second order in @xmath69 reads @xmath112 \right \rangle .\ ] ] the spin current ( rate of change of the spin angular momentum ) due to tunneling when @xmath113 excess electrons are on the island reads ( _ cf_. eq .
( [ dndt ] ) ) @xmath114 \nonumber \\ & + \frac{\hbar}{4\pi e^{2}}\sum_{\alpha s^{\prime \prime}}p_{\alpha}g_{\alpha } s^{\prime \prime}\left ( \overrightarrow{m}_{\alpha}\times \hat{s}\right ) \times \nonumber \\ & \left ( \int d\epsilon_{1}\int^{\prime}d\epsilon_{2}\frac{f\left ( \epsilon_{1}\right ) \left ( 1-f\left ( \epsilon_{2}\right ) \right ) } { \left ( \epsilon_{2}-\epsilon_{1}+e_{m-1}-e_{m}+\mu_{cf\alpha}-s^{\prime \prime}\frac{\delta \mu}{2}\right ) } \right .
\nonumber \\ & \left .
-\int d\epsilon_{1}\int^{\prime}d\epsilon_{2}\frac{f\left ( \epsilon_{2}\right ) \left ( 1-f\left ( \epsilon_{1}\right ) \right ) } { \left ( \epsilon_{2}-\epsilon_{1}+e_{m}-e_{m+1}+\mu_{cf\alpha}-s^{\prime \prime}\frac{\delta \mu}{2}\right ) } \right ) , \end{aligned}\ ] ] where the prime denotes a principal value integral .
here we used the relation : @xmath115 _ { s^{\prime}s^{\prime } } = \frac{1}{2}s^{\prime \prime}\hat{s } + \frac{1}{2}s^{\prime}\overrightarrow{m}_{\alpha}+\frac{1}{2}is^{\prime } s^{\prime \prime}\left ( \overrightarrow{m}_{\alpha}\times \hat{s}\right ) , \\
\text{with } s^{\prime},s^{\prime \prime } & \in \
{ \uparrow,\downarrow \}=\{+,-\}.\nonumber\end{aligned}\ ] ] to first order in the small induced energy shifts the exchange hamiltonian @xmath44 modifies the unpaired spins as @xmath116 , \ ] ] which results in a precession : @xmath117 where @xmath118 is the number of transport channels in the normal metal and the energy shifts @xmath119 are found in eq . ( [ shifts ] ) .
the conductance parameters of an f@xmath0n contact are@xcite @xmath120 here @xmath121 and @xmath113 denote the transport channels in the normal metal and @xmath122 and @xmath123 are the corresponding spin - dependent reflection coefficients .
the contact conductances for spin - up and spin - down electrons are @xmath124 and @xmath10 and the mixing conductance @xmath125 governs the transverse spin currents that are absorbed and reflected by the ferromagnet @xmath11 . the current polarized normal to the magnetization but in the plane of @xmath54 and @xmath126 is proportional to @xmath46 and describes the spin - transfer to the magnet , thereby dissipating the spin accumulation . in the case of tunnel junctions @xmath127 .
the out - of - the @xmath54,@xmath128 plane spin current is caused by reflection processes that make spins precess around @xmath128 and is proportional to @xmath129 .
this mixing conductance has been evaluated from first principles for various contact materials and is small for intermetallic interfaces because positive and negative contributions in the space spanned by the transport channels average out.@xcite however , there is no general reason that @xmath129 should be smaller than @xmath130 or @xmath131 it is known to be quite large for the fe@xmath0inas interface@xcite and found to be very significant for the magnetization dynamics of mgo magnetic tunnel junctions.@xcite for a simple model barrier discussed in appendix [ imgmodel ] , we find the value @xmath132 . using the relation between the reflection phases and the energy shifts as derived in eq .
( [ shifts ] ) , we can rewrite the contribution given in eq .
( [ precession ] ) in terms of the imaginary part of the mixing conductance as ( cf .
@xmath133 the spin accumulation can also be affected by a magnetic field @xmath134 , which can either be externally applied , a stray field from the ferromagnets , or an internal anisotropy field . the spin accumulation induced by this magnetic field
can safely be neglected , but the induced precession of the spin accumulation is relevant , and is given by @xmath135 finally , spin - flip relaxation in the normal metal is taken into account by spin - accumulation decay with a spin - flip relaxation time @xmath136 @xmath137 combining the terms in eq .
( [ dsdt ] ) , the spin accumulation should fulfil the stationary state condition:@xmath138 \nonumber \\ & + \frac{g\mu_{b}}{\hbar}\vec{b}_{\mathrm{eff}}\times \left \langle \vec { s}\right \rangle -\frac{\left \langle \vec{s}\right \rangle } { \tau_{\mathrm{sf}}}=0 , \label{squasieq}\ ] ] where @xmath139 the total effective magnetic field @xmath140 consists of the external magnetic field and contributions from the exchange effects x1 and x2 , and reads@xmath141 @xmath142 . \nonumber\end{aligned}\ ] ] here we introduced@xcite @xmath143 , \nonumber\end{aligned}\ ] ] where @xmath144 is the digamma function . in appendix [ virttun ]
we discuss the derivation of the expression for @xmath145 in more detail and comment on the differences compared to the case of a single - level quantum dot .
fig . [ exch ] shows the magnitude of the total effective magnetic field @xmath140 as a function of gate voltage ( solid line ) for a symmetric spin valve with parallel magnetizations ( it vanishes for the antiparallel configuration ) and a polarization @xmath146 .
the x1 term is a constant that does not depend on gate voltage ( dotted line ) .
@xmath147 vanishes when @xmath32 equals @xmath148 and @xmath149 . at these points , contributions from incoming and outgoing electrons
cancel each other ( see appendix [ virttun ] ) .
the curve repeats as a function of gate voltage with period @xmath150 . as a function of gate voltage ( solid line ) for a spin valve in the parallel configuration .
the parameters are @xmath151 @xmath152 , @xmath153 and @xmath154 .
the imaginary part of the mixing conductance gives a constant offset ( dotted ) .
the dot - dashed / dashed curves show the effective field for zero and one excess electron on the island .
, width=324 ] the spin accumulation on the island found from eq .
( [ squasieq ] ) tends to suppress the current through the system .
spin - flip and exchange effects that dissipate or dephase the spin - accumulation therefore increase the conductance . as a reference we list here the conductance @xmath155 for a spin valve _ without interaction _ , with equal conductance parameters for the left and the right tunneling barrier @xmath151 @xmath156 : @xmath157 the final result for the symmetric spin - valve _ with interaction _ can be obtained simply from this expression by the substitutions : @xmath158 for nonmagnetic contacts ( @xmath159 ) this result reduces to the known expression for normal metal single - electron transistors.@xcite for a symmetric sv - set with ratio @xmath160 ( solid ) , @xmath161 ( dashed ) and @xmath162 ( dotted ) in units of @xmath163 .
the polarization @xmath164 is @xmath165 .
( b ) conductance as a function of the angle for the same parameters as in a ) , with @xmath32 fixed to 0.5 .
( c ) conductance as a function of @xmath7 with @xmath166 .
results are shown for @xmath32 equal to @xmath167 ( dashed)@xmath168 ( solid ) and @xmath169 ( dotted).,width=324 ] as shown in fig .
[ cosc](a ) , changing the relative strengths of x1 and x2 , or , since the x2 contribution is proportional to the polarization of the leads , @xmath170 qualitatively modifies the current profile of the coulomb oscillations .
the constant offset given by @xmath171 skews the exchange field around @xmath102 , causing asymmetric conductance curves . when the offset starts to dominate the symmetry
gets restored .
the x2 contribution vanishes when the coulomb blockade is lifted @xmath172 , so the angular dependence of the conductance for different values of @xmath173 in fig .
[ cosc](b ) reflects only the x1 effect .
the curve is a simple cosine for @xmath174 , but is sharpened for larger @xmath175 because of the dephasing of the spin accumulation occurring for noncollinear angles . in fig .
[ cosc](c ) @xmath173 is fixed to @xmath161 and curves are plotted for different values of the gate voltage . it can be seen that the angular dependence differs because the x2 depends on @xmath30 in an asymmetric way around @xmath102 . as can be seen in fig .
[ highp ] , the shape of the coulomb oscillation can develop minima when the polarization is high and the magnetizations are nearly antiparallel . at the values of gate voltage where the x1 and x2 exchange effects cancel , the spin accumulation is not dephased and the conductance is suppressed . and @xmath176 .
curves are shown for @xmath146 ( solid ) , @xmath177 ( dashed ) @xmath178 ( dotted).,width=324 ] fig .
[ mu ] shows results for the conductance and spin accumulation as a function of applied magnetic field in the @xmath179 ( solid line ) , @xmath180 ( dashed ) and @xmath29 ( dotted ) directions .
the spin valve is again symmetric with @xmath146 and @xmath181 .
the angle @xmath7 is fixed to @xmath182 and @xmath102 .
the conductance then depends only on the @xmath179-component of the spin accumulation ( see eq .
( [ cond ] ) ) . without applied magnetic field , the spin accumulation has components in the @xmath179 and @xmath180 directions , while @xmath140 is in the @xmath180-direction .
the results can be understood in terms of the dephasing of the spin accumulation by the magnetic - field induced precession that , for sufficiently large and non - collinear magnetic fields , quenches the spin accumulation .
this hanle effect is responsible for the conductance minimum at negative applied magnetic field in the @xmath180 direction . in [ mu](c ) only two curves are visible because the curves for magnetic fields in the @xmath179 and @xmath180 direction overlap .
( solid ) , @xmath180 ( dashed ) or @xmath29 ( dotted ) direction in units of @xmath163 . the sv - set has symmetric junction parameters , with polarizations @xmath146 and @xmath181 . the magnetizations are fixed to @xmath183 , yielding an angle @xmath184 and @xmath102 .
( b , c , d ) the @xmath179 , @xmath180 and @xmath29 components of the spin accumulation for the same parameters . the curves in ( c ) for magnetic fields in the @xmath179 and @xmath180 directions overlap.,width=324 ]
we studied the transport properties of single - electron spin valve transistors as a function of the magnetization configurations in the orthodox model of the coulomb blockade .
two types of exchange effects between the spin accumulation on the island and the lead magnetizations play a role : a nonlocal interface exchange effect ( x1 ) and exchange due to virtual tunneling processes ( x2 ) . for metallic dots
these two effects are found to be of comparable magnitude .
we predict that a line shape analysis of the coulomb oscillation peaks should help to experimentally disentangle the two contributions .
additional information can be obtained by the hanle effect .
we would like to thank j. knig , j. martinek , m. braun and y.v .
nazarov for useful discussions .
this research was supported by the nwo and by the dfg via the sfb 689 , and supported in part by the national science foundation under grant no .
phy99 - 07949 .
let us consider a normal metal island in contact to a ferromagnet by a tunnel barrier ( see fig .
[ f_shifts ] ) without coulomb interaction .
we wish to calculate the spin - dependent shifts of the energy levels due to the presence of the f@xmath0n contact . in ref .
an analogous calculation was done for a ballistic one - dimensional quantum wire .
here we consider an island in the quasi - classical regime , _
i.e. _ the de broglie wavelength is much smaller than the size of the island .
the bohr - sommerfeld quantization rule @xcite @xmath185 can be used to find the energy shifts , where @xmath186 is the classical momentum for an electron in mode @xmath113 , and @xmath121 is an integer . the integral is over a whole period of the classical motion in the quasi - classical region .
the total phase shift due to the reflections at the turning points is @xmath187 , where @xmath188 is the spin - independent phase shift picked up during the reflections from the boundaries for an isolated island without contact to the ferromagnet .
the small spin - dependent phase shift @xmath189 arises from the weak coupling to the ferromagnet .
the phase shifts have to be computed quantum mechanically via the spin - dependent reflection coefficients @xmath190 for mode @xmath113 at an interface that is assumed to be specular ( see also appendix [ imgmodel ] ) . from eq .
( [ bohr - sommerfeld ] ) , we see that increasing the quantum number @xmath121 by one corresponds to introducing an extra phase period that increases the kinetic energy by @xmath191 where @xmath62 is the density of states of the island and @xmath118 is the number of modes .
the energy shift for an electron in mode @xmath113 is therefore , to linear order in @xmath192 , @xmath193 the effect of the interface on the island states can be taken into account by introducing an effective hamiltonian as in eq .
( [ hex ] ) . in the case of a spin - independent tunneling barrier to a ferromagnet ,
the spin - splitting of the energy levels is small , of the same order as the transmission probability ( see app .
[ imgmodel ] ) .
here we evaluate the spin - mixing conductance @xmath194 for a model barrier , giving more details of the results of ref . .
we consider a smooth rectangular barrier between a normal metal and a stoner - model ferromagnet .
the solution of the schrdinger equation for spin @xmath37 in the normal metal , @xmath195 can be used to determine the reflection coefficients @xmath190 for each mode @xmath113 .
it reads @xmath196 where @xmath197 is the transverse wave function and @xmath198 is the longitudinal wave number for mode @xmath113 in the normal metal . in terms of the wave numbers in the normal metal @xmath198 , barrier @xmath199 and ferromagnet @xmath200 for a given energy ,
the reflection coefficient for mode @xmath113 at the barrier reads @xmath201 where @xmath202 is the barrier thickness and @xmath203 for a tunneling barrier , @xmath199 is imaginary and the spin dependent correction to the reflection coefficient is exponentially small in the barrier thickness . for a numerical estimate we use a fermi energy in the normal metal of 2.6 ev ,
a barrier height of 3 ev and barrier thickness of @xmath204 .
the fermi momenta in the ferromagnet are taken to be @xmath205 and @xmath206 ( characteristic for fe , see ref . ) .
for the spin - mixing conductance @xmath207 , eq .
( [ gmix ] ) , we find that @xmath132 for this choice of parameters .
the effective field due the interface exchange effect is therefore not negligible compared to the conductance parameters .
more realistic electronic structure calculations@xcite should be carried out to obtain better estimates .
here we present more details concerning the derivation of eq .
( [ bx2 ] ) for the effective exchange field x2 in classical sv - set s , that complement the derivation in refs . for single - level quantum dots .
since the model is periodic in the gate voltage with period @xmath150 , we restrict our discussion to the range @xmath208 . from eq .
( [ dsdt ] ) , the contributions from virtual tunneling processes to the rate of change of @xmath54 then read : @xmath209 using the spin currents from eq .
( [ spincurr ] ) , we obtain , _ e.g. _ : @xmath210 the first term in brackets describes virtual processes in which an electron tunnels out of the island , and the second term corresponds to incoming electrons .
the expressions for the energy differences are given by @xmath211 because of the periodicity in the gate voltage @xmath212 we can now rewrite eq .
( [ dsdtx2 ] ) in terms of the function @xmath213 , defined in eq .
( [ defeta ] ) .
the probabilities @xmath103 and @xmath104 are taken from eq .
( [ pzero ] ) .
after linearization in @xmath214 and @xmath92 , we arrive at the expression eq .
( [ bx2 ] ) .
we note the differences with the results for single - level quantum dots;@xcite our expression includes an additional integral over the island states . for a single - level quantum dot
, x2 is active only when exactly one electron resides on the dot , since there is no unpaired spin in an empty or doubly occupied dot .
in contrast , a net spin accumulation can reside on our classical dot for all numbers of electrons .
the effective magnetic field is a sum weighted with the probabilities for @xmath215 and @xmath216 electrons on the dot , which leads to a partial cancellation of the contributions for different @xmath31 , as is shown in fig .
[ exch ] .
j. knig and j. martinek , phys
* 90 * , 166602 ( 2003 ) ; m. braun , j. knig , and j. martinek , phys .
b * 70 * , 195345 ( 2004 ) ; j. knig , j. martinek , j. barna , and g. schn , in _ cfn lectures on functional nanostructures _ , eds .
et al . _ , lecture notes in physics * 658 * ( springer ) , pp .
145 - 164 , ( 2005 ) ; m. braun , j. knig , and j. martinek , superl .
* 27 * , 333 ( 2005 ) .
a. brataas , yu . v. nazarov , and g. e. w. bauer , phys .
lett . * 84 * , 2481 ( 2000 ) ; a. brataas , y. v. nazarov , and g. e. w. bauer , eur .
j. b * 22 * , 99 ( 2001 ) ; a. brataas , g. e. w. bauer and p. j. kelly , phys . rep . * 427 * , 157 ( 2006 ) .
m. d. stiles , nanomagnetism : ultrathin films , multilayers and nanostructures ( contemporary concepts of condensed matter science , vol 1 ) , eds .
d. mills and j. a. c. bland , new york : elsevier , 2006 , pp 51 - 77 . | we study electron transport through single - electron spin - valve transistors in the presence of non - local exchange between the ferromagnetic leads and the central normal - metal island .
the coulomb interaction is described with the orthodox model for coulomb blockade and we allow for noncollinear lead magnetization directions .
two distinct exchange mechanisms that have been discussed in the literature are shown to be of comparable strength and are taken into account on equal footing .
we present results for the linear conductance as a function of gate voltage and magnetic configuration , and discuss the response of the system to applied magnetic fields . |
gauge / string duality@xcite provides us with new insights into gauge theories in a strong coupling regime .
there have been substantial progresses in studying strong coupling gauge theories by using such gauge / string duality .
a few years ago , polchinski and strassler@xcite studied the deep inelastic scattering on hadrons by using gauge / string duality , in which the spinless hadron and spin-@xmath4 hadron correspond to supergravity modes of dilaton and dilatino , respectively .
the usual structure functions @xmath5 and @xmath6 are calculated for both spinless and spin-@xmath4 hadrons when bjorken-@xmath2 is finite ( @xmath3 ) where supergravity approximation is valid .
furthermore , they also investigated the case at small-@xmath2 where the pomeron contribution with a trajectory of @xmath7 was found . in their work , since an infrared cutoff @xmath8 is introduced in order to generate confinement , the model is then called the hard wall model .
there are also some earlier studies@xcite on high energy scattering in gauge / string duality .
there have been a lot of further developments along this direction@xcite . a saturation picture based on deep inelastic scattering in ads
/ cft is developed@xcite afterwards and recently reviewed in ref .
in addition , the deep inelastic scattering off the finite temperature plasma in gauge / string duality is studied in refs .
@xcite .
recently , the above deep inelastic scattering ( dis ) calculation in gauge / gravity duality has been extended to the case of polarized dis off the charged dilatino in ref .
@xcite and obtained the spin - dependent structure functions @xmath9 and @xmath10 for a spin-@xmath4 hadron at finite @xmath2 . in ref .
@xcite , the small @xmath2 behavior of such spin - dependent structure functions at large coupling limit was analyzed .
furthermore , the nonforward compton scattering has been also investigated in refs .
other recent relevant work can be found in refs .
@xcite . in ref .
@xcite and ref .
@xcite , both the unpolarized and polarized structure functions when @xmath2 is finite ( @xmath3 ) are calculated with only minimal interaction . it is found that all the structure functions are power suppressed and vanish in the large @xmath11 limit . so it is worthwhile to investigate how the structure functions look from other possible interactions . also , since the minimal interaction is proportional to the charge that the dilatino carries , such interaction does not contribute to the structure functions of the neutral dilatino .
one of our main objects in this paper is to extend the calculation of the structure functions of the charged spin-@xmath4 hadron just call it `` proton''in polarized dis in ref .
@xcite to the neutral spin-@xmath4 hadron just call it `` neutron''through introducing a new interaction pauli interaction term in @xmath12 space . besides , in ref .
@xcite , the authors have obtained an interesting burkhardt - cottingham - like sum rule @xmath13 , which is completely independent of @xmath14 and @xmath11 , hence another main object of our paper is to investigate whether such sum rule still holds from other possible interaction terms such as pauli interaction here .
since there has been extensive study @xcite of the elastic form factors , we will not take this subject into account in our present work . for simplicity and also consistency with the previous work
, we still work in the hard wall model .
this paper is organized as follows . in sec.[deft ] , we recall the definitions for various structure functions as well as kinematic variables . in sec.[hard ]
, we calculate the structure functions of the `` neutron '' at finite @xmath2 from the pauli interaction term .
section [ disc ] is devoted to the discussions and comments on these structure functions and compare them with the structure functions of the charged dilatino which have been calculated in the previous work where only minimal interaction was considered .
we summarize our results in sec.[conc ] .
finally , in the appendix we will illustrate how a pauli interaction term in 5d can arise naturally from 6d fermion - graviton coupling through the usual kaluza - klein reduction .
.,width=377 ] deep inelastic scattering has played an important role in the history of investigating the internal structure of hadrons .
it is the study of lepton - hadron scattering in the limit that @xmath2 is fixed , and @xmath15 .
the basic diagram for such process is illustrated schematically in fig .
the structure of the hadron can be completely characterized by the hadronic tensor @xmath16 , which is defined as @xmath17 | p , s \rangle \ , , \label{dis3}\ ] ] with @xmath18 being the incident current . in our present work
, we will specify the hadron as the spin-@xmath4 hadron .
the hadronic tensor @xmath19 can be split as @xmath20 according to lorentz and @xmath21 invariance , the symmetrical and antisymmetrical parts can be expressed in terms of 8 independent structure functions as@xcite @xmath22\nonumber\\ & & - \frac{1}{p{\cdot}q}\left ( p_\mu - \frac{p{\cdot}q}{q^2 } \ , q_\mu \right)\left ( p_\nu - \frac{p{\cdot}q}{q^2 } \ , q_\nu \right ) \left[f_2(x , q^2)+\frac{m s\cdot q}{p\cdot q}g_4(x , q^2)\right]\nonumber\\ & & -\frac{m}{2p\cdot q}\left[\left ( p_\mu - \frac{p{\cdot}q } { q^2}q_\mu\right)\left(s_\nu - \frac{s{\cdot}q}{p{\cdot}q } \ , p_\nu\right ) + \left ( p_\nu - \frac{p{\cdot}q } { q^2}q_\nu\right)\left(s_\mu - \frac{s{\cdot}q}{p{\cdot}q } \ , p_\mu\right)\right]g_3(x , q^2),\nonumber\\ w_{\mu\nu}^\mathrm{(a ) } & = & -\frac{m \ , \varepsilon_{\mu\nu\rho\sigma } \ , q^\rho}{p{\cdot}q } \left\ { s^\sigma \ , g_1(x , q^2 ) + \left [ s^\sigma - \frac{s{\cdot}q}{p{\cdot}q } \ ,
p^\sigma \right ] g_2(x , q^2 ) \right\ } -\frac{\varepsilon_{\mu\nu\rho\sigma}q^\rho p^\sigma}{2p{\cdot}q } f_3(x , q^2 ) , \label{wmunu}\end{aligned}\ ] ] where @xmath23 is the mass of the hadron , @xmath24 is its polarization vector , @xmath25 is the momentum carried by the current and @xmath26 is the initial momentum of the hadron ( see fig .
[ dis ] ) . in deep inelastic scattering ,
we define the kinematic variables as the following @xmath27 the mass of the intermediate state after the scattering is defined as @xmath28 .
all the structure functions are only functions of @xmath2 and @xmath11
. we will use the most plus metric throughout this paper instead of the usual most minus metric in particle physics , so there are some sign changes in our definitions comparing to the usual definitions in @xcite .
according to the conjecture of ads / cft , at large t hooft parameter , the gauge theories have a dual string description , which can make accurate analytic calculations possible . for the @xmath29 dimensional conformal gauge theories , the dual string theory lives in the space @xmath30 .
the metric in @xmath30 space can be written as @xmath31 where @xmath32 are identified with the space - time coordinates in the gauge theory and @xmath33 denotes a five - dimensional compact space .
we will specify such compact space as @xmath34 in our work .
the conformal invariance can be broken through introducing a sharp cut - off @xmath35 , leading to the mass gap of hadrons .
this simple model is the so - called hard wall model . following the formalism proposed by polchinski and strassler in ref .
@xcite , the incident current is chosen to be the @xmath36-current which couples to the hadron as an isometry of @xmath34 .
according to the ads / cft correspondence , the current excites a nonnormalizable mode of a kaluza - klein gauge field at the minkowski boundary of the @xmath37 space @xmath38 where @xmath39 denotes a killing vector on @xmath34 with @xmath40 being the angular coordinates on @xmath34 .
@xmath41 is the external potential in the gauge theory corresponding to the operator insertion @xmath42 on the boundary of the fifth dimension of the @xmath43 space with the boundary condition @xmath44 this gauge field fluctuation @xmath45 can be viewed as a vector boson field which couples to the @xmath36-current @xmath46 on the minkowski boundary , and then propagates into the bulk as a gravitational wave , and eventually interacts with the supergravity modes of the dilatino or dilaton .
the gauge field satisfies maxwell s equation in the bulk , @xmath47 , which can be explicitly written as @xmath48 = 0,\]]where @xmath49 , @xmath50 , ... are indices on @xmath0 . in the lorentz - like gauge @xmath51the maxwell equation can be written as @xmath52the solutions to the above equations with the proper boundary conditions @xmath53 are given by , we will see it leads to unreasonable constraint @xmath54 on the boundary condition ( [ boundary - a ] ) . ]
@xmath55 , \nonumber \\
a_{z } & = & in\cdot qe^{iq\cdot y}z\left[\text{k}_{0}(qz)-c\text{i}_{0}(qz)\right],\end{aligned}\ ] ] where @xmath56 in the large @xmath11 regime , we can just simply identify @xmath57 as @xmath58 .
however , in the small @xmath11 regime , such a term contributes as much as the others .
this is just the reason the form factor gives rise to logarithmic divergent charge radii for the charged dilatino in the work @xcite by gao and xiao .
spin-@xmath4 hadrons corresponds to supergravity modes of the dilatino . in the conformal region
the dilatino field can be written as @xmath59where @xmath60 is an @xmath61 spinor on @xmath62 and @xmath63 is an @xmath64 spinor on @xmath65 .
the wave - function @xmath66 satisfies a five - dimensional dirac equation in @xmath62 space .
let us first review how to derive this five - dimensional dirac equation in the following . a convenient choice of vielbein is given by @xmath67where @xmath68 .
the levi - civita connection is given by @xmath69here we use @xmath70 to denote indices in flat space , and @xmath71 to denote indices in curved space ( @xmath62 space ) .
in addition , the greek indices @xmath72 are defined in minkowski space . from the metric
, one knows @xmath73it is straightforward to work out the levi - civita connection in @xmath62 space @xmath74from vielbein and levi - civita connection , we can have the spin connection @xmath75the only nonvanishing spin connections are@xmath76using above results , the operator @xmath77 can be cast into @xmath78with @xmath79 $ ] . the free dilatino field in @xmath62 space satisfies the dirac equation @xmath80 its normalizable solution is given by @xcite , @xmath81 u_{\sigma } \ ] ] where @xmath82 for the initial hadron , by assuming @xmath83 in the interaction region and expanding the bessel functions up to linear term in @xmath23 , one gets @xmath84.\ ] ] @xmath85.\ ] ] for the intermediate hadron , @xmath86 and @xmath87 u_{x\sigma } \ .\ ] ] @xmath88 \ .\ ] ] in ref .
@xcite and ref .
@xcite , the interaction between the kaluza - klein gauge field and charged dilatino is given by the minimal coupling @xmath89however , since we are now interested in the effect of the other possible interactions and the neutral dilatino in such @xmath90 current , where the above minimal interaction will not contribute , we need introduce a new interaction term the pauli interaction term such that @xcite @xmath91\psi .\end{aligned}\ ] ] actually such term can be derived very naturally from kaluza - klein reduction of higher dimensional fermion - graviton coupling .
we put such derivation into the final appendix in our present work . with the pauli interaction action at hand , following the same line in ref .
@xcite , we can compute the matrix element @xmath92p_{-}u_{i\sigma } + c_2 q\overline{u}_{f\sigma^{\prime } } [ \,q\hspace{-6pt}\slash , \gamma^\mu]p_{+}u_{i\sigma } \right.\nonumber\\ & & \left.+c_3 \overline{u}_{f\sigma^{\prime } } ( q^\mu q\hspace{-6pt}\slash - q^2\gamma^\mu ) p_{-}u_{i\sigma } + c_4 \overline{u}_{f\sigma^{\prime } } ( q^\mu q\hspace{-6pt}\slash - q^2\gamma^\mu ) p_{+}u_{i\sigma } \right\},\end{aligned}\ ] ] and its complex conjugate @xmath93p_{+}u_{f\sigma^{\prime } } -c_2 q\overline{u}_{i\sigma } [ \,q\hspace{-6pt}\slash , \gamma^\mu]p_{-}u_{f\sigma^{\prime } } \right.\nonumber\\ & & \left.+c_3 \overline{u}_{i\sigma } ( q^\mu q\hspace{-6pt}\slash - q^2\gamma^\mu ) p_{-}u_{f\sigma^{\prime } } + c_4 \overline{u}_{i\sigma } ( q^\mu q\hspace{-6pt}\slash - q^2\gamma^\mu ) p_{+}u_{f\sigma^{\prime } } \right\}.\end{aligned}\ ] ] the coefficients @xmath94 are given by @xmath95\gamma(\tau-1),\\ c_2&=&c_0\int d z z^{\tau+1 } \ ,
k_1(qz ) j_{\tau-1}(m_x z)\nonumber\\ & = & c_0 2^\tau m_x^{\tau-1}(m_x^2+q^2)^{-\tau-1}q \gamma(\tau+1),\\ c_3&=&\frac{c_0 m}{2(\tau-1)}\int d z z^{\tau+2 } \ , k_0(qz ) j_{\tau-1}(m_x z)\nonumber\\ & = & c_0 2^\tau m_x^{\tau-1}(m_x^2+q^2)^{-\tau-2}m\tau\left[\tau q^2-m_x^2\right]\gamma(\tau-1),\\ c_4&=&c_0 \int d z z^{\tau+1 } \ , k_0(qz ) j_{\tau-2}(m_x z)\nonumber\\ & = & c_0 2^\tau m_x^{\tau-2}(m_x^2+q^2)^{-\tau-1}\left[q^2(\tau-1)-m_x^2\right]\gamma(\tau),\end{aligned}\ ] ] where @xmath96 and @xmath97 . in the above calculation
, we have relaxed the upper limit of integration from @xmath98 to @xmath99 .
now let us continue to calculate the hadronic tensor @xmath100 in large @xmath11 limit , we can make the approximation @xmath101 together with @xmath102 and @xmath103 , we can write the symmetric @xmath104 and antisymmetric @xmath105 as , respectively , @xmath106 ^ 2 \left[1+\frac{2m(s\cdot q)}{2 p\cdot q}\right]\left(\eta_{\mu\nu}-\frac{q_\mu q_\nu}{q^2}\right)\nonumber\\ & & -\frac{8q^6}{x } ( 4c_2 ^ 2+c_4 ^ 2)\left[1+\frac{m(s\cdot q)}{p\cdot q}\right ] \frac{1}{p\cdot q}\left(p_\mu-\frac{p\cdot q}{q^2}q_\mu\right)\left(p_\nu-\frac{p\cdot q}{q^2}q_\nu\right)\nonumber\\ & & -\frac{8q^6}{x}\left[(4c_2 ^ 2+c_4 ^ 2)+(c_2 c_3-c_1 c_4)\frac{2q}{x m}\right]\frac{m}{2p\cdot q}\nonumber\\ & & \times\left[\left(p_\mu-\frac{p\cdot q}{q^2}q_\mu\right)\left(s_\nu-\frac{s\cdot q}{p\cdot q}p_\nu\right ) + \left(p_\nu-\frac{p\cdot q}{q^2}q_\nu\right)\left(s_\mu-\frac{s\cdot q}{p\cdot q}p_\mu\right)\right]\end{aligned}\ ] ] @xmath107 ^ 2 \frac{\epsilon_{\mu\nu\alpha\beta}q^\alpha p^\beta}{2p\cdot q } -\frac{4q^6}{x}\left[2c_2\left(\frac{1-x}{x}\right)^{1/2}+ c_4\right]^2\frac{m\epsilon_{\mu\nu\alpha\beta}q^\alpha s^\beta}{p\cdot q } \nonumber\\ & & + \frac{2q^6}{x^2}\left[(4c_2 ^ 2+c_4 ^ 2)+2(4c_1 c_2-c_3 c_4)\frac{q}{m } \left(\frac{1-x}{x}\right)^{1/2 } + 2(c_1 c_4+c_2 c_3)\frac{q}{m}\frac{2x-1}{x } \right]\nonumber\\ & & \hspace{1cm}\times\frac{m\epsilon_{\mu\nu\alpha\beta}q^\alpha}{p\cdot q } \left(s^\beta-\frac{s\cdot q}{p\cdot q}p^\beta\right)\end{aligned}\ ] ] to obtain the above final results , we have used the identity @xmath108= q^\mu\epsilon^{\nu\alpha\beta\gamma}p_\alpha q_\beta s_\gamma -q^\nu\epsilon^{\mu\alpha\beta\gamma}p_\alpha q_\beta s_\gamma -q^2\epsilon^{\mu\nu\alpha\beta}p_\alpha s_\beta\end{aligned}\ ] ] comparing with eq . ( [ wmunu ] ) , we can obtain all the structure functions of the `` neutron , '' @xmath109 ^ 2,\\ f_2^\mathrm{n}&=&g_4^\mathrm{n}=32\pi\kappa^2 a'\left({\lambda^2}/{q^2}\right)^{\tau-1 } x^{\tau+1}(1-x)^{\tau-2}\left[1-\tau x(2 - 4\tau + 3\tau x)\right],\\ g_2^\mathrm{n}&=&-8\pi \kappa^2 a'\left({\lambda^2}/{q^2}\right)^{\tau-1}x^{\tau}(1-x)^{\tau-2}\frac{1}{\tau-1}\nonumber\\ & & \times\left[\tau(2\tau-5)+2\tau x(\tau^2 - 10\tau + 8)+\tau x^2(7\tau^2 + 17\tau-6)-6\tau^2 x^3(\tau+1)-1\right],\\ g_3^\mathrm{n}&=&32\pi \kappa^2 a'\left({\lambda^2}/{q^2}\right)^{\tau-1}x^{\tau+1}(1-x)^{\tau-2}\frac{1}{\tau-1}\nonumber\\ & & \times\left[\tau(4x-3)-\tau^2(x^2 + 7\tau x^2 - 4x-6\tau x+2)-1\right].\end{aligned}\ ] ] it is straightforward to compute the moments of all the structure functions when the contributions from @xmath110 are negligible .
typically there are four different kinds of moments , e.g. , @xmath111\frac{\gamma(\tau)\gamma(\tau+n+1)}{\gamma(2\tau+n+2)}\\ \label{g2nn } \int_{0}^{1 } g_2^\mathrm{n}\left(x , q^2\right ) x^{n-1}\textrm{d}x & = & -8\pi\kappa^2 a^{\prime } ( \lambda^2/q^2)^{\tau-1 } \left[n^3(3\tau^2 - 4\tau+1)+2n^2\tau(10\tau^2 - 10\tau+1)\right.\nonumber\\ & & \left.+n(33\tau^4 - 31\tau^3 - 9\tau^2 - 1)+\tau(16\tau^4 - 13\tau^3 - 13\tau^2 - 6\tau-2)\right]\nonumber\\ & & \times\frac{\gamma(\tau)\gamma(n+\tau)}{(\tau-1)\gamma(2\tau+n+2)}\\ \label{g3nn } \int_{0}^{1 } g_3^\mathrm{n}\left(x , q^2\right ) x^{n-1}\textrm{d}x & = & 32\pi\kappa^2 a^{\prime } ( \lambda^2/q^2)^{\tau-1 } \left[n^2(\tau^2 - 1)+n(3\tau^3 + 3\tau^2 - 1)+2\tau(\tau+1)^3\right]\nonumber\\ & & \times\frac{\gamma(\tau)\gamma(n+\tau+1)}{(\tau-1)\gamma(2\tau+n+2)}\\ \label{f2nn } \int_{0}^{1 } f_2^\mathrm{n}\left(x , q^2\right ) x^{n-1}\textrm{d}x & = & 32\pi\kappa^2 a^{\prime } ( \lambda^2/q^2)^{\tau-1 } \left[n^2(\tau^2 - 1)+n(6\tau^2-\tau-1)+5\tau^3\right]\nonumber\\ & & \times\frac{\gamma(\tau)\gamma(n+\tau+1)}{\gamma(2\tau+n+2)}\end{aligned}\ ] ] in order to discuss these results further and make comparisons with the previous work by gao and xiao in ref .
@xcite , it is helpful to rewrite the structure functions of the `` proton '' that have been obtained in ref .
@xcite , @xmath112 typically there are just two different kinds of moments , e.g. , @xmath113 in all the above expressions , we have defined @xmath114 .
and @xmath115.,width=453 ] and @xmath116.,width=453 ] and @xmath117.,width=453 ] and @xmath118.,width=453 ] in this section , we focus on the interpretation of the structure functions of the `` neutron '' from pauli interaction that we obtained from last section using gauge / string duality and compare these results with the ones of the `` proton '' in ref .
@xcite , where only minimal interaction was included . * just like what we did in ref .
@xcite , only the linear term in @xmath23 is kept in the initial wavefunction and throughout the calculation , the results shown above , whether the `` neutron '' or the `` proton , '' are from leading order calculations .
the corrections are of order @xmath119 or @xmath120 . *
the power order @xmath121 of the structure functions of the `` neutron '' from only the pauli interaction are the same as that from only minimal interaction . ignoring the relative magnitude of @xmath122 and @xmath123 , but only from the naive dimensional analysis
, one might expect that the pauli interaction will lead to less power order @xmath121 by one than that from the minimal interaction .
however , when the warp factor @xmath124 is taken into account , the extra @xmath25 will combine with @xmath125 from @xmath124 and lead to @xmath126 , which results in the same power suppression as the minimal interaction . * the relations @xmath127 and @xmath128 still hold from the pauli interaction , but the relation @xmath129 from the minimal interaction
is broken .
the differences of the dependence on @xmath2 of all the structure functions are illustrated in figs.[g1][f2 ] , where the coefficient @xmath130 and @xmath131 .
* in qcd , there is an interesting inequality @xmath132@xcite . in ref.@xcite
, we have found that @xmath133 , i.e. the bound is saturated .
here we see that the saturation condition @xmath133 still holds , which indicates that initial hadron is completely polarized .
this implies that the struck dilatino just tunnels or shrinks to smaller size of order the inverse momentum transfer during the scattering . as a result ,
the structure function exhibits a power law behavior in terms of the @xmath11 dependence which comes from the tunneling probability@xcite . * for all the moments
( [ g1nn])eq .
( [ f2nn ] ) , eq .
( [ g1pn ] ) and eq .
( [ g2nn ] ) , we expect that the moments are correct at least for @xmath134 where the low - x contributions are negligible .
when one sets @xmath135 for @xmath136 , there is an interesting sum rule @xmath137 which is completely independent of @xmath14 and @xmath11 . in qcd , this sum rule is known as the burkhardt - cottingham sum rule@xcitein large @xmath138 limit . however
, this sum rule can be invalidated by non - regge divergence at low-@xmath2 .
now let us set @xmath135 for @xmath139,we can have @xmath140 it is obvious that such sum rule which holds for minimal interaction in the classic supergravity approximation is broken due to introducing the pauli interaction term . in this place , it is a good opportunity to compare the above conclusion with qed but with an extra nonrenormalizable pauli interaction term introduced , i.e. @xmath141\psi\right).\end{aligned}\ ] ] where we just specify @xmath142 as a quark field .
, where @xmath25 is the energy scale in which we are working and @xmath143 is planck energy scale .
hence such contribution is highly suppressed when @xmath25 is in the scale of gev or tev .
since we just want to show the pure effect of such pauli interaction term , we will neglect such realistic issues .
] it is easy to verify that , in the tree diagram level of @xmath123 or @xmath122 and twist-3 level of @xmath144 ( @xmath49 denotes the mass of the quark ) , the pure minimal interaction results in @xmath145 while the pure pauli interaction results in @xmath146 where @xmath147 , @xmath148 , @xmath149 , and @xmath150 are all positive coefficients , which is irrelevant with our current problem .
hence , it is obvious that , similar to the @xmath62 space , the pauli interaction term always makes the burkhardt - cottingham sum rule invalidated .
it should be clarified that the effective pauli interaction can be produced from high order contributions in usual qed or instantons @xcite in usual qcd without the pauli interaction term in the original lagrangian .
actually , from the spirit of the conjecture of ads / cft , the pauli interaction introduced in our present work in @xmath151 space is equivalent to summing over all the loop contributions in cft side . *
it is obvious that the moments of all the structure functions are power suppressed , for sufficiently large @xmath15 , all these integrals vanishes .
actually in all our calculations , the results are only valid at large t hooft coupling @xmath1 and finite @xmath2 with @xmath3 . in order to take into account the moments of the structure functions completely , we need to consider the very small @xmath2 case .
for example , the missing contributions from the pomeron exchanges to @xmath5 and @xmath6 peaks around @xmath152 .
@xmath153 where the correction to the pomeron intercept arises from the curvature of @xmath43 .
such pomeron contribution will survive in the large @xmath11 limit and give us a nonvanishing second moment of @xmath5@xcite , which make energy momentum conserved .
there is a similar contribution to @xmath154 at small-@xmath2 which yields a singular@xcite @xmath155 with @xmath156 when @xmath2 is extremely small .
this contribution will also survive in large @xmath11 limit and yield a finite first moment .
this may indicate that most of the hadron spin is carried by the small-@xmath2 constituents inside the hadron .
the detailed discussions on the small-@xmath2 limit of the @xmath154 structure function can be available in ref .
@xcite . * just repeat the arguments in ref .
@xcite on the parity violating structure functions @xmath157 ,
@xmath158 , @xmath159 , and @xmath160 .
these parity violating structure functions are as large as the @xmath6 structure function due to the reason that the dilatino is right - handed fermion in massless limit .
they are tightly related to the peculiar wavefunction of the dilatino .
however , we expect that @xmath154 and @xmath161 may exhibit some common features of the polarized structure functions of spin-@xmath4 hadrons in the nonperturbative region when the coupling is large .
* phenomenologically , we can just regard @xmath122 as a free parameter , which can be fixed by the experimental values of the `` neutron '' magnetic moments .
the detailed discussion and fitting results can be found in ref.@xcite .
through introducing the pauli interaction term in the action in the @xmath37 space , using gauge / string duality , we have calculated the structure functions of the `` neutron '' which is dual to a spin-@xmath4 dilatino which is neutral corresponding to the @xmath162 current we are considering .
we obtain both the unpolarized and polarized structure functions .
we find the structure functions of the `` neutron '' purely from pauli interactions are power suppressed at the same order as the ones of the `` proton '' purely from minimal interactions .
we also find that the burkhardt - cottingham sum rule for @xmath161 which is satisfied independent of @xmath14 and @xmath11 in the minimal interaction is broken due to such a pauli interaction term .
we acknowledge fruitful discussions with bo - wen xiao .
j. gao acknowledges financial support by the china postdoctoral science foundation funded project under contract no .
20090460736 . z. mou is supported by the national natural science foundation of china under the project nos .
10525523 and 10975092 , the department of science and technology of shandong province .
in this appendix , we will illustrate how a pauli interaction term can be produced from kaluza - klein reduction of higher dimensional fermion - graviton coupling .
since it is only an illustration , for simplicity , let us just consider the reduction from 6d to 5d .
we start with the simple example of a left chiral fermion field in 6d with the sixth dimension @xmath163 compactified .
the action reads @xmath164where we have suppressed the pure graviton self - interaction and @xmath165 it should be noted that the coefficient @xmath166 can be generally complex . in the following , we use the capital @xmath167 in the six - dimension space and the lower @xmath168 in the five - dimensional space .
we will use indices @xmath169 to denote all six general curved spacetime dimensions , and @xmath170 refer to all six local inertial spacetime dimensions , while @xmath171 denote five dimensions in @xmath43 and @xmath172 refer to five dimensions in the local flat five - dimensional spacetime . following the formalism of kaluza - klein ,
we write the vielbein field as @xcite @xmath173 where @xmath174 and we have set the scalar field simply as 1 and neglected the dimension of the field .
all the components in eq.([vielbein ] ) do not depend on the sixth dimension coordinate @xmath163 .
we make the split of @xmath175-matrices @xcite as @xmath176 where @xmath177 and @xmath178 are the d=5 @xmath179-matrices and usual pauli matrices , respectively .
we also decompose the spinor field as @xmath180 where @xmath181 is the spinor in d=5 , and @xmath182 is the spinor in d=2 . using the identity @xmath183 and the specific expressions in eq.([vielbein ] ) , we can have @xmath184 inserting it into the lagrangian in eq.([s ] ) , we can obtain @xmath185where we have taken the fermion to be in a charge eigenstate , @xmath186 in such a way , the action can be rewritten as , @xmath187where we have normalized @xmath188now we can see that the pauli interaction term in @xmath43 space has been produced from higher 6d fermion - graviton interaction by using kaluza - klein reduction . for the uncharged fermion where @xmath189 , only the pauli interaction term will contribute .
the reduction from 10d to 5d will be similar except for more possible complications involved dealing with more extra dimensions , which is beyond the scope of our present work .
e. witten , adv .
* 2 * , 253 ( 1998 ) [ arxiv : hep - th/9802150 ] .
s. s. gubser , i. r. klebanov and a. m. polyakov , phys .
b * 428 * , 105 ( 1998 ) [ arxiv : hep - th/9802109 ] .
j. polchinski and m. j. strassler , phys .
lett . *
88 * , 031601 ( 2002 ) [ arxiv : hep - th/0109174 ] .
j. polchinski and m. j. strassler , jhep * 0305 * , 012 ( 2003 ) [ arxiv : hep - th/0209211 ] .
r. c. brower and c. i. tan , nucl .
b * 662 * , 393 ( 2003 ) [ arxiv : hep - th/0207144 ] .
h. boschi - filho and n. r. f. braga , phys .
b * 560 * , 232 ( 2003 ) [ arxiv : hep - th/0207071 ] .
c. a. ballon bayona , h. boschi - filho and n. r. f. braga , jhep * 0809 * , 114 ( 2008 ) [ arxiv:0807.1917 [ hep - th ] ] . c. a. ballon bayona , h. boschi - filho and n. r. f. braga , jhep * 0810 * , 088 ( 2008 ) [ arxiv:0712.3530 [ hep - th ] ] .
c. a. ballon bayona , h. boschi - filho and n. r. f. braga , jhep * 0803 * , 064 ( 2008 ) [ arxiv:0711.0221 [ hep - th ] ] .
b. pire , c. roiesnel , l. szymanowski and s. wallon , phys .
b * 670 * , 84 ( 2008 ) [ arxiv:0805.4346 [ hep - ph ] ] .
@xcite l. cornalba , m. s. costa and j. penedones , jhep * 0709 * , 037 ( 2007 ) [ arxiv:0707.0120 [ hep - th ] ] .
y. hatta , e. iancu and a. h. mueller , jhep * 0801 * , 026 ( 2008 ) [ arxiv:0710.2148 [ hep - th ] ] .
y. hatta , prog .
phys . suppl .
* 174 * , 298 ( 2008 ) .
y. hatta , e. iancu and a. h. mueller , jhep * 0801 * , 063 ( 2008 ) [ arxiv:0710.5297 [ hep - th ] ] .
a. h. mueller , a. i. shoshi and b. w. xiao , arxiv:0812.2897 [ hep - th ] .
e. iancu and a. h. mueller , jhep * 1002 * , 023 ( 2010 ) [ arxiv:0912.2238 [ hep - th ] ] . c. a. b. bayona , h. boschi - filho and n. r. f. braga , arxiv:0912.0231 [ hep - th ] .
b. hassanain and m. schvellinger , arxiv:0912.4704 [ hep - th ] .
j. h. gao and b. w. xiao , phys .
d * 80 * , 015025 ( 2009 ) [ arxiv:0904.2870 [ hep - ph ] ] .
y. hatta , t. ueda and b. w. xiao , jhep * 0908 * , 007 ( 2009 ) [ arxiv:0905.2493 [ hep - ph ] ] . j.
h. gao and b. w. xiao , phys .
d * 81 * , 035008 ( 2010 ) [ arxiv:0912.4333 [ hep - ph ] ] .
g. f. de teramond and s. j. brodsky , phys .
lett . * 102 * , 081601 ( 2009 ) [ arxiv:0809.4899 [ hep - ph ] ] .
s. j. brodsky and g. f. de teramond , phys .
d * 78 * , 025032 ( 2008 ) [ arxiv:0804.0452 [ hep - ph ] ] .
s. j. brodsky and g. f. de teramond , phys .
d * 77 * , 056007 ( 2008 ) [ arxiv:0707.3859 [ hep - ph ] ] .
s. j. brodsky and g. f. de teramond , phys .
lett . * 96 * , 201601 ( 2006 ) [ arxiv : hep - ph/0602252 ] .
s. j. brodsky and g. f. de teramond , phys .
b * 582 * , 211 ( 2004 ) [ arxiv : hep - th/0310227 ] .
h. r. grigoryan and a. v. radyushkin , phys .
d * 76 * , 115007 ( 2007 ) [ arxiv:0709.0500 [ hep - ph ] ] .
h. r. grigoryan and a. v. radyushkin , phys .
d * 76 * , 095007 ( 2007 ) [ arxiv:0706.1543 [ hep - ph ] ] . h. r. grigoryan and a. v. radyushkin , phys .
b * 650 * , 421 ( 2007 ) [ arxiv : hep - ph/0703069 ] .
z. abidin and c. e. carlson , phys . rev .
d * 77 * , 115021 ( 2008 ) [ arxiv:0804.0214 [ hep - ph ] ] .
z. abidin and c. e. carlson , phys .
d * 77 * , 095007 ( 2008 ) [ arxiv:0801.3839 [ hep - ph ] ] .
d. k. hong , m. rho , h. u. yee and p.
yi , phys .
d * 77 * , 014030 ( 2008 ) [ arxiv:0710.4615 [ hep - ph ] ] . | we investigate deep inelastic scattering off the polarized `` neutron '' using gauge / string duality .
the `` neutron '' corresponds to a supergravity mode of the neutral dilatino . through introducing the pauli interaction term into the action in @xmath0 space
, we calculate the polarized deep inelastic structure functions of the `` neutron '' in supergravity approximation at large t hooft coupling @xmath1 and finite @xmath2 with @xmath3 . in comparison with the charged dilatino `` proton , '' which has been obtained in the previous work by gao and xiao , we find the structure functions of the `` neutron '' are power suppressed at the same order as the ones of the `` proton . ''
especially , we find the burkhardt - cottingham - like sum rule , which is satisfied in the work by gao and xiao , is broken due to the pauli interaction term .
we also illustrate how such a pauli interaction term can arise naturally from higher dimensional fermion - graviton coupling through the usual kaluza - klein reduction . |
the sensitivity of dark matter ( dm ) direct detection experiments is undergoing rapid progress and is expected to continue in the next decade .
there are a number of proposed experiments which will probe complementary aspects of dark matter properties with much better sensitivities than the existing ones : dm mass , spin - independent ( si ) and spin - dependent ( sd ) cross sections , the dependence of the cross sections on the target nuclei , directional information , etc .
the focus , rightly , is often on the detection of spin - independent dm interactions , because , due to a coherence effect , the si interaction cross section with heavy nuclei is enhanced by @xmath0 , the number of nucleons in a nucleus , and is , therefore , expected in many models to be the dominant interaction in dm detectors .
there is a good chance that , in the not - too - distant future , direct detection experiments will be able to extend their sensitivity to cover the full detectable parameter space for si cross sections , down to the @xmath1 level , below which atmospheric neutrinos constitute an irreducible background .
prior studies @xcite have considered the relationship between si and sd cross sections , concluding that the two are typically correlated when a viable dark matter candidate is present .
most of the discussions have been in the context of the minimal supersymmetric standard model .
( similar statements have been made about dm candidates in universal extra dimensions @xcite and little higgs models @xcite , as well . ) in general , the common wisdom is that si experiments have a much better chance of first direct detection discovery .
the generality of this conclusion can not be addressed by merely considering operators ; one must explore the underlying models which determine relationships between operator coefficients .
for example , the conclusions stated above ultimately stemmed from the assumption of dm with electroweak charges , which generically implies both mediators with at least weak - scale masses to justify null results thus far and couplings to the higgs leading to si signals .
once this condition is relaxed , the relationship between si and sd cross sections becomes weaker , and models in which sd interactions are more easily accessible , or even the only interaction accessible in direct detection experiments , become feasible . here
we point out that in order to impose the last condition , i.e. , uniquely sd detection , the consideration of subleading effects is crucial . since , due to coherence effects , si experiments are more sensitive than sd ones ( currently by 5 orders of magnitude ) , a loop - induced si process might be only marginally more difficult or possibly even as easy to detect than a tree - level sd one . upon considering these additional operators , we find that models with light pseudoscalars are uniquely capable of generically evading such detection modes .
although several ingredients of our analysis appear in the literature @xcite , the impact of light mediators on a general analysis of operators has not been heretofore discussed , and the effect of loop corrections on dm scattering has not been considered in this context . in sec .
[ sec : det ] , we review current bounds on si and sd cross sections and the expected improvements . sec .
[ sec : oper ] then constitutes the bulk of the paper .
we discuss operators relevant for the detection of dm particles , including operators which become important in the case of light mediators .
we then consider which models could generate exclusively sd interactions , and calculate the loop - induced interactions that would simultaneously be present . in sec .
[ sec : axion ] we construct a viable model achieving our goals , in which the si interaction is out of reach , but the sd interaction may be detected in future experiments .
[ sec : conc ] concludes .
the best si bounds come at present from xenon10 @xcite , cdms @xcite , and xenon100 @xcite , with the highest sensitivity from xenon100 near @xmath2 at @xmath3 . in general ,
optimal sensitivity is for dm masses of order the mass of the recoiling nucleus . at higher masses ,
the sensitivity decreases roughly as @xmath4 . within the coming years
, xenon100 , lux , and supercdms can improve these bounds down to the @xmath5 or possibly near the @xmath6 level .
ultimately multi - ton xenon or germanium experiments can achieve sensitivities to @xmath7 or maybe even @xmath1 , at which point atmospheric neutrinos form an irreducible background @xcite , and achieving sensitivity to lower si dm - nucleon interactions seem unfeasible . for sd detection ,
the best current limit for dm - proton interaction is near @xmath8 from simple @xcite , with slightly weaker bounds from coupp @xcite , kims @xcite , and picasso @xcite , at similar optimal masses as above . for dm - neutron cross sections the best bound
is from xenon10 @xcite at @xmath9 at optimal sensitivity near 30 . within the next few years , coupp @xcite , picasso @xcite , and
xenon100 should improve these to @xmath10 , for both protons and neutrons .
these limits could then be extended to near @xmath11 with experiments such as dmtpc , or to @xmath12 for a 500 kg extension of coupp @xcite .
bounds on direct detection cross sections can also come indirectly from other experiments .
one source is from dm annihilation signals from the sun .
the annihilation at equilibrium is proportional to the rate of dm capture , which is driven by the same interactions as direct detection . in this case
the si terms are not so strongly enhanced over the sd ones , since this capture is mostly due to light nuclei , almost entirely hydrogen and helium .
( some enhancement does occur due to small amounts of fe and o , but bounding the sd interaction neglecting the si contribution is conservative . )
super - kamiokande @xcite and icecube @xcite used this to place limits on sd proton interactions at around @xmath13 , assuming annihilations primarily to @xmath14 . above @xmath15 ,
icecube could even place a bound at @xmath16 if the dm annihilated to @xmath17 .
however , these indirect bounds do not apply in the case of light mediators which will be discussed below , since if the annihilations proceed through a light on - shell particle , decays to neither heavy quarks nor @xmath18 bosons occur .
other bounds can be placed from constraints on operators from collider searches @xcite . in cases
where the mediator can be integrated out , these searches place bounds on interactions of very light dark matter better than those of direct detection , while remaining competitive with them for sd interactions of dm that can be directly produced at the tevatron .
the expected lhc reach is expected to also remain competitive with direct detection sensitivites of upcoming expierments @xcite . however , for mediators light enough to be produced on - shell , the bound deteriorates rapidly @xcite , and is also not applicable for the class of models we discuss below .
.operators relevant for scalar dark matter detection .
the suppression factor given is for the relevant cross section .
operators @xmath19 and @xmath20 are only allowed for complex scalars . [ cols="^,^,^,^",options="header " , ] there are a large number of operators that could mediate sd interactions . however ,
for our purposes , some of these may be ignored right away .
for example , @xmath21 and @xmath22 lead to both sd and si interactions of comparable magnitudes .
it may naively seem that all operators that come with kinematic suppression factors can be dismissed just as easily .
after all , with dm in the galactic halo at such low velocities , the nonrelativistic limit is appropriate for detection , and traditionally such operators have indeed been neglected .
let us examine this assumption more carefully . within the dominant wimp paradigm
, the mediator has typically been assumed to be at the weak scale , with direct detection occurring with @xmath23 momentum transfers and @xmath24 recoil energies . in that case
, the integrated - out mediator sets the scale of the operators through a factor of @xmath25 . in the nonrelativistic limit ,
terms like @xmath26 are suppressed by factors of @xmath27 or @xmath28 , while others , like @xmath29 , have some components scale as @xmath30 .
operators with any of these factor can typically be dismissed , because they are suppressed by @xmath31 .
this means that , even if present , such interactions can be ignored .
for example , in the case of majorana fermion dark matter , such as the neutralino in supersymmetric models , the only two operators that need to be considered are scalar scalar and axial - vector axial - vector @xcite ; all others are highly suppressed .
however , recent interest in explaining various possibly dm - related anomalies have introduced models with @xmath32 mediator particles . in this case , if the leading operators were suppressed or forbidden for some symmetry reason , the traditionally subleading operators could lead to contributions of the correct magnitude to be accessible to current or future direct detection experiments . as pointed out in ref .
@xcite , these two statements may in fact be connected , since the spontaneous breaking of a symmetry forbidding the appearance of certain operators can provide for a natural explanation for the presence of light ( pseudo ) nambu goldstone scalars .
this opens up new possibilities .
if si operators without kinematic suppression factors are forbidden or highly suppressed for other reasons , the set of operators which may lead to a detectable sd signal becomes much larger .
if we wish to remain agnostic about the nature of the dm - nucleon interactions , we can say no more .
however , if a further step is to be taken , it seems most conservative to assume that the dm comes from some theory with renormalizable interactions in which the operators leading to direct detection come from heavy states that have been integrated out .
one can then ask what sort of renormalizable interactions could lead to the operators given above .
such a procedure was followed in ref .
@xcite . here
we quote their results , along with the additional possibilities afforded by interactions yielding kinematically suppressed operators .
for scalar dm , the only option for generating solely sd operators seems to be a @xmath33-channel exchange of a light pseudoscalar , which yields @xmath34 .
while such an interaction breaks parity , given that parity is badly broken already in the standard model ( sm ) , this is not a serious concern . for fermionic dm , several possibilities present themselves .
once again a @xmath33-channel light pseudoscalar exchange produces solely sd interactions via @xmath35 .
additionally , for majorana fermions , the @xmath33-channel exchange of a vector with axial couplings , either the sm @xmath36 or a new @xmath37 , will generate only a single kinematically unsuppressed operator , @xmath38 .
other options are an @xmath39 or @xmath40-channel coupling through either a scalar or vector , provided the couplings are chiral , in which case @xmath38 is generated again .
if the couplings are not chiral , @xmath41 is produced as well .
finally , for vector dm , a light pseudoscalar in the @xmath33 channel produces only @xmath42 , which breaks parity as in the scalar case .
alternatively an @xmath39 or @xmath40-channel coupling through a fermion makes @xmath43 the leading operator , if the coupling is chiral while the vector boson is real .
suppose that one is presented with a model in which one of the above sd interactions is the only one present , or dominant over other by many orders of magnitude .
does that mean that only an experiment sensitive to sd interactions would see a signal ?
not necessarily .
the bounds on si cross sections are currently 57 orders of magnitude higher than the sd ones , and this looks to continue to be the case in the future
. therefore if any of the sd interactions discussed above induce subleading si couplings , such an effect could potentially be visible in a si experiment .
there are two sources for such effects .
first , there are kinematically suppressed contributions of tree level scattering that were ignored above .
these are easily estimated from tables [ tab : soper][tab : voper ] given earlier .
second , the tree - level sd interactions can induce si couplings at loop level .
these are not as simple to estimate , and should be calculated to confirm their effect .
let us consider a @xmath36 ( or @xmath37 exchange ) with a majorana fermion , as in fig .
[ fig : zexchtree ] .
while the dominant contribution comes from @xmath38 , also present is @xmath21 , the anapole coupling .
we see that this gives rise to a si interaction suppressed by @xmath44 .
similarly , both the scalar exchange of fig .
[ fig : fexchtree ] and the equivalent diagram for vector exchange give an anapole coupling after using fiertz identities .
a fermion exchange of the same form in the case of vector dm produces @xmath22 as well as @xmath43 in the chiral limit , which again mediates a @xmath44 suppressed si coupling . in all of these cases
, there is a si scattering cross section no more than @xmath45 smaller than the sd one , independent of any other field content of a model .
this means that such interactions would be seen in si experiments simultaneously or in the next generation of experiements after they appear in sd ones .
only the pseudoscalar exchanges evade this , as they lead to no @xmath44 suppressed subleading contributions to dm - nucleon scattering at all .
all the aforementioned interactions should also be computed at the one - loop level .
while these will be suppressed by loop factors and extra couplings , they may also generate si interactions . for large enough couplings
, these loops might even give rise to interactions larger than the kinematically - suppressed ones discussed above , and so might be even more readily detectable . without making any further assumptions about the underlying model
, we can already identify diagrams which will produce si interactions at loop - level .
for sd interactions involving a @xmath33-channel exchange , at a minimum , exchanging two mediators in a box diagram will give rise to a si interaction .
for an @xmath39 or @xmath40-channel processes , a si loop level contribution can come from a loop with @xmath18 or @xmath36 bosons exchanged between the quarks .
consider the exchange of a @xmath36 with axial couplings to quarks .
( we will discuss the case of a @xmath37 shortly . ) in that case , the quark level operator for tree - level scattering ( fig .
[ fig : zexchtree ] ) is @xmath46 where @xmath47 is the coupling of the dm to the @xmath36 .
then the dm - proton sd cross section generated is ( see apps .
[ app : matel ] and [ app : xsec ] for details ) @xmath48 with the dm - neutron cross sections about 20% smaller . in this case ,
two one - loop processes lead to si effective interactions : one with two @xmath36 exchanges , and a higgs coupling through a @xmath36 loop to the dm ( fig .
[ fig : zexchloops ] ) .
we work in the limit @xmath49 .
( this limit is generally the one in which the dm has the correct relic abundance in models where the only coupling of the dm to the quarks is through electroweak bosons , while foregoing the last inequality only yields @xmath50 changes , see ref .
the si contribution to the effective coupling is then @xcite ] which all simplify to @xmath51 on shell , but can have different nuclear matrix elements , to their on - shell value .
in fact this seems to yield a conservative estimate , as out of the nuclear matrix elements known , the first one has the smallest value ( for a detailed discussion of these issues see ref .
@xcite ) . ]
@xmath52 m_q\ , \bar{\chi}\chi\ , \bar{q}q\,.\ ] ] taking a reference value of @xmath53 , this interactions will induce a si cross section of @xmath54 asking that the sd signal be just beyond current sd experimental bounds implies @xmath55 , giving a si cross section of @xmath56 .
this , while not detectable in experiments underway , is feasible with ones in preparation .
this result would make the @xmath44-suppressed constribution to si scattering dominant .
however , it is worth mentioning that this cross section acts as a lower bound it could be that the dm particle is part of larger representation of @xmath57 , in which case additional loops involving @xmath18s would also contribute .
generally , the size off the cross section grows as @xmath58 , with @xmath59 the dimension of the representation @xcite , making it possible for the loop contribution to be dominant , and not merely competitive with the kinetically suppressed contribution , and even being large enough to be discovered simultaneously with the sd signal .
if one wishes to consider models with a new @xmath37 , then the existence of a @xmath37 with higgs coupling becomes model dependent . to talk about a lower bound
, we can then ignore the contribution of the second term in the effective coupling .
the heavier mediator mass that such a model would entail would have to be offset with a larger coupling in order to be detectable .
thus , at loop level , one would generally expect the effective interaction to be of at least similar size , or possibly larger , due to the higher power of the coupling appearing in the loop - induced term
. if one considers the possibility of a light @xmath37 , which are not ruled out by collider constraints down to the gev range for gauge couplings smaller than the sm by @xmath60 , the situation discussed above would be reversed , and one would expect a smaller loop - induced contribution . however , the si contribution due to kinematically suppressed operators is insensitive to changes in the mediator mass , and would still be present . constructing a model without such operators and without significant
fine - tuning seems extremely difficult .
it is difficult to say more in generality , due to the large freedom in assigning masses and charges under a new gauge group .
now let us consider dm with chiral couplings to the sm via an @xmath39 or @xmath40-channel .
the most model - independent loop - level processes here come from box diagrams with the quarks exchanging a @xmath18 or @xmath36 boson , an example of which is given in fig .
[ fig : fexchloops ] . the contributions of the loops have completely different forms depending on whether the coupling of the dm is left- or right - handed .
however , in all cases the loop - level processes only give rise to suppressed sd contributions .
in addition to dm of the form in fig . [
fig : fexch ] , this is also true for the cases of fermionic dm with a vector mediator and vector dm with a fermion mediator of similar topologies . in this case , we find that the most reliable lower bound on a si cross section in this case comes from the @xmath44 suppressed contribution to the tree - level interaction discussed earlier .
finally , let us turn to the box diagrams induced in the cases of light pseudoscalar exchange , fig .
[ fig : pseudloops ] .
first , we consider the case of scalar dm . at tree level ,
the operator obtained after integrating out the pseudoscalar is @xmath61 where @xmath62 is the yukawa coupling of the quark , so @xmath63 absorbs both the coupling of the dm and mediator , and any scaling to yukawas of the mediator - quark coupling .
this leads to a a tree - level cross section of @xmath64 ( see app . [
app : xsec ] for the definition of the cross section in cases of kinematically suppressed operators . ) for a mediator with mass of a few gev and @xmath65 , this would be accessible to currently running searches
. the calculation of the loop diagram in the same limits as the previous @xmath36-mediated case does not give as compact of an answer , but can be expressed in closed form in terms of passarino veltman scalar integrals @xcite , computed with the use of feyncalc @xcite as @xmath66\ , \phi^\dagger \partial^\mu \phi\ , \bar{q}\gamma_\mu q. \end{split}\ ] ] a numerical evaluation of the coefficients show the @xmath67 and @xmath68 functions with these parameters to scale as @xmath69 and @xmath70 , respectively , beyond their overall @xmath71 dependence .
using a fiducial value of @xmath72 gives @xmath73 where @xmath74 .
note that if the dm were real , this operator vanishes identically , and there is no loop induced coupling at one loop order at all .
if present , the cross section induced is @xmath75 undetectable for any any choice of parameters that would make the sd cross section detectable .
the case of vector dm is very similar .
for @xmath76 the tree - level cross section takes the same value as eq .
( [ eq : scxsec ] ) . meanwhile , the loop induced coupling is @xmath77 \bigg\}\ , b_\nu^\dagger\ , \partial^\mu b^\nu\ , \bar{q}\gamma_\mu q\ , , \end{split}\ ] ] which numerically evaluates to @xmath78 with @xmath79 very close to the scalar case , giving a loop induced si cross section as in eq .
( [ eq : scloopxsec ] ) , and similarly giving no contribution if the dm were real .
the case of fermionic dm is slightly different .
this is because the tree - level operator responsible for scattering is @xmath80 and , therefore , is parametrically suppressed by @xmath81 , instead of the previous cases @xmath82 .
the tree level cross section then becomes @xmath83 we see that due to the greater momentum suppression , we require a lighter mediator mass and can not afford the coupling of the dm to be as small as in the bosonic case above . in this case a cross section detectable in current experiments would require , for example , a mediator with @xmath84 and @xmath85 .
meanwhile , the effective coupling from computing the loop diagram in the same limits as the other cases is @xmath86\ , \bar{\chi}\gamma^\mu\chi\ , \bar{q}\gamma_\mu q\\ & + \frac{3}{8 } \big[1 + b_0(m_\chi^2;0,m_\chi^2 ) - b_0(0;m_a^2,m_a^2 ) + 4 m_\chi^2\ , c_0(m_\chi^2,0,m_\chi^2;m_\chi^2,m_a^2,0 ) - m_\chi^2\ , c_0(m_\chi^2,m_\chi^2,0;m_a^2,m_\chi^2,m_a^2)\\ & + 3 m_a^2 m_\chi^2\ , d_0(m_\chi^2,m_\chi^2,0,0,0,m_\chi^2;m_a^2,m_\chi^2,m_a^2,0 ) \big]\ , \frac{m_q}{m_\chi}\ , \bar{\chi}\chi\ , \bar{q } q \bigg\}\ , \end{split}\ ] ] which numerically yields @xmath87 with @xmath88 and @xmath89 . the magnitudes of these coefficients can be understood as arising from the @xmath69 and @xmath70 behavior of @xmath67 and @xmath68 mentioned above .
the loop - level cross section is then @xmath90 we will confirm below in the explicit model of sec .
[ sec : axion ] that the loop induced coupling is indeed tiny , but it is simple to see here why this is generically so .
unlike in the massive mediator cases , there are two mass scales in the dark sector , that of the dm itself and that of the mediator . at tree level
, the lighter mediator mass is the one that appears in the denominator of the operator .
however , at loop level , the value of the loop integral is parametrically controlled by the mass of the dm , the heaviest particle in the loop .
additionally , a pseudoscalar which is the nambu
goldstone boson of a broken symmetry would be expected to couple to quarks proportional to the masses of the quarks .
thus , at loop level , the effective operator would be expected to be suppressed by extra factors of quark yukawa couplings .
together , both effects combine to make the loop - level coupling to be as many as 20 orders smaller than the tree level one , with higher order corrections to the non - relativistic scattering approximation coming at similar orders as @xmath91 , so that the si induced interaction is expected to be completely negligible .
we have just seen that without tuning of couplings , models with light pseudoscalar mediators provide the unique method of avoiding any si signal , while still producing a sd direct detection signature .
now we turn to the question of whether a viable model producing dm with the correct abundance can have these features .
coupling a light pseudoscalar to quarks is most efficiently achieved by adding a scalar field which spontaneously breaks a global symmetry , and which , by mixing with the higgs , gets a coupling to the sm . allowing this scalar to have a new global charge , while adding new fermions charged under the same symmetry , ensures that the new scalar field is the only method for the new fermions to interact with the sm . as a simple realization of such a mechanism , where the dominant interaction is @xmath92 via a pseudoscalar interaction , we introduce , following ref .
@xcite , a scalar field charged under a new global @xmath93 charge that is spontaneously broken to @xmath94 this scalar field is coupled to a new fermion , which is vector - like under the sm , through @xmath95 , so that after the scalar field acquires a vacuum expectation value , the fermion receives a mass of @xmath96 , allowing it to act as dark matter , with stability ensured by the remnant of @xmath93 after breaking . in order for the pseudoscalar to interact with the sm , some known particles must also carry charges under the new @xmath93 . in a two higgs doublet model ,
this can be accomplished by adding a term of the form @xmath97 by assigning the appropriate charges to the higgses and sm fermions , and promoting the @xmath93 to a peccei - quinn ( pq ) symmetry . for @xmath98 ,
this coupling is of the same form as in the case of the dfsz axion @xcite , while the @xmath99 case functions like that of the pq - symmetric limit of the next - to - minimal supersymmetric standard model @xcite we now have a dark matter candidate coupling to the sm though a massive scalar and an axionlike nambu goldstone boson .
the nambu
goldstone boson is assumed to get a small mass through an unspecified mechanism .
anticipating making the scalar heavy , by virtue of @xmath100 a choice of , say , @xmath101 and @xmath102 ( corresponding to @xmath103 ) yields a cross section of @xmath104 and so generates the correct order of magnitude for the relic abundance @xcite . for direct detection ,
two channels present themselves .
the scalar gives a si cross section through the operator @xmath105 , due to mixing of the scalar with the two @xmath106-even higgses , while the light axionlike state yields a sd interaction , @xmath92 , by a similar mixing with the @xmath106-odd higgs . for our purposes , we need check whether this tree - level si cross section can be small enough to be completely negligible .
the mixing of the scalar with the two @xmath106-even higgs has a lot of arbitrariness to it due to the 11 constants in the most general @xmath107-preserving two - higgs - doublet and one - singlet potential .
however , we can say that barring accidental cancellations , this mixing will be @xmath108 , so that we may write the tree - level si cross section as @xmath109 ( see app .
[ app : matel ] for a caveat on the values of the nuclear matrix elements in this calculation . ) in the model considered in ref .
@xcite , @xmath110 needed to be light , @xmath111 , in order to provide a mechanism for sommerfeld enhancement to explain astrophysical anomalies . in that case , the direct detection cross section was in tension with the si bound and could only be slightly beyond current limits , at @xmath112 .
however , if we impose no such condition , @xmath110 could be larger .
if it is at the electroweak scale , then the cross section is at most @xmath113 , smaller than the sensitivity of the next generation of direct detection experiments . if @xmath114 , a reasonable choice given the scale of @xmath115 in this setup , then the cross section becomes undetectably small , below the irreducible atmospheric neutrino limit .
let us next consider the pseudoscalar channel . with the interaction kinetically suppressed by the momentum transfer as @xmath81 , we can not merely compute the cross section in the limit of @xmath116 as we did in the scalar exchange case .
instead , we must define a cross section at a fixed momentum transfer .
( see app .
[ app : xsec ] for a more thorough discussion . )
we choose to do so at @xmath117 . because the signal is different from that of unsuppressed interactions relative to the expected recoil energies ,
the sensitivities of experiments are modified .
this was studied in ref .
@xcite , with the result that at the same reference momentum transfer , optimal sensitivities of sd experiments to pseudoscalars remained at the same order of magnitude as in the unsuppressed case , but with @xmath4 scaling of the limits . with this definition , we can compute the sd cross section for @xmath118 as @xmath119 where @xmath120 $ ] is the mixing of the @xmath39 with the higgses @xcite . from this
we see that given a dm mass @xmath102 , a pseudoscalar with a mass @xmath121 generates a cross section of @xmath122 , within the range of the next generation of direct sd detection experiments .
in fact , in a two higgs doublet model like this , the nuclear matrix element also has a dependence on @xmath123 as up - type quarks couple with a coefficient proportional to @xmath124 , while down type ones couple proportional to @xmath125 .
we have evaluated the matrix elements for the above cross section at @xmath126 . at large values of @xmath125
the cross section can rise by almost 2 orders of magnitude .
given the tiny size of the tree - level si cross section , and in keeping with the discussion of the previous section , we should confirm that the loop - induced couplings fail to produce a detectable si cross section .
the calculation mostly mirrors that of sec .
[ subsec : loops ] .
the only substantial difference is the aforementioned different coupling to up and down type quarks .
as before , we evaluate the nuclear matrix elements at @xmath126 , but this time , varying @xmath125 can not only modify the cross section by a factor of @xmath50 as the suppression of @xmath127 at high @xmath125 is too strong , so we find @xmath128 with no additional implicit @xmath125 dependence .
as the sensitivity of both si and sd direct dm detection experiments increases , it is worth asking to what extent the discovery potential of the two methods is complementary . in this work
we have pointed out that when one considers the full range of possible mediators , instead of being confined to new weak - scale particles , the range of possible viable interactions generating sd cross sections increases . at the same time , when one searches for interactions for which sd experiments are complimentary for discovery ones which could not be seen in any si experiments without the need for accidental cancellations or other tuning
it becomes necessary to take into account subleading contributions to scattering , such as suppressed operators and loop processes .
the outcome is that the traditional models considered also generically produce si interactions whose suppression is counterbalanced by the greater sensitivity of si experiments .
the list of viable candidates whose interaction with the sm can be described by tree - level mediators integrated out in a renormalizable model is then reduced to merely ones mediated by light pseudoscalars .
we have presented a realistic model of such interactions that generates the right dm abundance with a fermionic dm candidate without having other interactions generating detectable si interactions .
similar scenarios can also be considered with a scalar or vector dark matter candidate . just as in the case of fermionic dm
, @xmath129 gives the leading interaction in the nonrelativistic limit , while @xmath130 is kinematically suppressed .
the necessary couplings between the pseudoscalar and the scalars or vectors can not be generated in as simple a manner as those used above , so more model building will be required . however , the suppression is only by @xmath82 , so the mass differences between the scalar and pseudoscalar do not have to be quite as large , and the couplings themselves can be smaller , so that the parameter space of couplings and the pseudoscalar mass are not as tightly limited by experiment , potentially making the exercise worthwhile .
we thank jeremy mardon and michele papucci for helpful discussions , and tomer volansky for drawing our attention to the importance of loop contributions .
this work was supported in part by the u.s .
department of energy under contract de - ac02 - 05ch11231 .
here we summarize how to compute the dark matter - nucleon interaction cross sections from quark - level interactions .
much of this has been discussed in the dm literature , with the exception of the pseudoscalar matrix element , as it only plays a role in momentum suppressed cross sections . for a vector coupling ,
nuclear matrix elements are straightforward to compute , since a vector coupling to quarks is a conserved current , so the coupling to a nucleon is obtained from the sum of the currents of the valence quarks . in the case of a scalar coupling to quarks , we are interested in the effective nucleon coupling induced by a quark level coupling : @xmath131 we define the nuclear matrix elements conventionally by @xmath132 on including the coupling to gluons induced by integrating out heavy quark loops , @xmath133 is given by @xmath134 where @xmath135 .
unlike the @xmath40 and @xmath136 matrix elements , which can be extracted from @xmath137 scattering , the uncertainty associated with the strange quark matrix element @xmath138 is higher , which introduces a substantial uncertainty in the si coupling to nucleons .
most studies use numerical values @xmath139 based on older calculations . a representative set of values is that used by the darksusy package @xcite , wherein , @xmath140 these are the values used for the numerical estimates given above , and in most of the literature .
however , recent lattice qcd results give substantially smaller values , @xmath141 @xcite ( see also @xcite ) , and so the si cross section from scalar exchange ( if it couples proportionally to mass ) may be smaller by a factor of @xmath142@xmath143 than numerical results quoted by many calculations . for sd interaction
we need to consider the nuclear matrix elements induced by the quark level axial - vector and pseudoscalar couplings , @xmath144 and @xmath145 for the axial - vector current , defining @xmath146 where @xmath147 is the spin of the nucleon , we have @xmath148 the matrix elements coming from polarized deep inelastic scattering carry much smaller uncertainties than for the scalar si interaction above . for our numerical results ,
we use again the darksusy values , @xmath149 more recent determinations favor slightly different values , and the pdg quotes @xmath150 , @xmath151 , @xmath152 , with a 0.02 uncertainty for each @xcite ; the effect on our numerical results is negligible . for the pseudoscalar current in eq .
( [ sdpseudoscalar ] ) the nucleon - level coupling is determined by the same axial - vector matrix elements above .
the relationship is established through generalized goldberger treiman relations . while not normally considered in dark matter detection , it has been well - studied in the axion literature @xcite .
taking divergences of the axial currents and using the equations of motion for the quarks yields @xcite @xmath153 where @xmath154 , @xmath155 , and @xmath156 , while @xmath157 is the mean of the quark coupling coefficients .
due to uncertainties in the value of @xmath158 , the value of @xmath159 can vary by as much as a factor of @xmath142 .
in this appendix , we provide a summary of cross sections for dm - nucleon interactions relevant for calculating the various cross sections discussed above in the non - relativistic limit .
we first consider the unsuppressed operors in the limit of zero momentum transfer .
si cross sections can come from either scalar or vector quark couplings . effective dm - nucleon scalar interactions for fermions of the form @xmath160 which are derived from the quark - level couplings using nuclear matrix elements , as explained in app .
[ app : matel ] , lead to a dm - nucleus cross section @xmath161 ^ 2 , \ ] ] for majorana dm fermions .
( for dirac fermions , all results for majorana fermions are divided by 4 . ) here @xmath162 is the reduced mass of the dm - nucleus system .
the per - nucleon cross section , which is usually quoted for comparisons , is @xmath163 ^ 2,\ ] ] where @xmath164 is the reduced mass of the dm - nucleon system .
for scalar or vector dark matter , the relevant operators are ( we include the dm mass to give all operators the same dimension ) @xmath165 the nucleon cross section for either operator is @xmath166 ^ 2.\ ] ] vector interactions for fermions only exist in the case of dirac dm : @xmath167 where , @xmath168 and @xmath169 , due to vector current conservation , as discussed above in app .
[ app : matel ] . then @xmath170 ^ 2.\ ] ] for the operators
@xmath171 which only exist for complex scalars or vectors , the cross section is @xmath170 ^ 2 .\ ] ] unsuppressed sd interactions come solely from the quarks axial currents . in the case of @xmath172
the dm - nucleus cross section is @xmath173 and for a nucleon @xmath174 the only other unsuppressed sd interaction is for vector dm and comes from @xmath175 here , the dm - nucleon cross section is @xmath176 all of the above cross sections are quoted in the @xmath116 limit . in this limit ,
interactions mediated by light pseudoscalars are all zero , so we need another way of expressing such cross sections .
to do so , we will use the fact that while in the nonrelativistic limit @xmath177 , @xmath178 , so that using the results above we can write ( since @xmath179 in the nonrelativistic limit ) .
@xmath180 we then compute the cross section as above , and quote a result at a reference value of @xmath82 .
we have chosen @xmath181 since with @xmath182 , where @xmath183 is the recoil energy of the nucleus , this is a typical value for most sd detectors .
other momentum - suppressed operators can be handled the same way .
g. bertone , d. g. cerdeno , j. i. collar and b. c. odom , phys .
* 99 * , 151301 ( 2007 ) [ arxiv:0705.2502 [ astro - ph ] ] . v. barger , w. y. keung and g. shaughnessy , phys . rev .
d * 78 * , 056007 ( 2008 ) [ arxiv:0806.1962 [ hep - ph ] ] .
g. belanger , e. nezri and a. pukhov , phys .
d * 79 * , 015008 ( 2009 ) [ arxiv:0810.1362 [ hep - ph ] ] .
t. cohen , d. j. phalen and a. pierce , arxiv:1001.3408 [ hep - ph ] .
p. agrawal , z. chacko , c. kilic and r. k. mishra , arxiv:1003.1912 [ hep - ph ] .
j. fan , m. reece and l. t. wang , arxiv:1008.1591 [ hep - ph ] .
j. angle _ et al . _ [ xenon collaboration ] , phys .
* 100 * , 021303 ( 2008 ) [ arxiv:0706.0039 [ astro - ph ] ]
. z. ahmed _ et al .
_ [ the cdms - ii collaboration ] , science * 327 * ( 2010 ) 1619 [ arxiv:0912.3592 [ astro-ph.co ] ] .
e. aprile _ et al .
_ [ xenon100 collaboration ] , phys . rev .
lett . * 105 * , 131302 ( 2010 ) [ arxiv:1005.0380 [ astro-ph.co ] ] .
b. cabrera , l. m. krauss , f. wilczek , phys .
lett . * 55 * , 25 ( 1985 ) .
m. felizardo _ et al .
_ , phys .
* 105 * , 211301 ( 2010 ) [ arxiv:1003.2987 [ astro-ph.co ] ] ; t. girard _ et al . _
[ for the simple collaboration ] , arxiv:1101.1885 [ astro-ph.co ] .
e. behnke _ et al .
_ , arxiv:1008.3518 [ astro-ph.co ] . h. s. lee _ et al . _
[ kims collaboration ] , phys .
rev . lett .
* 99 * , 091301 ( 2007 ) [ arxiv:0704.0423 [ astro - ph ] ] .
s. archambault _ et al .
b * 682 * , 185 ( 2009 ) [ arxiv:0907.0307 [ hep - ex ] ] .
j. angle _ et al . _
[ xenon10 collaboration ] , phys .
* 101 * , 091301 ( 2008 ) [ arxiv:0805.2939 [ astro - ph ] ] .
e. behnke
_ et al . _
[ coupp collaboration ] , science * 319 * , 933 ( 2008 ) [ arxiv:0804.2886 [ astro - ph ] ] .
m. barnabe - heider _ et al . _
[ picasso collaboration ] , phys .
lett . * b624 * , 186 ( 2005 ) .
[ hep - ex/0502028 ] .
s. desai _ et al .
_ [ super - kamiokande collaboration ] , phys . rev .
d * 70 * , 083523 ( 2004 ) [ erratum - ibid .
d * 70 * , 109901 ( 2004 ) ] [ arxiv : hep - ex/0404025 ] .
r. abbasi _ et al .
_ [ icecube collaboration ] , phys . rev .
lett . * 102 * , 201302 ( 2009 ) [ arxiv:0902.2460 [ astro-ph.co ] ] .
j. goodman , m. ibe , a. rajaraman _ et al .
. lett . * b695 * , 185 - 188 ( 2011 ) .
[ arxiv:1005.1286 [ hep - ph ] ] .
a. kurylov and m. kamionkowski , phys .
d * 69 * , 063503 ( 2004 ) [ arxiv : hep - ph/0307185 ] .
m. w. goodman and e. witten , phys .
d * 31 * , 3059 ( 1985 ) .
i. wasserman , phys .
d * 33 * , 2071 ( 1986 ) .
s. chang , a. pierce and n. weiner , jcap * 1001 * , 006 ( 2010 ) [ arxiv:0908.3192 [ hep - ph ] ] .
r. essig , phys .
d * 78 * , 015004 ( 2008 ) [ arxiv:0710.1668 [ hep - ph ] ] .
m. cirelli , n. fornengo and a. strumia , nucl .
b * 753 * , 178 ( 2006 ) [ arxiv : hep - ph/0512090 ] .
m. drees and m. nojiri , phys .
d * 48 * , 3483 ( 1993 ) [ arxiv : hep - ph/9307208 ] .
g. passarino , m. j. g. veltman , nucl .
* b160 * , 151 ( 1979 ) .
m. freytsis , z. ligeti and j. thaler , phys .
d * 81 * , 034001 ( 2010 ) [ arxiv:0911.5355 [ hep - ph ] ] .
p. gondolo , j. edsjo , p. ullio , l. bergstrom , m. schelke and e. a. baltz , jcap * 0407 * , 008 ( 2004 ) [ arxiv : astro - ph/0406204 ] .
k. takeda , s. aoki , s. hashimoto , t. kaneko , t. onogi and n. yamada [ jlqcd collaboration ] , arxiv:1012.1907 [ hep - lat ] ; k. takeda , s. aoki , s. hashimoto , t. kaneko , j. noaki and t. onogi [ jlqcd collaboration ] , arxiv:1011.1964 [ hep - lat ] .
r. d. young and a. w. thomas , phys .
d * 81 * , 014503 ( 2010 ) [ arxiv:0901.3310 [ hep - lat ] ] .
d. toussaint and w. freeman [ milc collaboration ] , phys .
lett . *
103 * , 122002 ( 2009 ) [ arxiv:0905.2432 [ hep - lat ] ] .
k. nakamura _ et al .
_ [ particle data group ] , j. phys .
g * 37 * , 075021 ( 2010 ) ; see the axions and other similar particles " minireview . | with much higher sensitivities due to coherence effects , it is often assumed that the first evidence for direct dark matter detection will come from experiments probing spin - independent interactions .
we explore models that would be invisible in such experiments , but detectable via spin - dependent interactions .
the existence of much larger ( or even only ) spin - dependent tree - level interactions is not sufficient , due to potential spin - independent subdominant or loop - induced interactions .
we find that , in such a way , most models with detectable spin - dependent interactions would also generate detectable spin - independent interactions .
models in which a light pseudoscalar acts as the mediator seem to uniquely evade this conclusion .
we present a particular viable dark matter model generating such an interaction . |
the super - kamiokande ( sk ) collaboration has recently presented an updated set of atmospheric neutrino data for a detector exposure of 79.5 kty @xcite .
the corresponding statistics is about twice as large as compared to the one considered in our earlier published analysis of three - flavor mixing @xcite and of two - flavor mixing with nonstandard dynamics @xcite , and is about 12% larger than in our previous analysis of four - family schemes in @xcite .
in addition , the chooz collaboration presented in @xcite final _ spectral _ data , which we now include in @xmath0 analyses @xcite , improving the accuracy of our previous results @xcite based on the chooz _ total rate _ @xcite . therefore , we think it useful to present a concise update of our @xmath1 , @xmath0 , and @xmath2 oscillation studies , performed under the hypothesis of one mass - scale dominance , so as to elucidate the current status and implications of subdominant ( electron or sterile ) neutrino mixing in @xmath3 .
finally , we also discuss a specific @xmath0 example which maximally violates the assumption of one - dominant mass scale .
the standard @xmath1 case of @xmath4 oscillations can be parametrized by one squared mass difference between two states @xmath5 , @xmath6 and by one mixing angle @xmath7 , describing the flavor content of @xmath8 , @xmath9 and @xmath10 .
figure 1 shows our @xmath1 best fit to the latest sk data , reached at @xmath11 , and corresponding to @xmath12 for @xmath13 degrees of freedom a very good fit . with respect to the @xmath1 ( subcase ) analysis presented in @xcite is mainly due to a better agreement of the latest sk electron distributions with their no - oscillation expectations ( especially in the multi - gev sample ) . ]
the sk collaboration finds the @xmath14 minimum at a slightly different point , @xmath15 @xcite .
however , the difference is not statistically significant , since the @xmath14 function turns out to be rather flat around the minimum , and the `` distance '' between our best - fit point and the sk one is only about one unit in @xmath16 . concerning the @xmath1 bounds on @xmath17 from parameter estimation , they will be discussed later as limits of @xmath0 and @xmath2 cases . the striking evidence in favor of standard @xmath4 oscillations ( fig .
1 ) strongly constrains nonstandard explanations . by using the approach described in @xcite , we parametrize a wide class of scenarios involving nonstandard dynamics through three ( free ) parameters : an oscillation amplitude @xmath18 , an overall phase factor @xmath19 , and an energy exponent @xmath20 ,
the standard mass - mixing dynamics being recovered for @xmath21 .
figure 2 shows the results of such a three - parameter fit in terms of the projection @xmath22 .
the corresponding bounds on @xmath20 give @xmath23 at 90% c.l .
( @xmath24 for three free parameters , @xmath25 ) , in perfect agreement with the standard case .
such results strengthen our previous bounds obtained with smaller ( 45 kty ) sk statistics ( @xmath26 @xcite ) , and definitely exclude nonstandard dynamics with integer @xmath27 in the @xmath4 channel .
oscillations of three neutrinos @xmath28 , under the hypothesis of one mass scale dominance for atmospheric @xmath29 s ( equivalent to set @xmath30 ) , are characterized by one squared mass difference , @xmath31 ( the cases @xmath32 and @xmath33 being physically different @xcite ) , and by two mixing angles ( @xmath34 $ ] and @xmath35 $ ] ) , describing the flavor content of the state @xmath36 , latexmath:[\[\begin{aligned } pure @xmath4 case being recovered for @xmath38 ( see @xcite and references therein ) . in order to show the bounds in the mixing parameter space
, we used in @xcite a triangular representation @xcite , which is basically a linear mapping in the variables ( @xmath39 ) .
such a representation has the advantage of embedding unitarity by construction , but has the disadvantage of not showing in detail the phenomenologically interesting case of small @xmath40 .
we use here the alternative representation in terms of @xmath41 in logarithmic scale ( also introduced in @xcite ) , which expands the small @xmath40 region while preserving octant symmetry ( when applicable ) .
using sk data only , and assuming @xmath32 , we find the best fit @xmath42 at @xmath43 .
the slight deviation of the best fit mixing from the pure @xmath4 maximal mixing [ @xmath44 , although intriguing .
] , is unfortunately not statistically significant @xmath45 .
this also implies that there is no significant indication for possible matter effects related to @xmath46 mixing @xmath47 in the sk data .
figure 3 shows the @xmath0 volume allowed at 90% and 99% c.l .
( @xmath48 and 11.36 for @xmath25 , respectively ) in the @xmath49 parameter space , through its projections onto the coordinate planes .
the upper limit on @xmath50 ( @xmath51 at 90% c.l . ) improves the one found in @xcite ( @xmath52 at 90% c.l . from 33 kty sk data ) , showing the steady progress of sk in confirming dominant @xmath4 mixing and in constraining additional @xmath46 mixing . the 90% c.l .
range for @xmath53 is @xmath54@xmath55 ev@xmath56 .
the bounds on @xmath57 in fig
. 3 are octant - symmetric only in the @xmath1 limit @xmath58 ( as they should ) , and show a slight preference for @xmath7 in the second octant when @xmath59 .
correspondingly , slightly higher values of @xmath53 are preferred .
the ( weak ) positive correlation between @xmath50 and @xmath57 or @xmath53 , however , is largely suppressed by the inclusion of chooz data , as we now discuss . as described in @xcite
, we can now include the chooz reactor spectral data ( 14 bins minus one adjustable normalization parameter ) through a @xmath14 statistics reproducing the bounds of the so - called `` chooz analysis a '' @xcite .
this improvement provides more accurate bounds in the @xmath53 region of interest for atmospheric neutrinos .
our best fit to sk+chooz data ( @xmath60 ) is reached at @xmath61 . once again , the small deviation of the best fit mixing from @xmath62 is not statistically significant ( @xmath63 ) .
figure 4 shows the projections of the @xmath49 volume allowed by sk+chooz . by comparing fig . 3 with fig . 4 ,
the tremendous impact of chooz on @xmath46 mixing bounds becomes evident ( one order of magnitude difference in the upper bound on @xmath50 ) .
as expected , at the small values of @xmath50 allowed by the fit in fig . 4 ,
both the octant - asymmetry in @xmath7 and the upper limit on @xmath53 are reduced , and the 90% c.l .
range for @xmath53 becomes @xmath54@xmath64 ev@xmath56 .
we have also repeated the fit for the case @xmath33 ( not shown ) , corresponding to a state @xmath36 lighter than @xmath65 . for negative @xmath53
, we get somewhat weaker bounds on @xmath50 ( @xmath66 at 90% c.l . ) for the fit to sk data only , while the fit to sk+chooz data gives results almost identical to those in fig . 4 .
this fact shows that , unfortunately , current atmospheric+reactor data are basically unable to discriminate the sign of @xmath53 in @xmath0 scenarios , as it was the case for pre - sk and pre - chooz data @xcite
. finally , fig .
5 shows the sk zenith distributions computed for three representative cases at @xmath67 ( allowed at 90% c.l . by sk+chooz ) and for @xmath7 both maximal ( @xmath68 ) and nonmaximal ( @xmath69 and 2 ) . within statistical errors , the @xmath0 zenith distributions in fig .
5 are hardly distinguishable from the @xmath1 one in fig . 1 ( even more so for @xmath33 , not shown ) , the differences being at most @xmath70 in a few bins .
therefore , there is little hope to unambiguously discover @xmath71 ( i.e. , @xmath46 mixing ) from sk atmospheric data in the near future .
we consider the @xmath2 ( 3 active + 1 sterile ) scenario described in @xcite , characterized by a 2 + 2 mass spectrum with well - separated atmospheric and solar doublets .
we also make the simplifying assumption @xcite that the atmospheric doublet @xmath72 is almost decoupled from @xmath46 , and that the solar doublet @xmath5 is almost decoupled from @xmath73 .
the dominant mass scale for atmospheric neutrinos is then @xmath74 and two mixing angles ( @xmath34 $ ] and @xmath75 $ ] ) are sufficient to describe the flavor contents of the state @xmath76 , latexmath:[\[\begin{aligned } @xmath4 and @xmath78 cases being recovered for @xmath79 and @xmath80 , respectively . in such scenario , the cases @xmath32 and @xmath33 are not physically different ( being equivalent under octant inversion , @xmath81 @xcite ) , and we take @xmath32 .
figure 6 shows the bounds obtained by a fit to sk data at 90% and 99% c.l.(@xmath25 ) in the parameter space @xmath82 .
the best fit point @xmath42 is reached at @xmath83 but , once again , the preferred mixing differs from @xmath62 by less than one unit in @xmath84 .
notice that in the @xmath85 plane , the @xmath2 projected bounds of fig . 6 are very similar to the @xmath0 ones in fig . 4 , implying that the current limits on the `` @xmath1 '' subset of parameters @xmath85 are rather stable even by making allowance for additional @xmath86 or @xmath46 mixing . concerning the @xmath87 plane in fig
6 , the upper bounds on @xmath88 indicate that pure @xmath78 oscillations @xmath89 are disfavored as compared with pure @xmath4 oscillations @xmath90 , in agreement with @xcite , although large @xmath86 mixing is not excluded yet . in particular , the current upper limit from fig .
6 ( @xmath91 at 90% c.l . ) is even slightly _ weaker _ than the one we found with smaller statistics in @xcite ( @xmath92 ) .
the reason can be traced to a peculiar feature of the latest up@xmath93 data , namely , the flatness of the muon suppression pattern in the four up@xmath93 bins at @xmath94 $ ] , as described in fig .
figure 7 shows the sk zenith distributions for three representative @xmath2 cases with sizable @xmath86 mixing ( @xmath95 ) .
as it is well known ( and evident from a comparison of fig . 7 with fig . 1 ) , additional @xmath86 mixing for atmospheric neutrinos tends to reduce the muon suppression and , in particular , tends to flatten the normalized up@xmath93 distribution . although the sk up@xmath93 data do prefer a mean positive slope rather than a flat suppression , the four bins in the zenith range @xmath94 $ ] happen to favor a locally flat distribution
. this current feature might be just a statistical fluctuation but , at present , it plays some role in global fits , where it tends to weaken the rejection of `` flat '' distributions ( i.e. , of sizable @xmath86 mixing ) , as compared with previous up@xmath93 data @xcite .
the sk collaboration has also presented additional ( preliminary ) indication in favor of @xmath4 mixing coming from statistical @xmath96 appearance in selected event samples @xcite .
it seems possible to isolate an excess of about @xmath97 @xmath98-like events , to be compared with standard @xmath4 expectations of @xmath99 @xcite .
taken at face value , such numbers imply an additional @xmath100 evidence ( @xmath101 ) in favor of pure @xmath4 ( @xmath102 ) as compared with pure @xmath78 ( @xmath103 ) .
we have then roughly parametrized the sk `` tau appearance '' signal by adding a penalty function @xmath104 in the @xmath2 fit .
we get an `` improved '' upper bound @xmath105 at 90% c.l . ( @xmath25 ) , to be compared with @xmath91 in fig .
6 ( without penalty function ) . on the one hand
, this seems to indicate that there is certainly room to refine current bounds on additional @xmath86 mixing in sk ; on the other hand , our analysis shows that large @xmath86 mixing ( e.g. , a fifty - fifty admixture of @xmath96 and @xmath86 at @xmath106 ) is not yet excluded at present .
therefore , compatibility with complementary solar neutrino bounds on @xmath88 @xcite is still possible , as it was the case for previous sk data @xcite .
the bounds on dominant and subdominant mixing found in the previous sections can be conveniently summarized in one single plot , as shown in fig .
the left panel shows the bounds on the @xmath85 parameters for pure @xmath4 mixing ( equivalent to our @xmath0 scheme for @xmath107 or to our @xmath2 scheme for @xmath108 ) . as discussed before ,
such bounds are not significantly altered by additional @xmath46 mixing ( @xmath59 ) or by additional @xmath86 mixing ( @xmath109 ) , and thus they hold also in the global @xmath0 and @xmath2 fits with good accuracy . for such reason , the @xmath1 bounds in the left panel are formally obtained for @xmath110 , so as to match those in the middle and right panels .
the middle panel shows the @xmath0 bounds on additional @xmath46 mixing ( @xmath59 ) , with and without chooz .
finally , the left panel shows the @xmath2 bounds on additional @xmath86 mixing ( @xmath109 ) .
such synthetic figure represents the main result of our analysis .
the analyses in the previous sections are based on the assumption that atmospheric neutrino oscillations are driven by only one mass scale @xmath111 .
this is not necessarily the case , especially if one takes the `` solar '' squared mass difference in the upper range allowed by the data ( see , e.g. , @xcite ) , provided that one accepts an averaged or quasiaveraged solar neutrino survival probability @xcite . concerning atmospheric neutrinos , two mass scales in the range @xmath112 ev@xmath56
were shown to provide acceptable fits to previous sk+chooz data @xcite . here
we show , through a specific example , that such possibility is not yet excluded by the latest data .
let us consider the specific @xmath0 case shown in fig .
9 , characterized by two equal squared mass differences ( @xmath113 ) ev@xmath56 , and by mixing angles , @xmath114 , and @xmath115 . ]
@xmath116 , giving the flavor composition of mass eigenstates shown in the same figure .
the spectrum in fig .
9 might be called `` democratic '' , since it maximally violates the usual `` hierarchical '' approximation ( @xmath117 ) .
notice that @xmath46 oscillations are driven only by @xmath118 , which is purposely chosen just below the current chooz bounds @xcite .
the solar neutrino survival probability is then @xmath119 , up to small ( quasiaveraged @xcite ) corrections .
figure 10 shows the sk zenith distributions computed for the democratic scenario in fig . 9 , and corresponding to @xmath120 for the fit to sk data only ( @xmath121 if chooz data are also included ) .
although such value is significantly higher than in the best - fit @xmath1 case of fig .
1 , it is still acceptable from the point of view of goodness of fit .
all in all , the curves in fig .
10 provide a globally acceptable `` fit - by - eye '' , with moderate departures from the data along the horizontal direction ( @xmath122 ) for the mg@xmath93 and us@xmath93 samples .
therefore , if one accepts an almost constant @xmath123 suppression as explanation of the solar neutrino deficit , @xmath0 scenarios with two comparable mass scales appear to represent a viable possibility in the current atmospheric neutrino phenomenology .
needless to say , the most general analysis of such cases ( depending on all the @xmath0 mass - mixing parameters ) would be rather intricated , and is postponed to a future work .
in the context of the latest ( 79.5 kty ) sk atmospheric @xmath29 data , we have concisely reviewed the status of dominant @xmath4 oscillations ( including the case of nonstandard dynamics , fig . 2 ) and of subdominant @xmath46 and @xmath86 mixing ( in @xmath0 and @xmath2 schemes , respectively ) .
in the @xmath0 case we have applied an improved chooz analysis .
the main @xmath1 , @xmath0 , and @xmath2 results are discussed separately and then summarized in fig .
finally , we have shown ( through a specific example ) that current atmospheric neutrino data are also compatible with oscillations driven by two comparable mass scales .
super - kamiokande presentations in winter conferences : c. mcgrew in _ neutrino telescopes 2001 _ , 9th international workshop on neutrino telescopes ( venice , italy , march 2001 ) , to appear ; t. toshito in _
moriond 2001 _ , xxxvi rencontres de moriond on electroweak interactions and unified theories ( les arcs , france , march 2001 ) , to appear .
macro collaboration , talk by b.c .
barish in _ neutrino 2000 _ , proceedings of the 19th international conference on neutrino physics and astrophysics , edited by j. law , r.w .
ollerhead , and j.j .
simpson , nucl .
b ( proc . suppl . ) * 91 * , 141 ( 2001 ) .
fig . 1 .
super - kamiokande zenith distributions ( 79.5 kty @xcite ) used in the analysis , normalized to no - oscillation expectations .
the data set includes sub - gev electrons ( sg@xmath124 , 10 bins ) , multi - gev electrons ( mg@xmath124 , 10 bins ) , sub - gev muons ( sg@xmath93 , 10 bins ) , multi - gev muons ( mg@xmath93 , 10 bins ) , upward stopping muons ( us@xmath93 , 5 bins ) , and upward through - going muons ( ut@xmath93 , 10 bins ) , for a total of 55 data points .
the error bars are statistical only ( @xmath125 ) ; systematic ( correlated ) uncertainties are treated as in @xcite .
the solid line is our best fit for @xmath1 oscillations ( @xmath126 ) .
fig . 2 .
dependence of the @xmath14 function on the neutrino energy exponent @xmath20 , assuming an oscillation phase proportional to @xmath127 , with unconstrained factors for the overall phase and amplitude .
the only integer @xmath20 compatible with the sk data is @xmath21 , corresponding to standard mass - mixing dynamics .
fig . 3 .
projections of the regions allowed in the @xmath0 parameter space @xmath49 at 90% and 99% c.l .
( @xmath24 and 11.36 for @xmath25 ) onto the coordinate planes .
the fit includes sk data only ( 79.5 kty ) .
the pure @xmath1 case of @xmath4 oscillations is recovered for @xmath128 .
nonzero values of @xmath40 parametrize @xmath46 mixing . fig . 4 . as in fig . 3 , but including final chooz positron spectra @xcite ( 14 data points minus one adjustable normalization factor )
zenith distributions for three representative @xmath0 cases with @xmath129 , allowed at 90% c.l . by sk+chooz .
notice the distortion of the mg@xmath124 distribution .
such distortion would be somewhat smaller for negative @xmath130 ev@xmath56 ( not shown ) .
fig . 6 .
projections of the regions allowed in the @xmath2 parameter space @xmath82 at 90% and 99% c.l .
( @xmath24 and 11.36 for @xmath25 ) onto the coordinate planes .
the fit includes sk data only ( 79.5 kty ) . the pure @xmath4 case ( @xmath131 ) is clearly preferred over the pure @xmath78 case ( @xmath132 ) .
however , sizable @xmath86 mixing [ @xmath133 is allowed in addition to @xmath4 oscillations .
fig . 7 .
zenith distributions for three representative @xmath2 cases with sizable @xmath86 mixing ( @xmath95 ) , allowed at 90% c.l .
by sk data . notice the reduced suppression in the muon samples , as compared with pure @xmath4 mixing in fig . 1 .
fig . 8 .
summary of @xmath1 , @xmath0 , and @xmath2 bounds at 90% and 99% c.l . on the mass - mixing parameters from sk data . left panel : bounds on @xmath85 for pure @xmath4 mixing ( i.e. , for @xmath134 ) .
middle panel : bounds on additional @xmath46 mixing ( parametrized by @xmath59 ) in @xmath0 scenarios , both without chooz ( see also fig . 3 ) and with chooz ( see also fig . 4 ) .
right panel : bounds on additional @xmath86 mixing ( parametrized by @xmath109 ) in @xmath2 scenarios ( see also fig .
all the bounds are derived for @xmath25 , including those in the left panel .
fig . 9 .
a `` democratic '' ( i.e. , nonhierarchical ) @xmath0 scenario with two equal squared mass differences ( @xmath135 ev@xmath56 ) and with @xmath136 flavor content as follows : @xmath137 for @xmath138 , @xmath139 for @xmath8 , and @xmath140 for @xmath36 .
notice that @xmath46 disappearance is driven only by @xmath118 ( just below the chooz sensitivity @xcite ) , and gives @xmath141 for solar neutrinos ( up to small quasiaveraged oscillation corrections ) .
zenith distributions for the democratic scenario in fig . 9 , giving @xmath120 for the fit to sk data ( @xmath121 for the fit to sk+chooz ) . | in the context of the recent ( 79.5 kty ) super - kamiokande atmospheric neutrino data , we concisely review the status of muonic - tauonic flavor oscillations and of the subdominant electron or sterile neutrino mixing , in schemes with three or four families and one dominant mass scale . in the three - family case , where we include the full chooz spectral data ,
we also show , through a specific example , that `` maximal '' violations of the one - dominant mass scale approximation are not ruled out yet . |
vlbi imaging at millimeter wavelengths ( mm - vlbi ) gives highest angular resolution and facilitates direct studies of the self - absorbed jet base in agn . for 3c273 ( @xmath1 ) images with an angular resolution of up to @xmath2@xmath3as ( 1@xmath3as @xmath4arcsec )
are made at 86ghz .
this corresponds to a spatial scale of @xmath5 schwarzschild radii ( for m @xmath6 m ) .
the small observing beam allows to accurately determine positions of jet components .
this facilitates detailed studies of the jet structure and kinematics near the nucleus , in particular with regard to the broad - band ( radio to gamma - ray ) flux density variability and the birth of new ` vlbi components ' .
here we summarize new results from our multi epoch ( 1990 1997 ) and high observing frequency ( 15 , 22 , 43 and 86ghz ) vlbi monitoring .
at sub - milliarcsecond resolution , 3c273 shows a one sided core - jet structure of several milliarcseconds length .
the jet breaks up into multiple vlbi components , which when represented by gaussian components seem to separate at apparent superluminal speeds from the stationary assumed vlbi core .
the cross - identification of the model - fit components , seen at different times and epochs , is facilitated by small ( @xmath7mas ) , and to first order negligible , opacity shifts of the component positions relative to the vlbi - core .
quasi - simultaneous data sets ( cf .
1 ) demonstrate convincingly the reliability of the component identification , which results in a kinematic scenario , in which all detected jet components ( c6 c18 ) move steadily ( without ` jumps ' in position ) away from the core ( fig .
4 ) . for the components with enough data points at small ( @xmath8mas ) and large ( @xmath9mas ) core separations , quadratic fits to the radial motion r(t )
( but also for x(t ) right ascension , and y(t ) declination ) represent the observations much better than linear fits .
thus the components seem to accelerate as they move out .
the velocities range typically from @xmath10 ( for ) .
fig . 1 : 3c273 at 22ghz ( top )
, 43ghz ( center ) , and 86ghz ( bottom ) observed in 1995.15 1995.18 .
contour levels are -0.5 , 0.5 , 1 , 2 , 5 , 10 , 15 , 30 , 50 , 70 , and 90% of the peak of 3.0 ( top ) , 5.4 ( center ) , and 4.7jy / beam ( bottom ) . for the 22ghz map ,
the 0.5% contour is omitted .
the restoring beam is @xmath11mas in size , oriented at @xmath12 .
the maps are centered on the eastern component ( the core ) , the dashed lines guide the eye and help to identify corresponding jet components . fig . 2 : spectral index variations along the jet . for 1997 ( circles , solid lines ) ,
the spectral index gradient is calculated directly from the intensity profiles of the maps at 15 and 86ghz . for 1995 ( squares , dashed line ) ,
the spectral indices were derived from gaussian component model fits at 22 and 86ghz .
we note that during the time interval 1995 1997 , different jet components occupied this jet region . the spectral profile along the jet , however , did not change significantly .
fig . 3 : the motion of the mean jet axis in the inner jet of 3c273 at 15ghz .
symbols denote for different epochs : 1995.54 ( circles ) , 1996.94 ( squares ) , 1997.04 ( diamonds ) , and 1997.19 ( triangles ) .
the transverse oscillations of the ridge line are measured relative to a straight line oriented along @xmath13 .
we note the systematic longitudinal displacement of maxima and minima .
this corresponds to an apparent pattern velocity of @xmath14 , which is by a factor of up to 2 slower than the component motion .
dual - frequency maps obtained in 1995 ( 22/86ghz ) and 1997 ( 15/86ghz ) allow to measure the spectral index gradient along the jet ( see fig .
the spectrum oscillates between @xmath15 ( ) .
most noteworthy , the spectral gradients did not change significantly over the 2 year time period , although different jet components occupied this jet region .
thus , the geometrical ( eg .
relativistic aberration ) and/or the physical environment ( pressure , density , b - field ) in the jet must determine the observed properties of the vlbi components .
hence , the latter do not form ` physical entities ' , but seem to react to the physical conditions of the jet fluid .
further evidence for a fluid dynamical interpretation comes from a study of the mean jet axis and the transverse width of the jet .
both oscillate quasi - sinusoidally on mas - scales . at 15ghz
, the variation of the ridge - line with time could be determined from 4 vlba maps obtained during 1995 1997 ( fig .
the maxima and minima of the ridge - line are systematically displaced .
this ` longitudinal ' shift suggests motion with a pattern velocity of @xmath16 .
the sinusoidal curvature of the jet axis , however , is more indicative for jet rotation rather than for longitudinal waves .
helical kelvin - helmholtz instabilities propagating in the jet sheath could mimic such rotation , which , when seen in projection , would explain also the spectral index oscillation in figure 2 . for many agn , a correlation between flux density variability and ejection of vlbi components is suggested . in 3c273
we identified 13 jet components ( c6 c18 ) and traced their motion back to their ejection from the vlbi core . the typical measurement uncertainty for the ejection times @xmath17 ranges between 0.2 0.5yr . in figure 5 ( left )
we plot @xmath17 and the millimeter - variability ( 22 230ghz )
. we also add the gamma - ray detections of 3c273 from egret . in figure 5 ( right )
we plot the onset times of the mm - flares derived from these light curves ( trler et al .
1999 ) together with the vlbi ejection time ( @xmath17 ) and the gamma - ray fluxes . for each onset of a mm - flare
, we find that a new jet component was ejected .
although the time sampling of the gamma - ray data is quite coarse , a relation between component ejection and high gamma - ray flux appears very likely ( note that each gamma - detection already means higher than usual @xmath18-brightness ) . from a more detailed analysis ( krichbaum et al .
2001 ) we obtain for the time lag between component ejection and onset of a mm - flare : @xmath19yr .
if we assume that the observed peaks in the gamma - ray light - curve are located near the times @xmath20 of flux density maxima , we obtain @xmath21yr .
although the gamma - ray variability may be faster , this result is fully consistent with the more general finding of enhanced gamma - ray fluxes mainly during the rising phase of millimeter flares .
we therefore suggest the following tentative sequence of events : @xmath22 the onset of a millimeter flare is followed by the ejection of a new vlbi component and , either simultaneously or slightly time - delayed , an increase of the gamma - ray flux .
if we focus only on those vlbi components , which were ejected close to the main maxima of the gamma - ray light - curve in figure 5 , we obtain time lags of @xmath23 of @xmath24yr for c12 , @xmath25yr for c13 , @xmath26yr for c16 and @xmath27yr for c18 . in all cases
the gamma - rays seem to peak a little later than the time of component ejection . with @xmath28 near the core , the gamma - rays would then escape at a radius @xmath29mas .
this corresponds to @xmath30 schwarzschild radii ( for a @xmath31 mblack hole ) or @xmath32 cm , consistent with theoretical expectations , in which gamma - rays escape the horizon of photon - photon pair production at separations of a few hundred to a few thousand schwarzschild radii .
fig . 5 : left : flux density variations at 230 ghz ( filled diamonds ) , 86 ghz ( open circles ) and 22 ghz ( filled squares ) .
upward oriented triangles denote gamma - ray fluxes from egret , downward oriented triangles are upper limits .
the extrapolated ejection times of the vlbi components and their uncertainties are indicated by filled circles with horizontal bars along the time axis .
right : broad - band flux density activity and component ejection .
vlbi component ejection ( open circles ) , gamma - ray fluxes ( triangles , downward oriented for upper limits ) and onset - times for the millimeter flares ( open squares , from trler et al .
labels denote the component identification from vlbi .
* references : * + trler m. , courvoisier t.j .- l . , &
paltani s. , ` modelling the submillimeter - to - radio flaring behaviour of 3c273 ' , 1999 , _
a&a _ , * 349 * , p. 4554 .
+ krichbaum t.p .
, et al . , ` high frequency vlbi observations of 3c273 ' , 2001 , _ a&a _ , submitted . | we show results from a 7 year vlbi monitoring programme of 3c273 at millimeter wavelengths .
we find evidence for component acceleration , motion or rotation of fluid dynamical patterns , and an outburst - ejection relation between @xmath0ray flares and new jet components . |
rare decays induced by flavor changing neutral currents can be used as tests of the standard model ( sm ) , and are sensitive to new physics .
it is expected that at future b factories and fixed target machines , other rare decay channels of the bottom quark @xcite will be discovered in addition to the observed @xmath8 transition@xcite .
these processes also offer useful information for extracting the fundamental parameters of the sm , such as @xmath9 @xcite .
rare decays can also serve as alternative channels to measure some elementary hadronic parameters .
for instance , the decay constants @xmath10 , @xmath11 can be extracted from @xmath12 @xcite . in the present work ,
we investigate the possibility of using the channel , @xmath13 to determine the decay constants .
pure leptonic decays of heavy pseudoscalar mesons into light lepton pairs are helicity suppressed , their branching ratios are of the order of @xmath14 for @xmath15 , and @xmath16 for @xmath17 @xcite .
this makes it difficult to determine @xmath18 from these processes . for @xmath7
the situation is even worse due to the smaller ckm angle .
if a photon is emitted in addition to the lepton pair , then the mechanism of helicity suppression will not hold any longer and larger branching ratios are expected .
although the process @xmath19 is free from helicity suppression and its branching ratio is around @xmath20 in the sm @xcite , it is likely to be compatible with the decays into lepton pairs only when its efficiency is better than @xmath21 . in section 2
the relevant effective hamiltonian will be given in the sm , and the constituent quark model will be used to give the numerical predictions .
finally , we make some comments in section 3 .
the most important contributions to @xmath22 ( @xmath23 ) stem from the effective hamiltonian which induces the pure leptonic processes @xmath24 .
let us start with the short distance contributions to the latter one , which include box , @xmath25 and photon penguin diagrams .
the feynman diagrams are displayed in fig.1 .
the qcd corrected effective hamiltonian in the sm is @xcite : @xmath26,\label{1}\ ] ] with @xmath27 , @xmath28 , and @xmath29 is the momentum of the lepton pair .
the qcd corrected wilson coefficients @xmath30 , @xmath31 , and @xmath32 are given by misiak @xcite .
because of the lightness of the leptons @xmath33 and @xmath34 , the pure leptonic processes @xmath35 are suppressed by helicity .
if a photon line is attached to any of the charged lines in fig.1 , the situation will be different : helicity suppression is overcome .
when the photon is attached to internal lines , there will be a suppression factor of @xmath36 in the wilson coefficient , since the resulting operators(dimension-8 ) are two orders higher in dimensions than the usual ones(dimension-6 ) .
one may expect that the additional two diagrams displayed in fig.2 ( a ) and ( b ) might also contribute to the decay @xmath37 .
however , our calculation shows that the contribution of fig.2(b ) is quite small and its influence can be neglected in the numerical results .
fig.2(a ) involves a singularity , since the intermediate @xmath38 quark can be on its mass shell .
the influence of this singularity is an artifact of the constituent quark model employed here . if we calculate the contribution of fig .
2(a ) to the differential decay rate anywhere except near the singularity we find it to be negligible ( at most a few percent ) .
therefore we neglect it altogether .
there are also long distance contributions to @xmath39 .
for instance , in the @xmath6 decay , there are cascade processes at the hadronic level : @xmath40 these involve the on - shell @xmath41 and are similar to @xmath42 @xcite .
they are estimated to be of order of @xmath14 , which is about the same order of the short distance contribution .
all these long distance effects , which are shown as sharp peaks in the invariant mass spectrum of the lepton pair , will not be included in the following . now using a simple constituent quark model ( see , for example , @xcite )
, we calculate the amplitude for diagrams with photon emitted from external fermion lines .
first , for diagrams with photon attached to external lepton lines , we get the amplitude proportional to : @xmath43 l.\label{4}\ ] ] using the decay constant definition : @xmath44 we can easily calculate the results of eqn.([4 ] ) .
note that , there is no contribution from the operator @xmath45 in this circumstance , since the definition of the decay constant shows that : @xmath46 for the other two diagrams with photon emitted from the external quark lines ( @xmath47 or @xmath48 ) , the amplitude is proportional to : @xmath49 b \left [ c_9^{eff } \bar l \gamma_\mu l + c_{10 } \bar l \gamma_\mu \gamma_5 l ) \right]\ ] ] @xmath50 b \;(\bar l \gamma_\mu l ) .
\label{4.1}\ ] ] in the constituent quark model , @xmath51 , and @xmath52 .
then applying eqn.([dd ] ) , neglecting terms suppressed by @xmath53 , @xmath54 , we get : @xmath55 \left[\left(c_9^{eff}-\frac{2c_7 m_{b_q}^2}{p^2 } \right)\bar l \gamma ^\mu l+c_{10 } \bar l\gamma^\mu \gamma_5 l \right],\label{5}\ ] ] with @xmath56 after squaring the total decay amplitude , we find that the contribution of diagrams with photon attached to the lepton lines ( eqn.([4 ] ) ) is proportional to @xmath57 , which is negligible in numerical calculations . the main contribution to @xmath58 is thus from the square of eqn.([5 ] ) . performing the phase space integration over one of the two dalitz variables , we get the differential decay width versus @xmath59 : @xmath60 we see that the differential rate is proportional to @xmath61 . using @xmath62 , @xmath63 gev , @xmath64 gev and @xmath65
, the differential decay rate as a function of @xmath66 is displayed in fig.3 , which shows that the contribution from soft photons , corresponding to large @xmath66 region , is negligibly small .
after integration , we also get : @xmath67 for @xmath7 meson decay , we take @xmath68 and @xmath69 gev .
the decay widths are then : @xmath70 if the lifetimes are taken as @xmath71 , @xmath72 @xcite , and if @xmath73mev is used , they correspond to the branching ratios : @xmath74 note that the branching ratios we get for radiative decays are just the same order of the pure leptonic decay @xmath75 .
the decay rates for pure leptonic decays are : @xmath76 using the same parameters as in the radiative decay , the numerical results are : @xmath77 so for the @xmath78 channel , the radiative decay dominates over the dileptonic one , while for the @xmath79 channel , the branching ratio of the radiative decay is a little larger than that for the pure leptonic decay .
we predict the branching ratios in the sm for @xmath80 to be around @xmath1 for @xmath2 and around @xmath3 for @xmath4 , with slightly larger values for @xmath5 . with these predictions
, they will hopefully be detected at lhc - b .
thus provide alternative channels for measuring @xmath81 ( @xmath82 ) . in lhc - b ,
approximately @xmath83 ( @xmath84 ) @xmath7 ( @xmath6 ) mesons are expected per year , therefore there is a good chance of observing the decays considered here .
we conclude with comments on the experimental extractions of the decay constants . in the measurements of the pure leptonic decays ,
processes with additional soft photons emitted ( say , photons with energy less than 50 mev ) might contribute due to the inability of the detector to separate them . from our calculation , we find that these soft photon processes are not important in the radiative dileptonic decays , see fig.3
. since the efficiency of detecting photons may be low , it will be useful for experiments to have the sum of the radiative decay and the pure leptonic processes : @xmath85 the total @xmath86 branching ratios for @xmath87 and @xmath7 decay are almost the same as that of the @xmath88 case .
the research of d .- x . z. is supported in part by grant 5421 - 3 - 96 from the ministry of science and the arts of israel .
the research of g.e . is supported in part by the israel science foundation .
g.e . would like to thank d. atwood , t. nakada , a. soni and the late r. mendel , for very helpful discussions .
99 for review see a.j . buras , m.k .
harlander , heavy flavors " , p58 - 201 , eds . a.j .
buras , m. lindner , world scientific , singapore ; a. ali , nucl .
b , proc . suppl .
39bc , 408 - 425 ( 1995 ) ; s. playfer and s. stone , int
a10 , 4107 ( 1995 ) .
r. ammar , et al . ,
cleo collaboration , phys .
lett . * 71 * , 674 ( 1993 ) ; m.s.alam et al . ,
cleo collaboration , phys .
lett . * 74 * , 2885 ( 1995 ) .
z. ligeti and m. wise , preprint calt-68 - 2029 , hep - ph/9512225 .
l and d .- x .
zhang , preprint technion - ph-96 - 06 , hep - ph/9604378 , phys .
b in press .
a. ali , in `` b decays '' , p67 , ed . s. stone , world scientific , singapore .
g. buchalla and a.j .
buras , nucl .
b400 , 225 ( 1993 ) .
w.s hou and r.s .
willey , phys .
* b202 * ( 1988)591 ; r. grigjanis et al .
b213 * ( 1988 ) 247 ; phys . lett .
* 223 * ( 1989 ) 239 ; phys .
* d42 * ( 1990 ) 245 ; b. grinstein , m.j .
savage and m.b .
wise , nucl .
* b319 * ( 1989 ) 271 ; g. cella , g. ricciardi , a. vicere , phys . lett .
* b258 * ( 1991 ) 212 ; a.j .
buras , m. mnz , phys .
d52 , 186 ( 1995 ) .
m. misiak , nucl . phys . *
b393 * ( 1993 ) 23 ; ( e ) ibid , * b439 * ( 1995 ) 461 .
m. bander , d. silverman and a. soni , phys .
44 , 7 ( 1980 ) ; ibid , 962 ( e ) ; g. eilam , b. margolis , r.r .
mendel and a. soni , phys .
b185 , 145 ( 1987 ) ; d. atwood , g. eilam , and a. soni , hep - ph/9411367 , mod .
phys . lett . in press ; h.y .
cheng , et al .
d51 , 1199 ( 1995 ) .
particle data group , phys .
d50 , 1173 ( 1994 ) .
deshpande , j. trampetic and k. panose , phys .
d39 , 1461 ( 1989 ) ; c.s .
lim , t. morozumi and a.i .
sanda , phys .
b218 , 343 ( 1989 ) ; p.j .
odonnell , and h.k.k .
tung , phys .
d43 , 2067 ( 1991 ) ; n. paver and riazuddin , phys .
d45 , 978 ( 1992 ) .
fig.1 feynman diagrams in standard model for @xmath89 . | we investigate the radiative dileptonic decays @xmath0 within the standard model . using the constituent quark model ,
the branching ratios turn out to be around @xmath1 for @xmath2 and around @xmath3 for @xmath4 , with slightly larger values for @xmath5 .
the differential rate as a function of the dilepton invariant mass is given .
the possibility of using these processes to determine the decay constants of @xmath6 or @xmath7 is discussed .
= -0 cm = -1 cm = 0.2 cm ( 0,0)(0,0 ) ( 330,270)hep - ph/9606444 ( 330,250)technion - ph-96 - 12 ( 330,230)june , 1996 |
the theory of big - bang nucleosynthesis ( bbn ) has been successful in predicting the abundance of light elements in the universe from a single parameter , baryon - to - photon ratio @xmath2 .
the recent results of the wilkinson microwave anisotropy probe ( wmap ) experiment @xcite , however , put this theory into challenge .
the extraordinarily precise results from wmap are put together with the standard bbn ( sbbn ) to predict the abundance of @xmath0 to be @xmath3@xcite if we adopt @xmath4 ( 68 @xmath5 c.l . )
this prediction is inconsistent with the observation of metal - poor halo stars which implies @xmath6 @xcite reported by ryan _
et al_. @xcite .
the inconsistency persists even if we adopt the recent observations , which give the less restrictive constraint of @xmath7 @xcite and @xmath8 @xcite .
this discrepancy can be hardly attributed to the correction of the cross section of nuclear reaction @xcite , and astrophysical solutions are pursued @xcite .
another interesting approach to this problem would be to consider the effects induced by new physics beyond the standard model ( sm ) .
exotic particles which interact with nuclei will open new channels to produce and destroy the nuclei , giving a potential solution to the @xmath0 problem . in this paper
we investigate a possibility that the interaction is initiated by a formation of the bound state of exotic negatively charged massive particles ( champs ) and a nucleus .
( for the other solutions , see @xcite . )
so far the bound - state effects by champs have been attracting many interests @xcite .
for doubly charged particles , see also refs @xcite . in particular , a significant enhancement of a @xmath9li - production rate through @xmath10 by the bound state with @xmath11
was reported @xcite for the first time and recently confirmed @xcite .
this hinders the compatibility between particle physics models and bbn @xcite .
in addition , some nonstandard effects on the abundance of @xmath0 and @xmath1 were also considered in ref .
@xcite and more recently in ref .
@xcite . introducing the champs with the mass of electroweak scale , the authors in ref .
@xcite newly considered several destruction channels of @xmath1 nuclei through the trapping of the champs to show that the abundance of the champs needs to be larger than 0.02 per baryon and that their lifetime has an allowed window between @xmath12 and @xmath13 .
we put the champ bbn scenario in the minimal supersymmetric standard model ( mssm ) with the conservation of @xmath14 parity .
mssm doubles the particle content of the sm by introducing superparticles , which can accommodate the champs .
the champs need a lifetime long enough to sustain the sufficient abundance at the time of nucleosynthesis .
although the @xmath14 parity conservation stabilizes the lightest superparticles ( lsps ) , the observational constraints exclude charged superparticles as a candidate for lsps , which is usually considered to be neutralinos @xmath15 or gravitinos .
a possible candidate of champs is the next - lightest superparticle ( nlsp ) with electric charge , which can have a long lifetime by assuming a small mass difference from the lsp @xcite .
we assume in the present paper that the lsp is a neutralino and the nlsp is a stau @xmath16 , the superpartner of tau lepton @xmath17 .
the staus can decay into neutralino lsp with the hadronic current , through which they also interact with the nuclei .
the gravitino lsp , on the other hand , does not couple with hadronic current .
we consider the bound state of @xmath1 and @xmath18 in the early universe and the subsequent decay chain of nucleus @xmath19 due to the interactions of the two .
the @xmath20 nuclei rapidly decay into @xmath21 nuclei , which are effectively stable in the considered time scale . with the freedom of the mass of stau @xmath22 and its lifetime @xmath23
, we search for the possible solution to the @xmath0 problem that are phenomenologically acceptable .
this paper is organized as follows . in sec .
[ sec.channels ] we overview some new decay channels by stau and estimate their lifetimes . here
we will see that stau - nucleus bound states play an important role in @xmath24be/@xmath24li reducing processes . in sec .
[ sec.numercal_calculation ] we numerically calculate primordial abundances of light elements while taking into account the new channels .
then we will see the possible solution of the @xmath24li problem .
finally , summarization is made in sec .
[ sec.summary ] .
we consider a modification of the sbbn scenario under the mssm .
mssm introduces a set of superparticles as the partners of the particle appearing in the standard model ( sm ) .
the superparticles interact with the standard particles and thus introduce additional decay channels of @xmath1 and @xmath0 to the standard bbn theory .
the additional channels give a possible solution to the problem where the theoretical prediction of the abundance of @xmath1 and @xmath0 , or collectively @xmath25 , exceeds the observational results by a factor of @xmath26 .
we consider in the present paper that the destruction of primordial @xmath25 nuclei is due to their interaction with the negatively charged staus @xmath18 , the superpartner of the tau lepton @xmath27 , which we identify as the next - lightest superparticle ( nlsp ) .
the mass eigenstate of the stau is given by the linear combination of the left - handed stau @xmath28 and the right - handed staus @xmath29 as @xmath30 where @xmath31 is the left - right mixing angle and @xmath32 is the cp - violating phase .
staus have attractive features when considering the destruction of @xmath25 .
first , staus have a negative charge and can form a bound state with nuclei so that they interact efficiently .
second , staus couples with the hadronic current @xmath33 , through which they interact with nuclei as we see below .
third , staus can be abundant at the time of bbn .
they can acquire the sufficientlty long lifetime when the staus and lsps , which we assume as the nutralinos , have a mass difference tiny enough .
the interaction of staus is described by the lagrangian @xmath34 where @xmath35 is the fermi constant , @xmath36 and @xmath37 are the chiral projection operators , @xmath38 , and @xmath39 and @xmath40 are the coupling constants .
these coupling constants are written in terms of the @xmath41 gauge - coupling constant @xmath42 ; when the neutralino is bino like , for instance , they are given by @xmath43 where @xmath44 is the weinberg angle .
the interaction lagrangian ( [ stau_lagrangian ] ) give rise to the following decay channels ( see fig . [ free_diagrams ] ) : @xmath45 , ( b ) @xmath46 , and ( c ) @xmath47.,width=302 ] process ( [ eq : stau - decay-2body ] ) has a typical lifetime @xmath48 , process ( [ eq : stau - decay-3body ] ) has @xmath49 , and ( [ eq : stau - decay-4body ] ) has @xmath50 .
since the bbn takes place @xmath51 after the big bang , the staus will decay entirely before bbn unless the channel ( [ eq : stau - decay-2body ] ) is closed .
our scenario therefore requires @xmath52 .
note that the channel ( [ eq : stau - decay-3body ] ) also closes when @xmath53 is less than the pion mass @xmath54 .
although the required lsp - nlsp mass difference is small compared to the typical mass of lsp which is @xmath55 , it is preferable in attributing the dark matter ( dm ) to the neutralino lsps since it allows the lsp - nlsp coannihilation . with this tiny @xmath53 ,
the neutralino naturally becomes a cold dark matter instead of warm or hot dark matter @xcite even though it is produced non thermally .
hence our model is free of the constraints from the large - scale structure formation of the universe . in this section ,
we consider the stau - nucleus interaction processes that are relevant to the primordial bbn .
three processes are discussed : ( 1 ) the hadronic - current interaction , ( 2 ) stau - catalyzed fusion , and ( 3 ) internal conversion of stau - nucleus bound state .
we consider the lifetimes of each process because it is crucial to understand the impact upon the modification of bbn .
staus can interact with the nuclei through the hadronic current and thereby alter the bbn processes .
the abundances of @xmath56 are changed by the new decay channels : @xmath57 the process @xmath58 does not occur since @xmath24he is very unstable , while the pions can be either real or virtual ; here the virtual pion should actually be regarded as a hadronic current propagating between the stau and the nucleus .
the pions produced in the process ( [ eq : stau - decay ] ) also change the proton - neutron ratio and thereby change the primordial abundance of the light elements .
we present the lifetime of the free stau in fig .
[ free_lifetime ] as functions of @xmath53 @xcite . here
we take @xmath59 , @xmath60 , and @xmath61 . .
here we take @xmath59 , @xmath62 , and @xmath63 .
the hadronic decay is dominant for @xmath64 while the leptonic decay is exclusively allowed for @xmath65 .
, height=188 ] another process to destroy the @xmath0/@xmath1 is nuclear fusion catalyzed by staus .
a nucleus has a coulomb barrier which normally prevents the nuclear fusion , while the barrier is weakened when a stau is captured to a state bound to the nucleus .
the nuclear fusion is thus promoted by forming a stau - nucleus bound state .
the stau serves as a catalyst and is left out as the fusion proceeds through .
this stau - catalyzed fusion process provides the following decay channels : @xmath66 the lifetime of the stau - catalyzed fusion is estimated to be longer than @xmath67 @xcite .
we follow ref .
@xcite to calculate the stau - catalyzed fusion rate .
be ( @xmath24li).,height=188 ] the interaction between a stau and a nucleus proceeds more efficiently when they form a bound state ( see fig . [ bound_diagram ] ) @xcite due to two reasons : ( 1 ) the overlap of the wave functions of the two becomes large since the stau and particle are packed in the small space , ( 2 ) the small distance between the two allows virtual exchange of the hadronic current even if @xmath65 .
the stau - nucleus bound state decays through the following processes : @xmath68 where the parentheses denote the bound states .
we note that we introduce not only reaction ( [ internal_be ] ) , but also reaction ( [ internal_li ] ) .
the @xmath21 nucleus can also decay into @xmath69 via @xmath70 decay with the lifetime @xmath71 .
we do not take this process into account since this process is much slower than the scattering process ( [ he6_back_ground ] ) .
the lifetime of the internal conversion @xmath72 is obtained from the lagrangian ( [ stau_lagrangian ] ) as @xmath73 where @xmath74 is the overlap of the wave functions of the staus and the nucleus , @xmath75 and @xmath76 here @xmath77 for the process ( [ internal_be ] ) and @xmath78 for ( [ internal_li ] ) . the following approximations are applied to evaluate the lifetime further .
we estimate the overlap of the wave functions in eq .
( [ eq : tauic ] ) by assuming that the bound state is in the s state of a hydrogen like atom , and obtain @xmath79 where @xmath80 is the radius of the nucleus
. the matrix element of the nuclear conversion appearing in eq .
( [ eq : sigmav ] ) is evaluated by the @xmath81 value of the corresponding @xmath70 decay obtained from the experiments .
the experimental @xmath81 value is available for @xmath82 but not for @xmath83 , however .
we assume that the two processes have the same @xmath81 value .
as long as we consider the quantum numbers of the ground state of li7 and he7 we can expect a gamow - teller transition can take place since they are similar with those of he6 and li6 and we know that they make a gamow - teller transition .
the gamow - teller transition is superallowed and has a similar @xmath81 value to the fermi transition such as @xmath84 .
our new effects have been treated as if @xmath24li or @xmath24be in its bound state would have an effectively new lifetime which is caused by the virtual exchange of the hadronic current with a stau .
thus this new process is not the two - bodies scattering .
so , there is no corresponding astrophysical s factor in these processes .
the evaluated lifetimes of reactions ( [ internal_be ] ) and ( [ internal_li ] ) under these approximations are presented in fig .
[ bound_lifetime ] as functions of @xmath53 . there
we take @xmath59 , @xmath60 , and @xmath63 for both reactions . .
top panel : @xmath85 , bottom panel : @xmath86 .
we take @xmath87 = 300gev , @xmath88 , and @xmath61 in both figures . ,
title="fig:",width=302 ] .
top panel : @xmath85 , bottom panel : @xmath86 .
we take @xmath87 = 300gev , @xmath88 , and @xmath61 in both figures .
, title="fig:",width=302 ] we find that the lifetime of the internal conversion process is in the order of @xmath89 .
the lifetime of stau-@xmath0 bound state diverges around @xmath90 , below which the internal conversion is kinematically forbidden .
as we will see later , the internal conversion processes are dominant over the other processes .
in this section , we study the effectiveness of new decay channels on the @xmath24li problem by numerical calculation . to do this , we choose the abundance of stau @xmath91 at freeze out time and mass difference @xmath53 as free parameters , since these values are sensitive to the abundance of @xmath24be and @xmath24li . here
@xmath92 is entropy density .
the number of @xmath24li interacted with the stau is determined by the number density of the stau .
the hadronic decay rate of stau is mainly determined by the mass difference .
the decay rate is also determined by the stau mixing angle @xmath31 , cp violating phase @xmath32 , and neutralino mass . as we showed in @xcite , however , the effects by these parameters are much less than mass difference .
we estimate the number density of bound states by using saha equation , @xmath93 here , @xmath94 , @xmath95 , and @xmath96 denote the number densities of the bound state , the stau , and the nucleus , respectively ; @xmath97 , @xmath98 , and @xmath99 denote the nucleus mass , binding energy of the bound state , and the temperature of the universe , respectively . the saha equation is valid only when the expansion rate of the universe is much smaller than the formation rate of the bound state .
this condition is not quite satisfied in our case .
we will explain that our results are qualitatively acceptable at the end of this section . for more detailed discussion , see @xcite .
@xmath100 plane .
the white region is the parameter space which is consistent with all the observational abundance including @xmath101li / h=@xmath102 @xcite .
the regions enclosed by dotted ( green ) , dashed ( light blue ) , and dash - dotted ( purple ) lines are excluded by the observations on @xmath103he , d and @xmath9li , respectively .
the thick - dotted line represents a yield value of stau whose daughter particle , neutralino , accounts for all the dark matter component .
here we took @xmath104 , @xmath87 = 300 gev , @xmath88 and @xmath61 .
, width=321 ] in fig .
[ numerical_result ] we show the parameter region of @xmath53 and @xmath100 allowed by the observational light - element abundances , where we take @xmath104 , @xmath87 = 300 gev , @xmath88 and @xmath61 . the white region is the parameter space which is consistent with all the observational abundance including @xmath101li to hydrogen ratio ( @xmath101li / h ) .
the regions enclosed by dotted ( green ) , dashed ( light blue ) , and dash - dotted ( purple ) lines are excluded by the observations on the mass fraction of @xmath103he ( @xmath105 ) , deuterium to hydrogen ratio ( d / h ) and @xmath9li to @xmath101li ratio ( @xmath9li/@xmath101li ) , respectively .
we adopt the following observational constraints on primordial light element abundances according to a discussion in ref .
see also the other recent observational values of @xmath105 in ref .
@xcite . ] li/@xmath101li , see also the discussion in @xcite . ]
@xmath106 the thick - dotted line represents a yield value of stau whose daughter particle , neutralino , accounts for all the dark matter component .
this line is given by the yield value of dark matter @xmath107 with @xmath108 ( @xmath109 c.l . )
@xcite , here @xmath110 is @xmath87 .
@xmath111 must be smaller than this value , in order to prevent the overclosure of the universe .
the cosmologically interesting region is below the line .
we can find another white region even if we adopt the more restrictive value of @xmath101li / h shown in ref .
the upper central region is excluded by the observational constrains on d / h and @xmath112he due to charged pions emitted from decaying staus @xcite . in the current model ,
no hadrodissociation processes of light elements occur .
the qualitative feature of the allowed region is expalained from the following physical consideration .
first , staus need to have lifetimes @xmath113 longer than the time required to form the bound state of a stau and a @xmath24be .
the required time @xmath114 is estimated from the binding energy @xmath115 as @xmath116 .
figure [ free_lifetime ] shows @xmath117 for @xmath118 and hence the allowed region appears only in this region .
second , the yield value of staus @xmath100 needs to be large compared with that of @xmath24li which we denote by @xmath119 .
we estimate @xmath119 from the hydrogen to entropy ratio @xmath120 and the constraint ( [ li7/h ] ) as @xmath121 we thus need @xmath122 , and again the allowed region appears in this region .
third , the excessive destruction of @xmath24li by the process ( [ internal_li ] ) needs to be avoided due to the constraint ( [ li6/li7 ] ) .
this condition puts a limit to the formation rate of the bound state of a stau and a @xmath24li .
we then need either @xmath123 ( here we use @xmath124 @xcite ) , or @xmath100 to be small enough .
the former condition leads to @xmath125 , although this region is subject to the strong restriction considered in the previous paragraph .
the latter imposes an upper limit on @xmath100 in the region @xmath126 , and fig .
[ numerical_result ] suggests that this limit is in the order of @xmath127 .
the tininess of @xmath111 shows that the internal conversion processes ( [ internal_be ] ) and ( [ internal_li ] ) are dominant over other processes such as the pion exchange and the stau - catalyzed fusion .
we can confirm this dominance by an explicit calculation of the rates of these processes .
we used the saha equation in our calculation although the formation rate is comparable to the expansion rate .
calculation using the boltzmann equation will give a lower number density of the bound state @xmath94 and consequently shift the allowed region upward in @xmath128 in fig .
[ numerical_result ] .
we have investigated a possible solution of the @xmath24li problem in a framework of mssm , in which the lsp and the nlsp are neutralino and stau , respectively , and have a tiny mass difference of @xmath129 .
the staus then survive throughout the bbn era as shown in fig .
[ free_lifetime ] , and provide additional decay processes as mentioned in sec . [ interactions_of_staus ] to reduce the primordial @xmath24li abundance . taking the three new processes into account , we numerically calculated the primordial abundance of light elements varying the lsp - nlsp mass difference and the abundance of stau .
taking @xmath130 , we obtained the parameter region consistent to the observed @xmath24li abundance .
though we have shown that the internal conversion is very important for the calculation of the primordial abundance , we need to improve our calculation for better accuracy .
first , we need to include reaction processes such as @xmath131 .
this process can change the prediction of @xmath132li and hence change the allowed region .
second , we should calculate the number density of the bound states not by the saha equation ( [ saha_eq ] ) but by the boltzmann equation . at the formation temperature ,
the capture rate is less than the expansion rate of the universe .
therefore we will obtain a lower number density of the bound states and consequently the upward shift of the allowed region .
third , we should explore other values of the parameters @xmath31 , @xmath32 , and @xmath133 , which affect the lifetime of stau and also those of the bound states .
we leave these for our future works .
j.s . thanks f. takayama for valuable discussion .
the work of k.k . was supported in part by pparc grant no .
pp / d000394/1 , eu grant no .
mrtn - ct-2006 - 035863 , the european union through the marie curie research and training network universenet " , mrtn - ct-2006 - 035863 .
the work of j.s . was supported in part by the grant - in - aid for the ministry of education , culture , sports , science , and technology , government of japan ( no . 17740131 and 18034001 ) .
the work of m.y . was financially supported by the sasakawa scientific research grant from the japan science society .
1 d. n. spergel et al . , astrophys .
j. suppl .
* 170 * , 377 ( 2007 ) a. coc , e. vangioni - flam , p. descouvemont , a. adahchour and c. angulo , astrophys .
j. * 600 * , 544 ( 2004 ) . s. g. ryan , t. c. beers , k. a. olive , b. d. fields and j. e. norris , astrophys . j. lett . * 530 * , l57 ( 2000 ) .
r. h. cyburt , b. d. fields and k. a. olive , phys .
b * 567 * , 227 ( 2003 ) .
p. bonifacio et al .
, astron .
astrophys . *
390 * , 91 ( 2002 ) .
j. melendez and i. ramirez , astrophys .
j. * 615 * , l33 ( 2004 ) .
r. h. cyburt , b. d. fields and k. a. olive , phys . rev .
d * 69 * , 123519 ( 2004 ) c. angulo _ et al .
_ , astrophys .
j. * 630 * , l105 ( 2005 ) .
a. j. korn _ et al .
_ , nature * 442 * , 657 ( 2006 ) .
[ arxiv : astro - ph/0608201 ] .
k. ichikawa and m. kawasaki , phys .
d * 69 * , 123506 ( 2004 ) ; k. ichikawa , m. kawasaki and f. takahashi , phys . lett .
b * 597 * , 1 ( 2004 ) .
k. jedamzik , phys . rev .
d * 70 * , 063524 ( 2004 ) ; k. jedamzik , k. y. choi , l. roszkowski and r. ruiz de austri , jcap * 0607 * , 007 ( 2006 )
. k. kohri , t. moroi and a. yotsuyanagi , phys .
d * 73 * , 123511 ( 2006 ) .
r. n. cahn and s. l. glashow , science * 213 * , 607 ( 1981 ) .
r. n. boyd , k. takahashi , r. j. perry and t. a. miller , science .
vol . * 244 * , 1450 ( 1989 ) ; a. de rujula , s. l. glashow and u. sarid , nucl .
b * 333 * , 173 ( 1990 ) ; s. dimopoulos , d. eichler , r. esmailzadeh and g. d. starkman , phys . rev .
d * 41 * , 2388 ( 1990 ) .
m. pospelov , phys . rev
. lett .
* 98 * , 231301 ( 2007 ) .
k. kohri and f. takayama , phys . rev .
d * 76 * , 063507 ( 2007 ) .
m. kaplinghat and a. rajaraman , phys .
d * 74 * , 103004 ( 2006 ) ; r. h. cyburt , j. ellis , b. d. fields , k. a. olive and v. c. spanos , jcap * 0611 * , 014 ( 2006 ) ; f. d. steffen , aip conf .
proc . * 903 * , 595 ( 2007 ) .
k. hamaguchi , t. hatsuda and t. t. yanagida , arxiv : hep - ph/0607256 .
d. fargion and m. khlopov , arxiv : hep - ph/0507087 ; d. fargion , m. khlopov and c. a. stephan , classical quantum gravity * 23 * , 7305 ( 2006 ) ; k. m. belotsky , m. y. khlopov and k. i. shibaev , gravitation cosmol .
* 12 * , 93 ( 2006 ) .
k. hamaguchi , t. hatsuda , m. kamimura , y. kino and t. t. yanagida , phys .
b * 650 * , 268 ( 2007 ) .
m. kawasaki , k. kohri and t. moroi , phys .
b * 649 * , 436 ( 2007 ) . c. bird , k. koopmans and m. pospelov , arxiv : hep - ph/0703096 .
t. jittoh , j. sato , t. shimomura and m. yamanaka , phys .
d * 73 * , 055009 ( 2006 ) .
w. b. lin , d. h. huang , x. zhang and r. h. brandenberger , phys .
lett . *
86 * , 954 ( 2001 ) ; j. hisano , k. kohri and m. m. nojiri ; phys .
b * 505 * , 169 ( 2001 ) .
k. kohri and f. takayama in preparation .
m. fukugita and m. kawasaki , astrophys .
j. * 646 * ( 2006 ) 691 .
m. peimbert , v. luridiana and a. peimbert , arxiv : astro - ph/0701580 . m. h. pinsonneault , t. p. walker , g. steigman and v. k. narayanan , astrophys .
j. * 527 * , 180 ( 1999 ) . m. h. pinsonneault , g. steigman , t. p. walker and v. k. narayanans , astrophys .
j. * 574 * , 398 ( 2002 ) .
t. k. suzuki and s. inoue , astrophys .
j. * 573 * , 168 ( 2002 ) ; e. rollinde , e. vangioni and k. a. olive , astrophys .
j. * 651 * , 658 ( 2006 ) ; v. tatischeff and j. p. thibaud , astron . and astrophys . in press
[ arxiv : astro - ph/0610756 ] .
t. kanzaki , m. kawasaki , k. kohri and t. moroi , phys .
d * 75 * , 025011 ( 2007 ) .
y. i. izotov , t. x. thuan and g. stasinska , arxiv : astro - ph/0702072 . j. m. omeara , s. burles , j. x. prochaska , g. e. prochter , r. a. bernstein and k. m. burgess , astrophys . j. * 649 * , l61 ( 2006 ) .
m. asplund , d. l. lambert , p. e. nissen , f. primas and v. v. smith , astrophys .
j. * 644 * , 229 ( 2006 ) .
m. h. reno and d. seckel , phys .
d * 37 * , 3441 ( 1988 ) ; k. kohri , phys .
d * 64 * , 043515 ( 2001 ) ; m. kawasaki , k. kohri and t. moroi , phys .
d * 71 * , 083502 ( 2005 ) ; phys .
b * 625 * , 7 ( 2005 ) ; k. jedamzik , phys .
d * 74 * , 103509 ( 2006 ) . | modification of standard big - bang nucleosynthesis is considered in the minimal supersymmetric standard model to resolve the excessive theoretical prediction of the abundance of primordial lithium 7 .
we focus on the stau as a next - lightest superparticle , which is long lived due to its small mass difference with the lightest superparticle .
it provides a number of additional decay processes of @xmath0 and @xmath1 .
a particularly important process is the internal conversion in the stau - nucleus bound state , which destroys the @xmath0 and @xmath1 effectively .
we show that the modification can lead to a prediction consistent with the observed abundance of @xmath0 . |
very - long - baseline interferometry ( vlbi ) provides the most accurate astrometric measurements of celestial objects currently attainable .
vlbi astrometry has been used , among other applications , to : define the most nearly inertial reference frame available for the positions of celestial objects ( e.g. , * ? ? ?
* ; * ? ? ?
* ) ; tie an inertial ( extragalactic ) reference frame to a planetary ephemeris via observations of pulsars ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) ; characterize motions and other properties of the earth ( e.g. , * ? ? ?
* ; * ? ? ?
* ) ; study positions and motions of maser spots in galaxies as a means to estimate black - hole masses ( e.g. , * ? ? ?
* ; * ? ? ?
* ) and distances to other galaxies @xcite ; and test general relativity via measurements of solar gravitational deflection ( e.g. , * ? ? ?
* ; * ? ? ?
* ) . in this paper
we describe the analysis of astrometric vlbi observations of the rs cvn binary star i m pegasi ( i m peg ; ) , the radio - bright star which served as the `` guide star , '' and hence as the positional reference , for the nasa / stanford _ gravity probe b _ ( ) experiment .
was designed to measure the geodetic and frame - dragging effects predicted by general relativity ( see paper i , * ? ? ?
* ) , as manifested in secular changes in the spin - axis orientations of four gyroscopes placed within a spacecraft in a low - altitude , polar orbit about the earth .
the spacecraft also had an on - board telescope equipped with a tracking system designed to keep the guide star , , at the center of the telescope field of view , and thereby provide a directional reference : the relativistic effects were estimated by measuring the drift rates in the spin - axis directions of the four gyroscopes relative to the direction to .
thus , if any motions of on the sky relative to an inertial frame were not accounted for sufficiently accurately , they could map directly into the gyro drift - rate signals and thereby corrupt the relativistic measurements .
( ideally the guide star would have been a distant quasar , which could be treated as effectively motionless on the sky , rather than a star in our galaxy , but the on - board telescope could only track objects brighter than about 6th magnitude , so a relatively bright star was the only possible choice . a list of criteria that had to be met by the guide star , and a description of the selection process that led to the choice of , is in . )
our vlbi observations of were therefore undertaken to determine the motions of the star relative to an inertial frame so that the relativistic effects included in the gyro measurements could be properly separated .
the goal of our vlbi observations was to determine , relative to an inertial frame , the proper motion of with standard error 0.14 or less in each of the north - south and east - west directions .
this accuracy goal was based upon an error projection for the full experiment prior to launch in april 2004 and the desire at the time for the uncertainty in the proper motion of not to constitute a significant source of error for the experiment . in this paper
we present a new approach for astrometric vlbi data analysis that we developed to meet the accuracy goal for .
we then compare the results from this approach with the results from two well - established vlbi analysis techniques : phase - referenced mapping ( prm ) and `` phase connection '' followed by parametric model fitting ( pmf ) .
our new approach is a combination of these two standard approaches that provides many of the benefits of each .
all three approaches use _ differential _ vlbi astrometry , in which the target source ( i.e. , the source of astrometric interest ) is observed alternately with at least one extragalactic , compact reference source nearby on the sky to provide model corrections in the analysis of the target - source data . the way these corrections are applied differs among the approaches , but in all cases the astrometric measurements of the target source are relative to the position(s ) of the reference source(s ) . in [ vlbiobs ]
we briefly describe the vlbi observables used in our analysis .
we then discuss some basic aspects of the prm and pmf analysis techniques in [ prmaps ] and [ pmfit ] , respectively . in
[ merged ] we describe our new `` merged '' analysis technique and then our implementation of it .
we compare the results from the three techniques in
[ compare ] , and offer corresponding conclusions in [ conclusion ] .
in vlbi observations of a compact source , the total fringe phase ( or visibility phase , or interferometric phase , or simply `` phase '' ) , @xmath0 , is the phase at frequency @xmath1 and time @xmath2 that is associated with the difference in arrival times of signals received from the source at two antennas in a vlbi array .
we describe @xmath0 as : @xmath3 + 2\pi n(\omega , t ) \label{phi}\ ] ] where @xmath4 , the `` geometric delay , '' is the difference in the signal arrival times in vacuum at the two antennas ; @xmath5 represents the difference in the instrumental delays ( including clock behavior ) at the two antenna sites ; @xmath6 represents the difference in signal propagation times to the two antennas due to all atmospheric effects , including the contributions of the hydrostatic ( or `` dry '' ) atmospheric constituents , atmospheric water vapor , and the ionosphere ; @xmath7 is the delay contribution from source structure , i.e. , from the non - pointlike brightness distribution of the source ; @xmath8 represents the ( thermal ) noise contribution to the phase measurement ; and @xmath9 represents the integer number of `` ambiguities , '' or `` phase wraps , '' included in the measurement .
information about the number of phase wraps associated with a fringe phase is not inherent to a measurement of fringe phase , and in general a measured fringe phase can be defined such that @xmath10 , or @xmath11 . there is a fringe phase associated with each antenna pair , or `` baseline , '' within the vlbi array .
the `` fringe rate '' is the partial derivative of the fringe phase with respect to time .
the `` multiband delay '' is a measured approximation of the group delay , which is the partial derivative of the fringe phase with respect to frequency .
to lowest order ( i.e. , neglecting , e.g. , relativistic effects ) , the geometric delay is given by : @xmath12 \label{geom}\ ] ] where @xmath13 is the speed of light in vacuum , @xmath14 is the 3-dimensional vector between the two antennas of the baseline , and @xmath15 is the unit vector in the direction of the observed source .
all of the astrometric information in the measured fringe phase is contained in @xmath4 .
( while we do not show the relativistic contributions to @xmath4 in equation [ geom ] , we do include those contributions throughout our analyses .
we also account for earth motions relative to the solar - system barycenter during the time between signal arrival at the two antennas . )
the model we used for @xmath6 can be more explicitly described by : @xmath16 where the `` @xmath17 '' subscript refers to site a ; @xmath18 is the propagation delay through the atmosphere at zenith under the assumption that the atmosphere is in hydrostatic equilibrium ; @xmath19 is the additional propagation delay through the atmosphere at zenith due to tropospheric water vapor @xcite ; @xmath20 and @xmath21 are , respectively , the `` mapping functions '' that project ( i.e. , scale ) the `` dry '' and `` wet '' delays at zenith to the line - of - site elevation , @xmath22 , of the observed source ; and @xmath23 denotes the line - of - site contribution of the ionosphere to phase delays in the direction of the observed source .
the basic idea behind the prm technique is to use the data from the observed reference sources to account for otherwise unmodeled fluctuations in instrumental or atmospheric delays , as well as for other model errors .
specifically , the `` residual '' fringe phases and rates , i.e. , the differences between the measured and a priori model values of these quantities , are obtained for the reference sources and then temporally and sometimes also spatially interpolated to the observation time and sky position of the target source to estimate the effects of model errors on the target - source observables ( see , e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
typically the cycle time over which the reference sources and target source are observed is relatively short , from several seconds to several minutes . to the extent that reference - source structure
is properly accounted for and the reference - source residuals are properly interpolated to the observation time and sky position of the target source , the remaining residual components in the target - source data can be attributed primarily to measurement noise , unmodeled target - source structure , and a position offset of the target source relative to the a priori model position .
perhaps the most important feature of the prm technique is that the target - source data can be coherently integrated over the entire span of an observing session .
thus , even very weak sources with flux densities well under 1 mjy can be detected and imaged with vlbi via this technique .
henceforth in this paper , a `` scan '' refers to a single continuous observation of a particular source .
fringe - phase and fringe - rate measurements from successive reference - source scans can provide proper model fringe - phase adjustments for intervening target - source data only if the change in the model errors between the scans is @xmath24 , so that the number of @xmath25 phase wraps between successive reference - source scans can be accurately tracked .
for that reason , the cycle time over which the reference and target sources are observed should be as short as possible .
on the other hand , the reference - source scans must be long enough to provide reliable fringe - phase measurements from single scans , and the scan times for the target source should be long enough to ensure sufficient coherent integration time over the course of the observing session to produce an image with acceptably high signal - to - noise ratio ( snr ) .
these trade - offs must be balanced to determine the `` optimum '' scan and cycle times in the observation schedule .
another beneficial feature of the prm approach is that it is relatively quick and efficient for obtaining high - accuracy astrometry , in large part due to the software packages now readily available . in our analyses we used almost exclusively the astronomical image processing system ( aips ) provided by the national radio astronomy observatory ( nrao ) to produce our phase - referenced maps .
we followed the guidelines of @xcite .
specifically , we : 1 . incorporated the instrumental phase and amplitude calibrations routinely provided in log files from individual antennas .
we also unweighted faulty data based upon information in these files and from operator and correlator reports .
incorporated an additional constant ( over time ) phase adjustment to the calibration for each antenna based upon `` fringe fitting '' ( via aips task fring ) of data from a selected reference scan .
sometimes we used multiple reference scans to provide such phase calibration adjustments for all antennas .
3 . ran a `` global fringe fit '' ( via aips task fring ) on all reference - source data while applying the phase and amplitude calibrations obtained so far .
4 . ran aips task imagr , with the fits from step 3 as calibration , to generate `` self - cal '' maps of the reference sources ( see paper ii , * ? ? ?
we then used these self - cal maps to refine the amplitude and phase calibration as well as the data flagging .
in general we repeated steps 3 and 4 several times .
5 . used aips task bpass to further refine the phase and amplitude calibrations as a function of frequency .
. used aips task imagr with the final calibrations from steps 1 through 5 to obtain a phase - referenced map of the target source .
unfortunately , the use of the prm technique for astrometry also poses some challenges , especially for campaigns such as ours in which the observations are made over many years . for example , one must take great care to assure that model components such as site positions , antenna axis offsets , and earth orientation parameters ( eops , i.e. , x- and y- pole positions , ut1@xmath26utc , and nutation angles in longitude and obliquity ; see , e.g. , @xcite ) are modeled consistently and correctly throughout all observation sessions .
( for example , if updated values of axis offsets for antennas are used in processing data from later observing sessions , then the new values will have to be incorporated into a re - analysis of data from earlier observing sessions to avoid possible systematic errors . ) furthermore , post - processing adjustments to model parameters can be quite cumbersome in some analysis packages , including aips , especially if the adjustments have to be made to a large number of experiments .
in addition , without the use of multiple reference sources and special interpolation routines to handle the spatial ( in particular , elevation - angle ) dependencies of atmospheric delays , a simple temporal interpolation of reference - source fringe phases introduces model - correction errors that scale roughly with the angular separation between the reference source and the target source .
thus the basic prm technique commonly loses viability for an angular separation between reference and target sources larger than a few degrees .
finally , the prm technique offers no inherently good way to assess the effects of systematic errors ( such as those due to inaccurate modeling of the atmosphere ) .
reliable estimates of the true accuracy of the astrometric results can therefore be difficult to obtain ( although see @xcite for a comprehensive assessment specifically for the vlba and evn arrays ) . our development and use of the `` merged '' technique described in
[ merged ] of this paper was motivated in part by our desire to overcome these drawbacks .
the pmf technique for analysis of vlbi data is , in all implementations with which we are familiar , essentially the method of `` weighted - least - squares '' estimation . for each reference - source and target - source scan ,
the total measured fringe phase and fringe rate are estimated for each baseline at an epoch near the center of the scan period .
any @xmath25 `` phase jumps '' between successive phase measurements are then resolved ( or flagged as unresolvable ) for each source and baseline .
the overall integer number of @xmath25 differences between the phases from different sources is also resolved .
finally , the complete collection of `` phase - connected '' data is fit to a model in which corrections to a wide range of parameters including instrumental delays , propagation delays through the atmosphere , and source positions are simultaneously estimated . to avoid the often onerous task of determining the change in the number of @xmath25 phase wraps between successive scans prior to model fitting , multiband delays ( e.g. , * ? ? ?
* ) are commonly used in place of fringe phases as the principal observables with the pmf technique .
the use of multiband delays comes at a significant cost in statistical ( i.e. , snr - derived ) measurement accuracy relative to the use of fringe phases , but the use of only bright radio sources as well as observations over wide spectral bandwidths can reduce this cost .
the pmf approach with multiband delays is used by the geodetic vlbi community in virtually all of their data analyses , including their studies to define the international celestial reference frame and track apparent motions of extragalactic sources @xcite . in general , the radio sources chosen for such studies are the brightest and most compact known .
the pmf technique has several advantages over the prm technique .
for example , since total measured fringe phases and rates ( rather than residual fringe phases and rates ) are processed in the analysis , one can make improvements or changes to a priori models used with multi - year data sets much more easily .
furthermore , the use of data from multiple reference sources to account for spatial ( e.g. , elevation - angle - dependent ) as well as temporal phase variations is straightforward and readily implemented with existing software packages .
in addition , because the pmf technique is a method of weighted - least - squares estimation of parameters that are treated as gaussian random variables , one can readily obtain a measure of the sensitivity of one estimated parameter ( such as a source declination ) to the variation in another ( such as the propagation delay through a site atmosphere ) by evaluating the covariance matrix obtained with the parameter estimates .
the effects on estimated parameters of other model changes are also straightforward to evaluate , so various sensitivity studies to assess systematic errors can be readily implemented .
unfortunately , the pmf technique also has its limitations and problems .
unlike the prm technique , the pmf technique can make use of only those data for which fringe phases for a baseline are reliably detected in a single scan .
thus the technique is only well suited for relatively bright sources .
furthermore , the pmf technique offers no direct mechanism to account for source structure .
one can assess fringe amplitudes and `` closure '' phases @xcite to identify the presence of significant structure , but the model corrections for structure have to be generated in a separate process .
finally , when accuracy requirements dictate that the pmf technique be used with fringe phases rather than multiband delays , the integral number of @xmath25 phase wraps between neighboring scans must be determined , which can be a very labor - intensive process and sometimes is not even possible .
we desired the most accurate astrometry that we could attain from our vlbi observations of . unfortunately ,
neither the prm nor the pmf analysis technique adequately met our needs . the prm technique , as implemented with aips , relied upon insufficiently accurate models for our astrometric demands , in particular for @xmath6 and also for eops and other parameters .
these models were difficult to reliably correct properly within aips . also , the use of correct , consistent values for some model parameters , such as antenna coordinates and axis offsets , would have been somewhat burdensome to implement and assure over the 8.5 year span of our vlbi observations ( which are described further in [ obs ] ) . in addition , we wanted to make full use of observations of a second reference source to model the effects of atmospheric gradients on our measured fringe phases , and we had no satisfactory software tools with this capacity available to us with the prm technique .
finally our use of the prm technique made difficult a robust assessment of several possible sources of systematic error in our astrometric results , including , for example , our sensitivity to errors in our a priori eop values or in the various components of our atmospheric model .
we also could not use the pmf technique with all of our data , because during several observing sessions the radio emissions from were too weak ( @xmath27 mjy ) for the star to be detected in any single scan . in several other sessions
, the star was detected only intermittently in single scans and in those cases only on the most sensitive baselines .
furthermore , our principal reference source , , which we selected because of its close proximity on the sky to ( 0.7 angular separation ) and very high radio brightness , has a complex and evolving structure that made necessary the inclusion of structure corrections in our model .
the pmf technique can not provide such corrections , so if we had relied upon the pmf technique for our astrometry , then we still would have needed to image separately and generate structure models for it .
likewise , had highly time - variable and sometimes complex structure ( see paper vii , @xcite , and @xcite ) , so we would have had to image this source separately , too . our solution to overcome the drawbacks associated with each of the prm and pmf techniques was to merge the two approaches in a way that gave us the advantages and shed many of the disadvantages of both .
the basic approach behind the ma technique is first to follow the approach of the pmf technique to obtain model corrections .
one bases these corrections on data from only those radio sources usually the reference sources whose emissions are sufficiently bright that they can be reliably detected in a single scan on many baselines and hence are well suited for analysis with the pmf technique . for improved accuracy , structure corrections for these sources
can be obtained separately and included in the analysis .
the data from the sources usually the target sources that are not well suited for the pmf technique are left unweighted but used as `` placeholders '' so that model corrections corresponding to the observation times and positions of those sources are readily attainable via interpolation of the model results obtained from the data that are weighted . in effect , more accurate models for the target - source observables are obtained than with the conventional prm technique , because one has full use of the pmf - technique tools and models that are ( in general ) superior to those available with the prm technique . the corrected models from the pmf technique are then used with the prm technique , in place of the models and corrections that would have otherwise been used in a conventional implementation of the prm technique .
our ma technique thus combines the superior model - correction capabilities of the pmf technique with the superior sensitivity of the prm technique .
here we provide details about the way that we implemented the ma technique to analyze the data obtained for .
our observation strategy , as well as a list of the vlbi antennas we used , is provided in .
we repeat some relevant points about our observations here .
we had 35 sessions of vlbi observations of between january 1997 and july 2005 .
we made all of these observations over a continuum near 8.4 ghz ( @xmath28 cm ) .
typically we used data from 12 to 14 vlbi antennas located around the world .
most of the antennas we used are within the united states .
we sequenced our observations through either three or four sources in a repeating 5.57 minute cycle .
the three sources we observed in all 35 sessions were : our target source , ; our principal extragalactic reference source , ( b2251 + 158 ) ; and a secondary extragalactic reference source , .
we added the fourth source , , to our cycle for the final 12 sessions .
we treated this source as a second target source and used it to place bounds on the sizes of apparent changes in the positions of our reference sources due to structural evolution ( i.e. , changes in the brightness distributions of the reference sources ) .
we also used it to test our different astrometric analysis techniques : we expected ( unlike ) to be a stationary ( extragalactic ) object on the sky , so any estimated motions of the source would be a measure of the experimental errors inherent to our astrometric technique .
figure [ figskypos ] shows the relative positions on the sky of the four sources we observed .
note that is between , and approximately collinear with , the reference sources and .
we intentionally selected our reference sources with such alignment to obtain more accurate models for data from the interpolation of the models fit to our reference - source data : these models account approximately for the elevation - angle dependencies of our measured fringe phases .
we selected primarily on the basis of its close proximity on the sky to and secondarily for its approximate alignment , too , between and .
the cycle time , and the scan time on each source per cycle , were chosen with the intent that they would be : ( i ) long enough to get reliable fringe - phase measurements for both reference sources on all baselines within a single scan ; ( ii ) long enough to perhaps get reliable fringe - phase measurements of the guide star on at least the most sensitive baselines within a single scan , thereby allowing use of the pmf technique ; and ( iii ) short enough to determine ( without ambiguity ) the integral number of 2@xmath29 phase wraps between successive measurements of each of the sources .
we used two reference sources rather than one primarily so that we could distinguish model errors that have elevation - angle dependence ( e.g. , atmospheric delays ) from those that do not ( e.g. , station clock behavior ) and thereby improve our overall astrometric accuracy ( see , too , * ? ? ?
we also analyzed data from four sessions of vlbi observations of made by one of us and other colleagues @xcite between december 1991 and july 1994 in support of the _ hipparcos _ mission @xcite .
however , these four sets of observations had significant differences from the later 35 sessions of observations made specifically for . perhaps most significantly
, these earlier observations used only four antennas per session and a single reference source , . therefore we did not include the results from these four earlier sessions in the comparisons of the astrometric techniques that we present in this paper .
we obtained the amplitude and phase calibration of our data for the ma technique by following steps 15 outlined in [ prmaps ] for the prm technique .
we then used aips task fring to estimate the fringe phases , fringe rates , and group delays from the visibility ( i.e. , cross - correlation ) data output from the correlator .
we obtained these estimates on a scan - by - scan basis .
we also obtained these estimates one baseline at a time by specifying two antennas at a time with the antenna parameter of fring .
the visibility data from which these estimates for each baseline were derived were independent , and thus we followed the common ( and greatly simplifying ) practice of treating the estimates for each baseline as independent . to account for the non - pointlike brightness distribution ( i.e. , structure ) of our reference sources , we made images of the sources using the self - calibration ( or `` hybrid mapping '' ) scheme described by @xcite .
we generated one such image for each reference source for each session of observations .
we obtained structure corrections for our estimated fringe phases , fringe rates , and group delays by running aips task fring twice for each baseline : once with the self - calibrated clean map used as input calibration , and once with the source modeled as a point source ( which is the fring default ) .
we took the structure corrections for the observables to be the differences between the estimates with a point - source map and with the clean map .
an important consideration in the generation of the structure corrections is the choice within the clean map of the reference position that is defined to have zero structure correction ( i.e. , the effective point - source position of the source ) .
initially , we used the brightness peak of the image of each reference source as the reference position , because the brightness peak was always clearly defined and easily identifiable . however , as we discuss in paper iii ( @xcite ; see also paper ii ) , further studies of the collection of images of our primary reference source revealed that a different point within the source made a substantially better choice of presumed - stationary reference position .
we thus adopted this point , identified as `` c1 '' in papers ii and iii , as the reference position for our structure corrections .
a set of images that show the location of `` c1 '' within the brightness distribution of can be found in .
the results from aips task fring for fringe phase , fringe rate , and group delay are estimates of residuals to the model used by the correlator .
we used a customized version of aips task cl2hf to add these residual estimates to the a priori model values from the correlator and place the resultant `` total '' observables into aips tables .
we then `` exported '' these total observables from aips with aips task hf2sv .
the resultant directory of files of total observables was combined with files of eops , surface meteorological data , antenna and source coordinates , and additional calibration data ( see below ) to produce databases that we used for the next stage of data reduction .
we created these databases with the support of the vlbi group of the space geodesy program at nasa s goddard space flight center .
the theoretical observables and partial derivatives with respect to model parameters were computed with that vlbi group s calc software @xcite .
we used calc version 9.13 and the jet propulsion laboratory s de200 ephemeris @xcite for all of our a priori models .
along with the observable data , we also exported from aips a collection of files of the model fringe phases and rates used by the correlator . in the final stage of our data reduction ,
discussed in [ mergfin ] below , we effectively replaced these model phases and rates with improved models that we derived from the observable data that we exported .
the next step of our data reduction was to use the measured phases from the reference sources to improve the model estimates of the phases of the target source .
in particular , we used the measured phases from our two main reference sources , and , to separate a priori model errors that have an elevation - angle dependence ( as would be contained in @xmath6 in equation [ phi ] ) from those that do not ( as would be contained in @xmath5 in equation [ phi ] ) .
we also used these measured phases to identify time intervals when the phases can not be well tracked by existing models .
we unweighted the target - source data within these intervals , which could be as short as a single scan of target - source data .
`` phase connection '' refers to the determination of the correct integral number of @xmath25 phase wraps between successive measurements of fringe phase for a particular baseline and source ( see figure [ figphcon1 ] ) .
phase connection is possible in part because of our very accurate a priori models and in part because of the very stable time reference ( hydrogen - maser frequency standard ) used at each vlbi antenna .
our basic approach to phase connection was to start with the baselines and time segments of and data for which our a priori models were sufficiently accurate that the number of @xmath25 phase wraps between successive fringe - phase measurements was immediately evident ( see figure [ figphcon2 ] ) .
we then used this subset of phase - connected data to estimate adjustments to the parameters of our a priori model .
the resulting improvement in our model enabled us to phase connect more data , and so the process proceeded in an iterative manner until we could no longer reliably phase connect any additional data . for observations above 10 elevation , we reliably connected , in total , 83% of the and phases that we evaluated in this data - reduction stage .
the terms in equation [ phi ] that can be associated with difficulties in the phase - connection process are those for which the model errors can result in relatively large ( @xmath30 radian ) residual phase fluctuations in a random pattern over successive reference - source scans , i.e. , in just a few minutes of time .
our a priori values for parameters such as site coordinates and eops were sufficiently accurate that they could not cause such rapid residual phase fluctuations .
in fact , nearly all such phase fluctuations could be attributed to changes either in instrumental delays , @xmath5 , which include possible `` clock jumps '' in the frequency standards at the sites , or , most commonly , in propagation delays through the atmosphere , @xmath6 .
the data that we could not reliably phase connect tended to involve sites with large wet atmospheric delays ( the st .
croix vlba site was generally the most problematic for phase connection ) or be from observations just above the 10 elevation threshold .
we used surface meteorological data ( barometric pressure , temperature , and humidity or dew point ) along with the equations from @xcite to obtain a priori estimates of @xmath31 and @xmath32 . for the mapping functions @xmath33 and @xmath34 we used the formulas provided by @xcite .
our values for @xmath35 came from the united states air force s parameterized ionosphere model ( pim ; * ? ? ? * ) , which we adapted for use with vlbi observables @xcite . in fitting our data to a model
, we used a kalman - filter estimator that allowed us to model atmospheric delays at zenith and instrumental drifts as gauss - markov stochastic processes .
our vlbi data do not provide sufficient information to simultaneously estimate adjustments to the separate terms in equation [ tau_atm ] ; thus we adjusted only @xmath32 , the term we assumed to have the least accurate a priori model .
figure [ smzenadj ] shows the adjustments to @xmath32 that we obtained with our kalman - filter estimator for the data corresponding to figures [ figphcon1 ] and [ figphcon2 ] . through the use of data from two reference sources , and
, we were able to distinguish model adjustments to @xmath36 , which has elevation - angle dependencies , from model adjustments to other terms , e.g. , @xmath37 , which do not .
large phase jumps between successive reference - source scans were generally attributed to atmospheric fluctuations . since the size of atmospheric fluctuations roughly scaled with the signal path length through the atmosphere , phase connection was generally easiest when the observed sources were at high elevation angles . thus we generally started our phase - connection process with data obtained during the middle of an observing session , when the sources were at their highest elevation angles at antennas near the middle of the array .
we then worked our way `` outward '' towards the early and late scans of the session .
when we could not confidently determine the integral number of @xmath25 phase wraps between consecutive reference - source scans , we unweighted the target - source data between those scans .
we also inserted a `` break '' marker at the point of the undetermined phase jump ; the kalman - filter estimator takes account of such breaks in its model fit .
figure [ sm2252res ] illustrates target - source residuals to the phase models derived from the reference - source data .
the integral numbers of @xmath25 phase wraps between successive scans are reliably determined for all weighted data .
once we had phase - delay and fringe - rate models based on parameter fits to the reference - source data , we imported those models into aips to use as the basis for our phase - referenced maps of the target source , either or .
first we created files of fringe - phase and fringe - rate differences between the models derived from the reference - source data and the models ( from the correlator ) originally used in aips .
then we used aips task tbin to create an aips `` solution '' ( sn ) table with these differences .
we could then treat the change from the original models used in aips to the improved models derived from our reference - source data as a standard aips calibration step .
we used aips task clcal to incorporate these model adjustments .
we then made phase - referenced maps with aips task imagr following standard procedures ( e.g. , * ? ? ?
we obtained positions for our target sources from these imagr maps as described in paper vi @xcite .
we used two figures of merit to compare the results of our ma technique with the more conventional prm and pmf techniques : the quality , as described below , of the and images produced ( ma and prm techniques only ) ; and the level of consistency of the astrometric results for each of these two sources .
figure [ 2252compfig ] shows a representative map of produced via the conventional prm technique alongside the corresponding map generated via our ma technique .
table [ 2252comptab ] lists the brightness - peak flux density , the dynamic range ( i.e. , the ratio between the flux density of the brightness peak and the flux density of the image noise floor ) , and the ratio of the peak - brightness amplitude ( @xmath38 ) to the amplitude of the most negative peak ( @xmath39 ) that we obtained with the prm and ma techniques for each of our 12 sessions of observations of this source . if the extent of the brightness distribution of a source is comparable to , or larger than , the size of the restoring beam of its image , then the measured peak brightness and ( hence ) the dynamic range are functions of the size of the restoring beam . the size ( and orientation ) of the restoring beam
can in turn depend upon which data were weighted ( among other factors ) when the image was produced . since we employed different data - selection criteria with the prm and ma techniques and hence did not necessarily use identical data with the two techniques , we consider the @xmath38/@xmath39 ratio to be the most meaningful measure of image quality among the three table [ 2252comptab ] metrics .
c c c c c c c c c c c c c & & + & & & & + & & & & + session&&dyn.&&&dyn .
& & & dyn . & & & dyn . &
+ start date&@xmath38&range & & @xmath38&range&&@xmath38&range&&@xmath38&range & + 2002 - 11 - 20 & 12.2 & 64 & 9.7 & 14.2 & 68 & 13 .
& 15.2 & 165 & 37 .
& 15.5 & 170 & 31 .
+ 2003 - 01 - 26 & 11.3 & 53 & 5.6 & 14.1 & 70 & 8.3 & 17.0 & 140 & 29 .
& 17.9 & 149 & 38 .
+ 2003 - 05 - 18 & 12.5 & 62 & 7.8 & 16.6 & 60 & 8.7 & 19.1 & 187 & 47 .
& 19.8 & 173 & 44 .
+ 2003 - 09 - 08 & 11.7 & 64 & 13 . & 13.7 & 72 & 16 .
& 15.4 & 141 & 34 .
& 16.0 & 132 & 34 .
+ 2003 - 12 - 05 & 11.0 & 51 & 7.2 & 14.3 & 61 & 9.9 & 15.4 & 106 & 16 .
& 16.0 & 97 & 20 .
+ 2004 - 03 - 06 & 8.9 & 56 & 8.8 & 11.0 & 76 & 12 . & 12.1 & 124 & 30 .
& 12.4 & 110 & 25 .
+ 2004 - 05 - 18 & 6.5 & 37 & 7.3 & 8.8 & 48 & 8.8 & 10.0 & 77 & 18 .
& 10.5 & 68 & 18 .
+ 2004 - 06 - 26 & 7.0 & 54 & 11 .
& 9.0 & 70 & 14 .
& 9.8 & 109 & 25 .
& 10.1 & 103 & 22 .
+ 2004 - 12 - 11 & 9.0 & 58 & 8.9 & 12.0 & 94 & 14 .
& 12.2 & 128 & 34 . & 12.8 & 116 & 30 .
+ 2005 - 01 - 15 & 9.7 & 87 & 12 .
& 11.0 & 103 & 12 .
& 11.3 & 175 & 39 . & 11.6 & 149 & 38 .
+ 2005 - 05 - 28 & 11.7 & 48 & 10.0 & 15.5 & 69 & 15 .
& 16.3 & 125 & 36 .
& 16.5 & 114 & 33 .
+ 2005 - 07 - 16 & 12.5 & 64 & 11 .
& 15.4 & 80 & 14 .
& 16.0 & 127 & 27 .
& 16.5 & 131 & 28 .
+ mean & 10.3 & 58.1 & 9.3 & 13.0 & 72.6 & 12.0 & 14.2 & 133.6 & 31.0 & 14.6 & 125.9 & 30.0 + for both the prm and ma techniques , we show these metrics for images produced after we applied either a 10 or a 30 elevation - angle cutoff to the data ( i.e. , after we unweighted all data from each site whenever the observed source was below the specified elevation angle ) .
since the major source of phase - calibration error with the prm technique is often poor modeling of the atmosphere , and since ( as mentioned earlier ) atmospheric model errors tend to get larger at low elevation angles , the removal of low - elevation - angle data can potentially improve the quality of both the image and the astrometry obtained with the prm technique .
( note that , for the prm technique , the images when a 30 elevation - angle cutoff is used nearly always have higher dynamic ranges and @xmath38/@xmath39 ratios than the corresponding images when a 10 elevation - angle cutoff is used . ) for the ma technique , only those time spans with reliably phase - connected data are weighted , so we might expect a degradation of image quality ( due to the reduction of weighted data ) when a higher elevation - angle cutoff is imposed .
the results found for our images are mostly consistent with this expectation , although table [ 2252comptab ] shows that for , the higher elevation - angle cutoff did improve the @xmath38/@xmath39 ratio with the ma technique in four of the twelve sessions . in any event , even if , for the prm technique , we select for each session the elevation - angle cutoff in table [ 2252comptab ] with the highest dynamic range or @xmath38/@xmath39 ratio for each session , the corresponding ma images for a 10 elevation - angle cutoff have , on average , 1.9 times the dynamic range and 2.7 times the @xmath38/@xmath39 ratio of the prm images . the ma images , when compared with the prm images , also show significantly less variability in peak flux density for the two different elevation - angle cutoffs that are used . this superior consistency , too , suggests that the ma technique provides more accurate model calibration than the prm technique .
figure [ hr8703compfig ] shows representative images of produced with both the prm and ma techniques .
averaged across all of our observing sessions , the peak flux density we detected for was 16% higher with the ma technique than with the prm technique .
however , the significance of this result is unclear , because we used different restoring beams for each technique .
unfortunately , the data files used to make the images of with the prm technique were no longer available at the time of preparation of this publication , so we could not use restoring beams with the same size and orientation for both techniques .
we also did not have noise - floor or `` minimum peak '' information available for these prm images .
in addition , we know that can vary in total brightness and brightness distribution ( i.e. , radio structure ) over hour time scales ( paper vii , * ? ? ? * ; * ? ? ?
* ) , which further complicates comparisons between the techniques , especially since a constant flux density over the full duration of a session of observations is inherently assumed in the imaging process for both techniques .
thus , for comparison of image quality from the prm and ma techniques , we considered the results a better measure than the results .
figure [ 2252posplt ] compares the position estimates for ( relative to ) from the prm , pmf , and ma techniques .
( we used a 30 , rather than 10 , elevation - angle cutoff of data with the prm technique , because the higher cutoff yielded better results for this technique .
we used our nominal 10 elevation - angle cutoff of data with the pmf and ma techniques . ) table [ 2252postab ] shows the root - mean - square ( rms ) scatter about the mean position for each technique .
when , as with , the target source is commonly detected on many baselines in a single scan and structure corrections with the pmf technique are available , the ma technique shows no astrometric advantage over the pmf technique .
however , our ma technique yields position estimates with only about 60% of the rms scatter of the estimates from the prm technique , i.e. , removes the equivalent of a noise source of over 0.06 mas per coordinate estimate associated with the prm technique . , is a good approximation for the errors of the corresponding points shown in this plot .
such errors ignore possible systematic effects that could have shifted the mean positions for each of the three techniques .
, scaledwidth=85.0% ] l@c c@c c@c c & & & + & & & + technique & & & @xmath40 & @xmath41 & @xmath40 & @xmath41 + prm & @xmath260.039 & 0.021 & 0.082 & 0.092 & 0.025 & 0.028 + pmf & 0.012 & 0.019 & 0.048 & 0.053 & 0.014 & 0.016 + ma & 0.000 & 0.000 & 0.050 & 0.057 & 0.014 & 0.017 + the model for s position over time must include proper motion , parallax , and orbital motions associated with the binary system .
we used the preferred parametrization of paper v @xcite , one in which no proper acceleration is estimated , to obtain the results presented here . for seven of the 35 observing sessions of , including the most astrometrically important final three ,
the radio emissions from the star were insufficiently bright to obtain reliable detections of in single scans .
thus we could not obtain position estimates of with the pmf technique for any of these sessions .
we also found from images obtained with the prm and ma techniques that had relatively complex structure during several of our observing sessions , and the pmf technique by itself can not well identify or characterize such structure .
thus we deemed the pmf technique unsuitable for astrometric studies of with our data . in figure [ hr8703posplt ]
we show the post - fit residuals of the position estimates for from the prm and ma techniques , and in table [ hr8703postab ] the rms scatter of these same pos - tfit residuals . on average the post - fit residuals from our ma technique
had about 10% less scatter in right ascension and 7% less scatter in declination than the corresponding residuals from the prm technique .
the scatter from both techniques is likely dominated by the unmodeled intrinsic motions of the stellar radio emission relative to the star s primary component .
hence , since we have no model that can accurately describe the motion of the radio emission of relative to its primary , we can expect at most only a minor improvement in the rms scatter of position residuals from the use of any superior astrometric technique . indeed
the ma technique yielded just such a minor improvement .
.rms scatter of position residuals obtained from the prm and ma techniques.[hr8703postab ] [ cols="<,^,^ " , ]
in comparing the astrometric results from phase - referenced mapping ( prm ) , parametric model fitting ( pmf ) , and the merged analysis ( ma ) technique introduced in this paper , we found : 1 .
the prm technique can be used with relatively weak radio sources .
it is the least labor intensive among the techniques tested , but it also provides the least astrometric accuracy .
( with this technique , unlike with the pmf or ma techniques , we were unable to use multiple reference sources to better account for the elevation - angle dependence of model errors ; such use could improve the astrometric accuracy attainable with this technique . ) 2 .
the pmf technique is useful only when both reference and target sources are : ( i ) reliably detectable in short ( i.e. , few - minute ) scans of data ; and ( ii ) pointlike , unless structure corrections are available from other means . under these circumstances ,
the pmf technique yielded the highest astrometric accuracy among the techniques tested .
3 . our ma technique can provide astrometric accuracies nearly equal to those obtained with the pmf technique .
4 . very importantly
, our ma technique can be used with sources that are too weak for use of the pmf technique , as was during seven of our 35 vlbi sessions . in general , any source that is sufficiently bright for use of the prm technique is sufficiently bright for use of our ma technique .
the ma technique can also be used with sources that have complex brightness distributions and was shown to yield images with significantly higher dynamic ranges than comparable images from the prm technique .
our use of the ma technique enabled us to obtain more accurate astrometric measurements of , the guide star for the mission , than we could have otherwise obtained with the conventional prm or pmf techniques .
we thank n. nunes for developing the graphical software package we used to phase connect our vlbi data .
we thank the vlbi group of nasa s goddard space flight center , l. petrov in particular , for providing technical support and information .
we are also grateful to the many people involved in our campaign of vlbi observations who went above and beyond the call of duty , including s. dains , c. garcia miro , e. moll , and l. cameron .
we thank c. jacobs and o. sovers for information about the coordinates of the three 70 m nasa dsn antennas used in our observations , and r. c. walker for information about the vla coordinates .
we thank j. l. davis and s. s. shapiro for their information and insights on the use of our kalman - filter estimator ( solvk ) .
finally , we thank r. c. walker , k. desai , and e. greisen for their support in our use of aips . the national radio astronomy observatory ( nrao )
is a facility of the national science foundation operated under cooperative agreement by associated universities , inc .
this research made use of nasa s astrophysics data system , which was conceived , developed , and continues to be operated by the smithsonian astrophysical observatory at the harvard - smithsonian center for astrophysics .
our work was supported by nasa prime award nas8 - 39225 , stanford university sub - award pr6750 , the smithsonian institution , and york university .
, d. s. , ma , c. , & ryan , j. w. 1990 , nasa technical memorandum 100765 ( crustal dynamics project data analysis , 1990 ) , national aeronautics and space administration , goddard space flight center , greenbelt , md , tech . rep . , a. l. , gordon , d. , & jacobs , c. s. , ( eds . ) 2009 , iers technical note no .
35 , http://www.iers.org/iers/en/publications/technicalnotes/tn35.html ( frankfurt : verlag des bundesamts fr kartographie und geodsie ) , e. b. , & kogan , l. 2005 , aips memo 111 ( atmca : phase referencing using more than one calibrator ) , national radio astronomy observatory , socorro , nm , tech .
rep . , ftp://ftp.aoc.nrao.edu/pub/software/aips/text/publ/aipsmem111.ps , j. w. , ma , c. , & caprette , d. s. 1993 , nasa technical memorandum 104572 ( data analysis 1992 , final report of the crustal dynamics project vlbi geodetic results , 1979 - 1991 ) , national aeronautics and space administration , goddard space flight center , greenbelt , md , tech .
rep . , r. c. , & chatterjee , s. 1999 , vlba scientific memo 23 ( ionospheric corrections using gps based models ) , national radio astronomy observatory , socorro , nm , tech . | when vlbi observations are used to determine the position or motion of a radio source relative to reference sources nearby on the sky , the astrometric information is usually obtained via : ( i ) phase - referenced maps ; or ( ii ) parametric model fits to measured fringe phases or multiband delays . in this paper
we describe a merged " analysis technique which combines some of the most important advantages of these other two approaches . in particular , our merged technique combines the superior model - correction capabilities of parametric model fits with the ability of phase - referenced maps to yield astrometric measurements of sources that are too weak to be used in parametric model fits .
we compare the results from this merged technique with the results from phase - referenced maps and from parametric model fits in the analysis of astrometric vlbi observations of the radio - bright star i m pegasi ( ) and the radio source nearby on the sky . in these studies
we use central - core components of radio sources and as our positional references .
we obtain astrometric results for with our merged technique even when the source is too weak to be used in parametric model fits , and we find that our merged technique yields superior astrometric results to the phase - referenced mapping technique .
we used our merged technique to estimate the proper motion and other astrometric parameters of in support of the nasa / stanford _ gravity probe b _ mission . |
the effects of defects on the properties of graphene are of scientific and practical importance @xcite . for studying the effects of defects experimentally ,
ion irradiation has been used , which introduces point defects in graphene .
recently , in such an ion - irradiated graphene , the kondo effect is observed @xcite .
usually , the kondo effect indicates the presence of localized moments in the sample .
therefore the above result suggests that a point defect in graphene causes a localized moment .
the magnetic behavior of the defects in graphene is also observed in the recent magnetization measurement of ion - irradiated graphene @xcite . as a possible origin of this kondo effect , we proposed a localized moment on a @xmath3 orbital around the defect @xcite .
there are three @xmath3 orbitals around a point defect . with jahn - teller distortion ,
two of them form a covalent bond and remaining one behaves as a localized moment with spin - half . furthermore , with this lattice distortion , there is a hybridization between the localized @xmath3 orbital and conduction @xmath4 electrons , and the kondo effect can occur . on this scenario
, we introduced an effective model and analyzed it by the numerical renormalization group ( nrg ) method @xcite . as a result
, we showed that the experimentally observed gate voltage dependence of kondo temperature can be understood by assuming that the hybridization between the @xmath3 and @xmath4 orbitals is strong enough @xcite . in the above experiment ,
the negative magnetoresistance ( mr ) , which is the evidence of the kondo effect , was also observed .
although the characteristic magnetic field for this negative mr is found to be smaller by one order than the usual kondo - effect case , its origin has not been understood .
so far , the studies on the effect of magnetic field on the kondo effect in graphene are limited , and there are few studies on the effect of zeeman field @xcite .
motivated by this negative mr , in this paper , we construct an effective model of graphene with a point defect and study the magnetic field effect on the thermodynamics of the model on the basis of the nrg method . here , as a first step , we consider only the zeeman field and neglect the orbital magnetic effect on the conduction electrons .
the effective hamiltonian of the graphene with a point defect has been discussed in ref .
@xcite , and the one - dimensional energy representation for the conduction electrons @xcite which is necessary to apply the nrg method has been derived . in this representation ,
the effective hamiltonian is written as @xmath5 where @xmath6 , @xmath7 , and @xmath8 are hamiltonians for the conduction electrons , the defect @xmath3 orbital , and the hybridization , respectively , with @xmath9 here , @xmath10 is an annihilation operator of the conduction electrons with energy @xmath11 and spin @xmath12 , @xmath13 , which satisfies the anti - commutation relation @xmath14 @xcite .
@xmath15 is the dirac delta function .
@xmath16 is a cut - off energy that is the same order as the band width of the conduction electrons .
@xmath17 is an annihilation operator of an electron on the defect @xmath3 orbital with spin @xmath12 , @xmath13 , which satisfies @xmath18 .
@xmath19 is the chemical potential of electrons .
@xmath20 , @xmath21 , and @xmath22 are the energy level of the defect @xmath3 orbital measured from that of @xmath23 orbital on carbon atom , coulomb interaction , and zeeman field on the defect @xmath3 orbital , respectively .
@xmath24 is the amplitude of the hybridization between conduction electrons and the @xmath3 orbital .
this model of eqs .
( [ eq_hamiltonian])-([eq_hamiltonianhybridization ] ) is known as the single - channel pseudogap anderson model @xcite , where the density of states ( dos ) of the conduction electron is proportional to @xmath25 .
the kondo problem in this pseudogap system was studied in detail for the case with @xmath26 and @xmath27 @xcite .
the main feature in this case is as follows . in the particle - hole symmetric ( @xmath28 )
case , no kondo screening occurs , i.e. the localized moment is not screened by conduction electrons . on the other hand , in the particle - hole asymmetric ( @xmath29 ) case , the localized moment is screened when the coupling constant @xmath24 is larger than a critical value @xmath30 . in this paper
, we assume the particle - hole asymmetry and study the case with @xmath31 and both @xmath26 and @xmath32 .
we apply the nrg method for the model ( [ eq_hamiltonian])-([eq_hamiltonianhybridization ] ) , following the formalism of ref .
@xcite to investigate the electronic state at low temperatures .
the nrg method can include the effects of the pseudogap dos , a finite chemical potential , and zeeman field . in the nrg method ,
the effective hamiltonian is transformed into the one - dimensional form with a chain length @xmath33 , @xmath34 where @xmath35,\\ h_0&=&\lambda^{-1/2}\bigg[\sum_\sigma\epsilon_0c^\dagger_{0\sigma}c_{0\sigma } + \frac{1}{\alpha}(h_{\rm{def}}+h_{\rm{hyb}})\bigg],\\ h_{\rm{hyb}}&=&vf\sum_\sigma\left(c^\dagger_{0\sigma}d_\sigma+{\rm h.c.}\right).\end{aligned}\ ] ] here , @xmath36 is an annihilation operator of the conduction electrons on the @xmath37th site of the chain .
@xmath38 is a logarithmic discretization parameter and @xmath39 . in the following , @xmath40
is set to be @xmath41 .
@xmath42 is the hopping parameter `` , and @xmath43 is the on - site potential '' , and @xmath44 is a factor .
these parameters reflect the dos of the conduction electrons , and have numerical values of order of unity @xcite . by diagonalizing this hamiltonian iteratively @xmath33 times and by eliminating higher energy states
, we obtain the energy eigenvalues and the eigenstates . here
300 states are retained . by using these eigenvalues and eigenstates ,
thermodynamic quantities are calculated @xcite . in this paper ,
the localized moment contributions to the entropy , @xmath45 , and the expectation value of square of the @xmath46 component of localized moment , @xmath47 , are calculated .
orbital , @xmath45 , under the zeeman field @xmath48 at chemical potential @xmath26 for the hybridization @xmath49 , @xmath50 , and @xmath51 . here ,
boltzmann constant @xmath52 .
inset shows @xmath53 at @xmath54 for @xmath55 .
( b ) temperature dependence of expectation value of @xmath1 , @xmath47 , for the same parameters as in fig .
title="fig:",width=340][fig_thermodynamic ] in order to clarify the low - temperature states of this model , we calculate the temperature dependence of @xmath53 and @xmath47 in the nrg method .
the parameters @xmath56 and @xmath21 are assumed to be @xmath57 @xmath58 , @xmath59 @xmath58 , and @xmath60 @xmath58 .
thus we use the parameters @xmath61 , and @xmath62 , in the following . with these parameters , in the case of no zeeman field , @xmath63 , it has been shown @xcite that the critical hybridization is @xmath64 . at @xmath26 ,
the entropy of the defect is @xmath65 , @xmath66 , and @xmath67 , for @xmath68 , and @xmath69 , respectively at low temperatures .
these three cases correspond to the local moment state , the valence - fluctuation fixed point , and the kondo - yosida singlet state , respectively . here , the boltzmann constant is set to be @xmath52
. figure 1(a ) shows @xmath53 as a function of temperature for the hybridization @xmath49 , @xmath50 , and @xmath51 under the zeeman field of @xmath48 at @xmath26 . for the case with @xmath49 and @xmath51 , @xmath70 at low temperatures
however , their temperature dependences are different : at @xmath49 , for @xmath71 , @xmath53 changes from @xmath72 to @xmath73 .
thus , it is suggested that the ground state is a spin polarized state . at @xmath74 ,
on the other hand , @xmath45 changes from @xmath66 to @xmath75 for @xmath76 .
this temperature is much larger than the zeeman field ( @xmath48 ) .
thus , the ground state is considered to be the kondo - yosida singlet state . at a critical point @xmath54
, we find that @xmath45 approaches @xmath72 at the lowest temperature as shown in the inset of fig .
in order to obtain more information on the electronic states at low temperatures , we calculate @xmath47 .
figure 1(b ) shows @xmath47 as a function of temperature for the same parameters . for the case with @xmath49 and @xmath51
, we find @xmath77 and @xmath67 , respectively at low temperature , which confirms the ground states suggested above : the spin polarized state for @xmath49 , and the kondo - yosida singlet state for @xmath74 .
it is also found that @xmath47 suddenly changes from @xmath78 to @xmath67 at @xmath54 when we change @xmath24 .
thus , @xmath79 is a critical value which separates the spin polarized state and the kondo - yosida singlet state . at @xmath80 , @xmath81 . the critical value in the zeeman field , @xmath82 , is larger than the critical value @xmath30 at @xmath27 , because the kondo - yosida singlet state is suppressed by the zeeman field .
we suppose that @xmath82 is determined by the relation @xmath83 , where @xmath84 is a kondo temperature .
the values of @xmath65 and @xmath85 at @xmath82 imply that there are two degenerate states at this point .
we expect that these two states are up - spin state and empty ( unoccupied ) state , since @xmath85 can be interpreted as @xmath86/2 $ ] @xcite .
the detail of this unusual fixed point will be discussed elsewhere .
we also calculate the finite chemical potential case , @xmath32 , under the zeeman field .
@xmath45 is always zero at low temperatures . at low temperatures , @xmath47 ,
shows the change from @xmath87 to @xmath67 , as @xmath24 increases .
we do not find any unusual fixed point in this case .
in summary , the effect of the zeeman field on the defect - induced kondo effect in graphene has been investigated .
the effective model with the zeeman field has been derived and has been analyzed on the basis of numerical renormalization group method .
it has been found that , at @xmath26 , there is an unusual fixed point with the entropy of @xmath0 and the expectation value of @xmath1 of @xmath2 at the border of the spin polarized state and the kondo - yosida singlet state .
t. k. is supported by the global coe program the physical sciences frontier " of the ministry of education , culture , sports , science and technology , japan .
50 terrones h , lv r , terrones m and dresselhaus m s 2012 _ rep . prog . phys . _ * 75 * 062501 chen j - h , li l , cullen w g , williams e d and fuhrer m s 2011 _ nat . phys . _ * 7 * 535 nair r r , sepioni m , tsai i - l , lehtinen o , keinonen j , krasheninnikov a v , thomson t , geim a k and grigorieva i v 2012 _ nat .
phys . _ * 8 * 199 kanao t , matsuura h and ogata m 2012 _ j. phys . soc . jpn . _ * 81 * 063709 wilson k g 1975 _ rev .
phys . _ * 47 * 773 bulla r , costi t a and pruschke t 2008 _ rev .
phys . _ * 80 * 395 cornaglia p s , and usaj g and balseiro c a 2009 _ phys .
lett . _ * 102 * 046801 jones b a and varma c m 1987 _ phys .
* 58 * 843 sakai o and shimizu y 1992 _ j. phys .
jpn . _ * 61 * 2333 gonzalez - buxton c and ingersent k 1998 _ phys .
b * 57 * 14254 after the manuscript of this paper had been accepted , we became aware that this situation was discussed with perturbative renormalization group methods in fritz l and vojta m 2004 _ phys .
_ b * 70 * 214427 | the effect of the zeeman field on the defect - induced kondo effect in graphene is investigated .
the effective model of the kondo effect is derived , and is analyzed on the basis of the numerical renormalization group method .
it is found that , under the zeeman field , at the border of the spin polarized state and the kondo - yosida singlet state , there is an unusual fixed point where the entropy of the defect is @xmath0 and the expectation value of @xmath1 is @xmath2 . |
the discovery of graphene @xcite opens a new era of fundamental research and technological applications of carbon - based materials .
graphene is an elegant two - dimensional ( 2d ) material exhibiting exceptional electrical , thermal , mechanical and optical properties , and its low - energy behavior of electrons is described by a ( 2 + 1)-dimensional relativistic quantum theory @xcite .
stimulated by the tremendous interest of graphene , a great number of novel carbon structures were proposed and investigated in the past years ( e.g. refs . [ ] ) . among others , two intriguing @xmath4-@xmath0-bonded monolayers of elemental carbon , graphyne @xcite and
graphdiyne @xcite , and an insulating 2d hydrocarbon layer , graphane @xcite , are typical examples , all of which were experimentally obtained .
it appears that there is a growingly extensive interest to seek for more 2d allotropes of carbon with amazing physical and chemical properties as well as possibly diverse applications .
graphene has a perfect honeycomb lattice with @xmath5 symmetry .
recent studies @xcite show that the line defects with octagons and pentagons could be formed self - assembly in graphene .
the formation of small carbon molecules with various polygon rings has also been studied from experimental aspects @xcite .
inspired by these observations , we propose a stable 2d periodic atomic sheet consisting of carbon octagons , coined as _
octagraphene _ , which possesses intriguing properties and might be synthesized experimentally by line defects or acetylene scaffolding ways .
as shown in fig .
1(a ) , octagraphene comprises of carbon octagons and squares with two bond lengths that forms a square lattice with @xmath1 symmetry . by means of the first - principles calculations
, it is shown that octagraphene is energetically and kinetically stable , and is more favorable in energy than graphyne and graphdiyne , although it is metastable against graphene . by rolling octagraphene along specific directions
, we can get novel sawtooth and armchair carbon nanotubes [ fig .
1(b ) ] that are metallic , which can escape from the difficult separation of usual metallic and semiconducting carbon nanotubes @xcite ; by wrapping octagraphene and making use of pentagons , hexagons or heptagons as caps , many stable unconventional fullerenes such as c@xmath6 , c@xmath7 , c@xmath8 , c@xmath9 and so forth [ fig .
1(c ) ] can be constructed .
the electronic properties of octagraphene manifest that it is a semimetal with small density of states ( dos ) around the fermi level , whose fermi surface consists of one hole and one electron pockets centered at @xmath10 and @xmath11 points , respectively .
the low - energy physics of octagraphene can be well described by a tight - binding model of @xmath2 electrons .
octagraphene has the density of 0.68 @xmath12 , young s modulus of 306 @xmath3 , breaking strength of 34.4 @xmath3 and poisson s ratio of 0.13 .
it could be the strongest periodic carbon sheet after graphene till now .
an energy gap can be opened by doping boron nitrogen pairs .
the possible routes for obtaining octagraphene are suggested .
the ti - absorbed octagraphene can also be used for hydrogen storage with capacity of 7.76 wt% . with these amazing characters
, we anticipate that octagraphene , similar to its cousin graphene , once being obtained , will likewise have a crucial impact in physics , chemistry , materials and information sciences .
the rest of this article is organized as follows . in sec .
ii , the calculational methods are described . in sec .
iii , the geometrical structure , mechanical properties , electronic structures and low - energy physics of octagraphene are presented . in sec .
iv , new carbon nanotubes and unconventional fullerenes from octagraphene are given .
the band engineering and hydrogen storage of octagraphene are discussed in sec .
v. finally , a conclusion is briefly presented .
both the first - principles calculations within the framework of the density - functional theory ( dft ) @xcite and the tight - binding approximation ( tba ) @xcite are invoked to study this 2d system . for the first - principles method ,
primary calculations were performed within the vienna _ ab initio _ simulation package ( vasp ) @xcite with the projector augmented wave ( paw ) method @xcite . both the local density approximation ( lda ) in the form of perdew - zunger @xcite and generalized gradient approximation ( gga ) developed by perdew and wang @xcite were adopted for the exchange correlation potentials .
the plane - wave cutoff energy is taken as 400 ev . the monkhorst - pack scheme for k - point samplings with 32@xmath1332@xmath131 mesh for planar sheets and 1@xmath131@xmath13100 mesh for nanotubes
were used to sample the brillouin zone @xcite .
the supercells are used for calculations of isolated sheet structures , and the distance between two layers is about 10 to avoid interactions . the geometries were optimized when the remanent hellmann - feynman forces on the ions are less than 0.01 ev / .
phonon calculations with a k - mesh of 8@xmath138@xmath131 are performed using the density functional perturbation theory @xcite .
and @xmath14 ; ( b ) single - wall sawtooth ( 10,0 ) and armchair ( 8,8 ) carbon nanotubes rolled from octagraphene ; and ( c ) unconventional fullerenes of c@xmath6 , c@xmath7 , c@xmath8 and c@xmath9 obtained from octagraphene.,width=396 ] the optimized geometric structure of octagraphene is presented in fig .
carbon atoms in octagraphene form a square lattice with the lattice constant @xmath15 , where a unit cell contains four carbon atoms .
similar to graphene , every carbon atom in octagraphene has three @xmath16-bonded nearest neighbors , forming three @xmath17 bonds .
in addition , octagraphene has two types of bond lengths , @xmath18 of intra - squares and @xmath19 of inter - squares , and two bond angles @xmath20 and @xmath21 . for graphene , octagraphene , graphyne and graphdiyne , respectively , where @xmath22 is the optimized lattice constant .
( b ) phonon spectra of octagraphene . ] the cohesive energy @xmath23 per atom as a function of lattice constant ( @xmath24 ) is given in fig .
2(a ) for graphene , octagraphene , graphyne and graphdiyne , respectively .
it is clear that a single minimum of @xmath25 appears at @xmath26 for these 2d structures , suggesting that they are energetically stable .
note that the minimum of @xmath27 for octagraphene is smaller than those of graphyne and graphdiyne but larger than that of graphene , suggesting octagraphene is more energetically stable than graphyne and graphdiyne but metastable against graphene .
2(b ) gives the phonon spectra of octagraphene .
no imaginary phonon modes are found , implying that octagraphene is also kinetically stable .
we performed quantum molecular dynamics simulations and found that at temperature 500 k the planar structure of octagraphene still maintains .
considering that graphyne and graphdiyne were already obtained experimentally , it is reasonable to believe that octagraphene can be synthesized in lab or likely already exists in nature .
in fact , carbon rings with four or eight sides have already been observed in some experiments @xcite . for a comparison , the cohesive energy @xmath28 , density @xmath29 , bond length @xmath30 , energy gap @xmath31 , young s modulus @xmath32 , breaking strength @xmath17 and poisson s ratio @xmath33 for graphene , octagraphene , graphyne and graphdiyne , respectively ,
are collected in table i. notice that apart from the experimental data of graphene taken from literature , all data presented here were calculated by ourselves using the same method . as octagons in octagraphene
are larger in size than hexagons in graphene , octagraphene has the density ( 0.68 @xmath34 ) slightly smaller than that of graphene ( 0.77 @xmath34 ) .
now let us discuss the mechanical properties of octagraphene .
as it has a c@xmath35 symmetry , only three elastic constants @xmath36 are independent .
our dft calculations estimate that the values of @xmath37 , @xmath38 and @xmath39 of octagraphene are 296 , 46 , and 49 @xmath3 , respectively .
the young s modulus @xmath32 of octagraphene is obtained to be 306 and 168 @xmath3 along the sawtooth and armchair directions ( see below ) , respectively , in comparison to those of graphene 350 @xmath3 , graphyne 245 @xmath3 and graphdiyne 174 @xmath3 .
the stress - strain curves for graphene , octagraphene , graphyne and graphdiyne under a uniaxial tension along two non - chiral directions are illustrated in fig .
the ultimate tensile strength or the breaking strength @xmath17 of octagraphene is 34.4 and 27.3 @xmath3 along the sawtooth and armchair directions , respectively , which are quite comparable with those of graphene 38.4 @xmath3 ( zigzag direction ) and 34.8 @xmath3 ( armchair direction ) but larger than those of graphyne ( 23.4 and 19.5 @xmath3 ) and graphdiyne ( 16.8 , 13.4 @xmath3 ) .
the experimental values @xcite of @xmath32 and @xmath17 for graphene are 340 @xmath40 50 @xmath3 and 42 @xmath40 4 @xmath3 , respectively , showing that our dft calculations are quite reliable .
the poisson s ratio of octagraphene is 0.13 ( sawtooth direction ) and 0.47 ( armchair direction ) , which are comparable with 0.21 ( zigzag direction ) and 0.17 ( armchair direction ) of graphene
. the poisson s ratio of graphyne and graphdiyne can be found in table i. from above results we see that octagraphene has the mechanical properties very similar to graphene , and might be the strongest carbon atomic sheet after graphene till now . [ cols="^,^,^,^,^,^,^,^,^,^,^",options="header " , ] point , and the electron pocket ( blue ) is at @xmath11 point .
( c ) energy bands of octagraphene nanotubes ( 3,3 ) , ( 5,5 ) , ( 12,12 ) , ( 5,0 ) , ( 8,0 ) and ( 16,0).,width=396 ] the valence electron density , electronic structures and the dos of octagraphene are obtained by means of the dft and tba calculations , respectively . from fig .
3(b ) one may see that more electrons are distributed on the inter - square bonds than on the intra - square bonds in octagraphene , which is well consistent with its structural feature , showing an unequal @xmath0 hybridizing character .
the electronic structures in fig .
4(a ) show that octagraphene is a semimetal , as its top of valence band and the bottom of conduction band are located at @xmath10 and @xmath11 points in the brillouin zone , respectively , giving rise to an indirect _ negative _ band gap ( -4.03 ev ) , and leading to the fermi level passes through both the conduction and valence bands .
thus , in contrast to graphene , the charge carriers in octagraphene have both holes and electrons around the fermi level .
note that graphene is also a semimetal or a zero - gap semiconductor because its valence and conduction bands touch at @xmath41 points , while graphyne and graphdiyne are semiconductors with narrow band gap of 0.48 and 0.51 @xmath42 , respectively .
the projected dos of octagraphene shows that the energy bands near the fermi surface are predominantly contributed by the @xmath43 orbital of carbon atom , forming the same @xmath2-bond as that in graphene .
the fermi surface of octagraphene is plotted in fig .
4(b ) , which consists of a hole pocket at @xmath10 point and an electron pocket at @xmath11 point .
since each unit cell contributes four electrons with two bonding states and two antibonding states , each point in the brillouin zone should be filled with @xmath44 electrons , and the number of unoccupied states in valence band is just equal to that of occupied states in conduction band , resulting in that the hole and electron pockets share the same area . to describe the low - energy physics near the fermi surface of octagraphene , we propose a tight - binding model which contains only @xmath2 electrons with nearest neighbor hoppings .
the hamiltonian can be written as @xmath45 where @xmath46 and @xmath47 are the electron annihilation and creation operators at site @xmath48 with spin @xmath17 , @xmath49 runs over the nearest neighbors within the same square and @xmath50 is the corresponding hopping amplitude of electrons , @xmath51 runs over the nearest neighbors connecting different squares and @xmath52 is the corresponding hopping amplitude . in the matrix form , the hamiltonian reads @xmath53 where @xmath54 and @xmath55 are @xmath56 and @xmath57 components of momentum of @xmath2 electrons , and the lattice spacing @xmath58 is taken .
the dispersion @xmath59 of octagraphene within the tba is determined by @xmath60=0 .
\label{dispersion}\end{aligned}\ ] ] the tba results are in good agreement with those obtained by the dft calculations [ fig .
4(a ) ] , implying that the low - energy properties of @xmath2 electrons in octagraphene can be well described by the tba hamiltonian , where the fittings give @xmath61 @xmath42 and @xmath62 @xmath42 . for graphene , @xmath63 @xmath42 @xcite . to obtain the structure of octagraphene experimentally , we suggest that one may make the line defects periodically in graphene , and then connect properly the broken bonds of carbon atoms , as demonstrated in fig .
5 . in addition
, it may be possible to synthesize it through the acetylene scaffolding ways @xcite .
by rolling octagraphene , the novel single - walled carbon nanotubes can also be obtained . for this purpose ,
let us define the chiral vector @xmath64 of octagraphene carbon nanotube ( ocnt ) by @xmath65 , where @xmath66 and @xmath67 are integers with @xmath68 because of the @xmath1 symmetry .
we call @xmath69 the sawtooth ocnt ( @xmath70 , @xmath71 ) , and @xmath72 the armchair ocnt ( @xmath73 , @xmath74 ) , where the chiral angle is defined as @xmath75 .
the tube diameter is given by latexmath:[$d_t = sawtooth ocnt ( 10,0 ) and armchair ocnt ( 8,8 ) are presented in fig .
1(b ) as examples , whose structural stabilities were verified by the dft calculations .
the energy bands of ocnt ( 3,3 ) , ( 5,5 ) , ( 12,12 ) , ( 5,0 ) , ( 8,0 ) and ( 16,0 ) are given in fig .
4(c ) , which show that regardless of the curvature effect , the single - walled ocnts , either sawtooth or armchair , are all metallic . in addition , by wrapping octagraphene and using pentagons , hexagons or heptagons as caps , many structurally stable unconventional carbon fullerenes can be built .
the optimized examples like c@xmath6 , c@xmath7 , c@xmath8 and c@xmath9 are shown in fig .
these results show that octagraphene is indeed a versatile 2d crystalline allotrope of carbon after graphene .
) for different structures of ( a ) , where the doping concentration is from small to large in sequence.,width=321 ] similar to graphene and carbon nanotubes@xcite , the semimetallic octagraphene may be applicable in the band engineering via the chemical doping .
in fact our calculation reveals that a band gap opens if the octagraphene is substitutionally doped with boron and nitrogen ( b - n ) pairs , as shown in fig . [ gap ] , where eleven structures for different doping configurations and a pure b - n octa - square structure are considered .
all these doped structures are energetically favorable . in fig .
[ gap ] ( a ) , the structures of number 1 - 11 are those of octagraphene of 4@xmath134 supercell with substituting carbon ( blue ball ) by boron ( red ball ) and nitrogen ( green ball ) atom pairs , and the structure of number 12 is the pure b - n octagon - square structure of 3@xmath133 supercell . in the first structure two carbon neighbors are replaced by one b - n pair ; in the second ( third ) structure a square ( octagon ) is substituted by two ( four ) b - n pairs ; in the fourth ( sixth , seventh or ninth ) one a nanoribbon in octagraphene is substituted by four ( eight , twelve or sixteen ) b - n pairs ; in the fifth one a 2@xmath132 supercell is replaced by eight b - n pairs ; in the eighth one a cross is replaced by fourteen b - n pairs ; the tenth one is the dual structure of the eighth ; the eleventh is the dual structure of the fifth .
the band gap varies significantly for different doping configurations , as shown in fig .
[ gap ] ( b ) .
it is seen that the boron nitrogen pair doped octagraphene can open a gap from 0.009 ev ( structure 2 ) to 1.866 ev ( structure 11 ) , which can gain some insights for nanoelectronics .
it is noting that the band gaps of most of structures considered here are around 0.2 ev , even when the doping concentration is as large as 50.82% ( structure 9 ) , which is quite different from the doped graphene @xcite .
considering that the gga usually underestimates the band gap , the present doping study gives useful information for the band engineering in octagraphene . ) , two bridge sites on top of intra - square ( @xmath77 ) and inter - square ( @xmath78 ) c - c bond , respectively , and two hollow sites on top of square ( @xmath79 ) and octagon ( @xmath80 ) , respectively.,width=207 ] site .
reveal the single - side hydrogen storage configurations with one , two , three and four hydrogen molecules adsorbed on each ti atom , respectively ; ( e)-(h ) show the double - side hydrogen storage configurations with one , two , three and four h@xmath81 adsorbed on each ti atom , respectively ; where c , ti and h atoms are colored in dark blue , bluish green and brick red , respectively.,width=359 ] next , let us consider the possibility of hydrogen storage in octagraphene . in general , there are two ways to realize this purpose .
one is that hydrogen atoms can be absorbed on carbons via the chemical bond to form a new hydrocarbon sheet , similar to graphane @xcite ; the other is that the hydrogen molecules can be adsorbed physically on some metal atoms ( e.g. ti , ca , al , etc . )
that are deposited on the sheet of octagraphene . in the former case
, the hydrogen storage capacity of octagraphene can reach 7.7 wt% as graphane does @xcite . in the latter case ,
the hydrogen capacity depends on the adsorption configuration of metal atoms .
there are five possible adsorption positions for metal atoms ( see fig .
[ site ] ) , including one top site on c atoms ( @xmath82 ) , two bridge sites on top of intra - square ( @xmath77 ) and inter - square ( @xmath78 ) c - c bond , respectively , and two hollow sites on top of square ( @xmath79 ) and octagon ( @xmath80 ) , respectively . to analyze the stability of these five adsorption configurations , we calculate the cohesive energy @xmath83 with @xmath11 the adsorbed metal atom by @xmath84/n,\end{aligned}\ ] ] where @xmath85 , @xmath86 and @xmath87 are the total energies of pure octagraphene , @xmath11 atom and octagraphene with adsorbed metal @xmath11 , respectively . for ti doping ( @xmath11=ti ) ,
@xmath79 site is the most stable position with cohesive energy of 2.71 ev per ti atom
. figs .
[ hydrogen](a)-(d ) present single - side hydrogen storage configurations with one , two , three and four hydrogen molecules adsorbed on each ti atom .
the average adsorption energy per h@xmath81 is 0.207 ev , 0.361 ev , 0.336 ev and 0.325 ev , respectively
. figs . [ hydrogen ] ( e)-(h ) give double - side hydrogen storage configurations with one , two , three and four h@xmath81 adsorbed on each ti atom .
the average adsorption energy per h@xmath81 is 0.235 ev , 0.355 ev , 0.336 ev and 0.325 ev , respectively . in this case
, the hydrogen storage capacity could reach 7.76 wt% for ti [ as shown in fig .
[ hydrogen](h ) ] , which is very close to that of ca - adsorbed graphene , say 8.4 wt% @xcite .
we also perform the same calculation for al and ca atoms ( @xmath11=al or ca ) , respectively , and found that the adsorption energy of hydrogen molecule is too small .
the adsorption energy of ti - adsorbed octagraphene is quite promising , say , about 0.3 ev / h@xmath81 .
it appears that the hydrogen storage capacity of octagraphene is close to that of fullerenes @xcite , carbon nanotubes @xcite and graphene @xcite .
in conclusion , we propose a periodic carbon sheet by the first - principles calculations .
this novel carbon monolayer is a @xmath0-hybridized structure and coined as octagraphene . unlike graphene , octagraphene
is composed of carbon squares and octagons possessing @xmath1 symmetry , and is more favorable in energy than graphyne and graphdiyne .
its optimized structure has two bond lengths with 1.48 within each square and 1.35 between adjacent squares .
the band structure shows that it is a semimetal with fermi surface consisting of one hole pocket and one electron pocket . in order to describe the low - energy physics of octagraphene
, we also suggest a tight binding model of @xmath2 electrons which is uncovered to agree well with dft calculations , and is quite helpful for further exploring the properties of octagraphene in a magnetic field or emergent quantum phenomena such as quantum hall effect , kondo effect , and so on in octagraphene .
the intriguing mechanical properties of octagraphene are also obtained , which are observed close to those of graphene .
similar to graphene , we find that octagraphene can also be rolled seamlessly to form stable carbon nanotubes and unconventional fullerenes . by substitutionally doping boron nitrogen pairs
, we disclose that the semimetallic octagraphene can be turned into a semiconductor with a band gap , the property will be useful for possible applications in nanoelectronics .
this new carbon allotrope has also the possibility for hydrogen storage .
our calculation shows that for the 2d octagraphene sheet the hydrogen storage capacity could reach 7.76% via ti - adsorption that is close to that of graphene .
finally , we would like to mention that one might obtain this new kind of carbon atomic sheet via making line defects in graphene or through the acetylene scaffolding ways .
we are grateful to dr .
eric germaneau for useful discussions .
all calculations were completed in the shanghai supercomputer center , china .
this work is supported by the nsfc ( grants no .
90922033 , no . 10934008 , no .
10974253 , no . 10904081 and no .
11004239 ) , the most of china ( grant no .
2012cb932900 ) and the cas .
vincent h. crespi , lorin x. benedict , marvin l. cohen , and steven g. louie , phys .
b * 53 * , r13303 ( 1996 ) ; h. terrones , m. terrones , e. hern@xmath88ndez , n. grobert , j - c . charlier , and p. m. ajayan , phys .
lett . * 84 * , 1716 ( 2000 ) ; x. rocquefelte , g .-
rignanese , v. meunier , h. terrones , m. terrones , and j .- c .
charlier , nano lett .
* 4 * , 805 ( 2004 ) .
paolo giannozzi , stefano baroni , nicola bonini , matteo calandra , roberto car , carlo cavazzoni , davide ceresoli , guido l chiarotti , matteo cococcioni , ismaila dabo , andrea dal corso , stefano de gironcoli , stefano fabris , guido fratesi , ralph gebauer , uwe gerstmann , christos gougoussis , anton kokalj , michele lazzeri , layla martin - samos , nicola marzari , francesco mauri , riccardo mazzarello , stefano paolini , alfredo pasquarello , lorenzo paulatto , carlo sbraccia , sandro scandolo , gabriele sclauzero , ari p seitsonen , alexander smogunov , paolo umari and renata m wentzcovitch , j. phys .
matter * 21 * , 395502 ( 2009 ) . t. o. wehling , k. s. novoselov , s. v. morozov , e. e. vdovin , m. i. katsnelson , a. k. geim , and a. i. lichtenstein , nano lett . * 8 * , 173 ( 2008 ) ; m. deifallah , p. f. mcmillan , f. cora , j. phys . chem
. c , * 112 * , 5447 ( 2008 ) ; t. b. martins , r. h. miwa , ant@xmath89nio j. r. da silva , and a. fazzio , phys .
lett . * 98 * , 196803 ( 2007 ) ; f. cervantes - sodi , g. cs@xmath88nyi , s. piscanec , and a. c. ferrari , phys .
b * 77 * , 165427 ( 2008 ) ; matteo calandra and francesco mauri , phys .
b * 76 * , 161406(r ) ( 2007 ) ; r. rold@xmath88n , m. p. l@xmath90pez - sancho , and f. guinea , phys . rev .
b * 77 * , 115410 ( 2008 ) .
r. czerw , m. terrones , j. c. charlier , x. blase , b. foley , r. kamalakaran , n. grobert , h. terrones , d. tekleab , p. m. ajayan , w. blau , m. r@xmath91hle , and d. l. carroll , nano lett .
* 1 * , 457 ( 2001 ) ; kai xiao , yunqi liu , pingan hu , gui yu , yanming sun , and daoben zhu , j. am . chem . soc . * 127 * , 8614 ( 2005 ) . yufeng zhao , yong - hyun kim , a. c. dillon , m. j. heben , and s. b. zhang , phys .
lett . * 94 * , 155504 ( 2005 ) ; mina yoon , shenyuan yang , christian hicke , enge wang , david geohegan , and zhenyu zhang , phys .
lett . * 100 * , 206806 ( 2008 ) ; qiang sun , puru jena , qian wang , and manuel marquez , j. am .
soc . * 128 * , 9741 ( 2006 ) . | we study a versatile structurally favorable periodic @xmath0-bonded carbon atomic planar sheet with @xmath1 symmetry by means of the first - principles calculations .
this carbon allotrope is composed of carbon octagons and squares with two bond lengths and is thus dubbed as octagraphene .
it is a semimetal with the fermi surface consisting of one hole and one electron pocket , whose low - energy physics can be well described by a tight - binding model of @xmath2-electrons .
its young s modulus , breaking strength and poisson s ratio are obtained to be 306 @xmath3 , 34.4 @xmath3 and 0.13 , respectively , which are close to those of graphene .
the novel sawtooth and armchair carbon nanotubes as well as unconventional fullerenes can also be constructed from octagraphene .
it is found that the ti - absorbed octagraphene can be allowed for hydrogen storage with capacity around 7.76 wt% . |
for definiteness , consider a scalar field @xmath8 of mass @xmath6 minimally coupled to gravity . following the hamiltonian treatment of linear perturbations in cosmology @xcite , we can separate the homogeneous and the inhomogeneous degrees of freedom , and the classical dynamics for mode @xmath2 of @xmath8 is generated ( up to second order ) by the hamiltonian @xmath9\end{aligned}\ ] ] here , @xmath10 are the phase space variables representing mode @xmath2 , while @xmath11 are the conjugated variables representing the homogeneous degrees of freedom of gravity ( that is , the scale factor and its momentum ) .
@xmath12 is the part of the hamiltonian that accounts for the evolution of these gravitational degrees of freedom at the same order . at this order , and can thus be disregarded in light of the born - oppenheimer test field approximation ( see later ) . ] after formal quantization of matter and gravity , ( [ mat ] ) defines the following schroedinger - like equation @xmath13 \psi\end{aligned}\ ] ] where @xmath14 and @xmath15 , with @xmath1 being the hilbert space of quantum gravity . : _ any _ theory of quantum gravity will do .
] at this point , we take the _ test field _
approximation : we assume that the scalar field does not back - react on the gravitational part .
it is therefore allowed to retain only the 0th order in the born - oppenheimer expansion of @xmath16 : during the whole evolution @xmath17 , where @xmath18 and @xmath19 evolves via schroedinger - like equation @xmath20 .
this being the case , we can trace away the gravitational part in ( [ q - mat ] ) and obtain an equation for the matter part only : @xmath21 where @xmath22\end{aligned}\ ] ] the point first observed in @xcite and further analysed in @xcite , is that equation ( [ final ] ) resembles the schroedinger equation for a quantum field @xmath8 on a suitably defined _ classical _ spacetime .
let the spacetime be classically described by a metric @xmath23 of the robertson - walker type : @xmath24 construcing regular qft on such a curved spacetime , one obtains for mode @xmath25 of @xmath8 the following effective schroedinger equation : @xmath26 where @xmath27\end{aligned}\ ] ] in other words , we can replace the fundamental theory described by ( [ q - mat ] ) with regular qft on curved spacetime ( [ eff - geo ] ) , provided that the terms in the two hamiltonians ( [ funh ] ) and ( [ effh ] ) match .
this last requirement gives rise to a system of 2 equations for 2 unknowns : @xmath28 the solution of the system is @xmath29 where @xmath30 is the solution to the algebraic equation @xmath31 it is a non - trivial fact that this equation has a unique positive solution for every @xmath32 .
it is given explicitly by @xmath33 where @xmath34{\dfrac{\delta}{2 } - \dfrac{k^6}{27 m^6 } \pm \sqrt{\dfrac{\delta^2}{4 } - \dfrac{k^6}{27m^6}\delta}}\end{aligned}\ ] ] and @xmath35 the two functions in ( [ goodgod ] ) match continuously at @xmath36 , where @xmath37 is the unique positive solution to equation @xmath38 .
for @xmath39 we are in the first case , while for @xmath40 in the second .
let us expand ( [ goodgod ] ) for @xmath41 . up to order @xmath42
, we have @xmath43\end{aligned}\ ] ] with @xmath44{\frac{\langle \widehat{h_o^{-1 } { a^6 } } \rangle}{\langle \hat{h}_o^{-1 } \rangle } } , \ \ \ \ \ \beta : = \frac{\langle \widehat{h_o^{-1 } { a^4 } } \rangle}{\langle \hat{h}_o^{-1 } \rangle^{1/3 } \langle \widehat{h_o^{-1 } { a^6 } } \rangle^{2/3 } } - 1\end{aligned}\ ] ] from @xmath45 and @xmath46 we find @xmath47 and @xmath48 via the first equation in ( [ sol - c ] ) .
we can then identify two effective flrw metrics : the low - energy one @xmath49 and the @xmath2-dependent one @xmath50 we can interpret @xmath51 as the metric measured by a classical observer , while scalar field ( and especially its relativistic modes ) propagate on @xmath52 .
are meaningful .
we identify unambiguously this family as the cosmological ( i.e. , comoving ) observers of the metric @xmath51 .
the reason to consider such observers as `` classical '' is the following : if such an observer only performs measurement of geodesics of macroscopic bodies ( for which @xmath41 , @xmath2 being the momentum she measures ) , she will only investigate the regime in which @xmath53 , therefore concluding that the spacetime is described by @xmath51 . +
a mathematical description of the situation is the following : there exists a manifold whose homogeneous and isotropic metric is @xmath51 ; semiclassical observers coincide with cosmological observers of this metric ; if the momentum @xmath2 of a particle as measured by such observers is non - negligible compared to the mass @xmath6 , then such particle obeys a dispersion relation with @xmath52 . ]
suppose that an observer with 4-velocity @xmath54 detects a particle with 4-momentum @xmath55 .
the energy and ( norm of ) momentum measured by the observer are @xmath56 where we used the fact that @xmath57 to discover that @xmath58 . on the other hand ,
the particle satisfies the mass - shell relation in _ its _ metric ( [ metrick ] ) : @xmath59 having introduced the so - called `` rainbow functions '' @xcite @xmath60 from ( [ low - k - lim ] ) it is immediate to compute @xmath61 and @xmath62 , which explicitly depend on the physical momentum @xmath63 : @xmath64 we thus obtain a modified dispersion relation from ( [ massshell ] ) , @xmath65 as expected , the standard dispersion relation @xmath66 is recovered in the limit @xmath67 . the first correction in the case
@xmath68 is precisely @xmath7 , a quantity of exquisitely quantum gravitational origin .
note that contrary to the general belief no particular role is played by planck energy , @xmath69 .
in fact , for a highly quantum spacetime we have @xmath70 , and hence the particles probe the quantum structure of spacetime already at @xmath68 . for a proton
, this would correspond to mild energies of order @xmath71 . on the other hand ,
it is clear that @xmath72 for semiclassical states , and hence quantum gravity corrections are irrelevant for low - energy particles .
we should mention that a similar result was recovered in the semiclassical limit in @xcite .
having the dispersion relation , it is possible to compute the velocity of the mode : @xmath73 this expression only holds in the limit @xmath67 , but it is enough to show the deformation already at low energies . the exact dispersion relation ( obtained numerically from ( [ goodgod ] ) ) is represented in figures 1 and 2 , where the classical one ( @xmath72 ) is compared with the choice @xmath74 corresponding to @xmath75 ( a highly non - classical situation ) . in figure 2
we also show the dispersion relation for massless particles : there is no dependence of velocity of light on @xmath5 ( though there is still a dependence on @xmath7 , and hence on time ) . to see why this happens , consider the limit @xmath76 of ( [ goodgod ] ) .
it leads to @xmath77 , \ \
\ \bar{a}_\infty^2 : = \sqrt{\dfrac{\langle \widehat{h_o^{-1 } a^4 } \rangle}{\langle \hat{h}_o^{-1 } \rangle}}\end{aligned}\ ] ] the zeroth order coincides with the solution to system ( [ system ] ) with @xmath78 , that is , the system we would have obtained if we considered a massless scalar field from the start .
the massless scalar field case was first studied in @xcite , and @xmath79 of ( [ high - energy - metric ] ) coincides with the result therein .
now , since @xmath79 is independent of @xmath2 , the metric seen by particles with @xmath78 is @xmath2-independent ( though different from the semiclassical metric , @xmath46 ) .
it is therefore not surprising that no mode - dependence is found in the velocity of light particles .
this can be made explicit by computing the dispersion relation for such field : plugging @xmath80 and @xmath81 in the general formula ( [ disprel ] ) with @xmath78 , we derive @xmath82{\frac{\langle \hat{h}_o^{-1 } \rangle}{\langle \widehat{h_o^{-1 } { a^6 } } \rangle } } \ p = \sqrt{1 + \beta } \ p\end{aligned}\ ] ] thus , as far as massless particles are concerned , the effect of the quantum background amounts to a constant shift in the velocity of light .
while this effect can not be detected by any local measurement , we stress that the derivation was done for a massless scalar field , and might well be different in the case of photons . for a scalar field of mass @xmath6 .
red = classical spacetime ; blue = quantum spacetime ( eq .
( [ quantum - state])).,height=211 ] [ fig : vels1 ] of different modes of the massive field .
red = classical spacetime ; blue = quantum spacetime ( eq .
( [ quantum - state ] ) ) .
the dashed lines represent the speed of light in the semiclassical spacetime ( black ) and in the quantum spacetime ( green ) respectively.,height=211 ] [ fig : vels2 ] in conclusion , our analysis shows that the quantum nature of cosmological spacetime unavoidably affects the propagation of test particles , producing ( apparent ) lorentz - violating effects .
this has been shown by constructing an effective metric from general assumptions ( in particular , we did not need to restrict to a specific quantum theory of cosmology ) .
intuitively , the generality of the result can be understood by observing that independently of the chosen quantum theory
the quantum state @xmath0 of the homogeneous gravitational field is not an eigenstate of the `` scale factor operator '' in general .
hence , it is to be expected that matter particles of different momenta @xmath5 will couple differently to the quantum geometry , probing different aspects of it . in the present work ,
we have used a massive scalar field @xmath8 one of the simplest forms of matter but there is no reason not to expect the same qualitative behaviour for other species .
as for the characterization of this effect , we have shown that the only parameter governing the corrections is @xmath7 , a single function of @xmath0 .
this is rather striking , considering that infinitely many quantum states can be found that give the same value @xmath7 . on the other hand
, we should not be too surprised , since the same happens for photons propagating in refractive media : the scattering of light through a crystal is perfectly well described in terms of the refractive index @xmath83 , a single parameter in spite of the infinitely many possible microscopic configurations of the atoms .
moreover , we should stress that the current results are based on a homogeneous quantum background : it is possible that inclusion of inhomogeneities might invalidate this result , but it is also conceivable that it will make it even more interesting ( for instance , @xmath7 might depend on the position , as does the refractive index of a non - homogeneous medium ) . but
how strong is this effect ?
how big is @xmath7 ?
while there is no _ fundamental _ reason why @xmath7 should be small , it is an _ observational fact _ that the universe today is classical .
this means that the state @xmath0 describing the current geometry of spacetime must be a coherent state with @xmath7 extremely small @xcite .
our philosophy is therefore to use @xmath7 as a test for the soundness of coherent states within a specific quantum cosmology or in case no good coherent state can be found as an indicator that said theory of quantum cosmology is incorrect . while @xmath72 today , several quantum gravity and quantum cosmology theories maintain that in the early stages of its life , the universe should be described by a non - classical state . in this case @xmath70 , and the modified dispersion relations have to be taken into account when studying the behaviour of primordial matter .
we can expect that traces of such effects are left on the cmb , or even speculate that they might have played a role in inflation and subsequent formation of structures .
this work has been partially financed by the grants of polish narodowe centrum nauki nr 2011/02/a / st2/00300 and nr 2013/09/n / st2/04312 . | in this letter , we describe a general mechanism for emergence of a rainbow metric from a quantum cosmological model . this idea is based on qft on a quantum spacetime . under general assumptions
, we discover that the quantum spacetime on which the field propagates can be replaced by a classical spacetime , whose metric depends explicitly on the energy of the field : as shown by an analysis of dispersion relations , quanta of different energy propagate on different metrics , similar to photons in a refractive material ( hence the name `` rainbow '' used in the literature ) . in deriving this result
, we do not consider any specific theory of quantum gravity : the qualitative behavior of high - energy particles on quantum spacetime relies only on the assumption that the quantum spacetime is described by a wave - function @xmath0 in a hilbert space @xmath1 .
it has been argued @xcite that classical gravity could be a collective phenomenon emerging from quantum degrees of freedom , not unlike fluid dynamics emerges from microscopic molecular interactions .
it is often stated that such an effective spacetime should be described by a so - called rainbow metric " @xcite , i.e. , a metric that depends somehow on the energy of the particles propagating on it : it is not difficult to conceive that probing such an effective spacetime with high enough energies leads eventually to corrections due to the underlying fundamental quantum structure .
a fundamental origin for rainbow metrics has been identified ( in the principle of relative locality @xcite ) , their phenomenology has been studied @xcite , and tests ( based on the lorentz - violating nature of such energy - dependent metrics ) have been proposed @xcite .
what was missing until today as far as our knowledge goes is a general _ mechanism _ which produces an emergent rainbow metric from a quantum spacetime . indeed , while various proposals for quantum gravity can all be argued to reproduce classical gravity in the low energy limit , it seems to us that a clean procedure to extract this limit is yet to be formulated . in this letter
, we put forward such a proposal . in section 1 , we provide a mechanism for emergence of cosmological spacetime from quantum cosmology in complete generality ( we only require gravitational degree of freedom to be described in terms of a state @xmath0 in a hilbert space @xmath1 , and to be `` heavy '' compared to the matter degrees of freedom in the born - oppenheimer sense ) . the idea for this mechanism
is based on qft on quantum spacetime as first introduced in @xcite . with no _ ad - hoc _ input ,
we find that the effective metric describing the emergent spacetime is indeed of the rainbow type , as it depends on the wave - vector @xmath2 of the mode of the matter field . , and it does not lead to a @xmath2-dependent metric . whether the two approaches are physically equivalent is currently unknown , though the issue is being investigated by the authors . ] in section 2 we perform a low - energy expansion of this metric , and show that the first correction to the classical metric " @xmath3 is of order @xmath4 , where @xmath5 is the physical momentum of the mode , @xmath6 is the mass of the field , and @xmath7 is a simple function of @xmath0 .
it is rather surprising that the only information needed to reconstruct the effective metric from the quantum spacetime is just parameter @xmath7 .
finally , in section 3 , we study the modified dispersion relation of this emergent metric , and find that heavy particles as opposed to light ones behave in a different way than in classical gravity .
in particular , the velocity of light remains an upper bound , but is now dependent on @xmath7 . |
emerging patterns due to coarsening are encountered in a huge variety of systems @xcite , ranging from magnetic systems @xcite to complex fluids @xcite and from competition in bacterial colonies @xcite to opinion dynamics in social systems @xcite .
the general understanding of phase ordering kinetics @xcite is well developed for a range of systems , notably those that are curvature driven and where the size of a typical domain increases with the square root of time .
power - law growth is also observed in other instances , as for example in systems with a conserved order parameter undergoing ostwald ripening @xcite .
however , coarsening processes in more complex systems can yield a domain growth much slower than the algebraic growth observed in the simpler situations @xcite .
systems where a logarithmic growth of the form @xmath1 is expected include spin glasses and disordered ferromagnets
. in general , it is challenging , both in experimental and numerical studies , to reach the very long timescales needed to fully access the asymptotic regime .
this has been a vexing issue of many years for spin glasses . on the other hand , disordered ferromagnets , while still being challenging systems , are more easily dealt with in that regard , and over the years a range of studies focusing on elastic lines in disordered media @xcite and on disordered ising systems @xcite , including the one - dimensional random field ising model @xcite , have shown that the asymptotic growth regime is indeed non - algebraic and compatible with a logarithmic growth of the domains . a logarithmically growing length scale is not restricted to disordered systems , but is also encountered in some systems dominated by dynamical constraints rather than by quenched disorder @xcite .
anomalous coarsening with logarithmic domain growth has been identified in a variety of situations with dynamical constraints , as for example in the @xmath0 model introduced by evans et al @xcite , where three different particle types swap places asymmetrically , in the model discussed by lahiri and ramaswamy @xcite , where two sublattices are considered with two types of particles on each sublattice , or in the driven two - lane particle system studied by lipowski and lipowska @xcite . among these models ,
the one - dimensional @xmath0 model has enjoyed much attention in recent years @xcite , due to its unique properties combined with its amenability to exact calculations in some special cases .
thus the @xmath0 model provides an opportunity to study exactly the long - range correlations in a system undergoing a non - equilibrium phase transition as well as to elucidate ensemble inequivalence far from equilibrium . whereas most of the papers on the @xmath0 model focused on steady - state properties , only few studies directly addressed the coarsening process , the logarithmic growth and other issues related to this anomalous slow process . after having identified the anomalous slow domain growth in the @xmath0 model , evans et al @xcite proposed a simplified interface model where domains rather than individual sites are updated . in @xcite
slow coarsening was shown to also exist in a two - dimensional version of the @xmath0 model .
finally , in @xcite the focus was on the interface model in one dimension whose study provided some insights into aging and dynamical scaling in presence of a logarithmically growing length .
in this paper we aim at further elucidating slow coarsening and relaxation processes in both the one- and two - dimensional @xmath0 models . in order to do so
, we study a range of different quantities ( the time - dependent space - time correlation function and the length scale obtained from it , two - times quantities like the autocorrelation and response functions , as well as the mean displacement of individual particles during the coarsening process ) that allow us to gain a good understanding of the non - equilibrium processes taking place in this system .
we thereby discuss different regimes that can be identified during the coarsening process .
the effects of the system size and of the swapping rate on the domain ordering are discussed .
the paper is organized in the following way . in the next section we recall the model , both in one and two space dimensions , and introduce the quantities that we study in the remainder of the paper .
section iii is devoted to a discussion of the time - dependent lengths that govern the coarsening process .
whereas asymptotically the length in the horizontal direction increases as a logarithm of time , complicated early time regimes emerge , depending on the degree of asymmetry and the extent of the system .
additional insights are provided by the mean displacement of individual particles . in section
iv we study two - times quantities . whereas the autocorrelation exhibits obvious dynamical scaling properties , a complicated response of the system to perturbations ( realized through a sudden change of the swapping rates )
is observed .
we summarize our results in the final section .
we consider particles of three different types ( called @xmath2 , @xmath3 , and @xmath4 ) moving on a two - dimensional lattice with @xmath5 sites and periodic boundary conditions in both the @xmath6- and @xmath7-direction .
whereas in the @xmath7-direction all exchanges are symmetric and happen with rate 1 , in the @xmath6-direction a bias is introduced that yields an asymmetry in the exchanges and , concomitantly , an asymmetrical diffusion of the particles @xcite .
this is achieved by allowing exchanges between particles on neighboring sites in the horizontal direction to take place with the following rates : @xmath8 i.e. a @xmath3 particle and an @xmath2 particle to the right of the @xmath3 particle will swap places with rate 1 , whereas the same two particles will swap places with some rate @xmath9 in case their order is different .
unbiased exchanges are of course recovered when @xmath10 . the driven diffusive motion that results when @xmath9 entails phase separation and the formation of ordered domains in the form of vertical stripes @xcite that are arranged in repetitions of the sequence @xmath0 where @xmath2 stands for a domain occupied by a majority of @xmath2 particles . for comparison , we also simulate the @xmath0 model on a ring composed of @xmath11 sites @xcite where the same biased rates are used as for the horizontal direction in the two - dimensional model . most of the data discussed below have been obtained for systems with @xmath12 .
we carefully checked that this horizontal length is big enough so that it does not yield finite - size effects for the times accessed in our simulations . in all our simulations
we consider fully occupied lattices where initially every species occupies exactly one third of the lattice sites chosen at random .
this is in fact a special situation in one space dimension , as detailed balance is then fulfilled @xcite so that the system relaxes to an equilibrium steady state : in case of identical coverage of the lattice by each of the species , a hamiltonian with long - range interactions can be derived .
however , as soon as the particle densities are not the same for all three species , detailed balance is broken in one dimension and the steady state is a non - equilibrium steady state .
the phase transition between the ordered phase for @xmath13 and the disordered phase at @xmath10 ( in the large volume limit ) then becomes a non - equilibrium phase transition @xcite which has been at the center of some of the recent studies dealing with the @xmath0 model .
for the two - dimensional model , however , detailed balance is always broken , even when the three particle densities are identical , and the steady state is always a non - equilibrium steady state @xcite .
we are using a standard scheme for our simulations where we randomly select a neighboring pair of sites . in case
the particles occupying these sites belong to different species , they are exchanged with the direction - dependent rates given above .
we define a time step as @xmath14 such proposed updates , where @xmath15 is the total number of sites forming the lattice .
every lattice site @xmath16 can be characterized by a time - dependent three - state potts variable @xmath17 . using that description
, an exchange of particles occupying neighboring sites corresponds to updating the values of the potts variables characterizing these sites . in this way
it is also very easy to write down expressions for the quantities measured in our simulations .
a central quantity for our analysis is the equal - time spatial correlation @xmath18 where @xmath19 is the kronecker delta .
the sum is over all lattice sites , and @xmath20 stands for an ensemble average over both initial configurations and noise realizations .
the subtraction by @xmath21 assures that in an infinite system @xmath22 goes to zero when @xmath23 . the space - time correlation ( [ eq : spacecorr ] ) allows the extraction of a time - dependent correlation length .
as we use different swapping rates in horizontal and vertical directions , the spatial correlation , and therefore the correlation length , is direction dependent .
for example , in the horizontal direction a time - dependent length @xmath24 can be obtained in a standard way by determining the intersection of the normalized correlation @xmath25 with a constant function @xmath26 : @xmath27 we carefully checked that the qualitative features of @xmath24 discussed below do not depend on the chosen value of @xmath26 .
the data shown in the following have been obtained for @xmath28 .
alternatively , a time - dependent horizontal length can be obtained by measuring the average horizontal extent of a connected cluster formed by particles of the same species .
we carefully compared the two lengths and found that both lengths are completely equivalent and show the same qualitative features .
a vertical length can be obtained from the vertical extent of these clusters .
two - times quantities have been shown in many studies to be very valuable when investigating the non - equilibrium properties of systems relaxing to a steady state @xcite .
the two - times autocorrelation function @xmath29 compares the configurations at the waiting time @xmath30 with that at the observation time @xmath31 .
important insights can also often be obtained from two - times response functions . for the model discussed here the only control parameter that can be used to perturb
the system is the swapping rate @xmath32 .
therefore we use as our perturbation a sudden global change of @xmath32 from some initial value @xmath33 to some final value @xmath34 at time @xmath30 after preparing the system . for times
@xmath31 we then monitor how the system relaxes to the final steady state by measuring the time - dependent horizontal length @xmath35 of the perturbed system .
of particular interest is the difference @xcite @xmath36 between the length measured in the perturbed system and that measured in the system where @xmath37 at all times .
all the quantities mentioned up to now do not provide direct information on the behavior of individual particles . in order to probe the system at this more microscopic level , we tag individual particles and follow their motion . from the resulting individual trajectories we determine the displacement of each particle from its initial position as a function of time . in the two - dimensional system particles of course move in both dimensions , but the vertical motion is rather trivial as exchanges between neighboring particles of different types always take place .
for that reason we only discuss the mean displacement of the particles in the horizontal direction : @xmath38 where @xmath39 is the horizontal coordinate of the position vector @xmath40 of particle @xmath41 at time @xmath42 , whereas @xmath43 is the horizontal coordinate of the position of the same particle in the initial configuration .
as discussed in the previous section , in two space dimensions we can extract both a horizontal and a vertical length .
due to the anisotropy of the system , these lengths are not identical . in order to directly compare with the one - dimensional @xmath0 model , we mostly focus in the following on the length in the horizontal direction .
[ h ] ( color online ) time - dependent length @xmath24 for ( a ) @xmath44 and ( b ) @xmath45 .
systems with different extents in the @xmath7-direction are considered , ranging from 1 ( top curve for large @xmath42 ) to 100 ( bottom curve for large @xmath42 ) lines .
the length is obtained from the space - time correlation function @xmath46 in the horizontal direction .
some configurations are shown for the @xmath47 system .
for the larger systems the length is extracted after averaging over more than 1000 independent realizations , whereas for the one - dimensional system the ensemble average is performed over 16000 independent runs .
, title="fig : " ] + ( color online ) time - dependent length @xmath24 for ( a ) @xmath44 and ( b ) @xmath45 .
systems with different extents in the @xmath7-direction are considered , ranging from 1 ( top curve for large @xmath42 ) to 100 ( bottom curve for large @xmath42 ) lines .
the length is obtained from the space - time correlation function @xmath46 in the horizontal direction .
some configurations are shown for the @xmath47 system . for the larger systems the length is extracted after averaging over more than 1000 independent realizations , whereas for the one - dimensional system
the ensemble average is performed over 16000 independent runs .
, title="fig : " ] fig .
[ fig1 ] shows the horizontal length @xmath24 for systems of up to 100 lines and two very different values of @xmath32 , namely @xmath44 in fig .
[ fig1]a and @xmath45 in fig .
the value @xmath44 is close to the value @xmath10 for unbiased exchanges .
this closeness to @xmath10 gives raise to a very slow initial onset of the ordering process , and it takes for a system composed of 10 or more lines around 2000 time steps before a growth in @xmath24 can be observed .
this is different for the case of strong bias , @xmath45 , where ordering emerges already after less than 100 time steps .
once ordering sets in for @xmath44 , it rapidly yields the formation of ordered stripes ( see the configurations in fig .
[ fig1]a for the system with 100 lines ) .
these stripes initially contain a large number of particles from the minority species .
when the domains are well formed , a coarsening process sets in that reveals itself by a logarithmic growth of @xmath24 . for systems with @xmath45 and only a few lines ,
this behavior is qualitatively very similar to the behavior encountered for @xmath44 , the main differences being the earlier onset of ordering and the rapid formation of a large number of well ordered domains in form of stripes .
once these domains are formed , they coarsen very slowly and grow only logarithmically with time .
interestingly , for systems with a larger vertical extent ( see the data for the @xmath47 system in fig .
[ fig1]b ) an intermediate regime shows up after the onset of ordering but before the domain coarsening .
the nature of this regime , which is characterized by an accelerated increase of @xmath24 followed by a strong slowing down of the growth process , is revealed when inspecting the typical configurations shown in fig .
[ fig1]b and fig . [ fig2 ] .
in the early stages of the ordering process , clusters merge and grow in vertical direction until they reach from bottom to top .
this vertical growth , which gives raise to a transient algebraic growth regime , yields initially some defects that need to be eliminated before entering the logarithmic coarsening regime .
consequently , the asymptotic growth regime sets in much later than for the systems with only a few lines . [ h ] ( color online ) snapshots of a system with @xmath45 composed of @xmath47 sites for six different times between @xmath48 and @xmath49 . only one of the three particle types is shown for clarity . the times shown span the initial regime of domain formation , the regime where the domains merge and expand in vertical direction , and the regime where defects are eliminated and the straight stripes are forming , after which the system enters the logarithmic coarsening regime .
the circle highlights one instance where the misalignment of merging domains and the following expansion in the vertical direction yields bended domains as well as a @xmath50-shaped connection .
, title="fig : " ] in order to gain a more microscopic picture of the processes taking place in the pre - asymptotic regimes , we show in fig .
[ fig2 ] six different snapshots , with times between @xmath48 and @xmath49 , for a system with @xmath47 sites and @xmath45 .
only one of the three particle types is shown , as this allows to better see the processes taking place in the system .
the initial ordered clusters develop at random positions and rapidly merge into anisotropically ordered domains where the growth in vertical direction is much faster than in horizontal direction .
when merging , these vertically rapidly growing domains are not always well aligned , which yields defects as for example bended stripes or @xmath50-shaped connections . the red circles in fig .
[ fig2 ] indicate the formation and subsequent evolution of some defects .
once these imperfect stripes are formed , a regime sets in where the defects are eliminated . at @xmath49
only few defects are left and most of the stripes , which have become much straighter , extend through the whole system in vertical direction .
the asymptotic logarithmic growth regime is fully accessed only after all these defects have been eliminated .
[ h ] ( color online ) the same as in fig .
[ fig1 ] , but now for systems with a larger extent in @xmath7-direction .
the data result from averaging over at least 1000 independent runs .
the inset in ( b ) shows the same data as in the main panel , but now in a log - log plot .
the full solid line indicates the intermediate algebraic growth regime with an exponent 0.25(1 ) .
, title="fig : " ] + ( color online ) the same as in fig .
[ fig1 ] , but now for systems with a larger extent in @xmath7-direction .
the data result from averaging over at least 1000 independent runs .
the inset in ( b ) shows the same data as in the main panel , but now in a log - log plot .
the full solid line indicates the intermediate algebraic growth regime with an exponent 0.25(1 ) .
, title="fig : " ] fig .
[ fig3 ] shows the behavior of @xmath24 when further increasing the number of lines .
we first note from fig .
[ fig3]b that for @xmath45 the intermediate regime lasts longer the larger the vertical extent of the system is , as it takes longer to form the parallel stripes spanning the system .
the approximate power - law growth connected to this regime is revealed in the log - log plot shown in the inset .
the full solid line indicates that this growth is governed by an effective exponent of 0.25(1 ) .
for @xmath44 , see fig .
[ fig3]a , the intermediate regime also emerges for larger systems in @xmath7-direction .
in fact the same regimes exist for all values of @xmath32 and only the characteristic extent in @xmath7-direction needed for the existence of the intermediate regime depends on @xmath32 .
[ h ] ( color online ) time evolution of the vertical length @xmath51 for ( a ) @xmath44 and ( b ) @xmath45 .
the systems all have 600 sites in horizontal direction , but different extents in @xmath7-direction .
the different regimes indicated by numbers in ( b ) for the system with @xmath52 spins are discussed in the main text .
, title="fig : " ] + ( color online ) time evolution of the vertical length @xmath51 for ( a ) @xmath44 and ( b ) @xmath45 .
the systems all have 600 sites in horizontal direction , but different extents in @xmath7-direction .
the different regimes indicated by numbers in ( b ) for the system with @xmath52 spins are discussed in the main text .
, title="fig : " ] the different regimes identified from our analysis of @xmath24 also reveal themselves through characteristic features in the vertical length , as shown in fig .
[ fig4 ] where we plot the average vertical extent of connected clusters .
focusing first on fig .
[ fig4]b , we note that the initial domains grow rapidly in the early time regime ( 1 ) , before entering the intermediate regime ( 2 ) where the merging of ordered clusters yields defect structures that needs to be eliminated .
the growth regime ( 3 ) then corresponds to the elimination of these defects and the straightening of the stripes .
the asymptotic regime ( 4 ) , characterized by a logarithmic growth in horizontal direction , reveals itself in the following very slow increase of @xmath51 .
all these features are also observed in fig .
[ fig4]a for @xmath44 , albeit less pronounced and separated by more gradual transitions . in @xcite and @xcite interface models with simplified dynamics
were proposed both in one and two space dimensions in order to capture the long time behavior .
these models assume instantaneous moves of particles from one domain to another of the same type , with a rate that depends on @xmath32 as well as on the width of other stripes located between these two domains . in doing so
, one neglects that in the original @xmath0 model particles need a finite amount of time to cross these intermediate stripes .
[ h ] ( color online ) time - dependent length @xmath24 vs @xmath53 for various values of the bias @xmath32 in a system composed of @xmath54 sites .
the values of @xmath32 increase from top to bottom .
plotting @xmath55 on the @xmath7-axis yields in all cases curves with slopes @xmath56 .
, title="fig : " ] for the @xmath57 and @xmath54 systems we are able to access in the microscopic model the asymptotic growth regime for all values of @xmath32 between 0.9 and 0.1 . in agreement with what has been observed for the interface model @xcite , we find also for the microscopic model that the plot of @xmath24 versus @xmath53 yields a slop that depends on @xmath32 in the following way : @xmath58 with @xmath56 for both system sizes .
this is illustrated in fig .
[ fig5 ] for the system containing @xmath54 sites .
the value of @xmath59 is much smaller than the value @xmath60 found for the interface model in one dimension @xcite , in agreement with the expectation that the simplified dynamics in that model leads to a change in the value of the unit of time .
[ h ] ( color online ) mean displacement in horizontal direction of particles in a system composed of @xmath61 sites in ( a ) a log - log plot and ( b ) a log - linear plot .
the values of @xmath32 decrease from top to bottom .
results for different values of @xmath32 are shown .
for @xmath10 a square - root increase of @xmath62 is observed , as expected for an unbiased random walk . for @xmath13 a crossover to a regime where the displacement only grows logarithmically with time is encountered .
these data result from averaging over 16 independent runs . ,
title="fig : " ] taking a more microscopic point of view , we can record the trajectories of the individual particles and determine in this way their typical motion .
[ fig6 ] shows the mean displacement in horizontal direction of particles moving in systems with 100 lines . as is the case for @xmath24 ,
two well - separated regimes can also be identified for large @xmath32 values when studying this quantity , whereas for smaller @xmath32 values indications of an additional intermediate regime are seen .
initially , the mean displacement for every @xmath32 displays the same square - root behavior as for @xmath10 .
this indicates an unbiased diffusion regime at short times before the formation of stripes sets in .
for large values of @xmath32 , a crossover to a logarithmic time dependence takes place once the stripes are formed .
note that this transition is very sharp for @xmath44 .
for the smaller values of @xmath32 , see for example the data for @xmath45 , a more gradual transition is observed , in agreement with the more complicated behavior of @xmath24 for that case .
this gradual transition is related to the elimination of defects in the horizontal direction and the subsequent straightening of the stripes . during this
process particles can move more freely in vertical direction than in horizontal direction , as due to the defect structures not every particle in a stripe is surrounded in vertical direction by particles of the same type .
consequently , it happens that a particle at the borderline region of an imperfect stripe lands after moving only a few steps in vertical direction in another layer where it finds itself the midst of a domain of another species , thus allowing it to then move in the horizontal direction . as shown by our data , this process yields a displacement in horizontal direction that is slower than diffusive and faster than logarithmic .
[ h ] ( color online ) mean displacement of particles on the one - dimensional ring with 9000 sites in ( a , c ) a log - log plot and ( b , d ) a log - linear plot .
at very large times , the values of @xmath32 decrease from top to bottom . in ( a , b ) , following the rules of our model , a particle is not exchanged with a neighboring particle of the same type , yielding for unbiased exchanges with @xmath10 an effective sub - diffusive behavior @xmath63 .
when also allowing for exchanges between particles of the same type , see ( c , d ) , the square - root increase of a random walk prevails .
irrespective whether or not exchanges between like particles are allowed , a crossover to a logarithmic increase is observed for @xmath13 .
the curves result from an average over 200 independent runs .
, title="fig : " ] the data shown in fig .
[ fig7]a and [ fig7]b for the ring reveal a behavior that at first look seems odd . indeed for @xmath10 one does not observe the free diffusion of particle , but instead the particles move sub - diffusively , with a mean displacement @xmath63 .
in order to understand this , let us recall that in the definition of the abc model , see section ii , exchanges between particles of the same type are not assumed .
in fact , for all aspects discussed in previous papers as well as for all the time - dependent quantities studied in our work , with the exception of the mean displacement , it is completely irrelevant whether we allow exchanges of same particles or not . however , for individual trajectories this does matter , as without these exchanges particles get stuck behind others of the same type and can only move if a particle of a different type comes by and replaces the particle on the neighboring site . indeed , when allowing for these additional exchanges , we do recover free diffusion for @xmath10 also in one dimension , and for @xmath13 a crossover between that early time behavior and a logarithmic regime is observed , see fig .
[ fig7]c and [ fig7]d . in the two - dimensional system shown in fig .
[ fig6 ] particles get rarely stuck as they can diffuse both in @xmath6- and @xmath7-direction .
consequently , we have for @xmath10 the expected square - root behavior of a random walk even in absence of exchanges between particles of the same species .
[ h ] ( color online ) scaling of the equal - time correlation function for @xmath44 and different system sizes .
data ranging from @xmath64 to @xmath65 are included .
( a ) rescaled data for a system with @xmath47 sites .
the inset shows the unscaled data .
( b ) comparison of the scaling functions for three different system sizes .
the inset shows the scaling of the correlator for a ring with 600 sites and three different values of the swapping rate @xmath32 .
the values of the scaling exponent @xmath66 that achieve the best data collapse are gathered in table [ tab1 ] . for the larger systems these data
result from averaging over more than 1000 independent realizations , whereas for the one - dimensional system the ensemble average is performed over 16000 independent runs .
, title="fig : " ] + ( color online ) scaling of the equal - time correlation function for @xmath44 and different system sizes .
data ranging from @xmath64 to @xmath65 are included .
( a ) rescaled data for a system with @xmath47 sites .
the inset shows the unscaled data .
( b ) comparison of the scaling functions for three different system sizes .
the inset shows the scaling of the correlator for a ring with 600 sites and three different values of the swapping rate @xmath32 .
the values of the scaling exponent @xmath66 that achieve the best data collapse are gathered in table [ tab1 ] . for the larger systems
these data result from averaging over more than 1000 independent realizations , whereas for the one - dimensional system the ensemble average is performed over 16000 independent runs . , title="fig : " ] for systems governed by a time - dependent length scale @xmath24 one expects from general considerations @xcite that the single - time correlator @xmath67 exhibits a simple scaling behavior of the form @xcite @xmath68 for sufficiently large times @xmath69 , where @xmath70 is a microscopic reference time such that @xmath71 is larger than any microscopic length scale . in the present case
@xmath70 has to be larger than the typical time needed for the formation of ordered stripes . as shown in fig .
[ fig8]a , once the logarithmic growth regime has been fully accessed , dynamical scaling prevails and a data collapse is achieved when assuming the scaling ( [ eq : c_t_scal ] ) .
the values of the scaling exponent @xmath66 are gathered in table [ tab1 ] for three different vertical extents and three different values of the swapping rate @xmath32 .
inspection of the table reveals a weak dependence on the number of lines in the system .
in addition , the value of @xmath66 changes with @xmath32 . a similar dependence is also seen for
the scaling function itself , see the inset in fig .
it follows that for the @xmath0 model scaling functions and scaling exponents depend on system parameters .
this is similar to what is seen in other systems with logarithmic growth @xcite .
finally , we note from fig .
[ fig8]b that the scaling functions for different extents in vertical direction are not identical for the finite times accessed in our simulations .
this can be viewed as a finite - size effect , and the dependence of the scaling function ( and , of course , of @xmath24 ) on the vertical direction will vanish in the large volume limit and long time limit .
.values of the scaling exponent @xmath66 for three different systems and three different values of the swapping rate @xmath32 .
[ tab1 ] [ cols="^,^,^,^",options="header " , ]
two - times quantities are used in many studies in order to elucidate relaxation processes and aging phenomena far from stationarity @xcite . as the discussion of the equal - time correlation function has established the presence of dynamical scaling in our system , we expect the following scaling @xcite for the two - times autocorrelation function @xmath72 , see eq .
( [ eq : autocorr ] ) : @xmath73 where the exponent @xmath66 should be the same as for the equal - time correlator ( [ eq : c_t_scal ] ) .
this scaling form assumes that @xmath24 , @xmath74 as well as @xmath75 .
this last condition can be difficult to fulfill for a system with logarithmic growth , and sizeable finite - time corrections could spoil the expected scaling .
[ h ] ( color online ) scaling of the two - times autocorrelation function @xmath72 for a system composed of @xmath54 spins , with ( a ) @xmath45 , ( b ) @xmath76 , and ( c ) @xmath44 .
the insets show the autocorrelation as function of the time difference @xmath77 and reveal the aging properties of that quantity .
the waiting times decrease from top to bottom . in all cases a good data collapse
is achieved for the larger waiting times @xmath30 when assuming the scaling form ( [ eq : c_t_s_scal ] ) , with @xmath66 given in table [ tab1 ] .
these data result from averaging over 8000 runs .
, title="fig : " ] fig .
[ fig9 ] explores the scaling ( [ eq : c_t_s_scal ] ) for a system with @xmath54 sites and three different swapping rates @xmath32 .
these three panels are also representatives for other system sizes and other values of @xmath32 . besides large waiting times @xmath30 for which we clearly are in the logarithmic growth regime , we also use rather short ones where in some cases the intermediate regime has not yet been left
. we first remark that in all studied cases dynamical scaling ( [ eq : c_t_s_scal ] ) indeed prevails for the larger waiting times and @xmath78 not too close to 1 .
this is indeed achieved with the exponents listed in table [ tab1 ] . on the other hand , finite - time corrections , which become less important for increasing @xmath30 ,
are readily identified in the vicinity of @xmath79 . for the case
@xmath44 shown in panel ( c ) the shortest waiting time @xmath80 is obviously not yet in the scaling regime . doing the same analysis for larger systems , say ones composed of @xmath47 sites , we have a data collapse only for waiting times for which the system left the crossover regime and entered the asymptotic logarithmic regime with ordered stripes .
[ h ] ( color online ) time - dependent length when changing at the waiting time @xmath30 the swapping rate from @xmath81 to @xmath82 ( dashed colored lines ) and from @xmath83 to @xmath82 ( full colored lines ) .
the waiting times increase from left to right .
the system sizes are ( a ) @xmath57 , ( b ) @xmath54 and ( c ) @xmath47 . the full black lines indicate the growing length for a system where @xmath32 is fixed at 0.3 , 0.2 , or 0.1 ( from top to bottom ) .
these data result from averaging over at least 1600 independent runs .
, title="fig : " ] whereas the two - times correlation function contains information on the time evolution of a system by comparing configurations at different times since its initial preparation , a two - times response function provides insights in how a system that still evolves reacts to a perturbation
. a variety of protocols can be designed in order to probe the response of a system to a change of system parameters . as already discussed in section
ii , we consider in the following sudden changes in the biased exchange rate @xmath32 and monitor how the perturbed system relaxes to the same state as that reached by a control system that evolves all the time at that final value of @xmath32 . fig .
[ fig10 ] compares the length after the perturbation , @xmath35 , with that of the unperturbed system for two cases : ( 1 ) @xmath84 and @xmath85 ( dashed lines ) and ( 2 ) @xmath83 and @xmath86 ( full lines ) . for systems with @xmath57 and @xmath54 sites
the perturbations happen after the formation of the stripes .
on the other hand , for the system with @xmath47 sites shown in panel ( c ) the perturbations take place close to the end of the intermediate regime but before the logarithmic growth is fully established .
qualitatively the responses for all three system sizes are very similar , though . in all cases the perturbed length tends to the corresponding length of the control system .
this approach is rather fast for an increase of @xmath32 ( in the present case from 0.1 to 0.2 ) , but takes much longer for a decrease of @xmath32 .
when increasing @xmath32 the domains just after the change are too compact when compared to the domains that grow at the fixed value @xmath45 . on the other hand , when considering the case where @xmath32 is decreased , the domains after the change are less dense than what should be the case for a @xmath45 system . although @xmath87 is larger for @xmath88 than that for @xmath45 , @xmath24 does not decrease after the quench , see the dashed lines in fig .
[ fig10 ] that remain almost flat .
once the domains are formed , their size can not be reduced in a sizeable manner ; the domain growth will be slowed down until the time @xmath89 when @xmath90 for the system evolving at the fixed rate @xmath45 is comparable to @xmath87 for the @xmath88 system .
[ h ] ( color online ) response function @xmath91 for a system composed of @xmath54 particles after the diffusion bias has been changed from @xmath84 to @xmath86 at time @xmath30 .
the waiting times @xmath30 decrease from top to bottom .
the inset shows a reasonable scaling @xmath92 for the three largest waiting times with a scaling exponent @xmath93 .
, title="fig : " ] in principle , the two - times response of a system to an infinitesimal instantaneous perturbation should display a scaling behavior similar to that of the two - times correlation function , see eq .
( [ eq : c_t_s_scal ] ) . in our case
, however , the perturbation , i.e. the change in @xmath32 , is a large one , with @xmath94 .
one should also note that a scaling behavior can not be expected when the perturbation takes place in a regime where @xmath24 displays a crossover between different types of behaviors .
inspection of fig .
[ fig10]c , for example , reveals for the change from @xmath83 to @xmath86 a rather different behavior of @xmath35 for @xmath80 than for @xmath95 ( see full lines ) , and dynamical scaling can not be realized . for an increase of @xmath32
we were unable to achieve a data collapse using as scaling variable @xmath78 .
this is different from when @xmath32 is decreased , however , as dynamical scaling can be identified in this case . fig .
[ fig11 ] discusses as an example the change of @xmath32 from 0.3 to 0.2 for a system composed of @xmath54 sites .
the difference @xmath91 between the perturbed length @xmath35 and the length @xmath24 of the control system evolving with @xmath45 , see eq .
( [ eq : rep ] ) , provides a quantity that displays aging as it does not merely depend on the time difference @xmath77 .
@xmath91 goes to zero for long times when the ordered stripes of the perturbed system get similar to those of the unperturbed system .
inspection of fig .
[ fig11 ] reveals the emergence of two different regimes : a regime where @xmath91 is small ( i.e. the perturbed length is close to the unperturbed one ) and only depends on @xmath78 , and a regime close to @xmath79 where @xmath91 is large and shifts to higher values for increasing waiting times @xmath30 . in that regime a good scaling can be achieved when assuming that @xmath96 see the inset in fig .
[ fig11 ] .
we remark that immediately after the change of the value of @xmath32 the differences between the domains in the perturbed and in the control systems strongly depend on when the change took place , which gives raise to an increase of @xmath91 with the waiting time @xmath30 . however , once the ordered domains in the perturbed system
have undergone major changes , they become similar to those that are formed in the control system , which results in a weaker dependence on @xmath97 .
phase - ordering kinetics in systems with algebraic domain growth , especially in cases where the domain coarsening is curvature driven , is rather well understood .
still open problems are encountered in disordered systems where after a transient algebraic - like growth regime a non - algebraic , logarithmic domain growth prevails for long times . in this paper
we discussed some aspects of domain coarsening in systems of a different type that are also characterized by logarithmic growth , namely in systems dominated by dynamical constraints .
the best known representative of this class is the @xmath0 model @xcite where in a particular direction , say the horizontal direction , particles of three different types are exchanged asymmetrically .
this leads to the formation of ordered domains which in a two - dimensional system take the form of stripes oriented perpendicularly to the direction of biased exchanges @xcite . in the asymptotic long - time regime , these domains coarsen and their size increases logarithmically with time .
our emphasis was on the ordering process and the related aging phenomena . studying a time - dependent length derived from the single - time spatial correlator
, we identified different regimes that we discussed as a function of the vertical extent of the system and of the bias in the particle exchanges . whereas in the logarithmic growth regime dynamical scaling prevails , as demonstrated by the data collapse of the single - time correlator as well as of the two - times autocorrelation ( and also , to some extent , of the two - times response function ) ,
complicated crossover phenomena are observed during the formation of the well ordered stripes , especially in vertically extended systems where an intermediate phase is observed that is dominated by the vertical growth of the domains .
the attentive reader will have noticed that in the discussion of our results we never brought up the fact that for the one - dimensional system with even coverage for all three species the steady state is an equilibrium steady state , whereas in the two - dimensional version of the model detailed balance is always violated and the steady state is a non - equilibrium steady state . in fact , the equilibrium or non - equilibrium nature of the steady state does not seem to be relevant for the dynamical , far - from - stationarity regime probed in our study . while it seems reasonable that a system in the stages of domain formation and subsequent domain coarsening does not know anything about the character of the steady state emerging in the long time limit ( see @xcite for another example where the dynamic properties during relaxation do not depend on the equilibrium or non - equilibrium nature of the stationary state ) , there do exist well known examples ( many - particle systems without detailed balance @xcite , driven ising systems @xcite , driven striped structures @xcite ) where for cases with a non - equilibrium steady state dynamic properties during the relaxation process are different from those encountered in the corresponding systems with an equilibrium steady state .
additional studies will be needed in the future in order to fully comprehend these differences . | we discuss relaxation and aging processes in the one- and two - dimensional @xmath0 models . in
these driven diffusive systems of three particle types , biased exchanges in one direction yield a coarsening process characterized in the long time limit by a logarithmic growth of ordered domains that take the form of stripes . from the time - dependent length , derived from the equal - time spatial correlator , and from the mean displacement of individual particles different regimes in the formation and growth of these domains can be identified .
analysis of two - times correlation and response functions reveals dynamical scaling in the asymptotic logarithmic growth regime as well as complicated finite - time and finite - size effects in the early and intermediate time regimes .
# 1#1 # 1f |
intervening metal absorption lines have routinely been observed along the line of sight to quasars ( qsos ) . among these features ,
strong mgii @xmath82796 , @xmath82803 absorption doublets ( with equivalent width @xmath11 ) represent good indicators of metal - enriched gas associated with foreground galaxies .
this is because magnesium is an @xmath12 element produced by red giants and dispersed in the interstellar medium ( ism ) by supernovae and stellar winds ( bergeron & boisse 1991 , steidel et al .
in particular , such a strong doublet on a qso sightline witnesses the presence of @xmath13 galaxies within 60 kpc projected distance and with different morphological types ( churchill et al .
2005 , kacprzak et al .
in addition , the @xmath14 @xmath15 mgii intervening absorbers are hosted within dark matter halos with characteristic masses of @xmath16 ( bouche et al . 2006 ) and @xmath17 are associated with damped lyman - alpha systems ( dlas , rao et al .
2006 ) . for a few hours after their onset , gamma - ray burst ( grb ) afterglows are the brightest beacons in the far universe
: therefore they provide an alternative and complementary way to qsos to fully explore the properties of high - z galaxies in which they are hosted ( see savaglio 2006 ; prochaska et al .
2007 ) and of those along their sightlines . in principle , the sightlines of qsos and grbs are expected to be equivalent .
prochter et al .
( 2006 ) found @xmath18 strong ( @xmath19 ) mgii intervening absorbers in @xmath20 qso spectra from the sdss dr4 , which corresponds to a redshift number density of @xmath21 .
surprisingly , the same authors report the identification of 14 mgii systems along 14 grb sightlines , which translates into a redshift number density of @xmath22 ( 99% confidence interval ) , almost 4 times higher than the qso one .
on the other hand , the intervening civ absorbers for the two classes of sources do not show any statistical difference ( sudilovsky et al .
2007 , tejos et al . 2007 ) .
the reason for this discrepancy is still uncertain , and several scenarios have been widely discussed ( see e.g. porciani et al .
2007 and cucchiara et al . 2009 ) .
one of these interpretations requires a different size of the source , namely , that qso emitting regions are larger than grbs which in turn are comparable to the typical size of the mgii clouds ( frank et al .
2007 ) . in this scenario variability in the column densities of the mgii absorber
is expected , since the lorentz factor decreases because the fireball decelerates on the interstellar medium , and thus the grb emission regions increase .
a mgii variability in multi - epoch spectroscopy data on grb060206 was first claimed by the analysis by hao et al .
( 2007 ) , but then disproved by aoki et al .
( 2008 ) and thne et al .
( 2008 ) .
occasionally , extremely bright optical transient emission is associated with the grb event , offering the superb opportunity to take spectra of the afterglows with high resolution instruments . in a fraction of these cases , multi - epoch , high resolution spectroscopy with an adequate signal to noise ratio ( s / n ) can be obtained , allowing to make detailed studies of the host galaxy ism and to put strong constraints on its physical parameters ( see vreeswijk et al .
2007 for grb060418 , and delia et al .
2009a for grb080319b ) . here
we take advantage of the high quality data collected for grb083019b to make a systematic study of this grb sightline and to search for variability of the mgii intervening absorbers .
the paper is organized as follows .
section @xmath23 presents a short summary of the grb080319b detection and observations ; section @xmath24 describes the datasets and data reduction ; section @xmath25 presents the full analysis of the intervening systems identified in the spectra ; finally in section @xmath26 the results are discussed and conclusions are drawn .
grb080319b was discovered by the burst alert telescope ( bat ) instrument on board swift on 2008 , march 19 , at 06:12:49 ut .
swift slewed to the target in less than 1 minute and a bright afterglow was found by both the x - ray telescope ( xrt ) and uv - optical telescope ( uvot ) at ra = @xmath27h @xmath28 m @xmath29s , dec = + 36@xmath30 @xmath31 @xmath32 ( racusin et al .
2008a ) with observations starting 60.5 and 175 s after the trigger , respectively .
the field of grb080319b was imaged by the `` pi of the sky '' apparatus located at las campanas observatory before , during and after the grb event ( cwiok et al .
the field was also targetted by the robotic telescope rem just 43 s after the bat trigger ( covino et al .
2008a , b ) . the tortora wide - field optical camera ( 12 cm diameter , 20@xmath3325 deg fov , tv - ccd , unfiltered ) mounted on rem also imaged the field before , during and after the grb event with good temporal resolution ( karpov et al .
these observations show that the grb reached the magnitudes @xmath34 about @xmath35 s and @xmath36 about @xmath37 s after the trigger .
this makes grb080319b the brightest grb ever recorded at optical wavelengths ( racusin et al .
2008b , bloom et al . 2009 ) . [ obs_log ]
[ cols="<,^,^,^,^,^,^,^,^,^",options="header " , ] all values are logarithmic @xmath38 cc cc cc
in this paper we present high resolution ( r=40000 , corresponding to 7.5 km / s ) spectroscopy of the optical afterglow of the `` naked - eye '' gamma ray burst grb080319b ( @xmath2 ) , observed by uves at the vlt in three different epochs , starting @xmath39 minutes , @xmath40 and @xmath41 hours after the trigger , respectively .
we concentrate here on the intervening absorbers along the grb sightline , since the absorption features in the vicinity of the afterglow have already been studied and presented in delia et al .
( 2009a ) .
our spectral coverage allows to analyze a redshift path for the mgii @xmath82796 in the range @xmath42 .
we are sensitive to lines with ew down to @xmath43 ( at the @xmath44 confidence level ) , thanks to the extremely high s / n of the first rrm observation .
at least four intervening systems between z = 0.8 and z = 0.5 have been identified .
most of them show a complex structure constituted by several components , similar to that of the intervening absorbers along the qso sightlines ( churchill & vogt 2001 ) .
all systems feature the mgii @xmath82796 , @xmath82803 doublet .
feii , mgi and mnii lines are detected in four , two and one systems , respectively .
prochter et al .
( 2006 ) claimed that the incidence of strong mgii absorbers along grb sight lines is nearly four times higher than that along the line of sight to qsos .
they analyzed the spectra of 14 grb optical afterglows , finding on average one intervening system per afterglow with equivalent width @xmath45 .
the grb080319b sightline confirms this trend , since the rest frame ew of the mgii @xmath82796 absorption line is @xmath45@xmath15 in one system .
structured intervening systems are expected if the discrepancy between qsos and grbs intervening absorbers is due to a different size of the source , namely , if qso emitting regions are larger than grbs which in turn are comparable to the typical size of themgii clouds ( frank et al .
according to their estimation , the qso beam size must be larger than 3 @xmath33 10@xmath46 cm while for the grb beam they suggest a size around a few @xmath33 10@xmath47 cm . to observe this effect ,
the mgii absorber should then be patchy with a clump size around 10@xmath46 cm .
nevertheless , in this scenario , variability in the column densities of the mgii absorbers along the grb sightlines is expected .
mgii variability in multi - epoch spectroscopy data on grb060206 was first claimed by the analysis by hao et al .
( 2007 ) , but then disproved by aoki et al .
( 2008 ) and thne et al .
this work confirms the lack of variability in the column densities of the grb intervening absorbers , in the specific case of grb080319b . in particular
, the ew of the mgii @xmath82796 absorption line is consistent within @xmath44 in the three uves observations .
given the s / n ratio of our data , we can conclude that a variability of this feature , if present , is less than 10% at the @xmath48 confidence level .
this upper limit is three times smaller than the ones that can be estimated for grb060206 using the data by aoki et al .
racusin et al .
( 2008b ) modeled the radiation of grb080319b as coming from a structured jet constituted by an inner , narrow jet ( @xmath49 ) and an outer , wider jet ( @xmath50 ) .
they interpreted the optical emission at @xmath51s ( @xmath52 being the burst detection time ) as produced by the reverse shock of the inner jet , while that at @xmath53 as the signature of the forward shock of the outer jet .
since the first uves observation starts at @xmath54 while the second and the third ones start more than @xmath55hr later , a strong increase in the size of the emitting region is expected from this scenario . in more detail ,
the dimension of the emitting region is @xmath56 cm at the beginning of the grb event and @xmath57 cm a few hours later ( pandey et al .
2009 , racusin et al . 2008b and kumar & panaitescu 2008 ) .
although we can not be more quantitative , a variation of a factor of ten in the dimension of the emitting region would imply a considerable reduction of the intervening column densities , but it is not detected .
indeed , time resolved , high resolution spectroscopy allows us for the first time to exclude a significant ( @xmath58 ) variability even by considering the single components of each intervening system alone .
porciani et al .
( 2007 ) investigated several possible explanations for the strong mgii excess , namely , dust obscuration bias , clustering of the absorbers , different beam sizes of the sources , multiband magnification bias of grbs , association of the absorbers with the grb event or with the circumburst environment .
they concluded that none of these effects alone can explain the observed difference , but maybe the combination of two or more of them can reduce the significance of the discrepancy .
we can take advantage of the dimensions of the emitting source in grb080319b and the lack of variability in the intervening absorbers to characterize the absorbers along this grb sightline .
since these dimensions are in the range @xmath59 cm , the mgii ( and other species ) clouds can not have clumps or structures smaller than @xmath60 pc , otherwise we should observe variability in their absorption .
this is not obvious , since ding et al .
( 2003 ) show that smaller intervening absorbers are present along the line of sight to qsos .
cucchiara et al .
( 2009 ) compared 81 qsos and 6 grb lines of sight obtained with uves .
they found no significant evidence to support a difference between the two absorber populations , concluding that a possible explanation for the mgii excess could be intrinsic to the grb environment . a similarity between the narrow lines of the grb intervening absorbers and that produced by the material ejected in the accretion disk winds of qsos has been pointed out
nevertheless , the high velocities required by the intervening redshifts ( @xmath61 ) and the lack of fine structure features in these systems represent a strong weakness of this picture .
in addition , tejos et al .
( 2009 ) recently found no evidence for a similar excess considering the weak ( @xmath62 ) mgii systems along the line of sight of 8 grb observed with echelle spectrographs . these authors tend to exclude an intrinsic nature of the discrepancy , since it would result in an excess in the weak systems too
they suggest that the best explanation available at present could be gravitational lensing bias due to lensing by the host galaxy of the absorber .
indeed the strong mgii system along the grb080319b sightline has a velocity of @xmath63 km / s with respect to the grb .
the data do not show evidence either of broad profiles and/or partial coverage ( see e.g. dodorico et al .
2004 ) , so this absorber is likely an intervening system . combining the data by prochter et al .
( 2006 ) with the results from the analysis in this paper and that in delia et al .
( 2009b ) regarding the line of sight to grb080330 , we obtain 16 strong ( ew@xmath45 ) mgii intervening absorbers along 16 grb sightlines .
the total redshift path becomes @xmath64 , and this results in a @xmath65 .
this surprising excess of strong mgii absorbers with respect to qso sightlines remains a matter of debate and a satisfactory explanation is still missing . clearly more observations and
analysis are needed in order to solve this issue .
churchill , c.w . ,
kacprzak , g.g .
& steidel , c.c . , 2005 , in iau colloq .
199 , probing galaxies through quasar absorption lines , ed .
williams , c. shuu & b. menard ( cambridge : cambridge univ . press ) , 24 | the aim of this paper is to investigate the properties of the intervening absorbers lying along the line of sight of gamma - ray burst ( grb ) 080319b through the analysis of its optical absorption features . to this purpose
, we analyze a multi - epoch , high resolution spectroscopic observations ( r=40000 , corresponding to @xmath0 kms@xmath1 ) of the optical afterglow of grb080319b ( @xmath2 ) , taken with uves at the vlt .
thanks to the rapid response mode ( rrm ) , we observed the afterglow just 8m:30s after the grb onset when the magnitude was r @xmath3 .
this allowed us to obtain the best signal - to - noise , high resolution spectrum of a grb afterglow ever ( s / n per resolution element @xmath4 ) .
two further rrm and target of opportunity observations were obtained starting @xmath5 and @xmath6 hours after the event , respectively .
four mgii absorption systems lying along the line of sight to the afterglow have been detected in the redshift range @xmath7 , most of them showing a complex structure featuring several components .
absorptions due to feii , mgi and mnii are also present ; they appear in four , two and one intervening absorbers , respectively .
one out of four systems show a mgii @xmath82796 rest frame equivalent width larger than 1 .
this confirms the excess of strong mgii absorbers compared to quasars , with @xmath9 , @xmath10 times larger than the one observed along quasar lines of sight .
in addition , the analysis of multi - epoch , high - resolution spectra allowed us to exclude a significant variability in the column density of the single components of each absorber . combining this result with estimates of the size of the emitting region , we can reject the hypothesis that the difference between grb and qso mgii absorbers is due to a different size of the emitting regions .
[ firstpage ] gamma - rays : bursts ism : abundances line : profiles atomic data . |
in the last few years there has been an increasing interest for single - molecule and , more in general , single - system spectroscopy @xcite .
usually these systems are characterized by intermittent fluorescence , namely , by sudden jumps between two states , a _
light on _ and _ light off _ state .
the probability density function ( pdf ) of the times of sojourn or waiting times ( wts ) @xmath0 in these two states are denoted by the symbols @xmath1 and @xmath2 . in the case of blinking quantum dots ( bqds ) @xcite there is a general agreement that these are renewal processes , namely , that the jumps between the two states have the effect of resetting to zero the system s memory .
the pdfs @xmath1 and @xmath2 are found to be inverse power laws , whose indexes are , according to ref .
@xcite , @xmath3 and @xmath4 .
the authors of ref .
@xcite afford directions on how to distinguish a renewal non poisson process ( rnpp ) , which is a homogeneous in time process from the apparently equivalent condition generated by a non - homogeneous poisson process ( nhpp ) , _ i.e. _ , with a time dependent rate of event productions @xmath5 .
let us recall that a ( homogeneous ) poisson process is a particular renewal process where the wt - pdf is an exponential function whose decay rate coincides with the ( constant ) rate of event productions .
the physical origin of rnpps is still the object of research and debate , whereas that of nhpps requires however the identification of dynamics responsible for the rate time dependence . in this letter
we consider the recent work of ref .
@xcite , which establishes the influence of the catalytic activity of a fluctuating enzyme on the intermittent fluorescence of the fluorogenic substrate . in this case
the function @xmath1 is exponential , and the complexity of the system is signaled by the function @xmath2 , which turns out to be a stretched exponential @xmath6 with @xmath7 .
we argue that this is a nhpp , namely , an example of the second class of processes .
according to the earlier work @xcite a non - exponential function can be expressed as the sum of infinitely many exponential functions @xcite . in refs .
@xcite this condition was realized by means of the stochastic modulation of the event production rate @xmath5 . in the present case , we adopt a different kind of modulation , which is quasi - periodic , and this allows us to establish an evident connection with the popular stochastic resonance ( sr ) effect @xcite .
sr is a phenomenon produced by making the event production rate depend harmonically on time , a condition generating a periodic wt reordering : the length of the wts as a function of the time order reproduces the rate time dependence @xcite . in this letter
the quasi - periodic modulation has an internal origin and is based on the enzyme s conformation changes .
the ultimate theoretical justification for the adoption of the periodic assumption is given by the work of winfree @xcite , who depicts the fundamental biological processes as operating through cyclical processes .
thus , we assume that the enzyme s conformation changes are periodic , although non - harmonic , functions of time .
we assume that the time duration of the _ on _ state is negligible , so that the system , after jumping from the _ off _ to the _ on _ state , jumps back to the _ off _ state almost immediately .
thus , the _ on _ state is a kind of flashing snapshot , and the distance between two consecutive snapshots is described by @xmath2 .
let us omit for simplicity s sake the subscript _ off _ and let us assign to @xmath8 the following form : @xmath9 being @xmath10 the exponential wt - pdf related to the poisson process with event production rate @xmath11 and @xmath12 the normalization constant .
the quantity @xmath13 represents the statistical weight of each @xmath14 .
we imagine @xmath5 to be a slow function of time , so as to produce many poissonian events with the same rate @xmath11 .
this is called slow modulation condition , and it makes it possible to define @xmath13 on the basis of the enzyme s fluctuations .
each enzyme conformation yields a given @xmath11 and the slower the enzyme s conformation change , the larger the statistical weight of this rate .
it is thus obvious that in the slow modulation limit the statistical weight @xmath13 is inversely proportional to the modulus of @xmath15 : @xmath16 the parameter @xmath17 defines a time scale , driving the slope of q(t ) , which can be properly adjusted to fit the experimental results . in order to derive a periodic behavior for @xmath5
, we assume the rate @xmath5 to increase in time from a minimum value @xmath18 to a maximum and very large value @xmath19 , from which it decays slowly to the minimum value again , and so on , thereby confining the stretched exponential distribution to a limited time range , which can , however , include several decades , depending on the choice of the parameters @xmath18 , @xmath19 and @xmath17 .
the switches between increasing and decreasing patterns of @xmath5 correspond to abrupt slope changes at times @xmath20 , which make @xmath5 piecewise smooth , and exactly periodic as well .
however , once a wt is selected we leave it to stretch beyond the exact time period @xmath17 , thereby making the exactly periodic slope change produce random effects and quasi - periodicity . according to eq .
( [ inverse ] ) , @xmath8 is the laplace transform of the function @xmath21 .
thus , from the short - time limit , where the stretched exponential reads @xmath22 using the tauberian theorem , we find for @xmath13 , at @xmath23 , the following asymptotic expression @xmath24 by plugging eq .
( [ useit ] ) within eq.([slowrate ] ) , we obtain the following equations : @xmath25 the integration of eq .
( [ slowrate2 ] ) from @xmath26 to @xmath27 yields two different results depending on the sign : @xmath28 referring to the cases of increasing ( @xmath29 sign ) and decreasing ( @xmath30 sign ) rate , respectively . the times @xmath31 and @xmath32 are defined by @xmath33 and the parameter @xmath34 , determining the coefficient @xmath35 of the stretched exponential is given by @xmath36 these analytical results correspond to the correct derivation of the stretched exponential under the limiting condition of very slow modulation , namely , under the condition that @xmath17 is large enough .
a periodic behavior of @xmath5 is defined by applying suitable time - shifts of the kind @xmath37 to the rates given in eqs .
( [ solution_rate2 ] ) and ( [ solution_rate ] ) .
the purpose of this section is to assess by means of artificial sequences how large @xmath17 must be for the wt - pdf @xmath8 to fit the stretched exponential form of eq .
( [ stretched ] ) . on the other hand
, we note that the experimental results of ref .
@xcite do not yield an exactly periodic time ordering .
thus we plan to establish if it is possible to find values of @xmath17 that make @xmath8 fit the stretched exponential condition without producing a too markedly periodic time ordering , which would depart qualitatively from the experimental results of ref .
@xcite .
the numerical algorithm is based on the jump probability in the time interval @xmath38 $ ] : @xmath39 , @xmath40 being small enough to satisfy the conditions @xmath41 and @xmath5 nearly constant on the interval @xmath42 .
we used the following values of the parameters : @xmath43 , @xmath44 and @xmath45 . thanks to eq .
( [ inverse ] ) the only fitting parameter remains @xmath46 . as a function of @xmath0 .
( a ) @xmath47 , ( b ) @xmath48 , and ( c ) @xmath49 .
the full line is the stretched exponential of eq .
( [ stretched ] ) , with @xmath43 and @xmath50 determined by a fitting procedure yielding : ( a ) @xmath51 , ( b ) @xmath52 , and ( c ) @xmath53 . ]
figures [ fig1](a ) to [ fig1](c ) show the wt - pdf @xmath8 for different values of the parameter @xmath17 . fig .
[ fig1](a ) refers to @xmath47 . in this case
the slow modulation condition is violated .
this is the case where the process can be classified as renewal .
we see that for both fig .
[ fig1](b ) and fig .
[ fig1](c ) , referring to @xmath48 and @xmath49 , respectively , the agreement between the numerical result and the stretched exponential prediction of eq .
( [ stretched ] ) is accurate throughout the whole time regime
. both conditions , therefore , are in principle compatible with the experiment of ref .
@xcite . to select which of the two conditions is closer to the experiment of ref .
@xcite , let us plot the wts @xmath54 according to the chronological order , i.e. , with respect to their order of appearance @xmath55 .
the results are illustrated in fig .
[ chr_ord ] , for different values of the parameter @xmath17 .
we make the comparison among different values of @xmath17 with two distinct criteria : in the left column we adopt different numbers of wts , namely we refer to different chronological times , so as to produce the same number of abrupt slope changes of @xmath5 ; in the right column , on the contrary , the different plots contain the same number of wts , namely range from @xmath56 to the same maximum chronological time .
the results of the right column allow us to establish a qualitative comparison with the experimental results of ref .
furthermore , we see that in accordance with the stochastic resonance phenomenon @xcite , increasing @xmath17 has the effect of making more evident the resonant character of this effect .
we see that the condition corresponding to @xmath57 , where the resonant effect is much less marked than in the case @xmath58 , produces the best qualitative agreement with the experimental result of ref .
it is worth remarking that in this condition the wt - pdf @xmath8 still retains a markedly stretched exponential behavior so as to ensure a fully qualitative agreement with the experimental results of ref .
@xcite . .
on the left side we arranged the same number of peaks ( fixed number of enzyme fluctuations ) , while on the right side we compare the number of peaks per fixed number of wts ( or system s jumps ) .
@xmath59 from the top to the bottom .
, width=340,height=188 ]
to draw final conclusions we apply to the artificial sequences @xmath60 of section [ numerical ] the aging experiment ( ae ) described in the earlier publication of ref .
@xcite . here
we briefly recall that the method is based on the computation of the survival probabilities @xmath61 and @xmath62 .
both survival probabilities are aged and derived from time sequence by delaying the beginning of the waiting process for an event until a distance @xmath63 from each event of the sequence .
@xmath64 is derived by applying this procedure to the time sequence , and @xmath65 is derived from the experimental histogram by assuming that the renewal condition holds true . upon increase of @xmath66 , both @xmath67 and @xmath68 can become much slower than @xmath69 .
if @xmath70 , and both are slower than @xmath71 , this is the sign that the process is renewal and non - poissonian , insofar as in the poisson case , a renewal process does not yield aging . in fig .
[ aging1 ] and fig .
[ aging100 ] , we show a pair of dotted curves , illustrating @xmath67 at two different ages , and a pair of dashed curves , illustrating @xmath72 for the corresponding ages . in both cases
the two curves of the latter pair generate a decay significantly slower than @xmath71 .
this is only a consequence of the fact that @xmath71 is a stretched exponential .
we see , however , that in fig .
[ aging1 ] both dotted curves are slower than @xmath71 , although faster than the corresponding dashed curves , thereby indicating that a form of aging may exist , albeit significantly reduced .
this is an indication that many renewal events are present @xcite . in the case of fig .
[ aging100 ] , on the contrary , the two dotted curves virtually coincide with @xmath71 , if we neglect the short - time region , thereby implying that the deviation from exponential relaxation is caused by a so slow modulation as to cancel all the renewal events that may be non - poissonian .
we note that the condition of reduced aging of fig .
[ aging1 ] corresponds to @xmath48 , which , as shown by fig .
[ chr_ord ] , generates for the chronological order a structure qualitatively similar to that of the experimental results of ref .
+ we thus conclude that the sequence of wts emerging from the experimental results of ref .
@xcite may be a nhpp , generated by a stochastic resonance produced by the enzyme s conformation changes .
the debate on whether non - exponential relaxation is due to inhomogeneities or cooperative molecular motion is of wide interest in biophysics ( see , for instance ref .
we note that the recent letter of ref .
@xcite confirms the stretched nature of the wt - pdf of ref .
@xcite while demanding further theoretical progress to understand the correlation among events
. then we notice that experimental evidence for a periodic behavior of intermittent fluorescence can also be found in @xcite .
we think that the theory developed in this letter , along the lines of the earlier work of @xcite , may turn out to be useful to relate the experimental results to the enzyme dynamics .
o. flomembon , k. velonia , d. loos , s. masuo , m. cotlet , y. engelborghs , j. hofkens , a. e. rowan , r. j. m. nolte , m. van der auweraer , f. c. de schyver , and j. klafter , proc . natl
usa * 102 * , 2368 ( 2005 ) .
c. beck , physica a , * 344 * , 393 - 402 ( 2006 ) ; arxiv : cond - mat/0510841 .
beck discussed energy rather than time distributions and , more important , he focused on the range of small rates , corresponding to the limit of large energies ( or times ) . here , on the contrary , we focus on the short - time behavior . | we study the intermittent fluorescence of a single molecule , jumping from the `` light on '' to the `` light off '' state , as a poisson process modulated by a fluctuating environment .
we show that the quasi - periodic and quasi - deterministic environmental fluctuations make the distribution of the times of sojourn in the `` light off '' state depart from the exponential form , and that their succession in time mirrors environmental dynamics . as an illustration , we discuss some recent experimental results , where the environmental fluctuations depend on enzymatic activity . .
] pacs : 05.40.fb , 05.45.tp .
78.20.bh , 78.67.hc |
in our previous article @xcite , we considered the construction of the form factors of the trace operator in the massless flows @xcite from the uv coset model @xcite @xmath9 , with central charge @xmath10 to the ir coset @xmath11 .
the latter model is the unitary minimal model @xmath1 with central charge @xmath12 the flow is defined in uv by the relevant operator of conformal dimension @xmath13 ; it arrives in the ir along the irrelevant operator @xmath14 .
+ these flows include in particular for @xmath15 the famous massless flow from the tricritical ising model to the ising model .
the latter flow was studied by delfino , mussardo and simonetti in @xcite using a massless version of the form factor approach , originally developed for the massive case in @xcite . for this purpose
, the authors of @xcite used the scattering data proposed by al.b .
zamolodchikov in @xcite .
let us mention that the notion of massless scattering was first introduced and discussed in this latter paper . in @xcite , beside the trace operator , some form factors for the magnetization operator and the energy operator were constructed .
+ in this article we will try to generalize this construction to the whole family of flows introduced above .
+ we do not intend to repeat here the construction of the form factors in the sine - gordon model , and would like to refer the reader to our previous paper @xcite for notations and formulae , and to @xcite for complementary information on the global formalism .
+ let us recall that in the massless case , the dispersion relations read @xmath16 for right movers and @xmath17 for left movers , where @xmath18 is some mass - scale in the theory , and @xmath19 the rapidity variables .
zero momentum corresponds to @xmath20 for right movers and @xmath21 for left movers .
+ the @xmath22-matrices for the three different scatterings were found in @xcite : the @xmath23 and @xmath24 @xmath22-matrices describe the ir cft @xmath1 , and are thus given by the rsos restriction of the sine - gordon @xmath22-matrix @xcite .
+ we introduce by anticipation the minimal form factor in the sg model : @xmath25 where the parameter @xmath26 is related to the parameter @xmath27 by @xmath28 .
+ as for the @xmath29 scattering , it is given by and @xmath30 , the scattering data were first proposed in @xcite and @xcite respectively.]@xcite : @xmath31 in the ir limit , the scattering becomes trivial @xmath32 , and the two chiralities decouple .
+ the minimal form - factor in the rl channel satisfies the relation : @xmath33 and its explicit expression is given by @xmath34 its asymptotic behaviour in the infrared is : @xmath35 , where @xmath36 0.4 cm the plan of the paper is the following : in section 1 , we generalize the construction of the form factors of the magnetization operator in the flow tim @xmath2 i m , _ i.e. _ we construct form factors of an operator that flows to @xmath3 in the ir .
then we make numerical checks involving the @xmath4-sum rule , in order to identify the conformal dimension of the operator in the uv .
section 2 is devoted to the generalization of the form factors of the energy operator in the flow tim @xmath2 i m : we construct form factors of an operator which flows in the ir to the operator @xmath6 in @xmath1 ; an analysis of the two point correlator for such an operator can also be found .
finally , we give some concluding remarks .
a few form factors of the magnetization operator were obtained in @xcite in terms of symmetric polynomials for the case @xmath15 ( @xmath39 ) corresponding to the massless flow from tim to i m . due to the invariance of the theory under spin reversal , the order operator has non vanishing form factors only on a odd number of particles , whereas the disorder operator only on an even one .
we shall consider the case of the disorder operator with even number of right and left particles .
+ we recall that the form factors of this operator satisfy the following residue equation at @xmath40 @xcite : @xmath41 and a similar equation in the @xmath24 channel . the first step of recursion is given by @xmath42 .
+ in @xcite it was observed that all form factors of this operator could be rewritten as : @xmath43 where @xmath44 and @xmath45 as well as : @xmath46 it was found in @xcite that for two right and two left particles , the expression of the form factor is given by the following expression : @xmath47 the following relation holds when @xmath48 which defines the ir region @xcite : @xmath49 where @xmath50 are the form factors of the disorder operator @xmath51 in the thermal ising model @xcite : @xmath52 by analogy with the previous section , we shall now look for a solution to the following problem at @xmath40 : @xmath53 where @xmath54 .
a similar equation holds in the @xmath24 channel .
+ let us note that in the ir limit , given that @xmath55 , the latter relation becomes @xmath56 which is the residue equation satisfied by ( amongst others ) the operator @xmath3 in @xmath38 .
+ we construct a solution to the residue equation ( [ res ] ) with the initial condition @xmath57 , and with the following condition in the ir limit : @xmath58 in other words , we want to construct form factors of an operator that renormalizes in the ir on the operator @xmath3 in @xmath38 .
+ we make the following ansatz for the solution , to be compared with the one obtained for the trace operator in @xcite ( once again , we refer the reader to @xcite for further explanations on the construction of form factors in the sg model and basic notations ) : @xmath59 we introduced @xmath60 , which is the @xmath26-function of the operator @xmath3 in the minimal model @xmath38 .
we use the identification @xmath61 , followed by a modification of the multiparticles state @xcite : the modified bethe ansatz state @xmath62 is related to the usual bethe ansatz state by the relation @xcite @xmath63 we will use the @xmath26-function of the exponential fields @xmath64 in the sg model for the particular value @xmath65 ( the form factors of @xmath64 were first constructed in @xcite ; we use here the conventions and notations of @xcite ) : @xmath66 we introduced the scalar function ( completely determined by the @xmath22-matrix ) @xmath67 with @xmath68 and @xmath69 are normalization constants . + finally , the integration contours @xmath70 consist of several pieces for all integration variables @xmath71 : a line from @xmath72 to @xmath73 avoiding all poles such that @xmath74 and clockwise oriented circles around the poles ( of the @xmath75 ) at @xmath76 , @xmath77 .
the integration contours @xmath78 are similarly defined .
the function @xmath79 remains to be determined .
+ let us motivate our ansatz ( [ fphi ] ) : it is a slight modification of the results of @xcite that the form factors of the operator @xmath3 in the minimal model @xmath38 are written : @xmath80 when @xmath81 , each of the @xmath82 integration contours gets pinched at @xmath83 , and we have to take the sum of these three contributions . due to symmetry
, it is enough to consider the contribution of one of them ( _ e.g. _ @xmath84 ) , and multiply the result by @xmath82 .
the following computation is detailed in @xcite : + at @xmath85 , we have : @xmath86 whereas at @xmath87 : @xmath88 and at @xmath89 : @xmath90 remembering the relation : @xmath91 we see that the ansatz ( [ fphi ] ) will satisfy the residue equation ( [ res ] ) at the condition that the function @xmath92 satisfies at @xmath81 : * @xmath85 : @xmath93 * @xmath94 : @xmath95 and similar relations in the @xmath24 channel .
we introduce the sets @xmath96 and @xmath97 , as well as @xmath98 the subsets of @xmath22 , such that @xmath99 with @xmath100 and @xmath101 .
these subsets have number of elements : @xmath102 , @xmath103 , @xmath104 .
@xmath105 the subsets of @xmath106 : @xmath107 and @xmath108 are defined similarly .
+ we conjecture the following expression : @xmath109 let us give a sketch of the proof that this function satisfies the relations ( [ rel1 ] ) and ( [ rel2 ] ) : * when @xmath40 and @xmath85 , we can not have @xmath110 , otherwise the term @xmath111 becomes equal to zero .
consequently we should have @xmath112 or @xmath113 .
it follows from a simple inspection of the other cosines in the numerator that the only possibility in order to have the function @xmath114 different from zero is to have @xmath113 and @xmath115 .
equation ( [ rel1 ] ) follows from the use of simple trigonometric identities . *
conversely , when @xmath40 and @xmath94 , we find that we should have @xmath110 and @xmath116 .
equation ( [ rel2 ] ) follows .
moreover , it was checked with mathematica ( for a small number of particles ) that the relation ( [ ir ] ) holds . by analogy with what was done in @xcite for the flow tim @xmath2 i m
, one can think of using the @xmath4-sum rule in order to compute the variation of the conformal dimension and hence identify the conformal dimension of the operator in the uv .
we recall the relation @xcite : @xmath117 where @xmath118 is the vacuum expectation value of the operator @xmath37 , and @xmath119 the trace operator . + for our purpose , we will need the expression of the 4-particles form factor of @xmath119 found in @xcite : @xmath120 as well as the 4-particles form factor of the operator @xmath37 .
the formula presented in the previous section considerably simplifies for 4 particles : @xmath121 and this leads to the following expression for the form factor : @xmath122 the normalization of the latter form factor was chosen in order to ensure the initial condition @xmath123 .
this amounts to setting the constant : @xmath124 in formula ( [ fphi ] ) , where @xmath125 is defined by the asymptotic behaviour when @xmath126 of the minimal form factor : @xmath127 explicitly : @xmath128 the four particle form factor ( [ fphi1 ] ) at @xmath39 should be compared with equation ( [ disorder ] ) ; in this case we have : @xmath129 and consequently the two expressions are identical .
+ for an arbitrary number of particles , our general construction ( [ fphi ] ) for @xmath39 should reduce to the formula presented in section 1.1 .
this looks quite a non trivial task to be performed , and we hope we can return to this issue in the future ( a similar situation has already occured in @xcite and @xcite ) . +
let us note finally that the v.e.v @xmath118 does not need to be known exactly , as it does not enter the numerical estimation .
+ in order to apply the @xmath4-sum rule test , it is important to have in mind a uv operator that could be a good candidate : we recall that the numerical tests on the central charge in @xcite already showed quite a large discrepancy with respect to the exact results , so we do not expect the sum rule for the conformal dimension to give particularly accurate results .
+ the minimal model @xmath1 is nothing but the coset model @xmath11 , and we will denote the latter @xmath130 . the operator @xmath3 in @xmath130 has conformal dimension @xmath131 which is the same as @xmath5 in the coset model , the primaries @xmath132 with @xmath133 have conformal weight @xmath134 , where @xmath135 $ ] for @xmath136 even [ odd ] and @xmath137 .
] @xmath138 . from the perturbative rg calculations
@xcite for the @xmath139 induced flow from @xmath140 to @xmath1 , and the landau - ginzburg representation @xcite , it is well known that one expects @xmath5 in @xmath140 to flow to @xmath5 in @xmath1 for @xmath27 odd . for @xmath15 , @xmath5 in @xmath141
has conformal dimension @xmath142 , and @xmath143 in @xmath144 has conformal dimension @xmath145 .
this flow was later confirmed by @xcite by guessing massless tba equations for the first excited state in the theory @xmath141 , via a suitable modification of the massless tba equations for the ground state found in @xcite .
+ then , using the form factors obtained in @xcite for the flow tim @xmath2 i m , the authors of @xcite successfully identified the conformal dimension of @xmath5 in the uv thanks to the @xmath4-sum rule .
+ some results are also available about the massless flows between n=1 unitary superconformal models , thanks to perturbative rg analysis @xcite .
in particular , the results obtained in this article indicate that ( at least for @xmath136 large ) , there exist flows @xmath146 from @xmath147 to @xmath148 , ( the case we are particularly interested in is given by @xmath149 , which is not covered by rg analysis ) .
+ consequently , we find it natural enough -let us recall that the operator @xmath5 in the coset @xmath150 , which conformal weight is @xmath151 , is the fundamental landau - ginzburg field- to conjecture that we are dealing with @xmath152 in @xmath153 , flowing to @xmath5 in @xmath138 .
the conformal dimension of the uv operator @xmath5 is : @xmath154 and we conjecture that the variation of the conformal dimension along the flow is given by : @xmath155 in table [ tab1 ] we present the numerical estimation versus the exact result : we find a good agreement between them , at least within the precision of the four - particle approximation ( the discrepancies given in the last column of table 1 are compatible with such an approximation ) .
let us note also that the accuracy of the results observed here is actually better than the one we had obtained for the @xmath156-theorem in @xcite , where it had appeared that the precision of the numerical results diminished as one increased @xmath27 . in our previous paper , we had linked this phenomenon to the fact that the conformal weight of the trace operator @xmath157 gets closer to one as @xmath27 increases , spoiling thus the convergence in the uv of the integral expressing the variation of the central charge .
the funny pattern observed in table 1 is a bit more difficult to explain ; we believe it could be caused by a combined effect of @xmath158 when @xmath159 , as well as the non monotonic behaviour of @xmath160 , as @xmath161 for both @xmath162 and @xmath159 .
in any case , we have little doubt that the @xmath4-sum rule supports our conjecture .
still , it would be interesting to confirm our result by means of other methods ( we have in mind the tba analysis , in a similar way to what has been done in @xcite ) .
.four particle approximation - comparison between @xmath163 and @xmath164 ( conjectured variation of the conformal dimension ) .
[ cols="^,<,<,>,<,<",options="header " , ]
we recall that for the massless flow between tim and i m , the asymptotic states consist of right and left majorana fermions , with lorentz spin @xmath166 .
the form factors of the energy operator @xmath167 have non zero matrix elements for an odd number of right particles and an odd number of left particles .
they satisfy the residue equation in the @xmath23 channel at @xmath168 ( we recall that @xmath169 ) : @xmath170 and a similar relation in the @xmath24 channel .
it was noticed in @xcite that they could be written as : @xmath171 the expression for @xmath172 being given by : @xmath173 where the functions @xmath174 and @xmath175 are defined in ( [ phi ] ) and ( [ grandphi ] ) above , and the first recursion step is @xcite : @xmath176 in the ir , @xmath177 , with conformal dimension @xmath178 .
the authors of @xcite checked numerically that the power law behaviour in the uv of the two point correlation function truncated to one right and one left particle agreed with the expected conformal dimension @xmath179 of the field @xmath3 in tim .
we generalize the previous results for any value of @xmath26 : the important observation is that the primary field @xmath6 in @xmath38 has conformal dimension @xmath180 : for @xmath39 it is nothing but the energy operator @xmath181 in the ising model , and for @xmath30 , it coincides with the spinon field of the @xmath182 wznw model .
it is thus natural to think of each chiral components as a generalized fermion for @xmath26 arbitrary ( in other words , its chiral components generate the asymptotic states ) @xcite .
+ we call @xmath183 and @xmath184 the holomorphic components of @xmath6 : @xmath185,@xmath186 have lorentz spin @xmath187 , whereas @xmath188,@xmath189 have lorentz spin @xmath190 .
these operators are topologically charged : as @xmath185,@xmath189 create asymptotic particles , they have topological charge @xmath191 .
analogously , @xmath188,@xmath186 have topological charge @xmath192 .
the problem of the construction of form factors of the `` parafermionic '' operators @xmath193 with a non zero topological charge in the rsos restriction of the sine - gordon model was first addressed in @xcite , the form factors of the components of the fermi field in sine - gordon were constructed in @xcite , and quite generally , form factors of topologically charged operators can be found in @xcite . *
@xmath194 topological charge : + we shall look now at form factors of an operator @xmath165 which flows to @xmath195 . the topological charge is equal to @xmath196 in both the right and left sectors .
the first step of the recursion relation is given by : @xmath197 such that in the ir limit @xmath198 : @xmath199 we make the following ansatz for the solution : @xmath200 in the formula above : * * we introduced the modified bethe ansatz states @xmath201 one should pay attention that under a lorentz transformation , this multiparticule state possesses a lorentz spin @xmath202 when @xmath203 .
* * the @xmath26-function for @xmath185 with topological charge @xmath196 is : @xmath204 under a lorentz transformation , this @xmath26-function possesses a lorentz spin @xmath205 , such that together with the bethe ansatz state , the total lorentz spin is equal to @xmath206 . * * we took into account the relation between the number of particles @xmath207 , the topological charge @xmath208 and the number of integration variables @xmath209 : @xmath210 .
+ let the set @xmath211 , and @xmath212 and @xmath213 .
these subsets have the number of elements : @xmath214 , @xmath215 , @xmath216 the sets @xmath217 and @xmath218 are similarly defined .
we propose : @xmath219 the function @xmath220 satisfies the properties at @xmath221 : * * @xmath222 : @xmath223 * * @xmath224 : @xmath225 + and similar relations in the @xmath24-channel . in particular @xmath226 .
we have in the ir ( checked for a small number of particles with mathematica ) : @xmath227 let us note that the minus sign in the equation ( [ q1 ] ) is related to the fact that the term @xmath228 in the residue equation gives an extra minus sign in the ir limit . *
@xmath229 topological charge + this case is similar to the previous one , at the condition of changing the number of integration variables @xmath230 into @xmath231 , introducing a new function @xmath232 : @xmath233 that satisfies similar equations to ( [ q1 ] ) and ( [ q2 ] ) .
there are some obvious modifications in the @xmath26-functions to be made , that we think are needless to make more explicit here . in the flow tim @xmath2 i m
, the uv operator @xmath3 with conformal dimension @xmath179 flows in the ir to @xmath6 with conformal dimension @xmath234 .
+ let us notice the operators @xmath6 in @xmath235 with conformal dimension @xmath236 coincide with the operators @xmath237 in the coset @xmath138 ( for @xmath15 , it is @xmath6 ) . + consequently , by analogy with the case @xmath15
, we are tempted to conjecture that the uv operator @xmath165 we are looking for is @xmath7 in the coset model @xmath153 .
this operator has conformal dimension in the coset model @xmath238 has conformal dimension @xmath239 , where @xmath240 ( @xmath241 ) is the conformal dimension of @xmath6 in the minimal model @xmath1 ( @xmath242 ) .
this operator is the generalization of @xmath6 in minimal models . ] : @xmath243 in the framework of form factors we can try to give some evidence in favour of such a conjecture by means of the analysis of the @xmath244 correlation function . as usual , we will write down the leading contribution to the spectral expansion for the correlator given by the two - particles form factor ( [ firststep ] ) @xmath245 where @xmath246 is the modified bessel function of order zero . the next step is then to compare the approximated correlator with the expected power - law behaviour in both the ir and uv , being @xmath247 and @xmath248 respectively . in other words , we are interested in comparing the slope of the approximated correlator with that predicted by cft at the critical points ( we will use the log - log plane to plot them ) .
the inspection of the diagrams ( see figures [ kp_1 ] , [ kp_2 ] , [ kp_3 ] , [ kp_10 ] , obtained for @xmath15 , @xmath249 , @xmath250 , @xmath251 respectively ) shows first of all the expected agreement in the ir limit : it is worth recalling that as in the case of the trace operator @xcite , the leading contribution of the spectral series is enough to obtain the exact ir power - law .
it is instead quite surprising that such an approximation is able to give a qualitative good agreement also in the uv ( again similar to the case of the stress - energy tensor ) when ( [ psiuv ] ) is conjectured to give the correct power - law behaviour . on the one hand
, the previous result can be considered as a qualitative evidence that the operator @xmath7 in the uv actually flows to the operator @xmath6 in the ir theory . on the other hand
, we would like to stress that such a result , being qualitative , is far from being neither conclusive nor satisfactory in the perspective of the identification of the uv operator @xmath7 .
at best it can be viewed as an indication to stimulate the research of other , more reliable , methods to face the problem of the identification of operators in massless flows . a final remark about the diagrams : since we are interested in the comparison of the slope of the curves , all the normalizations have been fixed in order to make such a comparison as clear as possible .
in this work , we constructed form factors of operators in the massless flow from the coset model @xmath0 to the @xmath252 minimal model , mimicking the construction done in @xcite of form factors of the magnetization operator and the energy operator in the massless flow + tim @xmath2 i m . what we did in the section 1.2 is to look for solutions of the residue equation that are obtained replacing the @xmath23 and @xmath24 @xmath22-matrices @xmath253 by the @xmath254-matrix , given that they should reproduce in the ir limit in both the right and left channel the form factors of the operator @xmath3 in the minimal model @xmath1 .
then we made a numerical check on the variation of conformal weight along the flow thanks to the @xmath4-sum rule , and found that , if not excellent , it is compatible with the hypothesis that we are dealing with the operator @xmath5 in the uv .
the situation is slightly more complicated in the section 2.2 , because the asymptotic particles possesses generalized statistics for @xmath255 : we took into account the ir properties only , and constructed form factors which in the ir limit reproduce the form factors of the parafermionic operators with conformal dimension @xmath256 and @xmath257 . notwithstanding this ,
the approximation of the two point function with the lowest form factor with one right and one left particle is enough to give , at least at a qualitative level , a good agreement with the power - law behaviour expected in the uv if we conjecture that the corresponding operator is @xmath258 .
we are not saying that we constructed all the possible flows of operators , but only those which have an obvious counterpart in the flow tim @xmath2 i m .
it would be interesting to know what else could be constructed .
the results obtained in both @xcite and the present work show that form factors in integrable massless models can provide important non perturbative information , even in more complicated cases than the flow tim @xmath2 i m .
obviously , we are in a privileged situation : having a one parameter family of flows at hand certainly allows us to understand better the loss of precision in the numerical tests for the @xmath156- and @xmath4-sum rules as one increases the parameter @xmath27 . had we worked on the flow @xmath259 only , the discrepancy of @xmath260 with respect to the exact value of the central charge @xcite would have probably led us to conclude that our 4-particle form factor for the trace operator was wrong !
interestingly enough , the results of the present work as well as those of @xcite show that whether in the massive or the massless case , the truncation to the lowest form factor does not systematically give a very accurate approximation of the correlation function .
one can really wonder up to what point it can be unaccurate ; certainly , this means that one has to be rather cautious when interpreting the numerical results , in the case where a large discrepancy with respect to the expectations is observed .
we are aware of the fact that our task was considerably simplified as the flow is along @xmath14 . for other integrable
massless flows with a non diagonal @xmath29 scattering , the situation is far more involved , both theoretically and numerically : in most cases , we do not expect that the lowest form factors can be nicely written as an explicit product of simple functions in the right and left channel as it is the case here ; likely , even with the lowest number of particles one might not get rid of the integration variables , thus the integrals for the form factors should be evaluated numerically first .
the work of p.g . was supported by the euclid network hprn - ct-2002 - 00325 .
b.p . was supported by a linkage international fellowship of the australian research council . | massless flows from the coset model @xmath0 to the minimal model @xmath1 are studied from the viewpoint of form factors .
these flows include in particular the flow from the tricritical ising model to the ising model . by analogy with the magnetization operator in the flow tim @xmath2 i m
, we construct all form factors of an operator that flows to @xmath3 in the ir .
we make a numerical estimation of the difference of conformal weights between the uv and the ir thanks to the @xmath4-sum rule ; the results are consistent with the conformal weight of the operator @xmath5 in the uv . by analogy with the energy operator in the flow tim @xmath2 i m
, we construct all form factors of an operator that flows to @xmath6 .
we propose to identify the operator in the uv with @xmath7 . * form factors in the massless coset models + *
@xmath8 * * + + + + pacs : 11.10.-z 11.10.kk 11.55.ds |
in reliability engineering two crucial objectives are considered : ( 1 ) to maximize an estimate of system reliability and ( 2 ) to minimize the variance of the reliability estimate . because system designers and users are risk - averse
, they generally prefer the second objective which leads to a system design with a slightly lower reliability estimate but a lower variance of that estimate , ( eg , @xcite ) .
it provides decision makers efficient rules compared to other designs which have a higher system reliability estimate , but with a high variability of that estimate . in the case of
parallel series and/or by duality series
parallel systems , the variance of the reliability estimate can be lowered by allocation of a fixed sample size ( the number of observations or units tested in the system ) , while reliability estimate is obtained by testing components , see berry @xcite .
allocation schemes for estimation with cost , see for example @xcite , lead generally to a discrete optimization problem which can be solved sequentially using adaptive designs in a fixed or a bayesian framework . based on a decision theoretic approach ,
the authors seek to minimize either the variance or the bayes risk associated to a squared error loss function .
the problem of optimal reliability estimation reduces to a problem of optimal allocation of the sample sizes between bernoulli populations .
such problems can be solved _ via _ dynamic programming but this technique becomes costly and intractable for complex systems . in the case of a two components series or parallel system , optimal procedures can be obtained and solved analytically when the coefficients of variation of the associated bernoulli populations are known , cf .
eg , @xcite .
unfortunately , the coefficients of variation are not known in practice since they depend themselves on the unknown components reliabilities of the system . in @xcite
, the author has defined a sequential allocation scheme in the case of a series system and has shown its first order asymptotic optimality for large sample sizes with comparison to the balanced scheme . in @xcite , a reliability
sequential schemes ( r - ss ) was applied successfully to parallel series systems , when the total number of units to be tested in each subsystem was fixed .
recently , in @xcite , a two stage design for the same purpose was presented and shown to be asymptotically optimal when the subsystems sample sizes are fixed and large at the same order of the total sample size of the system .
the problem considered in this paper is useful to estimate the reliability of a parallel - series and/or by duality a series - parallel system , when the components reliabilities are unknown as well as the total numbers of units allowed to be tested in each subsystem .
this work improves the results in @xcite by developing a hybrid two stage design to get a dynamic allocation between the sample sizes allowed for subsystems and those allowed for their components .
for example , consider a parallel system of four components ( 1),(2),(3 ) and ( 4 ) , with reliabilities 0.05 , 0.1 , 0.95 and 0.99 , respectively , under the constraint that the total number of observations allowed is @xmath0 .
then , the sequential scheme given in @xcite suggests to test , respectively , 10 , 10 , 28 and 52 units and produces a variance of the system reliability estimate equal @xmath1 , approximately . this is visibly better ,
compared to the balanced scheme which takes an allocation equal 25 in each component and produces a variance ten times greater then the former .
the hybrid sequential scheme proposed in this paper is a tool to solve the same problem when the components are replaced by subsystems .
more precisely , it combines the schemes developed for parallel and/or series systems in order to obtain approximately the best allocation at subsystems level as well as at components level . in section [ prelim ] , definitions and preliminary results
are presented accompanied by the proper two stage design for a parallel subsystem just as was defined in @xcite and its asymptotic optimality is proved for a fixed and large sample size . in section
[ lower ] , a parallel - series system is considered and it is shown that the variance of its reliability estimate has a lower bound independent of allocation .
this leads , in section [ overlapping ] , to the main result of this paper which lies in the hybrid two stage algorithm and its asymptotic optimality for a fixed and large sample size allowed for the system . in section [ monte carlo ] ,
the results are validated _ via _ monte carlo simulation and it is shown that our algorithm leads asymptotically to the best allocation scheme to reach the lower bound of the variance of the reliability estimate . the last section is reserved for conclusion and remarks .
consider a system @xmath2 of @xmath3 subsystems @xmath4 connected in series , each subsystem @xmath5 contains @xmath6 components @xmath7 connected in parallel .
the system should be referred as parallel - series system .
assume s - independence within and across populations , then the system reliability is @xmath8 where @xmath9is the reliability of the parallel subsystem @xmath5 and @xmath10 the reliability of component @xmath11 .
an estimator of @xmath12 is assumed to be the product of sample reliabilities@xmath13 where @xmath14 and @xmath15 is the sample mean of functioning units in component @xmath16,@xmath17@xmath15 is used to estimate @xmath10 where @xmath18 is the sample size and @xmath19 is the binary outcome of the unit @xmath20 in component @xmath21 .
it should be pointed that a unit is not necessarily a physical object in a component , but it represents just a bernoulli observation of the functioning / failure state of that component .
hence , for each subsystem @xmath5 , one must allocate@xmath22units such that the estimated reliability of the system is based on a total sample size@xmath23as in the series case , with the help of s - independence and the fact that a sample mean is an unbiased estimator of a bernoulli parameter , see @xcite , the variance of the estimated reliability @xmath24 incurred by any allocation scheme can be obtained , @xmath25where @xmath26 \label{vp}\]]is given as a function of the allocation numbers @xmath18 and the coefficients of variation of bernoulli populations@xmath27we have found convenient to work with the equivalent expression of ( [ vp ] ) , @xmath28,\ ] ] where@xmath29is a sum over all the products of at least two of its arguments . the problem is to estimate @xmath12 when components reliabilities are unknowns and a total number of @xmath30 units must be tested in the system at components level .
the aim is to minimize the variance of @xmath24 .
hence , the problem can be addressed by developing allocation schemes to select @xmath18 , the numbers of units to be tested in each component @xmath31 in the subsystem @xmath32 , under the constraint@xmath33such that the variance of @xmath24 is as small as possible .
reliability sequential schemes ( r - ss ) exist for the series , parallel or parallel - series configurations when the sample sizes @xmath34 of the subsystems are fixed . therefore , one can fully optimize the variance of @xmath24 just by applying the ( r - ss ) to find the best partition @xmath35 of @xmath30 .
unfortunately , a full sequential design can not be used in practice for large systems since the number of operations will growth dramatically . for this reason
, we reasonably propose a hybrid two stage design which is shown to be asymptotically optimal when @xmath30 is large .
for the asymptotic optimization of the variance of the estimated reliabilities , we make use of the well - known lagrange s identity which can be written in the form : let @xmath36 , @xmath37 for @xmath38 and @xmath39 , then the following identity holds.@xmath40\ ] ] [ th0]denote by @xmath41then@xmath42 the proof is a direct consequence of the previous identity ( [ lem1 ] ) .
indeed@xmath43 following the expansion ( [ var rj ] ) and since @xmath44 contains second order terms ( see later ) , one gives interest to the numbers @xmath45 which minimize the expression@xmath46thus @xmath18 must verify for @xmath47@xmath48which implies that@xmath49 if one assumes that @xmath34 is fixed then a proper two stage scheme can be used to determine @xmath18 , just as was defined in @xcite , as follows : choose @xmath50 as a function of @xmath34 such that : 1 .
@xmath50 must be large if @xmath34 is large , 2 .
@xmath51 3 .
@xmath52 .
one can take for example @xmath53 $ ] , where @xmath54 $ ] denotes the integer part .
: : sample @xmath50 units from each component @xmath31 in the subsystem @xmath32 , estimate @xmath55 by its maximum likelihood estimator ( m.l.e ) @xmath56 and define the predictor , according to ( [ mij]),@xmath57,~i=1,\ldots , n_{j}-1\ ] ] stage 2 .
: : sample @xmath58 units for which @xmath59 are units from component @xmath31 in the subsystem @xmath32 where @xmath18 is the corrector of @xmath60 defined by @xmath61 [ th1 ] choosing the @xmath18 according to the previous two stage sampling scheme , one obtains @xmath62 from relation ( [ var rj ] ) , one can write@xmath63when @xmath34 is large enough , condition ( iii ) gives @xmath64 for @xmath65 .
so , the strong law of large numbers with the integer part properties give , when @xmath66,@xmath67for @xmath47 .
hence,@xmath68 and on the other hand@xmath69 which achieves the proof .
we consider now the parallel series system @xmath2 . from expression ( [ g ] ) , one can write@xmath70\ ] ] the following theorem gives a lower bound for the variance of @xmath24 .
[ th2 ] denote by@xmath71 ^{2}\]]then@xmath72 expanding the right hand side of ( [ g ] ) and using ( [ r ] ) , one obtains@xmath73,\ ] ] which gives with the help of theorem [ th0]@xmath74this last expression has the form @xmath75which can be expanded , thanks to identity ( [ lem1 ] ) , as follows@xmath76 ^{2}\nonumber \\ & + & r^{2 } t^{-1}\sum\limits_{i=1}^{n-1}\sum\limits_{j = i+1}^{n}\frac{\left ( t_{i}\frac{1-r_{j}% } { r_{j}}\sum\limits_{k=1}^{n_{j}}c_{kj}^{-1}-t_{j}\frac{1-r_{i}}{r_{i}}% \sum\limits_{k=1}^{n_{i}}c_{ki}^{-1}\right ) ^{2}}{t_{i}t_{j } } \label{vr2e}\end{aligned}\]]and as a consequence @xmath77 ^{2}=q,\]]which achieves the proof .
similarly to the case of a subsystem an from expressions ( [ vr1 ] ) and ( [ vr2e ] ) , one gives interest to the numbers @xmath34 which minimize the quantity @xmath78and obtains the asymptotic optimality criteria @xmath79for all @xmath80 , which gives the rule@xmath81 we can now implement a hybrid two stage design for the determination of the numbers @xmath34 as well as @xmath18 as follows : stage 1 : : choose @xmath82 $ ] : one applies the two stage scheme given in subsection [ par ] for each subsystem @xmath5 with @xmath83 and @xmath53 $ ] .
next , obtain the predictor , according to the rule ( [ tj]),@xmath84,~j=1,\ldots , n-1.\ ] ] stage 2 : : define the corrector @xmath85and take back the two stage scheme for each subsystem @xmath5 to calculate @xmath86 with the sample size equals @xmath34 .
now , the main result of this paper is given by the following theorem .
[ th3]choosing the @xmath34 and @xmath18 according to the hybrid two stage design , one obtains @xmath87where @xmath88 is defined in theorem [ th2 ] .
the relation ( [ vrj ] ) implies that@xmath89as a consequence of the hybrid two stage design and the strong law of large numbers , @xmath90 and @xmath91 remain bounded for all @xmath92 as @xmath93 .
it follows that , as @xmath94,@xmath95and @xmath96thanks to ( [ eq1 ] ) and ( [ eq2 ] ) .
thus,@xmath97which implies that@xmath98 as a consequence,@xmath99\]]now , expanding the product within the limit and applying identity ( [ lem1 ] ) , after having replaced @xmath100 by its expression ( [ qj ] ) , one obtains @xmath101 \\ & = & q+r^{2 } \left ( a+b\right ) , \end{aligned}\ ] ] where @xmath102 once more , the hybrid two stage allocation scheme and the strong law of large numbers provide@xmath103and@xmath104which achieves the proof .
let us remark first that the lower bound @xmath88 is a first order approximation of the optimal variance of the reliability estimate under the constraint ( [ constr ] ) when @xmath30 is large . in the first experiment
, we will validate the fact that the hybrid scheme provides the best allocation at system level . as in figure
[ fig : f1 ] , we consider a simple parallel - series system of two subsystems each one , with varying reliabilities and a fixed sample size @xmath105 .
for each situation a , b , c and d and for each partition sample size @xmath106 where @xmath107 varies from @xmath108 $ ] to @xmath109 $ ] , by step one , we have applied the proper two stage design for each parallel subsystem and reported in a bar diagram @xmath110 as a function of @xmath111 , see figure [ fig : f2 ] . on the other hand , in table [ tab:1 ] , we have reported the expected value of @xmath112 given by the hybrid two stage design . as expected
, our scheme gives the best allocation for each situation .
the second experiment deals with a non trivial parallel - series system just as in @xcite , where subsystems are composed , respectively , of 2,3,4 and 5 components , see figure [ fig : f3 ] .
the partition total numbers @xmath34 to test in each subsystem are evaluated systematically by the hybrid two stage design while their sum @xmath30 is incremented from 100 to 10000 by step of 100 .
figure [ fig : f4 ] shows the rate of the excess of variance @xmath113 at logarithmic scale as a function of the sample size @xmath30 .
the asymptotic optimality of the hybrid scheme is validated . as a function of @xmath114 for each case a , b , c and d :
@xmath115 shows the minimum of @xmath116,scaledwidth=75.0% ] .expected value of @xmath112 given by the hybrid two stage design [ cols="<,<,<,<,<,<",options="header " , ]
the proof of the first order asymptotic optimality for the proper two stage design for a parallel subsystem as well as for the hybrid two stage design for the full system has been obtained mainly through the following steps * an adequate writing of the variance of the reliability estimate , * a lower bound for this variance , independent of allocation , * the allocation defined by the hybrid sampling scheme and the strong law of large numbers . with a straightforward but tedious adaptation ,
the above study can be namely extended to deal with complex systems involving a multi - criteria optimization problem under a set of constraints such as risk , system weight , cost , performance and others , in a fixed or in a bayesian framework .
this work is supported with grants by the national research project ( pnr ) and the l.a.a.r laboratory of the department of physics in the university mohamed boudiaf of oran .
m. terbeche , o. broderick , two stage design for estimation of mean difference in the exponential family , advances and applications in statistics 5(3 ) ( 2005 ) 325339 .
m. woodroofe , j. hardwick , sequential allocation for an estimation problem with ethical costs , annals of statistics 18(3 ) ( 1990 ) 13581377 . | we give a hybrid two stage design which can be useful to estimate the reliability of a parallel series and/or by duality a series parallel system , when the component reliabilities are unknown as well as the total numbers of units allowed to be tested in each subsystem .
when a total sample size is fixed large , asymptotic optimality is proved systematically and validated _ via _ monte carlo simulation .
* keywords . * asymptotic optimality ; hybrid ; reliability ; parallel - series ; two stage design . |
the study of recursive program schemes is one of the classic topics in programming language semantics .
( see , e.g. , @xcite for some of the early references . )
one of the main goals of this line of research is to define the semantics of systems of recursive equations such as @xmath0 in the above recursion scheme , the symbols ifzero , add , pred , @xmath1 and @xmath2 denote given function symbols ; these are used to define the derived unary function @xmath3 , which we will refer to as a functor variable . interpreting ifzero as the function over the set of triples of @xmath4 that returns its second argument when the first is zero and the third otherwise , mult as multiplication and pred as the predecessor function , intuitively one would expect the above recursion scheme to describe the function over the natural numbers that , given an input @xmath5 , returns @xmath6 .
this expectation has been formalized in several ways in the literature on recursive program schemes .
a classic answer can be summarized by the ` motto ' of the initial - algebra - semantics approach : ` the semantics of a recursive program scheme is the infinite term tree ( or ranked tree ) that is the least fixed point of the system of equations associated with the program scheme . ' in the light of the role that infinite term trees play in defining the semantics of recursive program schemes , it is not surprising that the study of infinite term trees has received a lot of attention in the research literature . here
we limit ourselves to mentioning courcelle s classic survey paper @xcite , which presents results on topological and order - theoretic properties of infinite trees , notions of substitutions for trees as well as regular and algebraic term trees .
_ algebraic term trees _ are those that arise as solutions of recursive program schemes that , like ( [ algschema ] ) , are ` first order ' .
on the other hand , _ regular term trees _ are the solutions of systems of equations like @xmath7 which define parameterless functions @xmath8 and @xmath9 .
regular term trees arise naturally as the unfoldings of flowcharts , whereas algebraic term trees stem from the unfoldings of recursion schemes that correspond to functional programs @xcite . in this paper
, we study recursion schemes from a process - algebraic perspective and investigate the expressive power of algebraic recursion schemes over the signatures of basic ccs @xcite and of basic process algebra ( bpa ) @xcite as a way of defining possibly infinite synchronization trees @xcite .
both these signatures allow one to describe every finite synchronization tree and include a binary choice operator + .
the difference between them is that the signature for basic ccs , which is denoted by @xmath10 in this paper , contains a unary action prefixing operation @xmath11 for each action @xmath12 , whereas the signature for bpa , which we denote by @xmath13 , has one constant @xmath12 for each action that may label the edge of a synchronization tree and offers a full - blown sequential composition , or sequential product , operator .
intuitively , the sequential product @xmath14 of two synchronization trees is obtained by appending a copy of @xmath15 to the leaves of @xmath16 that describe successful termination of a computation . in order to distinguish successful and unsuccessful termination , both the signatures @xmath10 and @xmath13 contain constants @xmath17 and @xmath1 , which denote unsuccessful and successful termination , respectively . as an example of a regular recursion scheme over the signature @xmath13 , consider @xmath18 on the other hand , the following recursion scheme is @xmath10-algebraic , but not @xmath10-regular : @xmath19 it turns out that both these recursion schemes define the infinitely branching synchronization tree depicted on figure [ fig : anyfinitedepth ] . in the setting of process algebras such as ccs @xcite and acp @xcite ,
synchronization trees are a classic model of process behaviour .
they arise as unfoldings of labelled transition systems that describe the operational semantics of process terms and have been used to give denotational semantics to process description languages see , for instance , @xcite .
regular synchronization trees over the signature @xmath10 are unfoldings of processes that can be described in the regular fragment of ccs , which is obtained by adding to the signature @xmath10 a facility for the recursive definition of processes .
on the other hand , regular synchronization trees over the signature @xmath13 are unfoldings of processes that can be described in basic process algebra ( bpa ) @xcite augmented with constants for the deadlocked and the empty process as well as recursive definitions .
as is well known , the collection of regular synchronization trees over the signature @xmath13 strictly includes that of regular synchronization trees over the signature @xmath10 even up to language equivalence .
therefore , the notion of regularity depends on the signature .
but what is the expressive power of algebraic recursion schemes over the signatures @xmath10 and @xmath13 ?
the aim of this paper is to begin the analysis of the expressive power of those recursion schemes as a means for defining synchronization trees , and their bisimulation or language equivalence classes . in order to characterize the expressive power of algebraic recursion schemes defining synchronization trees , we interpret such schemes in continuous categorical @xmath10- and @xmath13-algebras of synchronization trees .
continuous categorical @xmath20-algebras are a categorical generalization of the classic notion of continuous @xmath20-algebra that underlies the work on algebraic semantics @xcite , and have been used in @xcite to give semantics to recursion schemes over synchronization trees and words .
( we refer the interested reader to @xcite for a recent discussion of category - theoretic approaches to the solution of recursion schemes . ) in this setting , the @xmath10-regular ( respectively , @xmath10-algebraic ) synchronization trees are those that are initial solutions of regular ( respectively , algebraic ) recursion schemes over the signature @xmath10 .
@xmath13-regular and @xmath13-algebraic synchronization trees are defined in similar fashion .
our first contribution in the paper is therefore to provide a categorical semantics for first - order recursion schemes that define processes , whose behaviour is represented by synchronization trees .
the use of continuous categorical @xmath20-algebras allows us to deal with arbitrary first - order recursion schemes ; there is no need to restrict oneself to , say , ` guarded ' recursion schemes , as one is forced to do when using a metric semantics ( see , for instance , @xcite for a tutorial introduction to metric semantics ) , and this categorical approach to giving semantics to first - order recursion schemes can be applied even when the order - theoretic framework either fails because of the lack of a ` natural ' order or leads to undesirable identities . as a second contribution
, we provide a comparison of the expressive power of regular and algebraic recursion schemes over the signatures @xmath10 and @xmath13 , as a formalism for defining processes described by their associated synchronization trees .
we show that each @xmath13-regular tree is @xmath10-algebraic ( theorem [ thm : delta2gamma ] ) by providing an algorithm for transforming a @xmath13-regular recursion scheme into an equivalent @xmath10-algebraic one that involves only unary functor variables .
in addition , we prove that every synchronization tree that is defined by a @xmath10-algebraic recursion scheme of a certain form that involves only unary functor variables can be transformed into an equivalent @xmath13-regular recursion scheme .
we provide examples of @xmath10-algebraic synchronization trees that are not @xmath13-regular and of @xmath13-algebraic trees that are not @xmath10-algebraic , not even up to bisimulation equivalence .
in particular , in proposition [ prop : bagnotalg ] we prove that the synchronization tree associated with the bag over a binary alphabet ( which is depicted on figure [ fig : bag ] on page ) is not @xmath10-algebraic , even up to language equivalence , and that it is not @xmath13-algebraic up to bisimilarity .
these results are a strengthening of a classic theorem from the literature on process algebra proved by bergstra and klop in @xcite .
since each @xmath10-algebraic synchronization tree is also @xmath13-algebraic , we obtain the following strict expressiveness hierarchy , which holds up bisimilarity @xcite : @xmath10-regular @xmath21 @xmath13-regular @xmath21 @xmath10-algebraic @xmath21 @xmath13-algebraic . in the setting of language equivalence ,
the notion of @xmath10-regularity corresponds to the regular languages , the one of @xmath13-regularity or @xmath10-algebraicity corresponds to the context - free languages , and @xmath13-algebraicity corresponds to the macro languages @xcite , which coincide with the languages generated by aho s indexed grammars @xcite .
( 0,-0.2 ) ellipse ( 1.65 * 0.90 and 1.8 ) ; ( 0,1.3 ) ellipse ( 3 * 0.90 and 3.3 ) ; ( 0,2.3 ) ellipse ( 3.5 * 0.90 and 4.3 ) ; at ( 0,3.2 ) @xmath10-alg .
@xmath22 @xmath23 @xmath24 @xmath13-reg .
; at ( 0,5.8)[text width=1.4 cm , text centered ] @xmath25 @xmath26 @xmath13-alg .
; at ( 0,0 ) @xmath27 @xmath28 @xmath29 @xmath30 @xmath28 @xmath31 @xmath28 @xmath10-reg . ; at ( 0,-2.5 ) ( a ) ; @xmath32 ( 0,-0.2 ) ellipse ( 1.65 * 0.90 and 1.8 ) ; ( 0,2 ) ellipse ( 3.25 * 0.90 and 4 ) ; ( 0,1.3 ) ellipse ( 3 * 0.90 and 3.3 ) ; ( 0,2.3 ) ellipse ( 3.5 * 0.90 and 4.3 ) ; ( 0,0.3 ) ellipse ( 2.15 * 0.90 and 2.3 ) ; at ( 0,2.1 ) @xmath13-reg . ; at ( 0,3.7 ) @xmath10-alg .
@xmath22 @xmath23 ; at ( 0,5.3 ) @xmath13-alg . ; at ( 0,6.3 ) @xmath33 ; at ( 0,0 ) @xmath27 @xmath28 @xmath29 @xmath30 @xmath28 @xmath31 @xmath28 @xmath10-reg . ; at ( 0,-2.5 ) ( b ) ; @xmath32 ( 0,-0.2 ) ellipse ( 1.65 * 0.90 and 1.8 ) ; ( 0,2 ) ellipse ( 3.25 * 0.90 and 4 ) ; ( 0,1.3 ) ellipse ( 3 * 0.90 and 3.3 ) ; ( 0,2.3 ) ellipse ( 3.5 * 0.90 and 4.3 ) ; ( 0,0.8)[rotate=30 ] ellipse ( 2.15 * 0.94 and 2.8 ) ; ( 0,0.8)[rotate=-30 ] ellipse ( 2.15 * 0.94 and 2.8 ) ; at ( -1.7,2.5 ) @xmath34 ; at ( 1.7,2.5 ) @xmath13-reg . ; at ( 0,4 ) @xmath10-alg . ; at ( 0,5.3 ) @xmath13-alg . ;
at ( 0,6.3 ) @xmath33 ; at ( 0,0 ) @xmath30 @xmath28 @xmath31 @xmath28 @xmath10-reg . ; at ( 0,-2.5 ) ( c ) ; in order to obtain a deeper understanding of @xmath10-algebraic recursion schemes , we characterize their expressive power by following the lead of courcelle @xcite . in those references
, courcelle proved that a term tree is algebraic if , and only if , its branch language is a deterministic context - free language . in our setting , we associate with each synchronization tree with bounded branching a family of branch languages and we show that a synchronization tree with bounded branching is @xmath10-algebraic if , and only if , the family of branch languages associated with it contains a deterministic context - free language ( theorem [ thm - dcfl ] ) . in conjunction with standard tools from formal language theory ,
this result can be used to show that certain synchronization trees are not @xmath10-algebraic . as a final main contribution of the paper
, we compare the expressiveness of those recursion schemes to that of the low levels in caucal s hierarchy .
this hierarchy is already known to include the term trees defined by safe higher - order recursion schemes when interpreted over the free continuous algebra @xcite .
unsurprisingly , we show that the classes of @xmath10-algebraic and @xmath13-algebraic synchronization trees belong to the third level of the caucal hierarchy .
we provide a more detailed comparison , which is summarized in figure [ fig : hierarchy ] . as a benefit of the comparison with the caucal hierarchy ,
we obtain structural properties and decidability of the monadic second - order theories of @xmath13-algebraic synchronization trees @xcite . by contraposition
, this implies that a synchronization tree with an undecidable monadic second - order theory can not be @xmath13-algebraic .
this allows us to show that @xmath10-algebraic ( and hence @xmath13-algebraic trees ) are not closed under minimization with respect to bisimulation equivalence ( proposition [ prop : quotientnotcaucal ] ) .
the technical developments in this paper make use of techniques and results from a variety of areas of theoretical computer science .
we employ elementary tools from category theory and initial - algebra semantics to define the meaning of recursion schemes over algebras of synchronization trees .
tools from concurrency and formal - language theory are used to obtain separation results between the different classes of synchronization trees we consider in this paper .
the proof of theorem [ thm - dcfl ] , characterizing the expressive power of @xmath10-algebraic recursion schemes in the style of courcelle , uses a mezei - wright theorem for categorical algebras @xcite as well as tools from monadic second - order logic @xcite .
overall , we find it pleasing that methods and results developed by different communities within theoretical computer science play a role in the study of natural structures like the synchronization trees that arise from the solution of algebraic recursion schemes .
the paper is organized as follows . in section [
sect : definitions ] , we introduce ( first - order ) recursion schemes and the synchronization trees they define using continuous categorical algebras . in section [ sect : comparison ] , we compare the classes of @xmath10-regular , @xmath13-regular , @xmath10-algebraic and @xmath13-algebraic synchronization trees up to isomorphism , bisimulation and language equivalence . in section [
sect : caucal ] , we compare our classes of synchronization trees to the first levels of the caucal hierarchy . in section [
sect : branchlang ] , following courcelle , we characterize the expressive power of @xmath10-algebraic recursion schemes by studying the branch languages of synchronization trees whose vertices have bounded outdegree . in section [
sect : treesandlogic ] , we show how the previous results can be used to prove that some synchronization trees can not be defined using algebraic recursion schemes . section [ sect : concl ] concludes the paper and lists topics for future research .
in this section , we provide a categorical semantics for first - order recursion schemes using continuous categorical algebras , and introduce several notions and results that will be used throughout the paper .
section [ sect : contalg ] introduces the necessary categorical background .
the continuous categorical algebras associated with synchronization trees are introduced in section [ sect : syntrees ] .
sections [ sect : edge - labelled - graphs][sect : graph - transformations ] present edge - labelled graphs , monadic second order logic over such structures and the kinds of graph transformations used in the paper . the definition of the synchronization tree defined by a recursion scheme is given in section [ sec - algebraic objects ] .
sections [ sect : coa ] and [ sec : morphism ] describe the essentially unique morphisms from the initial continuous ordered @xmath10-algebra ( resp .
@xmath13-algebra ) of @xmath10-term trees ( resp .
@xmath13-term trees ) to the continuous @xmath10-algebra ( resp .
@xmath13-algebra ) of synchronization trees . these morphisms play an important role when comparing the synchronization trees defined by recursion schemes to the low levels of the caucal hierarchy in section [ sect : caucal ] .
we conclude by giving some basic properties of synchronization trees defined by first - order recursion schemes in section [ sect : closure ] . throughout the paper ,
when @xmath5 is a non - negative integer , we denote the set @xmath35 by @xmath36 $ ] . in this section ,
we recall the notion of continuous categorical @xmath20-algebra .
these structures were used in @xcite to give semantics to recursion schemes over synchronization trees and words .
let @xmath37 be a ranked set ( or ` signature ' ) .
a * categorical @xmath20-algebra * is a small category @xmath38 equipped with a functor @xmath39 for each @xmath40 , @xmath41 .
a * morphism * between categorical @xmath20-algebras @xmath38 and @xmath42 is a functor @xmath43 such that for each @xmath40 , the diagram @xmath44 \efig\end{aligned}\ ] ] commutes _ up to a natural isomorphism_. here , the functor @xmath45 maps each object and morphism @xmath46 in @xmath47 to @xmath48 in @xmath49 .
suppose that @xmath38 is a categorical @xmath20-algebra .
we call @xmath38 * continuous * if @xmath38 has a distinguished initial object ( denoted @xmath50 or @xmath51 ) and colimits of all @xmath52-diagrams @xmath53 . moreover , each functor @xmath54 is continuous , i.e. , preserves colimits of @xmath52-diagrams .
thus , if @xmath55 , say , and if @xmath56 are @xmath57-diagrams in @xmath38 with colimits @xmath58 and @xmath59 , respectively , then @xmath60 has colimit @xmath61 a morphism of continuous categorical @xmath20-algebras is a categorical @xmath20-algebra morphism which preserves the distinguished initial object and colimits of all @xmath52-diagrams . in what follows , we will often write just @xmath62 for @xmath63 , in particular when @xmath38 is understood . for later use
, we note that if @xmath38 and @xmath42 are continuous categorical @xmath20-algebras then so is @xmath64 .
moreover , for each @xmath65 , the category @xmath66 $ ] of all continuous functors @xmath67 is also a continuous categorical @xmath20-algebra , where for each @xmath40 , @xmath68}(f_1,\ldots , f_n ) = \sigma^{c}\circ \langle f_1,\ldots , f_n \rangle\ ] ] with @xmath69 standing for the target tupling of the continuous functors @xmath70 . on natural transformations , @xmath71}$ ]
is defined in a similar fashion . in @xmath66
$ ] , colimits of @xmath52-diagrams are formed pointwise . in the rest of the paper , we will assume , without loss of generality , that the signature @xmath20 contains a special symbol denoted @xmath72 or @xmath17 of rank @xmath17 , interpreted in a continuous categorical @xmath20-algebra as its initial object . a * synchronization tree * @xmath73 over an alphabet @xmath74 of ` action symbols ' consists of * a finite or countably infinite set @xmath75 of ` vertices ' and an element @xmath76 , the ` root ' ; * a set @xmath77 of ` edges ' ; * a ` labelling function ' @xmath78 where @xmath79 is a label used to denote successful termination .
these data obey the following restrictions .
* @xmath80 is a rooted tree : for each @xmath81 , there is a unique path @xmath82 . * if @xmath83 and @xmath84 , then @xmath85 is a leaf , and @xmath86 is called an * exit vertex*. a synchronization tree is _ deterministic _ if no node in the tree has two equally - labelled outgoing edges .
a * morphism * @xmath87 of synchronization trees is a function @xmath88 which preserves the root , the edges and the labels , so that if @xmath89 is an edge of @xmath16 , then @xmath90 is an edge of @xmath15 , and @xmath91 .
in particular such a morphism maps the root of @xmath16 to the root of @xmath15 .
morphisms are therefore functional * simulations * @xcite .
it is clear that the trees over @xmath74 and tree morphisms form a category , which we denote by @xmath92 .
the tree that has a single vertex and no edges is initial .
it is known that the category of trees has colimits of all @xmath52-diagrams , see @xcite .
( it also has binary coproducts . ) in order to make the category of trees small , we may require that the vertices of a tree form a subset of some fixed infinite set .
suppose that @xmath93 is an @xmath52-diagram in @xmath92 , where @xmath94 for each @xmath95
. then the colimit @xmath96 can be constructed in the expected way .
first , we define for each @xmath97 the tree morphism @xmath98 as the composition of the morphisms @xmath99 or the identify morphism if @xmath100 .
then we take the disjoint union @xmath101 of the sets @xmath102 , @xmath41 , and define a vertex @xmath103 equivalent to a vertex @xmath104 if and only if there is some @xmath105 with @xmath106 .
the set of vertices of the colimit tree will be the equivalence classes of @xmath101
. the root will be the equivalence class containing the roots @xmath107 of the trees @xmath108
, @xmath41 .
when @xmath109 and @xmath110 are equivalence classes , the pair @xmath111 is an edge in the colimit tree exactly when there exist @xmath112 and @xmath113 for some @xmath5 such that @xmath114 contains the edge @xmath115 .
the label of this edge is @xmath116 .
note that , in this case , for each @xmath117 we have that @xmath118 , and that the label of this edge in @xmath119 is equal to @xmath120 .
this defines the object part @xmath16 of the colimit .
the canonical morphism @xmath121 maps a vertex @xmath122 to its equivalence class .
in the particular case when the morphisms @xmath123 are injective , we may usually assume that @xmath124 for each @xmath5 and that the morphism @xmath123 is the inclusion of @xmath102 into @xmath125 . in this case
the colimit is simply the ` union ' of the trees @xmath108 .
consider the sequence @xmath126 , @xmath41 of synchronization trees , where @xmath127 and @xmath128 maps each edge in @xmath114 to the action symbol @xmath12 .
let @xmath129 be the inclusion map from @xmath130 into @xmath131 .
then @xmath132 is an @xmath52-diagram and its colimit is @xmath133 , where @xmath134 and @xmath135 maps each edge to @xmath12 .
pictorially , we have : = [ circle , draw = black , fill = black , inner sep = 0.25 mm , outer sep= 0.5 mm ] = [ inner sep = 0 mm , outer sep= 0 mm ] at ( 0,0.75 ) @xmath136 ; ( n0 ) at ( 0,0 ) ; at ( 0,0.75 ) @xmath137 ; ( n0 ) at ( 0,0 ) ; ( n1 ) at ( 0,-0.75 ) ; ( n0 ) to[right ] node[yshift=1 mm ] @xmath12 ( n1 ) ; at ( 0,0.75 ) @xmath138 ; ( n0 ) at ( 0,0 ) ; ( n1 ) at ( 0,-0.75 ) ; ( n2 ) at ( 0,-1.5 ) ; ( n0 ) to[right ] node[yshift=1 mm ] @xmath12 ( n1 ) ; ( n1 ) to[right ] node[yshift=1 mm ] @xmath12 ( n2 ) ; at ( 0,0.75 ) @xmath139 ; ( n0 ) at ( 0,0 ) ; ( n1 ) at ( 0,-0.75 ) ; ( n2 ) at ( 0,-1.5 ) ; ( n3 ) at ( 0,-2.25 ) ; ( n0 ) to[right ] node[yshift=1 mm ] @xmath12 ( n1 ) ; ( n1 ) to[right ] node[yshift=1 mm ] @xmath12 ( n2 ) ; ( n2 ) to[right ] node[yshift=1 mm ] @xmath12 ( n3 ) ; ( 0,-0.5 ) ( 1,-0.5 ) ; at ( 0,0.75 ) @xmath140 ; ( n0 ) at ( 0,0 ) ; ( n1 ) at ( 0,-0.75 ) ; ( n2 ) at ( 0,-1.5 ) ; ( n3 ) at ( 0,-2.25 ) ; ( n4 ) at ( 0,-3 ) ; ( n0 ) to[right ] node[yshift=1 mm ] @xmath12 ( n1 ) ; ( n1 ) to[right ] node[yshift=1 mm ] @xmath12 ( n2 ) ; ( n2 ) to[right ] node[yshift=1 mm ] @xmath12 ( n3 ) ; ( n3 ) to[right ] node[yshift=1 mm ] @xmath12 ( n4 ) ; ( n4 ) ( 0,-4 ) ; the category @xmath92 of synchronization trees over @xmath74 is equipped with two binary operations : @xmath141 ( sum ) and @xmath142 ( sequential product or sequential composition ) , and either with a unary operation or a constant associated with each letter @xmath143 .
the * sum * @xmath144 of two trees is obtained by taking the disjoint union of the vertices of @xmath16 and @xmath15 and identifying the roots .
the edges and labelling are inherited . the * sequential product * @xmath14 of two trees
is obtained by replacing each edge of @xmath16 labelled @xmath145 by a copy of @xmath15 . with each letter @xmath143
, we can either associate a constant , or a unary * prefixing operation*. as a constant , @xmath12 denotes the tree with vertices @xmath146 and two edges : the edge @xmath147 , labelled @xmath12 , and the edge @xmath148 , labelled @xmath145 . as an operation
@xmath149 is the tree @xmath150 , for each tree @xmath16 .
let @xmath17 denote the tree with no edges and @xmath1 the tree with a single edge labelled @xmath145 . on morphisms ,
all operations are defined in the expected way .
for example , if @xmath151 and @xmath152 , then @xmath153 is the morphism that agrees with @xmath154 on the non - root vertices of @xmath16 and that agrees with @xmath155 on the non - root vertices of @xmath156 .
the root of @xmath157 is mapped to the root of @xmath158 . in the sequel we will consider two signatures for synchronization trees , @xmath10 and @xmath13 .
the signature @xmath10 contains @xmath141 , @xmath159 and each letter @xmath143 as a _ unary _ symbol .
in contrast , @xmath13 contains @xmath160 , @xmath159 and each letter @xmath161 as a _ nullary _ symbol .
it is known that , for both signatures , @xmath92 is a continuous categorical algebra .
see @xcite for details . in what follows , for each @xmath143 and term / tree @xmath16 , we write @xmath162 for @xmath149 .
we shall also abbreviate @xmath163 to @xmath12 , and write @xmath164 for @xmath165 note that @xmath142 is associative and has @xmath1 as unit , at least up to isomorphism , and that for all trees @xmath166 and @xmath156 , @xmath167 and @xmath168 up to isomorphism .
[ rem : delta - implies - gamma ] a continuous categorical @xmath13-algebra @xmath109 naturally induces an associated @xmath10-algebra @xmath169 . for each letter
@xmath143 , the functor @xmath170 associated to unary symbol @xmath12 is defined for each object and morphism @xmath171 by @xmath172 .
the functors for the other symbols are inherited ( i.e. , for all @xmath173 , @xmath174 ) .
it is easy to see that @xmath169 is indeed a continuous categorical @xmath10-algebra .
in particular the @xmath13-algebra of synchronization trees induces in this sense the @xmath10-algebra of synchronization trees .
[ def : bisimilarity - languageeq ] two synchronization trees @xmath73 and @xmath175 are _ bisimilar _
@xcite if there is some symmetric relation @xmath176 , called a _ bisimulation _ , that relates their roots , and such that if @xmath177 and there is some edge @xmath178 , then there is an equally - labelled edge @xmath179 with @xmath180 .
the _ path language _ of a synchronization tree is composed of the words in @xmath181 that label a path from the root to the source of an exit edge .
two trees are _ language equivalent _ if they have the same path language .
[ lem : bisim - det - iso ] if @xmath16 and @xmath15 are deterministic , bisimilar synchronization trees , then they are isomorphic .
the above lemma fails up to language equivalence .
for example , the trees and are language equivalent but not isomorphic .
we conclude this section by defining an operation on synchronization trees over @xmath182 that contracts all edges labelled by some symbol in @xmath183 .
the definition of this operation is illustrated in figure [ fig : contraction - synchronization - trees ] .
this notion will , in particular , be used to characterize @xmath10-algebraic trees in the caucal hierarchy .
let @xmath73 be synchronization tree over an alphabet @xmath182 .
the @xmath183-contraction of @xmath16 is the synchronization tree @xmath184 defined by : * @xmath185 , * a pair @xmath186 is an edge in @xmath187 if there exists a path in @xmath16 from @xmath85 to @xmath188 labelled by a word in @xmath189 , * for each @xmath190 , @xmath191 is the unique letter @xmath192 such that the path from @xmath85 to @xmath188 in @xmath16 is labelled in @xmath193 . = [ circle , draw = black , fill = black , inner sep = 0.25 mm , outer sep= 0.5 mm ] = [ inner sep = 0 mm , outer sep= 0 mm ] = [ font=,label distance=-1 mm ] = [ level distance=0.85cm , sibling distance=1 cm ] = [ level distance=0.85cm , sibling distance=1.25 cm ] child node[n , label=[l]below:@xmath194 edge from parent node[draw = none , above left ] @xmath12 child node[n , label=[l]right:@xmath195 child node[n , label=[l]left:@xmath196 child node[n , label=[l]below:@xmath197 edge from parent node[draw = none , left ] @xmath198 edge from parent node[draw = none , left ] @xmath199 child node[n , label=[l]right:@xmath200 child node[n , label=[l]below:@xmath201 edge from parent node[draw = none , right ] @xmath145 edge from parent node[draw = none , right ] @xmath202 edge from parent node[draw = none , right ] @xmath199 child node[n , label=[l]right:@xmath203 child node[n , label=[l]right:@xmath204 child node[n , label=[l]below:@xmath205 edge from parent node[draw = none , right ] @xmath198 edge from parent node[draw = none , right ] @xmath199 edge from parent node[draw = none , above right ] @xmath199 ; = [ level distance=0.85cm , sibling distance=0.85 cm ] child node[n , label=[l]below:@xmath194 edge from parent node[draw = none , above left ] @xmath12 child node[n , label=[l]below:@xmath197 edge from parent node[draw = none , left ] @xmath198 child node[n , label=[l]right:@xmath200 child node[n , label=[l]below:@xmath201 edge from parent node[draw = none , right ] @xmath145 edge from parent node[draw = none , right ] @xmath202 child node[n , label=[l]below:@xmath205 edge from parent node[draw = none , above right ] @xmath198 ; an _ edge - labelled graph _ ( or simply graph ) , whose edges are labelled by letters in a finite alphabet @xmath74 , is a pair @xmath206 where @xmath75 is a finite or countable set of vertices and @xmath207 is a set of labelled edges .
the edge @xmath208 has source @xmath86 , target @xmath85 and label @xmath12 .
the existence of an edge @xmath208 in @xmath209 is denoted by @xmath210 or simply @xmath211 when @xmath212 is clear from the context .
this notation is extended to words in @xmath213 in the usual way . for a language @xmath214 over @xmath74 , we write @xmath215 if there is some word @xmath216 such that @xmath217 . a graph @xmath206 is _ deterministic _ if for all @xmath218 and @xmath161 , if @xmath219 and @xmath220 then @xmath221 .
a _ root _ of a graph @xmath206 labelled by @xmath74 is a vertex @xmath222 such that for all @xmath223 , there exists a path from @xmath222 to @xmath85 .
if this path is unique for each vertex @xmath85 , the graph @xmath206 is a _ tree _ with @xmath222 as its root .
a graph @xmath206 labelled in @xmath224 represents a synchronization tree if @xmath206 is a tree with root @xmath222 and if the target of an @xmath145-labelled edge is always a leaf of the tree .
we will identify such a graph with the synchronization tree @xmath225 where @xmath226 and for all @xmath227 , @xmath228 is the unique symbol in @xmath224 such that @xmath229 belongs to @xmath209 .
the definition of @xmath183-contraction for labelled graphs generalizes the case of synchronization trees introduced in section [ sect : syntrees ] .
the @xmath183-contraction of a graph @xmath212 labelled in @xmath182 from one of its root vertices @xmath230 is the graph labelled by @xmath74 whose vertices are the targets ( in @xmath212 ) of edges labelled in @xmath74 together with the vertex @xmath230 .
there is an edge from a vertex @xmath86 to a vertex @xmath85 if there is a path from @xmath86 to @xmath85 in @xmath212 labelled by a word in @xmath231 .
_ monadic second - order logic ( mso ) _ on graphs is the extension of first - order logic with the ability to quantify over sets of vertices .
we use monadic - second order logic over edge - labelled graphs with the standard syntax and semantics ( see e.g. @xcite for a detailed presentation ) . _ monadic second - order formulas _ are built using _ first - order variables _ , which are interpreted as vertices of the graph and are denoted by lower case letters @xmath232 and _ monadic second - order variables _ , which are interpreted as sets of vertices of the graph and are denoted by upper case letters @xmath233 atomic formulas are of the form @xmath234 , @xmath235 and @xmath236 where @xmath171 and @xmath237 are first - order variables , @xmath8 is a second - order variable and @xmath12 is an edge label .
complex formulas are built as usual from atomic ones using boolean connectives and quantifiers .
free and bound occurrences of variables in a formula are defined in the standard fashion .
we write @xmath238 to denote that the formula @xmath239 has free variables in @xmath240 .
a _ closed formula _ has no free variables . for a formula @xmath238 , a graph @xmath212 , vertices @xmath241 of @xmath212 and sets of vertices @xmath242 of @xmath212 , we write @xmath243 $ ] to denote that @xmath212 _ satisfies _
@xmath239 when the free variable @xmath244 , @xmath245 $ ] , is interpreted as the set of vertices @xmath246 and the free variable @xmath247 , @xmath248 $ ] , is interpreted as the vertex @xmath249 . for a closed formula @xmath239
, we simply write @xmath250 .
the _ mso - theory _ of @xmath212 is the set of closed formulas satisfied by @xmath212 .
for example , the mso formula @xmath251 \rightarrow y \in x\ ] ] expresses that there exists a path from @xmath171 to @xmath237 whose edges are labelled @xmath12 . in the remainder of the paper
, we will use several transformations of edge - labelled graphs , namely the unfolding of a graph , mso - interpretations and mso - transductions .
the _ unfolding _ @xmath252 of a graph @xmath212 from one of its vertices @xmath230 is the tree whose vertices are the paths in @xmath212 starting from @xmath230 and with an edge labelled @xmath12 between two such paths @xmath253 and @xmath254 if @xmath254 extends @xmath253 by exactly one edge labelled @xmath12 .
an _ mso - interpretation _ is a graph transformation specified using mso - formulas .
an mso - interpretation is given by a tuple of mso - formulas of the form @xmath255 , where @xmath183 is a set of labels .
this interpretation when applied to a graph @xmath212 will produce a graph , denoted @xmath256 , whose edges have labels in @xmath183 .
the set of vertices of @xmath256 is the subset of the vertices of @xmath212 satisfying the formula @xmath257 . for each edge label @xmath258
, there is an edge from @xmath86 to @xmath85 labelled by @xmath198 in @xmath256 if and only if @xmath212 satisfies @xmath259 $ ] . more formally , the graph @xmath260 obtained by applying @xmath261 to a graph @xmath206 is such that : @xmath262 \ } \\ e ' & = & \ { ( u , b , v ) \in v ' \times b \times v ' \mid g \models \varphi_b[u , v]\}.
\\ \end{array}\ ] ] mso - interpretations can not increase the number of vertices of a graph . to overcome this weakness the notion of a transduction
was introduced by courcelle in @xcite .
the idea is to perform an operation that will increase the number of vertices before applying the mso - interpretation .
let @xmath263 be a finite set of edge labels .
a _ @xmath264-copying operation _ applied to a graph @xmath212 adds , to every vertex of @xmath212 , @xmath265 outgoing arcs respectively labelled by @xmath266 and @xmath267 all going to fresh vertices .
an _ mso - transduction _ is a @xmath264-copying operation followed by an mso - interpretation .
we start by introducing the types of recursion schemes studied in this paper .
[ def : recursion scheme ] let @xmath20 be a signature . a * @xmath20-recursion scheme * , or * recursion scheme over @xmath20 * , is a sequence @xmath209 of equations @xmath268 where each @xmath269 is a term over the signature @xmath270 in the variables @xmath271 , and @xmath272 contains the symbols @xmath273 ( sometimes called ` functor variables ' or function variables " ) of rank @xmath274 , @xmath275 $ ] .
a @xmath20-recursion scheme is * regular * if @xmath276 , for each @xmath245 $ ] .
suppose that @xmath109 is a continuous categorical @xmath20-algebra .
define @xmath277\times \cdots \times [ c^{k_n } { \rightarrow}c].\end{aligned}\ ] ] then @xmath278 is a continuous categorical @xmath20-algebra , as noted in section [ sect : contalg ] .
when each @xmath273 , @xmath245 $ ] , is interpreted as a continuous functor @xmath279 , each term over the extended signature @xmath280 in the variables @xmath281 induces a continuous functor @xmath282 that we denote by @xmath283 .
in fact , @xmath284 is a continuous functor @xmath285.\end{aligned}\ ] ] more precisely , we define @xmath284 as follows .
let @xmath286 denote continuous functors @xmath287 , @xmath288 $ ] , and let @xmath289 be a natural transformation @xmath290 for each @xmath288 $ ] . when @xmath16 is the variable @xmath291 , say , then @xmath283 is the @xmath292th projection functor @xmath282 , and @xmath293 is the identity natural transformation corresponding to this projection functor
. suppose now that @xmath16 is of the form @xmath294 , where @xmath295 and @xmath296 are terms
. then @xmath297 and @xmath298 , where @xmath299 and @xmath300 for all @xmath301 $ ] . ( here , we use the same notation for a functor and the corresponding identity natural transformation . ) finally , when @xmath16 is of the form @xmath302 , then @xmath303 , and the corresponding natural transformation is @xmath304 , where the @xmath305 and @xmath306 , @xmath307 $ ] , are defined similarly as above .
note that if each @xmath308 is an identity natural transformation ( so that @xmath309 , for all @xmath245 $ ] ) , then @xmath293 is the identity natural transformation @xmath310 . in any continuous categorical @xmath20-algebra @xmath109 , by target - tupling the functors @xmath311 associated with the right - hand sides of the equations in a recursion scheme @xmath209 as in ( [ eq : scheme ] ) , we obtain a continuous functor @xmath312 indeed , @xmath313 $ ] , for @xmath245 $ ] , so that @xmath314 thus , @xmath315 has an initial fixed point in @xmath278 , unique up to natural isomorphism , that we denote by @xmath316 so that , in particular , @xmath317 at least up to isomorphism , for each @xmath245 $ ] . it is well known that the initial fixed point @xmath318 can be ` computed ' in the following way .
let @xmath319 , where , for each @xmath320 $ ] , @xmath321}$ ] is the constant functor @xmath322 determined by the object @xmath323 .
then , for each @xmath324 , define @xmath325 , where @xmath326 for all @xmath327 $ ] .
next , let @xmath328 , where @xmath329 , @xmath327 $ ] , is the unique natural transformation @xmath330 } { \rightarrow}g_{1,j}$ ] .
moreover , for each @xmath324 , define @xmath331 as the natural transformation @xmath332 .
then @xmath318 is the colimit of the @xmath52-diagram @xmath333 in the continuous functor category @xmath278 .
suppose that @xmath109 is a continuous categorical @xmath20-algebra .
we call a functor @xmath334 * @xmath20-algebraic * , if there is a recursion scheme @xmath209 such that @xmath335 is isomorphic to @xmath336 , the first component of the above - mentioned initial solution of @xmath209 .
when @xmath337 , we identify a @xmath20-algebraic functor with a * @xmath20-algebraic object*. last , a * @xmath20-regular object * is an object isomorphic to the first component of the initial solution of a @xmath20-regular recursion scheme .
in particular , we get the notions of @xmath10-algebraic and @xmath10-regular trees , and @xmath13-algebraic and @xmath13-regular trees .
suppose that @xmath109 is the continuous categorical @xmath20-algebra @xmath92 , where @xmath20 is either @xmath10 or @xmath13 , and consider a recursion scheme @xmath209 of the form ( [ eq : scheme ] ) .
if the continuous functors @xmath338 , @xmath339 , @xmath340 preserve injective synchronization tree morphisms ( in each variable ) , then so does each @xmath341}$ ] , where @xmath342 .
it follows by induction that the functors @xmath343 preserve injective synchronization tree morphisms . moreover , the components of the natural transformations @xmath329 are injective synchronization tree morphisms
thus , when @xmath344 , so that the recursion scheme defines a tree , each @xmath345 is an embedding of the tree @xmath346 into @xmath347 . if we represent the trees @xmath346 so that these embeddings are inclusions , then the colimit @xmath348 of the @xmath52-diagram @xmath349 becomes the union of the trees @xmath346 , @xmath324 .
[ rem : arbitrary - rank - sum ] it is sometimes convenient to add to the signature @xmath10 sums of arbitrary nonzero rank .
for this , we consider the signature @xmath350 where the rank of each symbol @xmath351 is @xmath5 .
the synchronization tree associated with a @xmath352-term tree is defined similarly as the synchronization tree associated to a @xmath10-term tree .
the @xmath352-algebraic synchronization trees are @xmath10-algebraic .
indeed we can transform an algebraic @xmath352-scheme into a @xmath10-scheme defining the same synchronization tree by replacing every occurrence of a subterm of the form @xmath353 on the right - hand side of an equation by @xmath354 for @xmath355 , every subterm of the form @xmath356 by @xmath357 , and finally every subterm of the form @xmath358 by @xmath359 .
= [ circle , draw = black , fill = black , inner sep = 0.25 mm , outer sep= 0.5 mm ] = [ inner sep = 0 mm , outer sep= 0 mm ] = [ font=,label distance=-1 mm ] = [ level distance=0.85cm , sibling distance=1 cm ] = [ level distance=0.85cm , sibling distance=1.25 cm ] child node[n ] child node[n ] edge from parent node[draw = none , left ] @xmath145 edge from parent node[draw = none , above left ] @xmath12 child node[n ] child node[n ] child node[n ] edge from parent node[draw = none , left ] @xmath145 edge from parent node[draw = none , left ] @xmath12 edge from parent node[draw = none , left ] @xmath12 child node[n ] child node[n ] child node[n ] child node[n ] edge from parent node[draw = none , left ] @xmath145 edge from parent node[draw = none , left ] @xmath12 edge from parent node[draw = none , left ] @xmath12 edge from parent node[draw = none , left ] @xmath12 child node[n ] child node[n ] child node[n ] child node[n ] child node[n ] edge from parent node[draw = none , left ] @xmath145 edge from parent node[draw = none , left ] @xmath12 edge from parent node[draw = none , left ] @xmath12 edge from parent node[draw = none , left ] @xmath12 edge from parent node[draw = none , left ] @xmath12 child node[e ] edge from parent[dashed,- ] ; [ ex : anyfinitedepth ] the following @xmath13-regular recursion scheme @xmath360 has the infinitely branching tree @xmath361 depicted on figure [ fig : anyfinitedepth ] as its initial solution .
that tree is therefore @xmath13-regular .
note that the tree @xmath361 is also @xmath10-algebraic because it is the initial solution of the following @xmath10-algebraic recursion scheme @xmath362 ( recall that we use @xmath12 as an abbreviation of @xmath163 . )
the path language associated with the tree @xmath361 is @xmath363 .
note that the subtrees of that tree whose roots are children of the root of @xmath361 are pairwise inequivalent modulo language equivalence .
so @xmath10-algebraic recursion schemes can be used to define infinitely branching trees that have an infinite number of subtrees , even up to bisimilarity @xcite or language equivalence .
[ [ further - examples - of - algebraic - synchronization - trees ] ] further examples of algebraic synchronization trees + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + consider the labelled transition system ( lts ) on figure [ fig : dyck - like ] .
the synchronization tree associated with that lts is @xmath10-algebraic because it is defined by the recursion scheme below .
@xmath364 the idea underlying the above definition is as follows . at
any given vertex in the lts on figure [ fig : dyck - like ] , the argument @xmath85 of @xmath365 denotes the subtree obtained by unfolding the lts from that vertex _ and not _ taking the @xmath12-labelled edge as a first step .
@xmath366 denotes the tree obtained by unfolding the lts at that vertex .
this algebraic recursion scheme over @xmath10 corresponds to the regular recursion scheme over @xmath13 , @xmath367 @xmath368 & { * + [ o][f=]{\makebox(4,4 ) { } } } { } \ar@<1ex>[r]^a & { * + [ o][f-]{\makebox(4,4 ) { } } } { } \ar@<1ex>[r]^a \ar@<1ex>[l]^b & { * + [ o][f-]{\makebox(4,4 ) { } } } { } \ar@<1ex>[r]^a \ar@<1ex>[l]^b & { * + [ o][f-]{\makebox(4,4 ) { } } } { } \ar@<1ex>[r]^a \ar@<1ex>[l]^b & \cdots \ar@<1ex>[l]^b } \ ] ] as another example , consider the lts on figure [ fig : non - bpa2 ] . this lts is not expressible in bpa modulo modulo bisimilarity see ( * ? ? ?
* , example ( c ) ) .
on the other hand , the synchronization tree associated with that lts is @xmath10-algebraic because it is the unique solution of the @xmath10-algebraic recursion scheme below .
@xmath369 the idea underlying the above definition is as follows . at any given vertex in the lts on figure [ fig : non - bpa2 ] , the argument @xmath370 of @xmath371 denotes the subtree obtained by unfolding the lts from the vertex _ and not _ taking the @xmath12-labelled or @xmath202-labelled edges as a first step .
the argument @xmath372 of @xmath371 instead encodes the number of @xmath198-labelled edges that one must perform in a sequence in order to terminate .
@xmath373 denotes the tree obtained by unfolding the lts from that vertex .
@xmath368 & { * + [ o][f-]{\makebox(4,4 ) { } } } { } \ar@<1ex>[r]^a \ar[d]^c & { * + [ o][f-]{\makebox(4,4){}}}{}\ar@<1ex>[l]^b \ar@<1ex>[r]^a \ar[d]^c & { * + [ o][f-]{\makebox(4,4 ) { } } } { } \ar@<1ex>[r]^a \ar@<1ex>[l]^b \ar[d]^c & \cdots \ar@<1ex>[l]^b \\ & { * + [ o][f=]{\makebox(4,4 ) { } } } { } & { * + [ o][f-]{\makebox(4,4 ) { } } } { } \ar[l]^b & { * + [ o][f-]{\makebox(4,4 ) { } } } { } \ar[l]^b & \ar[l]^b \cdots } \ ] ] in our arguments , we will often make use of the following mezei - wright theorem @xcite : [ thm - mezei ] suppose that @xmath374 is a morphism of continuous categorical @xmath20-algebras .
if @xmath202 is an algebraic ( or regular ) object of @xmath109 , then @xmath375 is an algebraic ( or regular ) object of @xmath169 .
moreover , for every algebraic object @xmath376 of @xmath169 there is an algebraic object @xmath202 of @xmath109 such that @xmath376 is isomorphic to @xmath375 , and similarly for regular objects .
an important subclass of continuous categorical algebras is formed by the continuous _ ordered _ algebras , which constitute the classical framework for algebraic semantics @xcite .
we say that a continuous categorical @xmath20-algebra @xmath38 is ordered if its underlying category is a poset category , so that there is at most one morphism between any two objects .
the objects of a continuous ordered algebra are usually called its elements and , as usual , the categorical structure is replaced by a partial order @xmath377 .
the assumptions that @xmath38 has initial object and colimits of @xmath52-diagrams correspond to the requirements that there is a least element @xmath72 and each @xmath52-chain @xmath378 has a supremum , denoted @xmath379 .
thus , @xmath38 is an _
@xmath52-complete poset _ as is each finite power @xmath47 of @xmath38 , equipped with the pointwise order .
a continuous functor @xmath380 is simply a function that preserves the suprema of @xmath52-chains .
every such continuous function preserves the partial order .
continuous functions are ordered pointwise . as is well known , if @xmath38 and @xmath42 are @xmath52-complete posets ,
then so is the poset @xmath381 $ ] of all continuous functions @xmath382 .
suppose that @xmath74 is an alphabet and recall the definition of the signatures @xmath10 and @xmath13 .
an example of a continuous ordered @xmath13-algebra is the language semiring @xmath383 of all subsets of @xmath181 , ordered by set inclusion , where each letter @xmath161 denotes the set @xmath384 and @xmath17 and @xmath1 are interpreted as the empty set and the set @xmath385 containing only the empty word @xmath386 , and where @xmath387 and @xmath388 is the concatenation of @xmath389 and @xmath390 , for all @xmath391 .
alternatively , we may view @xmath383 as a continuous ordered @xmath10-algebra with @xmath392 for all @xmath161 and @xmath393 .
the initial continuous @xmath20-algebra may be described as the algebra of _
@xmath20-term trees_. a term tree may be represented as ( the isomorphism class of ) a possibly infinite directed ordered tree whose vertices are labelled with the letters in @xmath20 such that a vertex labelled in @xmath394 has exactly @xmath5 successors . in particular , vertices labelled in @xmath395 are leaves .
finite term trees over @xmath20 may be identified with variable - free @xmath20-terms . as usual , we identify each symbol in @xmath395 with a term .
let @xmath396 denote the set of all @xmath20-term trees .
we equip @xmath396 with a partial order defined by @xmath397 iff @xmath15 can be constructed from @xmath16 by replacing some leaves labelled @xmath72 by some term trees in @xmath396 .
it is well - known ( see e.g. @xcite ) that this relation turns @xmath396 into an @xmath52-complete partial order with least element @xmath72 .
we may further turn @xmath396 into a continuous ordered @xmath20-algebra . for each @xmath40 and term trees
@xmath398 , the tree @xmath399 is defined as usual as the tree whose root is labelled @xmath62 which has @xmath5 subtrees isomorphic in order to @xmath398 .
the following fact is known , see e.g. @xcite .
[ prop - freeness ] for every continuous categorical @xmath20-algebra @xmath38 there is , up to natural isomorphism , a unique continuous categorical @xmath20-algebra morphism @xmath400 .
sometimes it is convenient to represent a term tree in @xmath396 as a partial function @xmath401 whose domain is a prefix closed nonempty subset of @xmath402 , where @xmath403 denotes the set of positive integers , such that whenever @xmath404 , then for each @xmath405 , @xmath406 is defined if and only if @xmath245 $ ] .
the operations are defined in the usual way . for trees
@xmath407 , we have @xmath397 if and only if for all @xmath408 , either @xmath409 or @xmath410 .
see @xcite for details .
yet another representation of a term tree in @xmath396 is by edge - labelled trees . to this end , let @xmath411 denote the ordinary alphabet whose letters are the symbols @xmath412 where @xmath413 , @xmath414 and @xmath245 $ ] .
when @xmath16 is a @xmath20-term tree , each vertex @xmath86 may be addressed " by a word @xmath415 which encodes the unique path from the root to that vertex .
the edge - labelled tree corresponding to @xmath16 has as its vertex sets the addresses of its vertices , together with a vertex @xmath416 whenever @xmath417 is the address of a leaf labelled @xmath62 .
the edges are given as follows .
suppose that @xmath86 and @xmath85 are vertices of @xmath16 with associated addresses @xmath417 and @xmath418 , respectively .
if @xmath85 is the @xmath292th successor of @xmath86 and @xmath86 is labelled @xmath40 ( so that @xmath414 and @xmath245 $ ] ) , then there is an edge from @xmath417 to @xmath418 labelled @xmath412 .
moreover , if @xmath86 is a leaf vertex of @xmath16 , labelled @xmath419 , then there is an edge labelled @xmath62 from @xmath417 to @xmath416 .
consider , for example , the following term tree over @xmath10 .
= [ circle , draw = black , fill = black , inner sep = 0.25 mm , outer sep= 0.5 mm ] = [ inner sep = 0.25 mm , outer sep= 0.5 mm ] = [ inner sep = 0 mm , outer sep= 0 mm ] = [ font=,label distance=-1 mm ] = [ level distance=0.85cm , sibling distance=1 cm ] = [ level distance=0.85cm , sibling distance=1.75 cm ] = [ level distance=0.85cm , sibling distance=1.75 cm ] child node[m , label=[l]left:@xmath420 @xmath12 child node[m , label=[l]left:@xmath421 @xmath17 child node[m , label=[l]right:@xmath422 @xmath141 child node[m , label=[l]right:@xmath423 @xmath12 child node[m , label=[l]left:@xmath424 @xmath17 child node[m , label=[l]right:@xmath425 @xmath12 child node[m , label=[l]right:@xmath426 @xmath1 ; the resulting edge - labelled tree is = [ circle , draw = black , fill = black , inner sep = 0.25 mm , outer sep= 0.5 mm ] = [ inner sep = 0.25 mm , outer sep= 0.5 mm ] = [ inner sep = 0 mm , outer sep= 0 mm ] = [ font=,label distance=-1 mm ] = [ level distance=1cm , sibling distance=5 cm ] = [ level distance=1cm , sibling distance=3 cm ] = [ level distance=1cm , sibling distance=2.5 cm ] child node[m ] @xmath427 child node[m ] @xmath428 child node[m ] @xmath429 edge from parent node[l , left ] @xmath17 edge from parent node[l , left ] @xmath430 edge from parent node[l , above , xshift=-1 mm ] @xmath427 child node[m ] @xmath431 child node[m ] @xmath432 child node[m ] @xmath433 child node[m ] @xmath434 edge from parent node[l , left ] @xmath17 edge from parent node[l , left ] @xmath430 edge from parent node[l , above , xshift=-1 mm ] @xmath427 child node[m ] @xmath435 child node[m ] @xmath436 child node[m ] @xmath437 edge from parent node[l , left ] @xmath1 edge from parent node[l , left ] @xmath430 edge from parent node[l , above , xshift=1 mm ] @xmath431 edge from parent node[l , above , xshift=1 mm ] @xmath431 ; suppose that @xmath109 is a continuous categorical @xmath20-algebra and @xmath154 is the essentially unique continuous categorical @xmath20-algebra morphism @xmath438 .
then , for each finite tree @xmath439 , @xmath440 is uniquely determined up to isomorphism , since @xmath154 preserves the operations . moreover , when @xmath441 are finite with also the unique morphism @xmath442 in @xmath396 seen as a category . ]
@xmath397 , @xmath443 is determined by the conditions that @xmath444 is initial and @xmath154 preserves the operations .
when @xmath16 is an infinite tree , @xmath16 is the supremum of an @xmath52-chain @xmath445 of finite trees
say @xmath108 is the approximation of @xmath16 obtained by removing all vertices of @xmath16 at distance greater than or equal to @xmath5 from the root ( and relabeling vertices at distance @xmath5 by @xmath72 ) .
then @xmath440 is the colimit of the @xmath52-diagram @xmath446 , where @xmath447 . when @xmath448 or @xmath449 , and @xmath450 , the morphisms @xmath123 are injective , and thus the colimit @xmath440 is the ` union ' of the @xmath451 . by proposition [ prop - freeness ]
, there is an ( essentially ) unique morphism of continuous categorical @xmath10-algebras @xmath452 , as well as an essentially unique continuous categorical @xmath13-algebra morphism @xmath453 . in the first part of this section
, we provide a combinatorial description of ( the object part ) of these morphisms . in the second part of the section
, we will consider morphisms from @xmath92 to @xmath383 , seen as a @xmath10-algebra or a @xmath13-algebra .
we start by describing the ( object part of the ) essentially unique continuous categorical @xmath13-algebra morphism @xmath453 .
we will call the image @xmath15 of a term tree @xmath16 under this morphism the _ synchronization tree denoted by _ @xmath16 , or the _ synchronization tree associated with
_ @xmath16 .
suppose that @xmath16 is a @xmath13-term tree and @xmath86 and @xmath85 are leaves of @xmath16 , labelled in @xmath454 .
let @xmath455 ( resp .
@xmath456 ) denote the sequence of vertices along the unique path from the root to @xmath86 ( resp .
@xmath85 ) , including @xmath86 ( resp .
@xmath85 ) .
we say that @xmath457 if @xmath455 and @xmath456 are of the form @xmath458 and @xmath459 with @xmath460 labelled by @xmath142 and successors @xmath461 , ordered as indicated , such that * every vertex appearing in @xmath462 which is different from @xmath86 is either labelled @xmath141 , or if it is labelled @xmath142 then its second successor belongs to @xmath462 , and * every vertex appearing in @xmath463 which is different from @xmath85 is either labelled @xmath141 , or if it is labelled @xmath142 then its first successor belongs to @xmath463 .
we say that @xmath464 if each vertex appearing in @xmath455 which is different from @xmath86 is either labelled @xmath141 , or if it is labelled @xmath142 , then its first successor belongs to @xmath455 .
finally , we say that @xmath465 if each vertex appearing in @xmath455 which is different from @xmath86 is either labelled @xmath141 , or if it is labelled @xmath142 , then its second successor appears in @xmath455 .
note that if @xmath466 then there is no @xmath460 such that @xmath467 , and similarly , if @xmath468 then @xmath469 .
now form the edge - labelled directed graph @xmath470 whose vertices are a new _ root vertex _ @xmath471 , the leaves of @xmath16 labelled in @xmath472 , and the _ exit vertex _ denoted @xmath473 .
the edges of @xmath470 are the following : * for each @xmath464 , there is an edge from the root @xmath471 to @xmath86 , labelled by the label of @xmath86 in @xmath16 .
* for all leaf vertices @xmath86 and @xmath85 of @xmath16 labelled in @xmath472 such that @xmath474 , there is an edge in @xmath470 from @xmath86 to @xmath85 whose label is the same as the label of @xmath85 in @xmath16 .
* whenever @xmath86 belongs to @xmath475 , there is an edge in @xmath470 labelled @xmath145 from @xmath86 to @xmath473 .
note that @xmath470 does not contain any cycle .
we denote by @xmath476 the synchronization tree obtained by unfolding @xmath470 from @xmath222 and then contracting edges labelled by @xmath1 . in proposition [ prop - tau ]
below , we will prove that @xmath477 , the synchronization tree denoted by @xmath16 .
however , before doing so , we find it instructive to give several examples . as a first example , consider the term @xmath478 as seen in figure [ fig : example - morphism-1 ] , the graph @xmath470 contains 8 vertices , the root @xmath471 , the exit vertex @xmath473 , and vertices @xmath479 corresponding to the @xmath480 leaves of @xmath16 .
there is an @xmath12-labelled edge from @xmath471 to @xmath370 , edges from @xmath370 to @xmath372 , @xmath481 and @xmath482 labelled @xmath1 , an edge from @xmath370 to @xmath483 labelled @xmath12 , and edges from @xmath481 and @xmath482 to @xmath484 labelled @xmath198 .
finally , there is an edge labelled @xmath145 from each of @xmath372 , @xmath483 and @xmath484 to @xmath473 .
it is clear that @xmath476 is the synchronization tree denoted by @xmath16 .
= [ circle , draw = black , fill = black , inner sep = 0.25 mm , outer sep= 0.5 mm ] = [ inner sep = 0.25 mm , outer sep= 0.5 mm ] = [ inner sep = 0 mm , outer sep= 0 mm ] = [ font=,label distance=-1 mm ] = [ level distance=0.85cm , sibling distance=1 cm ] = [ level distance=0.85cm , sibling distance=1.75 cm ] = [ level distance=0.85cm , sibling distance=1.75 cm ] child node[m , label=[l]below:@xmath485 @xmath12 child node[m ] @xmath141 child node[m ] @xmath141 child node[m , label=[l]below:@xmath486 @xmath1 child node[m , label=[l]below:@xmath487 @xmath12 child node[m ] @xmath142 child node[m ] @xmath141 child node[m , label=[l]below:@xmath488 @xmath1 child node[m , label=[l]below:@xmath489 @xmath1 child node[m , label=[l]below:@xmath490 @xmath198 ; ( v0 ) at ( 0,0 ) ; ( v1 ) at ( 0,-0.85 cm ) ; ( v2 ) at ( -1.2cm,-1.7 cm ) ; ( v3 ) at ( -0.4cm,-1.7 cm ) ; ( v4 ) at ( 0.4cm,-1.7 cm ) ; ( v5 ) at ( 1.2cm,-1.7 cm ) ; ( v6 ) at ( 0.8cm,-2.55 cm ) ; ( etoile ) at ( 0,-3.4 cm ) ; ( v0 ) edge[-latex ] node[left ] @xmath12 ( v1 ) ( v1 ) edge[-latex ] node[above left ] @xmath1 ( v2 ) ( v1 ) edge[-latex ] node[left ] @xmath12 ( v3 ) ( v1 ) edge[-latex ] node[right ] @xmath1 ( v4 ) ( v1 ) edge[-latex ] node[above right ] @xmath1 ( v5 ) ( v4 ) edge[-latex ] node[left ] @xmath198 ( v6 ) ( v5 ) edge[-latex ] node[right ] @xmath198 ( v6 ) ( v2 ) edge[-latex , bend right=20 ] node[left ] @xmath145 ( etoile ) ( v3 ) edge[-latex , bend right=10 ] node[left ] @xmath145 ( etoile ) ( v6 ) edge[-latex , bend left=10 ] node[below right ] @xmath145 ( etoile ) ; = [ level distance=0.85cm , sibling distance=0.85 cm ] = [ level distance=0.85cm , sibling distance=0.85 cm ] child node[n ] edge from parent node[draw = none , left ] @xmath145 child node[n ] child node[n ] edge from parent node[draw = none , left ] @xmath145 edge from parent node[draw = none , left ] @xmath12 child node[n ] child node[n ] edge from parent node[draw = none , right ] @xmath145 edge from parent node[draw = none , right ] @xmath198 child node[n ] child node[n ] edge from parent node[draw = none , right ] @xmath145 edge from parent node[draw = none , right ] @xmath198 ; in the second example illustrated in figure [ fig : example - morphism-2 ] , consider the term tree @xmath16 defined by the scheme @xmath491 now the root @xmath222 of @xmath470 has two outgoing edges leading to different vertices @xmath492 and @xmath493 , say . these edges are labelled @xmath12 and @xmath376 , respectively . in addition , there is a sequence of vertices , call them @xmath494 , such that for each @xmath495 there is an edge from @xmath496 to @xmath291 , which is labelled @xmath1 , and there is a @xmath202-labelled edge from @xmath372 to @xmath370 .
last , @xmath473 is a vertex , and there exist edges from @xmath493 and @xmath370 to @xmath473 labelled @xmath145 .
@xmath470 is infinite , but the tree @xmath476 constructed from it is finite since @xmath497 is the synchronization tree denoted by @xmath16 .
= [ circle , draw = black , fill = black , inner sep = 0.25 mm , outer sep= 0.5 mm ] = [ inner sep = 0.25 mm , outer sep= 0.5 mm ] = [ inner sep = 0 mm , outer sep= 0 mm ] = [ font=,label distance=-1 mm ] = [ level distance=0.85cm , sibling distance=1 cm ] child node @xmath142 child node @xmath142 child node[label=[l]below:@xmath498 @xmath12 child node @xmath142 child node @xmath142 child node edge from parent[dashed,- ] child node[label=[l]right:@xmath487 @xmath1 child node[label=[l]right:@xmath486 @xmath1 child node[label=[l]right:@xmath485 @xmath202 child node[label=[l]below:@xmath499 @xmath376 ; ( v0 ) at ( 0,0 ) ; ( u1 ) at ( -0.75cm,-0.85 cm ) ; ( u2 ) at ( 0.75cm,-0.85 cm ) ; ( v1 ) at ( 0cm,-0.85 cm ) ; ( v2 ) at ( 0cm,-1.7 cm ) ; ( v3 ) at ( 0cm,-2.55 cm ) ; ( v4 ) at ( 0cm,-3.4 cm ) ; ( etoile ) at ( 0.75cm,-1.7 cm ) ; ( v0 ) edge[-latex ] node[left ] @xmath12 ( u1 ) ( v0 ) edge[-latex ] node[right ] @xmath376 ( u2 ) ( v2 ) edge[-latex ] node[right ] @xmath202 ( v1 ) ( v3 ) edge[-latex ] node[right ] @xmath1 ( v2 ) ( v4 ) edge[dashed ] ( v3 ) ( v1 ) edge[-latex ] node[above right , xshift=-1 mm ] @xmath145 ( etoile ) ( u2 ) edge[-latex ] node[right ] @xmath145 ( etoile ) ; = [ level distance=0.85cm , sibling distance=1.5 cm ] child node[n ] edge from parent node[draw = none , left ] @xmath12 child node[n ] child node[n ] edge from parent node[draw = none , right ] @xmath145 edge from parent node[draw = none , right ] @xmath376 ; in the last example illustrated in figure [ fig : example - morphism-3 ] , consider the term @xmath500 .
the vertices of @xmath470 are the root @xmath471 , vertices @xmath501 corresponding respectively to the leaves of @xmath16 with nonzero label , and the exit vertex @xmath473 .
there are @xmath502 edges , an edge labelled @xmath1 from @xmath471 to @xmath370 , an edge labelled @xmath1 from @xmath372 to @xmath483 , and an edge labelled @xmath145 from @xmath483 to @xmath473 .
now @xmath476 contains only the root and no edges , so that @xmath476 is the synchronization tree @xmath17 denoted by @xmath16 .
= [ circle , draw = black , fill = black , inner sep = 0.25 mm , outer sep= 0.5 mm ] = [ inner sep = 0.25 mm , outer sep= 0.5 mm ] = [ inner sep = 0 mm , outer sep= 0 mm ] = [ font=,label distance=-1 mm ] = [ level distance=0.85cm , sibling distance=1 cm ] child node [ ] @xmath142 child node[label=[l]left:@xmath485 @xmath1 child node [ ] @xmath142 child node [ ] @xmath17 child node[label=[l]below:@xmath486 @xmath198 child node[label=[l]right:@xmath487 @xmath1 ; ( v0 ) at ( 0,0 ) ; ( v1 ) at ( -1.35cm,-0.85 cm ) ; ( v2 ) at ( -0.45cm,-0.85 cm ) ; ( v3 ) at ( 0.45cm,-0.85 cm ) ; ( etoile ) at ( 1.35cm,-0.85 cm ) ; ( v0 ) edge[-latex ] node[above left ] @xmath1 ( v1 ) ( v2 ) edge[-latex ] node[above ] @xmath198 ( v3 ) ( v3 ) edge[-latex ] node[above ] @xmath145 ( etoile ) ; we still need to prove that the definition of @xmath476 is correct .
[ prop - tau ] [ prop : constructing - delta - trees ] for every @xmath13-term tree @xmath16 , the image @xmath15 of @xmath16 with respect to the essentially unique continuous @xmath13-algebra morphism @xmath453 is @xmath476 .
suppose first that @xmath16 is finite .
we will prove the claim by induction on the structure of @xmath16 . if @xmath503 , then @xmath470 has two vertices , the root and the exit vertex , and no edges .
if @xmath16 is a symbol in @xmath454 , then @xmath470 has three vertices , the root @xmath471 , a vertex @xmath370 and the vertex @xmath473 .
there is an edge from @xmath471 to @xmath370 and an edge from @xmath370 to @xmath473 .
the first edge is labelled @xmath12 if @xmath504 , and @xmath1 if @xmath505 .
the second edge is labelled @xmath145 . in either case , @xmath506 . in the induction step
, first suppose that @xmath507 for some terms @xmath166 denoting the synchronization trees @xmath508 and @xmath509 , respectively .
then @xmath470 is isomorphic to the disjoint union of @xmath510 and @xmath511 with roots and exit vertices merged .
thus , @xmath512 .
suppose next that @xmath513 with @xmath137 denoting @xmath508 and @xmath138 denoting @xmath509 .
then @xmath470 can be constructed from @xmath510 and @xmath511 as follows .
for all edges from the root @xmath372 of @xmath511 to a vertex @xmath85 of @xmath511 ( which necessarily corresponds to a leaf vertex of @xmath138 in @xmath514 ) , and for all @xmath515 , add a new edge from @xmath86 to @xmath85 labelled by the symbol which is the label of the edge from @xmath372 to @xmath85 in @xmath511 ( i.e. , the label of @xmath85 in @xmath138 ) . then remove the edge from @xmath86 to the exit vertex of @xmath510 . finally remove all edges originating in the root of @xmath511 .
the vertices of @xmath470 are the non - exit vertices of @xmath510 and the non - root vertices of @xmath511
. it should be clear that @xmath476 is the sequential product of @xmath516 and thus @xmath517 by the induction hypothesis .
suppose now that @xmath16 is infinite . for each @xmath95 ,
let @xmath108 denote the approximation of @xmath16 obtained by relabelling each vertex of @xmath16 of depth @xmath5 by @xmath17 and removing all vertices of depth greater than @xmath5 . for each leaf vertex @xmath86 of @xmath16
there is some @xmath518 such that @xmath86 is a vertex of @xmath108 with the same label for all @xmath519 .
moreover , for any two leaf vertices @xmath520 of @xmath16 with a nonzero label , @xmath457 iff there is some @xmath518 such that @xmath521 for all @xmath519 .
similarly , for each leaf @xmath86 of @xmath16 with a nonzero label , we have @xmath464 ( @xmath468 ) iff there is some @xmath518 such that @xmath522 ( @xmath523 , resp . ) for all @xmath519 .
this implies that @xmath470 is the union ( colimit ) of the @xmath524 and then it follows that @xmath476 is also the union ( colimit ) of the @xmath525 .
we conclude that @xmath526 , where for each @xmath5 , @xmath527 is the synchronization tree denoted by @xmath108 .
next we describe the essentially unique morphism of continuous categorical @xmath10-algebras @xmath528 .
fix a @xmath10-term tree @xmath529 .
we define a synchronization tree , denoted @xmath530 , which is essentially a representation of @xmath16 as a synchronization tree in which leaves labelled by @xmath1 ( in @xmath16 ) are added a dangling outgoing edge labelled by @xmath145 .
the vertices of the synchronization tree @xmath530 are the vertices of @xmath16 together with a fresh vertex @xmath531 for each leaf @xmath86 labelled by @xmath1 .
if @xmath86 is labelled by @xmath141 in @xmath16 then there is in @xmath530 an edge labelled by @xmath532 from @xmath86 to its first successor and an edge labelled by @xmath533 from @xmath86 to its second successor .
if @xmath86 is labelled by @xmath143 in @xmath16 , then there is in @xmath530 an edge labelled by @xmath12 from @xmath86 to its unique successor .
finally , for each leaf of @xmath86 labelled by @xmath1 , there is an edge in @xmath530 from @xmath86 to @xmath531 labelled by @xmath145 .
the synchronization tree defined by @xmath16 is the @xmath534-contraction of @xmath530 .
this construction is illustrated for the @xmath10-term @xmath535 in figure [ fig : example - morphism-4 ] .
= [ circle , draw = black , fill = black , inner sep = 0.25 mm , outer sep= 0.5 mm ] = [ inner sep = 0.25 mm , outer sep= 0.5 mm ] = [ inner sep = 0 mm , outer sep= 0 mm ] = [ font=,label distance=-1 mm ] = [ level distance=0.85cm , sibling distance=1 cm ] child node[label=[l]left:@xmath485 @xmath12 child node[label=[l]left:@xmath487 @xmath141 child node[label=[l]left:@xmath488 @xmath202 child node[label=[l]left:@xmath490 @xmath17 child node[label=[l]right:@xmath489 @xmath376 child node[label=[l]right:@xmath536 @xmath1 child node[label=[l]right:@xmath486 @xmath1 ; child node[n , label=[l]left:@xmath485 child node[n , label=[l]left:@xmath487 child node[n , label=[l]left:@xmath488 child node[n , label=[l]left:@xmath490 edge from parent node[draw = none , left ] @xmath202 edge from parent node[draw = none , left ] @xmath427 child node[n , label=[l]right:@xmath489 child node[n , label=[l]right:@xmath536 child node[n , label=[l]below:@xmath537 edge from parent node[draw = none , right ] @xmath145 edge from parent node[draw = none , right ] @xmath376 edge from parent node[draw = none , right ] @xmath431 edge from parent node[draw = none , left ] @xmath12 edge from parent node[draw = none , above left ] @xmath427 child node[n , label=[l]right:@xmath486 child node[n , label=[l]below:@xmath538 edge from parent node[draw = none , right ] @xmath145 edge from parent node[draw = none , above right ] @xmath431 ; child node[n , label=[l]left:@xmath487 child node[n , label=[l]left:@xmath490 edge from parent node[draw = none , left ] @xmath202 child node[n , label=[l]right:@xmath536 child node[n , label=[l]below:@xmath537 edge from parent node[draw = none , right ] @xmath145 edge from parent node[draw = none , right ] @xmath376 edge from parent node[draw = none , left ] @xmath12 child node[n , label=[l]below:@xmath538 edge from parent node[draw = none , right ] @xmath145 ; let @xmath539 denote the function that maps @xmath16 to the @xmath534-contraction of @xmath530 .
[ prop - gamma - hom ] for every @xmath540 , @xmath541 is the synchronization tree in @xmath92 denoted by @xmath16 .
we omit the proof which is essentially a simplification of that of proposition [ prop - tau ] .
the function which maps a synchronization tree @xmath542 to its path language in @xmath383 is the ( object part ) of a categorical @xmath10-algebra , as well as @xmath13-algebra , morphism @xmath543 .
it is clear that the empty tree is mapped to the empty language and the operations are preserved . when there is a morphism @xmath544 for synchronization trees @xmath545 , then the path language of @xmath16 is included in the path language of @xmath15 .
finally , suppose that @xmath546 is an @xmath52-diagram in @xmath92 with colimit @xmath547 .
then for every branch of @xmath16 ending in an edge labelled @xmath145 there is some @xmath518 such that the branch is the image of a corresponding branch of @xmath548 with respect to the morphism @xmath549 , and then the same holds for each @xmath519 .
using this fact , it follows easily that the path language of @xmath16 is the union of the path languages of the @xmath108 , proving that colimits of @xmath52-diagrams are preserved . in this section , we give some basic properties of synchronization trees defined by first - order recursion schemes .
first , we remark that @xmath10-regular and @xmath10-algebraic functors are closed under the sequential product .
this closure property is immediate for @xmath13-regular and @xmath13-algebraic functors as the sequential product is an operation of the continuous @xmath13-algebra @xmath92 ( which is not the case for the continuous @xmath10-algebra @xmath92 ) . if @xmath550 are @xmath10-algebraic ( resp .
@xmath10-regular ) , then so is @xmath551 .
suppose that @xmath335 and @xmath552 are the first components of the initial solutions over @xmath92 of the @xmath10-recursion schemes @xmath209 and @xmath187 .
without loss of generality , we may assume that @xmath209 and @xmath187 have disjoint sets of functor variables .
let @xmath553 and @xmath554 denote the left - hand sides of the first equations of @xmath209 and @xmath187 . then replace each occurrence of the symbol @xmath1 on the right - hand - side term of each equation of @xmath209 with @xmath555 , and consider the recursion scheme consisting of these equations together with the equations of @xmath187 .
the component of the initial solution of this scheme which corresponds to @xmath556 is @xmath557 .
[ ex : seqcomp ] the tree @xmath558 determined by single infinite branch with edges labelled @xmath198 is @xmath10-regular since it is the initial solution of the following @xmath10-regular recursion scheme @xmath559 we have already seen in example [ ex : anyfinitedepth ] on page that the tree @xmath560 is @xmath10-algebraic . according to the above proposition , the tree @xmath561 is also @xmath10-algebraic .
indeed , it is the initial solution of the recursion scheme @xmath562 all our classes of synchronization trees are also closed under the contraction operation for each non - empty subset of @xmath74 . [
prop : contr - alg ] the classes of @xmath10-regular , @xmath10-algebraic , @xmath13-regular and @xmath13-algebraic synchronization trees in @xmath92 are closed under contraction for any non - empty subset of @xmath74 .
let @xmath183 be a non - empty subset of @xmath74 .
let @xmath16 be a synchronization tree defined by an algebraic @xmath10-scheme @xmath209 and let @xmath15 be the synchronization tree obtained by contracting @xmath16 with respect to @xmath183 .
a @xmath10-scheme @xmath187 defining @xmath15 is easily obtained from @xmath209 by replacing each equation of the form @xmath563 by the equation @xmath564 where @xmath565 is inductively defined by @xmath566 , @xmath567 , @xmath568 , @xmath569 , @xmath570 if @xmath571 and @xmath572 for @xmath573 .
note that @xmath187 is @xmath10-regular , if so is @xmath209 .
the @xmath10-term defined by @xmath187 is the contraction of the @xmath10-term defined by @xmath209 from its root with respect to @xmath183 .
it then follows , from the definition of the mapping @xmath539 in [ sec : morphism ] and proposition [ prop - gamma - hom ] , that the synchronization tree defined by @xmath187 is the contraction of the synchronization tree defined by @xmath209 with respect to @xmath183 .
let @xmath16 be a synchronization tree defined by an algebraic @xmath13-scheme @xmath209 and let @xmath15 be the synchronization tree obtained by contracting @xmath16 with respect to @xmath183 . consider the @xmath13-scheme @xmath187 obtained by replacing in @xmath209 all occurrences of a constant in @xmath183 by the constant @xmath1 .
( note that @xmath187 is @xmath13-regular , if so is @xmath209 . )
the @xmath13-term defined by @xmath187 is the @xmath13-term defined by @xmath209 in which each occurrence of a constant in @xmath183 is replaced by the constant @xmath1 .
it follows from the definition of the mapping @xmath574 in [ sec : morphism ] and proposition [ prop - tau ] , that the synchronization tree defined by @xmath187 is the contraction of the synchronization tree defined by @xmath209 with respect to @xmath183 .
finally , the @xmath10-regular synchronizations trees can be characterized by a syntactic subfamily of the @xmath13-regular recursion schemes .
this characterization is similar in spirit to the characterization of regular languages by right - linear context - free grammars .
a @xmath13-regular scheme is said to be _ right - linear _ if the right - hand side of each equation is of the form @xmath575}t_i\cdot g_i\ ] ] up to commutativity and associativity of sum , where the @xmath576 , @xmath577 $ ] , are ( constant ) functor variables and the @xmath578 , @xmath579 are terms over the signature @xmath13 not containing variables .
( the empty sum stands for @xmath17 . )
it is rather standard to prove the following fact : [ prop : right - linear - implies - regular ] a synchronization tree in @xmath92 is @xmath10-regular iff it can be defined by a right - linear @xmath13-regular scheme .
it is clear how to transform a @xmath10-regular recursion scheme into a right - linear @xmath13-scheme by turning each prefixing operation into a sequential product .
we show how to transform a @xmath13-term @xmath580 with @xmath16 containing no functor variables into a corresponding @xmath10-term @xmath581 , possibly containing @xmath212 .
we proceed by induction on the structure of @xmath16 .
when @xmath503 , let @xmath582 , and when @xmath505 let @xmath583 . when @xmath584 where @xmath143 , define @xmath585 .
suppose now that @xmath507 .
then let @xmath586 .
finally , consider the case when @xmath587 . in this case
define @xmath581 as the term obtained by substituting @xmath588 for each occurrence of @xmath212 in @xmath589 .
a routine calculation shows that for each evaluation of @xmath212 in @xmath92 , the terms @xmath580 and @xmath581 yield the same synchronization tree .
for an example of such calculation , we refer the reader to the proof of theorem [ thm : delta2gamma ] on page .
using this fact , we may transform a right - linear @xmath13-scheme into a regular @xmath10-scheme by changing the right - hand side @xmath590}t_i\cdot g_i$ ] of each equation to @xmath591}t_i ' ( g_i)$ ] , where @xmath592 is the term obtained by substituting @xmath1 for @xmath212 in @xmath593 .
in this section , we interpret recursion schemes over the continuous categorical algebra @xmath92 , viewed either as a @xmath10-algebra or a @xmath13-algebra . we compare the resulting classes of synchronization trees with respect to language equivalence , bisimulation equivalence and isomorphism equivalence .
first for language equivalence , we show in section [ ssec : up - to - languages ] that the following hierarchy holds .
@xmath594 up to bisimulation or isomorphism , we show in section [ ssec : up - to - bisimulation - iso ] that the following hierarchy holds .
@xmath595 we conclude the section by a comparison with the classes of synchronization trees defined by bpa and bpp .
as already mentioned in the introduction , the path languages of the different classes can be characterized as follows .
the following properties holds . 1
. the path languages of the @xmath10-regular trees are the regular languages .
the path languages of the @xmath13-regular trees and of the @xmath10-algebraic trees are the context - free languages
the path languages of the @xmath13-algebraic languages are the indexed languages . by the mezei - wright theorem , the path languages of the @xmath10-regular trees and the @xmath13-regular trees in @xmath92 are just the @xmath10-regular and @xmath13-regular elements ( objects ) of @xmath383 , seen as a continuous @xmath10-algebra or @xmath13-algebra . by classic results ( e.g. ( * ? ? ?
* theorem 1.21 , page 116 ) ) , these in turn are the regular and context - free languages similarly , the indexed languages ( or oi - macro languages ) are exactly the @xmath13-algebraic elements of @xmath383 , cf .
@xcite , i.e. , the path languages of the @xmath13-algebraic trees by the mezei - wright theorem . by theorem [ thm : delta2gamma ] , every @xmath13-regular tree is @xmath10-algebraic .
thus , to complete the proof , it remains to show that the path language of a @xmath10-algebraic tree is context - free .
suppose that @xmath214 is the path language of a @xmath10-algebraic tree in @xmath92 , and let @xmath209 denote a @xmath10-algebraic scheme defining it .
then consider the term tree @xmath529 defined by @xmath209 . by the mezei - wright theorem ,
@xmath214 is the image of @xmath16 with respect to the unique continuous @xmath10-algebra morphism @xmath596 .
suppose that @xmath85 is a leaf of @xmath16 labelled @xmath1 , and consider the branch of @xmath16 from the root to @xmath85 .
the label of this branch is a word @xmath597 over the alphabet @xmath598 .
let us consider the image @xmath599 of @xmath597 under the erasing morphism @xmath600 that removes the letters @xmath427 and @xmath431 .
then @xmath601 is the set of all such words @xmath599 obtained by considering all leaves @xmath85 of @xmath16 labelled @xmath1 .
it follows from courcelle s characterization of the algebraic term trees by deterministic context - free languages ( * ? ? ?
* theorem 5.5.1,page 157 ) ( see also @xcite ) that the set of all words @xmath597 , where @xmath85 is a leaf of @xmath16 labelled @xmath1 , is a deterministic context - free language .
since the image of a ( deterministic ) context - free language with respect to a homomorphism is context - free , we conclude that @xmath214 is a context - free language .
the aim of this section is to establish the strict inclusions stated in equation .
as noted in remark [ rem : delta - implies - gamma ] on page , a @xmath10-term can be transformed into an equivalent @xmath13-term by replacing for each letter @xmath143 , all occurrences of a subterm @xmath149 by @xmath150 . in particular , every @xmath10-regular tree is @xmath13-regular and every @xmath10-algebraic tree is @xmath13-algebraic .
this establishes the first and third inclusions of .
these inclusions are strict up to bisimulation and up to isomorphism as they are already strict with respect to language equivalence ( cf . ) .
it only remains to establish the second strict inclusion of equation .
first we establish the inclusion up to isomorphism in theorem [ thm : delta2gamma ] . in corollary
[ cor : charac - delta - regular - in - gamma - algebraic ] , we characterize a syntactical subfamily of the @xmath10-algebraic schemes which captures exactly the @xmath13-regular schemes .
[ thm : delta2gamma ] every @xmath13-regular tree is @xmath10-algebraic .
consider a regular @xmath13-recursion scheme @xmath209 , @xmath602 which defines the @xmath13-regular tree @xmath603 .
let @xmath604 and @xmath605 , where each @xmath273 is of rank @xmath17 , @xmath606 is of rank @xmath17 , and each @xmath576 with @xmath607 is of rank @xmath1 . given a variable - free @xmath608-term @xmath16 , we define its translation @xmath15 to be a @xmath609-term in the variable @xmath370
. * if @xmath610 , for some @xmath245 $ ] , then @xmath611 . *
if @xmath503 then @xmath612 .
* if @xmath505 then @xmath613 . * if @xmath584 then @xmath614 .
* if @xmath507 then @xmath615 .
* if @xmath616 then @xmath617 , the term obtained by substituting @xmath509 for each occurrence of @xmath370 in @xmath508 . note that @xmath618 and @xmath619^n { \rightarrow}[{\mathsf{st}(a)}{\rightarrow}{\mathsf{st}(a)}]\ ] ] the two functors are related . for a tree @xmath620 ,
let @xmath621 denote the functor ` left composition with @xmath230 ' , @xmath622 in @xmath623 $ ] .
_ claim 1 .
_ for each variable - free @xmath624-term @xmath16 and its translation @xmath15 , and for each sequence of trees @xmath625 in @xmath92 , it holds that @xmath626 indeed , when @xmath627 , for some @xmath245 $ ] , then @xmath15 is @xmath628 and both sides are equal to the functor @xmath629 .
if @xmath630 , then both sides are equal to the constant functor @xmath631 determined by the tree @xmath632 , and when @xmath505 , both sides are equal to the identity functor @xmath631 . indeed , seen @xmath633 , left composition with @xmath634 is the identity functor as is @xmath635 .
suppose now that @xmath584 for some @xmath12 .
then both sides are equal to the functor @xmath636 , left composition with @xmath12 .
next let @xmath507 , and suppose that the claim holds for @xmath137 and @xmath138 .
then @xmath637 last , suppose that @xmath616 , and that the claim holds for both terms @xmath137 and @xmath138 .
then @xmath638 is the composition of the functors @xmath639 and @xmath640 ( where the second functor is applied first ) , as is the functor @xmath641 _ claim 2 .
_ for each variable - free @xmath624-term @xmath16 and its translation @xmath15 , and for each sequence of trees @xmath625 in @xmath92 , it holds that @xmath642 indeed , by claim 1 , we have @xmath643 let @xmath187 denote the @xmath10-algebraic scheme @xmath644 we claim that @xmath187 is equivalent to @xmath209 , i.e. , @xmath187 also defines @xmath156 . to this end , let @xmath212 denote the recursion scheme consisting of the last @xmath5 equations of @xmath187 .
in order to compare the schemes @xmath209 and @xmath212 , define @xmath645},\ldots,0^{[{\mathsf{st}(a)}{\rightarrow}{\mathsf{st}(a)}]})\\ g_{i+1 } & = & ( g_{i+1,1},\ldots , g_{i+1,n } ) = ( t_1'^{{\mathsf{st}(a)}}(g_i),\ldots , t_n'^{{\mathsf{st}(a)}}(g_i ) ) \end{aligned}\ ] ] for each @xmath292 , define @xmath646 .
we prove by induction on @xmath292 that @xmath647 .
this is clear when @xmath648 , since for each @xmath320 $ ] , @xmath649 } = \overline{0^{{\mathsf{st}(a ) } } } = \overline{s_{0,j}}$ ] . to prove the induction step ,
suppose that we have established our claim for some @xmath650 .
then for all @xmath327 $ ] , @xmath651 for each @xmath327 $ ] , let @xmath652 denote the unique morphism @xmath653 . then define @xmath654 next , let @xmath655 denote the unique natural transformation @xmath656 } { \rightarrow}g_{1,j}$ ] , for each @xmath327 $ ] .
define @xmath657}(\psi_i),\ldots , t_n'^{[{\mathsf{st}(a)}{\rightarrow}{\mathsf{st}(a)}]}(\psi_i))\end{aligned}\ ] ] thus , each @xmath658 is a natural transformation from @xmath659 to @xmath660 , and each @xmath329 is a morphism @xmath661 .
define @xmath662 to be the natural transformation @xmath663 such that for any tree @xmath335 , the corresponding component of @xmath662 is @xmath664 .
let @xmath665 . _ claim 3 .
_ for each @xmath292 , it holds that @xmath666 .
the proof is similar to the above argument .
using this claim , it follows that @xmath667 for each @xmath668 , since @xmath669 it is now easy to complete the proof . by the beki identity
, @xmath670 suppose that @xmath209 is given by the single equation @xmath671 then @xmath187 is @xmath672 both of them define the tree depicted on figure [ fig : anbn ] .
@xmath673[f-]{\makebox(3,3 ) { } } } { } \ar[d]_{{\mathsf{ex } } } \ar[r]_a & { * + [ o][f-]{\makebox(3,3 ) { } } } { } \ar[d]_b \ar[r]_a & { * + [ o][f-]{\makebox(3,3 ) { } } } { } \ar[d]_b \ar[r]_a & { * + [ o][f-]{\makebox(3,3 ) { } } } { } \ar[d]_b \ar[r]_a & \cdots \\ { * + [ o][f-]{\makebox(3,3 ) { } } } { } & { * + [ o][f-]{\makebox(3,3 ) { } } } { } \ar[d]_{{\mathsf{ex } } } & { * + [ o][f-]{\makebox(3,3 ) { } } } { } \ar[d]_b & { * + [ o][f-]{\makebox(3,3){}}}{}\ar[d]_b & \cdots \\ & { * + [ o][f-]{\makebox(3,3 ) { } } } { } & { * + [ o][f-]{\makebox(3,3 ) { } } } { } \ar[d]_{{\mathsf{ex } } } & { *
+ [ o][f-]{\makebox(3,3 ) { } } } { } \ar[d]_b & \cdots \\ & & { * + [ o][f-]{\makebox(3,3 ) { } } } { } & { * + [ o][f-]{\makebox(3,3 ) { } } } { } \ar[d]_{{\mathsf{ex } } } & \cdots \\ & & & { * + [ o][f-]{\makebox(3,3 ) { } } } { } & \cdots \\ } \ ] ] the translation given in the proof of theorem [ thm : delta2gamma ] allows us to characterize @xmath13-regular trees by a syntactic restriction on @xmath10-algebraic schemes .
[ cor : charac - delta - regular - in - gamma - algebraic ] a tree is @xmath13-regular if and only if it is defined by an algebraic recursion scheme @xmath212 over @xmath10 of the form @xmath674 where @xmath606 has rank @xmath17 , each @xmath576 with @xmath607 has rank @xmath1 , and where none of the terms @xmath675 has an occurrence of the constant @xmath1 . in the light of the proof of theorem [ thm : delta2gamma
] , it is enough to show that any such scheme can be obtained from some regular recursion scheme @xmath209 over @xmath13 .
let @xmath676 and @xmath604 , where each @xmath273 has rank @xmath17 .
we give a transformation of a @xmath677-term @xmath455 in the variable @xmath370 , which contains no occurrence of the constant @xmath1 , into a variable - free @xmath678-term @xmath679 such that @xmath680 , where @xmath681 is defined in the proof of theorem [ thm : delta2gamma ] . 1 .
if @xmath682 then @xmath683 .
2 . if @xmath684 then @xmath685 .
3 . if @xmath686 then @xmath687 .
4 . if @xmath688 then @xmath689 . 5 . if @xmath690 then @xmath691 . _ claim .
_ let @xmath455 be a @xmath677-term in the variable @xmath370 containing no occurrence of the constant @xmath1 .
then @xmath680 .
we prove this claim by induction on the structure of @xmath455 . when @xmath692 , @xmath683 and @xmath693 , and when @xmath684 , @xmath685 and @xmath694 .
suppose that @xmath455 is of the form @xmath695 and that the claim holds for @xmath696 and @xmath697 . in this case
@xmath698 and @xmath699 , by the induction hypothesis .
next let @xmath688 where @xmath143 and suppose that the claim holds for @xmath696
. then @xmath700 and @xmath701 .
last , suppose that @xmath690 for some @xmath245 $ ] and @xmath696 satisfying the claim
. then we have @xmath702 and @xmath703 .
this completes the proof of the claim .
now consider the regular @xmath13-scheme @xmath209 @xmath704 by the proof of theorem [ thm : delta2gamma ] , @xmath212 corresponds to @xmath209 .
thus , @xmath209 and @xmath212 define the same tree .
the restriction on the use of the constant @xmath1 in corollary [ cor : charac - delta - regular - in - gamma - algebraic ] is necessary . indeed
the proof of proposition [ prop : gamma - algebraic - not - delta - regular ] to follow gives an example of @xmath10-algebraic scheme of rank 1 generating a tree that is not @xmath13-regular even up to bisimulation equivalence . to establish equation ,
it remains to show the strictness of the second inclusion up to bisimulation equivalence .
[ prop : gamma - algebraic - not - delta - regular ] there exists a @xmath10-algebraic synchronization tree that is not bisimilar to any @xmath13-regular tree .
let @xmath705 , and consider the synchronization tree @xmath706 , defined by the @xmath10-algebraic recursion scheme : @xmath707 the tree @xmath708 , depicted in figure [ fig : tree - strict ] , has a single infinite branch whose edges are labelled @xmath12 , and the out - degree of each vertex on this branch is @xmath502 , since each such vertex is the source of an edge labelled @xmath12 , an edge labelled @xmath198 , and an edge labelled @xmath145 . since @xmath708 is deterministic , its vertices may be identified with the words in the prefix closed language @xmath709 a key feature is that every vertex is the source of an edge labelled @xmath145 .
= [ circle , draw = black , fill = black , inner sep = 0.25 mm , outer sep= 0.5 mm ] = [ inner sep = 0 mm , outer sep= 0 mm ] = [ font=,label distance=-1 mm ] = [ level distance=0.95cm , sibling distance=1.75 cm ] child node[n , label=[l]above left:@xmath710 child node[n , label=[l]above left:@xmath711 child node[e ] edge from parent[dashed,- ] child node[n , label=[l]left:@xmath712 child node[e ] edge from parent[draw = none ] child node[n , label=[l]left:@xmath713 child node[e ] edge from parent[draw = none ] child node[n , label=[l]left:@xmath714 child[sibling distance=0.75 cm ] node[n ] edge from parent node[draw = none , right ] @xmath145 edge from parent node[draw = none , left ] @xmath198 child[sibling distance=0.75 cm ] node[n ] edge from parent node[draw = none , right ] @xmath145 edge from parent node[draw = none , left ] @xmath198 child[sibling distance=0.75 cm ] node[n ] edge from parent node[draw = none , right ] @xmath145 edge from parent node[draw = none , left ] @xmath198 child[sibling distance=0.75 cm ] node[n ] edge from parent node[draw = none , right ] @xmath145 edge from parent node[draw = none , above left ] @xmath12 child node[n , label=[l]left:@xmath715 child node[e ] edge from parent[draw = none ] child node[n , label=[l]left:@xmath716 child[sibling distance=0.75 cm ] node[n ] edge from parent node[draw = none , right ] @xmath145 edge from parent node[draw = none , left ] @xmath198
child[sibling distance=0.75 cm ] node[n ] edge from parent node[draw = none , right ] @xmath145 edge from parent node[draw = none , left ] @xmath198 child[sibling distance=0.75 cm ] node[n ] edge from parent node[draw = none , right ] @xmath145 edge from parent node[draw = none , above left ] @xmath12 child node[n , label=[l]left:@xmath717 child[sibling distance=0.75 cm ] node[n ] edge from parent node[draw = none , right ] @xmath145 edge from parent node[draw = none , left ] @xmath198 child[sibling distance=0.75 cm ] node[n ] edge from parent node[draw = none , right ] @xmath145 ; we are going to show that every @xmath13-regular scheme @xmath209 defining a synchronization tree bisimilar to @xmath708 is equivalent to a right - linear one , modulo bisimilarity . by proposition [ prop : right - linear - implies - regular ] , this would imply that @xmath708 is @xmath10-regular , which yields a contradiction .
indeed , the tree @xmath708 has a countably infinite set of subtrees that are pairwise non - bisimilar . without loss of generality
, we may assume that the equations of @xmath209 are of one of three forms : 1 .
@xmath718 , 2 .
@xmath719 , and 3 .
@xmath720 for @xmath721 . in the following ,
we are interested in a particular family @xmath722 of subtrees " of @xmath708 , containing , for all @xmath65 , the subtree @xmath723 rooted at @xmath724 , and the trees obtained from @xmath723 by removing the exit edge originating in the root together with its target , or the edge labelled @xmath198 originating in the root together with all vertices and edges accessible from the end vertex of that edge , or both .
we denote these synchronization trees by @xmath725 and @xmath726 , respectively .
[ lem - lem2 ] suppose that a tree @xmath727 is bisimilar to a tree @xmath728 , where neither @xmath729 nor @xmath730 is bisimilar to @xmath1 .
then for some @xmath731 , @xmath732 , @xmath729 is bisimilar to @xmath733 , and @xmath730 is bisimilar to @xmath734 .
suppose that @xmath156 is bisimilar to @xmath728 and neither @xmath729 nor @xmath730 is bisimilar to the tree @xmath1 .
note that @xmath730 is not @xmath17 .
clearly , each vertex of @xmath729 , except possibly the root , must be the source of an exit edge .
suppose that @xmath156 contains and edge labelled @xmath12 from @xmath735 to @xmath736 such both @xmath735 and @xmath736 are sources of an exit edge .
then in the tree @xmath737 , they have successor vertices @xmath738 and @xmath739 such that the subtrees rooted at @xmath738 and @xmath739 are isomorphic ( and thus bisimilar ) and contain at least one edge . but @xmath708 does not have such vertices connected by an edge labelled @xmath12 and therefore neither does @xmath156 .
for this reason , @xmath729 can not have two consecutive edges labelled @xmath12 either .
this in turn yields that @xmath730 has at least one edge labelled @xmath12 and therefore @xmath729 can not have an edge labelled @xmath198 . we conclude that @xmath729 is bisimilar to the tree @xmath733 and then @xmath730 is bisimilar to @xmath734 for some @xmath731 .
now , by lemma [ lem - lem2 ] , we may transform @xmath209 into a right - linear scheme defining @xmath708 up to bisimilarity .
first mark all those variables @xmath212 such that the corresponding component in the initial solution of @xmath209 over @xmath92 has an infinite branch .
the first variable is clearly marked .
suppose that @xmath212 is marked .
if the equation for @xmath212 is @xmath740 , then @xmath741 or @xmath742 is marked .
if one of them is not marked , then it can be replaced by a variable - free term .
if the equation for @xmath212 is @xmath743 and the component in the initial solution of @xmath209 over @xmath92 corresponding to one of the @xmath576 is bisimilar to @xmath1 , then we may simply remove it .
otherwise we apply lemma [ lem - lem2 ] and replace @xmath741 by @xmath12 and mark @xmath742 if it is not yet marked .
eventually , we keep only the marked functor variables and obtain a right - linear scheme defining @xmath708 up to bisimilarity . the @xmath13-regular trees that can be defined using regular @xmath13-recursion schemes that do not contain occurrences of the constants @xmath17 and @xmath1 correspond to unfoldings of the labelled transition systems denoted by terms in basic process algebra ( bpa ) with recursion , see , for instance , @xcite .
indeed , the signature of bpa contains one constant symbol @xmath12 for each action as well as the binary @xmath141 and @xmath142 operation symbols , denoting nondeterministic choice and sequential composition , respectively . in the remainder of this paper
, we write bpa for ` bpa with recursion ' . alternatively , following @xcite ,
one may view bpa as the class of labelled transition systems associated with context - free grammars in greibach normal form in which only leftmost derivations are permitted .
the class of basic parallel processes ( bpp ) is a parallel counterpart of bpa introduced by christensen @xcite .
bpp consists of the labelled transition systems associated with context - free grammars in greibach normal form in which arbitrary derivations are allowed .
we refer the interested readers to @xcite for the details of the formal definitions , which are not needed to appreciate the results to follow , and further pointers to the literature .
[ prop : moreexpthanbpa ] 1 .
every synchronization tree that is the unfolding of a bpa process is @xmath10-algebraic .
2 . there is a @xmath10-algebraic synchronization tree that is neither definable in bpa modulo bisimilarity nor in bpp modulo language equivalence .
the former claim follows easily from theorem [ thm : delta2gamma ] . in order to prove the latter statement ,
consider the lts depicted on figure [ fig : algebraic - not - bpa ] .
this lts is not expressible in bpa modulo modulo bisimilarity and is not expressible in bpp modulo language equivalence ( if the states @xmath456 are @xmath230 are the only final states in the lts)see ( * ? ? ?
* , example ( f ) ) . on the other hand ,
the synchronization tree associated with that lts is @xmath10-algebraic because it is the unique solution of the recursion scheme below .
@xmath744 so non - regular @xmath10-algebraic recursion schemes are more expressive than bpa modulo bisimilarity and can express synchronization trees that can not be defined in bpp up to language equivalence , and therefore up to bisimilarity .
@xmath745[f=]{\makebox(3,3 ) { } } } { } & { * + [ o][f-]{\makebox(3,3 ) { } } } { } \ar[l]^b & { * + [ o][f-]{\makebox(3,3 ) { } } } { } \ar[l]^b & \ar[l]^b \cdots \\ \ar[r ] & { * + [ o][f-]{\makebox(3,3 ) { } } } { } \ar[u]^b \ar[r]^a \ar[d]^c & { * + [ o][f-]{\makebox(3,3){}}}{}\ar[u]^b \ar[r]^a \ar[d]^c & { * + [ o][f-]{\makebox(3,3){}}}{}\ar[u]^b \ar[r]^a \ar[d]^c & \cdots \\ & { * + [ o][f=]{\makebox(3,3 ) { } } } { } & { * + [ o][f-]{\makebox(3,3 ) { } } } { } \ar[l]^c & { * + [ o][f-]{\makebox(3,3 ) { } } } { } \ar[l]^c & \ar[l]^c \cdots } \ ] ]
in this section , we compare the expressiveness of recursion schemes to that of the low classes in the caucal hierarchy @xcite .
section [ sect : definition - caucal ] gives a general overview of the caucal hierarchy .
section [ sect : first - levels ] presents in more details the properties of the graphs and trees sitting in the first levels of the hierarchy .
section [ sect : treesinch ] shows that the classes of synchronization trees we introduced belong to the caucal hierarchy .
finally section [ sec : charac - gamma - algebraic ] characterises the @xmath10-algebraic synchronization trees as contractions of the synchronization trees in the class @xmath746 .
the caucal hierarchy ( also known as the pushdown hierarchy ) is a hierarchy of classes of edge - labelled graphs .
following @xcite , the caucal hierarchy is @xmath747 where @xmath748 and @xmath749 denote the classes of finite , edge - labelled trees and graphs , respectively .
moreover , for each @xmath95 , @xmath750 stands for the class of trees isomorphic to unfoldings of graphs in @xmath751 , and the graphs in @xmath752 are those that can be obtained from the trees in @xmath750 by applying a monadic interpretation ( or transduction ) @xcite .
as shown by caucal in @xcite , every graph in the caucal hierarchy has a decidable theory for monadic second order logic .
[ rem : rooted - graphs ] for a graph @xmath753 and a vertex @xmath85 of @xmath212 , the graph @xmath754 obtained by restricting @xmath212 to the set of vertices reachable from @xmath85 is also a graph in @xmath755 @xcite . in particular
, we can always assume that a tree in @xmath750 is obtained by unfolding a graph in @xmath755 from one of its root .
= [ circle , draw = black , fill = black , inner sep = 0.25 mm , outer sep= 0.5 mm ] = [ inner sep = 0 mm , outer sep= 0 mm ] = [ font=,label distance=-1 mm ] = [ draw , rectangle ] = [ thick , decoration = markings , mark = at position 1 with , double distance=1.4pt , shorten > = 5.5pt , preaction = decorate , postaction = draw , line width=1.4pt , white , shorten > = 4.5pt ] ; ( h0 ) edge[- > ] node[right ] @xmath199 ( h1 ) ( h1 ) edge[- > ] node[below ] @xmath12 ( h2 ) ( h2 ) edge[- > ] node[below ] @xmath145 ( h3 ) ( h0 ) edge[- > ] node[above ] @xmath199 ( g0 ) ( g0 ) edge[- > ] node[right ] @xmath199 ( g1 ) ( g0 ) edge[loop above ] node[above ] @xmath199 ( g0 ) ; ; ( h1 ) edge[- > ] node[above ] @xmath199 ( h2 ) ( h2 ) edge[- > ] node[above ] @xmath199 ( h3 ) ( h1 ) edge[- > ] node[right ] @xmath199 ( h7 ) ( h2 ) edge[- > ] node[right ] @xmath199 ( h8 ) ( h3 ) edge[- > ] node[right ] @xmath199 ( h9 ) ( h7 ) edge[- > ] node[below ] @xmath12 ( h6 ) ( h6 ) edge[- > ] node[below ] @xmath145 ( h5 ) ( h3 ) edge[dashed ] ( h4 ) ; ; ( h1 ) edge[- > ] node[above ] @xmath199 ( h2 ) ( h2 ) edge[- > ] node[above ] @xmath199 ( h3 ) ( h1 ) edge[- > ] node[right ] @xmath199 ( h7 ) ( h2 ) edge[- > ] node[right ] @xmath199 ( h8 ) ( h3 ) edge[- > ] node[right ] @xmath199 ( h9 ) ( h7 ) edge[- > ] node[below ] @xmath12 ( h6 ) ( h6 ) edge[- > ] node[below ] @xmath145 ( h5 ) ( h3 ) edge[dashed ] ( h4 ) ( h8 ) edge[- > ] node[below ] @xmath12 ( h7 ) ( h9 ) edge[- > ] node[below ] @xmath12 ( h8 ) ; ; ( h1 ) edge[- > ] node[above ] @xmath199 ( h2 ) ( h2 ) edge[- > ] node[above ] @xmath199 ( h3 ) ( h1 ) edge[- > ] node[right ] @xmath199 ( g1 ) ( h2 ) edge[- > ] node[right ] @xmath199 ( g2 ) ( h3 ) edge[- > ] node[right ] @xmath199 ( g3 ) ( g1 ) edge[- > ] node[right ] @xmath12 ( i1 ) ( g2 ) edge[- > ] node[right ] @xmath12 ( i2 ) ( g3 ) edge[- > ] node[right ] @xmath12 ( i3 ) ( i1 ) edge[- > ] node[right ] @xmath145 ( j1 ) ( i2 ) edge[- > ] node[right ] @xmath12 ( j2 ) ( i3 ) edge[- > ] node[right ] @xmath12 ( j3 ) ( j2 ) edge[- > ] node[right ] @xmath145 ( k2 ) ( j3 ) edge[- > ] node[right ] @xmath12 ( k3 ) ( k3 ) edge[- > ] node[right ] @xmath145 ( l3 ) ( h3 ) edge[dashed ] ( h4 ) ; ; ( h1 ) edge[- > ] node[above left ] @xmath12 ( g1 ) ( h1 ) edge[- > ] node[left ] @xmath12 ( g2 ) ( h1 ) edge[- > ] node [ left ] @xmath12 ( g3 ) ( g1 ) edge[- > ] node[left ] @xmath145 ( i1 ) ( g2 ) edge[- > ] node[right ] @xmath12 ( i2 ) ( g3 ) edge[- > ] node[right ] @xmath12 ( i3 ) ( i2 ) edge[- > ] node[right ] @xmath145 ( j2 ) ( i3 ) edge[- > ] node[right ] @xmath12 ( j3 ) ( j3 ) edge[- > ] node[right ] @xmath145 ( k3 ) ( h1 ) edge[dashed ] ( d ) ; ( 2.5cm,-0 cm ) edge[vecarrow ] node[above ] @xmath756 ( 4.5cm,-0 cm ) ; ( 4.5cm,-1 cm ) edge[vecarrow ] node[above left ] @xmath757 ( 2.5cm,-2.5 cm ) ; ( 2.5cm,-3 cm ) edge[vecarrow ] node[above ] @xmath756 ( 4.5cm,-3 cm ) ; ( 4.5cm,-4 cm ) edge[vecarrow ] node[above left ] @xmath758 ( 2.5cm,-5.5 cm ) ; figure [ fig : caucal ] shows a sequence of transformations constructing the synchronization tree @xmath560 starting from a finite graph .
the vertices from which the graphs are unfolded are signalled by an incoming arrow .
the mso - interpretation @xmath757 adds an @xmath12-labelled edge from a vertex @xmath86 to a vertex @xmath85 if @xmath86 has no outgoing edges and there exist two vertices @xmath156 and @xmath16 with @xmath759 , @xmath760 , @xmath761 .
the mso - interpretation @xmath758 is the @xmath762-contraction operation .
the class @xmath763 contains the regular trees of finite outdegree ( i.e. , the trees of finite degree having finitely many non - isomorphic subtrees ) .
it is well known that @xmath764 is the set of all prefix - recognizable graphs @xcite .
a _ prefix recognizable relation _ over a finite alphabet @xmath109 is a finite union of relations of the form denotes the relation @xmath765 . ]
@xmath766 , for some nonempty regular languages @xmath767 .
prefix recognizable graph _ over an alphabet @xmath183 is an edge - labelled graph that is isomorphic to a graph of the form @xmath768 , where for some alphabet @xmath109 , @xmath75 is a regular subset of @xmath769 and the edge relations @xmath770 are prefix recognizable relations over @xmath109 .
( of course , the alphabet @xmath109 may be fixed to be a 2-element alphabet . )
[ prop : level - one ] for a labelled graph @xmath212 , the following statements are equivalent : * @xmath212 belongs to @xmath771 , * @xmath212 is isomorphic to a prefix - recognizable graph @xcite , * @xmath212 can be mso - interpreted in the full binary tree @xmath772 @xcite .
[ prop : det - closure ] let @xmath212 be a graph in @xmath771 labelled in @xmath74 , and let @xmath230 be a root of @xmath212 .
the graph @xmath212 is isomorphic to the @xmath183-contraction of a deterministic graph @xmath773 labelled in @xmath182 from one of its root @xmath774 .
moreover @xmath754 can be chosen such that for any @xmath143 and any two vertices @xmath86 and @xmath85 belonging to the @xmath183-contraction from @xmath774 , there exists at most one path from @xmath86 to @xmath85 labelled by a word in @xmath193 .
let @xmath775 be a prefix - recognizable graph labelled in @xmath74 .
we may assume , without loss of generality , that the vertices in @xmath75 are words over the alphabet @xmath776 .
moreover we may assume that the relations @xmath777 , @xmath143 , can be expressed as the disjoint union of relations @xmath778 .
in addition , we may require that @xmath779 , for all @xmath780 $ ] where @xmath781 ( * ? ? ?
* proposition 2.1 ) .
these assumptions guarantee that if @xmath782 belongs to @xmath777 , then there exists a unique @xmath780 $ ] and a unique decomposition @xmath783 and @xmath784 such that @xmath785 and @xmath786 . before proceeding with the construction ,
we need to introduce some notations .
let @xmath143 and let @xmath780 $ ] .
we take @xmath787 to be a complete dfa accepting the reverse of @xmath788 and @xmath789 to be a complete dfa accepting @xmath790 .
we assume that the sets of states of these automata are pairwise disjoint and we take @xmath791 } q_{a , i}$ ] and @xmath792 } q_{a , i}'$ ] .
we are now going to define a prefix - recognizable graph @xmath793 satisfying the properties stated above .
the set of labels @xmath183 is @xmath794 \}$ ] .
the vertices of @xmath754 are words over the alphabet @xmath795 \ } \cup \ { \star \}$ ] .
intuitively , the graph @xmath754 is constructed in such a way that a path labelled by a word in @xmath796 from a vertex @xmath797 to a vertex @xmath798 simulates the relation @xmath799 for @xmath143 and @xmath800 $ ] .
for a fixed @xmath143 and @xmath780 $ ] , the simulation is done using two sets of vertices @xmath801 and @xmath802 . starting from a vertex @xmath797 ,
the vertices in @xmath788 are used to remove a suffix @xmath85 of @xmath171 belonging to @xmath788 and the vertices in @xmath803 are used to add a suffix @xmath460 in @xmath790 .
when moving from the vertices in @xmath788 to the vertices in @xmath804 , we check that the remaining prefix @xmath86 belongs to @xmath805 .
this guarantees that @xmath171 and @xmath237 can be respectively written as @xmath806 and @xmath807 with @xmath785 and @xmath786 . formally , from a vertex @xmath797 ,
there is an edge labelled @xmath808 to the vertex @xmath809 ( cf .
( 1 ) below ) . for all words @xmath810 and for all @xmath811 , @xmath812 ( cf . ( 2 ) below ) . if @xmath813 belongs to @xmath805 and @xmath814 , @xmath815 ( cf .
( 3 ) below ) . for all words @xmath810 and for all @xmath816 , @xmath817 ( cf . ( 4 ) below ) . finally , for @xmath813 and @xmath818 , we have @xmath819 ( cf .
( 5 ) and ( 6 ) below ) .
the set of vertices @xmath820 is taken to be @xmath821 .
the edges of @xmath754 are defined , for all @xmath143 , @xmath780 $ ] , @xmath822 , @xmath823 and @xmath824 , by : @xmath825 u\,b q ( a , i ) & { \overset{e_{0}}{\underset{h}{\longrightarrow } } } & & u \,\delta_{a , i}(q , b ) & & \text{for $ b \in \{0,1\}$ } & ( 2 ) \\[1em ] u\,q ( a , i ) & { \overset{e_{1}}{\underset{h}{\longrightarrow } } } & & u \,q_{a , i } ' ( a , i ) & & \text{if}\ ; q \in f_{i , a } { \;\text{and}\;}u \in u & ( 3 ) \\[1em ] u\ , q ' ( a , i ) & { \overset{e_{0}}{\underset{h}{\longrightarrow } } } & & ub \,\delta_{a , i}'(q , b ) ( a , i ) & & \text{for $ b \in \{0,1\}$ } & ( 4 ) \\[1em ] u \,q ' ( a , i ) & { \overset{e_{2}}{\underset{h}{\longrightarrow } } } & & u \ , \star ( a , i ) & & \text{if}\ ;
q ' \in f_{a , i } ' & ( 5)\\[1em ] u \,\star ( a , i ) & { \overset{a}{\underset{h}{\longrightarrow } } } & & u & & \text{for $ u \in v$ } & ( 6)\\ \end{array}\ ] ] the graph @xmath754 is a deterministic prefix - recognizable graph .
we have already seen that : for all @xmath826 , if @xmath827 then there exists a path in @xmath754 labelled in @xmath796 from @xmath171 to @xmath237 .
it remains to show the other direction . for all @xmath826
, if there exists a path from @xmath171 to @xmath237 in @xmath754 labelled in @xmath193 then this path is unique and @xmath827 .
let @xmath826 be such that there exists a path from @xmath171 to @xmath237 in @xmath754 labelled in @xmath193 . by the construction of @xmath754 , there exist @xmath780 $ ] , and @xmath520 and @xmath828 with @xmath783 and @xmath829 such that the path is of the form @xmath830 with @xmath831 , @xmath832 and @xmath833 .
it follows that @xmath782 belongs to @xmath799 . as remarked before ,
the index @xmath292 is unique and so is the decomposition into @xmath520 and @xmath460 .
hence the uniqueness of the path follows . from the above two claims
, it follows that the @xmath183-contraction of @xmath754 is equal to @xmath212 .
as first remarked in @xcite , the graphs in @xmath771 can be obtained by a form of contraction - contraction keeps the vertices that are the source of an edge in @xmath74 and adds an edge labelled by @xmath143 between two such vertices whenever there is a path labelled in @xmath834 .
] of the configurations graph of pushdown automata @xcite .
let us now consider the class @xmath746 .
the main property of the class @xmath746 is that it contains the @xmath20-term trees defined by any @xmath20-algebraic recursion scheme for any signature @xmath20 .
as @xmath746 only contains labelled trees , we need to fix a representation of @xmath20-term trees as labelled trees .
we follow the one from @xcite which is slightly different from the one presented in section [ sec : morphism ] ( see also remark [ rem : ht - caucal ] ) . a @xmath20-term tree @xmath16 is represented by a tree @xmath15 labelled by @xmath835 where @xmath265 is the maximum rank of a symbol of @xmath20 .
the vertices of @xmath15 are prefixes of sequences of the form @xmath836 where @xmath837 is the root of @xmath16 , @xmath838 is a leaf of @xmath16 and for all @xmath245 $ ] , @xmath839 is the @xmath840-th successor of @xmath841 .
let @xmath86 be a vertex of @xmath16 labelled by @xmath842 , if @xmath460 and @xmath843 are vertices of @xmath15 then there is an edge from @xmath460 to @xmath843 labelled by @xmath12 .
let @xmath844 $ ] , if @xmath460 and @xmath845 are vertices of @xmath15 then there is an edge from @xmath460 to @xmath845 labelled by @xmath846 .
this representation is illustrated in figure [ fig : representation - caucal ] for the @xmath13-term @xmath847 .
= [ circle , draw = black , fill = black , inner sep = 0.25 mm , outer sep= 0.5 mm ] = [ inner sep = 0 mm , outer sep= 0 mm ] = [ font=,label distance=-1 mm ] = [ level distance=0.85cm , sibling distance=1 cm ] = [ level distance=0.85cm , sibling distance=1.25 cm ] child node[label=[l]left:@xmath485 @xmath141 child node[label=[l]left:@xmath487 @xmath12 child node[label=[l]right:@xmath488 @xmath17 child node[label=[l]right:@xmath486 @xmath202 ; child node[n , label=[l]right:@xmath848 child node[n , label=[l]left:@xmath849 child node[n , label=[l]left:@xmath850 child node[n , label=[l]left:@xmath851 child node[n , label=[l]left:@xmath852 edge from parent node[draw = none , left ] @xmath12 edge from parent node[draw = none , left ] @xmath853 child node[n , label=[l]right:@xmath854 child node[n , label=[l]right:@xmath855 edge from parent node[draw = none , right ] @xmath17 edge from parent node[draw = none , right ] @xmath856 edge from parent node[draw = none , left ] @xmath141 edge from parent node[draw = none , left ] @xmath853 child node[n , label=[l]right:@xmath857 child node[n , label=[l]right:@xmath858 edge from parent node[draw = none , right ] @xmath202 edge from parent node[draw = none , right ] @xmath856 edge from parent node[draw = none , left ] @xmath142 ; [ theorem : caucal ] let @xmath20 be a signature .
a @xmath20-term tree is defined by some @xmath20-algebraic recursion scheme if and only if the tree representing it belongs to @xmath746 .
[ rem : ht - caucal ] it immediately follows that the synchronization tree @xmath530 associated to a @xmath10-term tree @xmath16 defined in section [ sec : morphism ] belongs to @xmath746 ( resp .
@xmath763 ) whenever @xmath16 is defined by a @xmath10-algebraic ( resp .
@xmath10-regular ) recursion scheme .
indeed the transformation going from the presentation of @xcite to ours is an mso - interpretation that commutes with the unfolding operation .
we conclude with some properties of the trees of @xmath746 and their contractions .
[ prop : trees - in - trees2 ] the following properties hold : 1 .
the contraction of a tree in @xmath746 is bisimilar to some tree in @xmath746 .
every tree in @xmath746 can be obtained as the contraction of a deterministic tree in @xmath746 . as illustrated in figure [ fig : counter - example ]
, the contraction operation does not in general commute with the unfolding operation .
however , it is easy to show that the contraction of the unfolding from a root @xmath230 and the unfolding of the contraction from the same root are bisimilar .
= [ circle , draw = black , fill = black , inner sep = 0.25 mm , outer sep= 0.5 mm ] = [ inner sep = 0 mm , outer sep= 0 mm ] = [ font=,label distance=-1 mm ] ; \(r ) edge[-latex ] node[above left ] @xmath199 ( s ) ( r ) edge[-latex ] node[above right ] @xmath199(t ) ( s ) edge[-latex ] node[below left ] @xmath12(u ) ( t ) edge[-latex ] node[below right ] @xmath12(u ) ( u ) edge[-latex ] node[right ] @xmath199 ( r ) ; = [ level distance=0.85cm , sibling distance=1 cm ] child node[n ] child node[n ] child node[e ] edge from parent[dashed,- ] edge from parent node[left ] @xmath12 edge from parent node[left ] @xmath12 ; = [ level distance=0.85cm , sibling distance=0.85 cm ] = [ level distance=0.85cm , sibling distance=1.75 cm ] child node[n ] child node[n ] child node[e ] edge from parent[dashed,- ] child node[e ] edge from parent[draw = none ] edge from parent node[draw = none , above left ] @xmath12 child node[n ] child node[e ] edge from parent[draw = none ] child node[e ] edge from parent[dashed,- ] edge from parent node[draw = none , above right ] @xmath12 edge from parent node[draw = none , above left ] @xmath12 child node[n ] child node[n ] child node[e ] edge from parent[dashed,- ] child node[e ] edge from parent[draw = none ] edge from parent node[draw = none , above left ] @xmath12 child node[n ] child node[e ] edge from parent[draw = none ] child node[e ] edge from parent[dashed,- ] edge from parent node[draw = none , above right ] @xmath12 edge from parent node[draw = none , above right ] @xmath12 ; for the first property , let @xmath16 be a tree in @xmath746 which is obtained by unfolding a graph @xmath212 in @xmath771 from one of its roots @xmath230 .
let @xmath15 be the @xmath183-contraction of @xmath16 from its root .
the @xmath183-contraction @xmath859 of @xmath212 from @xmath230 also belongs to @xmath771 ( as the @xmath183-contraction is a particular case of mso - interpretation ) . from the previous remark , we have that @xmath15 is bisimilar to @xmath860 which belongs to @xmath861 as @xmath212 belongs to @xmath764 . for the second property ,
let @xmath16 be a tree in @xmath746 which is obtained by unfolding a graph @xmath212 in @xmath771 from one of its root @xmath230
. furthermore assume that both @xmath16 and @xmath212 are labelled by @xmath74 . by proposition [ prop : det - closure ]
, @xmath212 can be obtained by @xmath183-contraction of a deterministic graph @xmath773 from one of its roots @xmath230 . for the @xmath183-contraction to commute with the unfolding ,
it is enough that the graph @xmath754 satisfies , for any label @xmath143 and any two vertices @xmath86 and @xmath85 of @xmath754 belonging to the @xmath183-contraction , that there exists at most one path labelled in @xmath193 from @xmath86 to @xmath85 .
this is the case for the graph @xmath754 obtained from proposition [ prop : det - closure ] . hence @xmath16 is isomorphic to the @xmath183-contraction of @xmath862 , which is a deterministic tree in @xmath746 . the @xmath10-regular synchronization trees , being the regular trees with potentially infinite degree , belong to @xmath771 .
as @xmath763 only contains the regular trees of finite degree , it does not contain the class of @xmath10-regular synchronization trees . in this section , we show that the classes of @xmath13-regular and @xmath10-algebraic synchronization trees are included in @xmath863 but not in @xmath746 .
similarly , we establish that the class of @xmath13-algebraic trees is included in @xmath864 but not in @xmath865 .
[ prop : carac ] the @xmath10-algebraic ( and hence the @xmath13-regular ) synchronization trees are contractions of trees in @xmath746 and hence belong to @xmath863 .
let @xmath209 be a @xmath10-algebraic recursion scheme defining a @xmath10-term tree @xmath16 and a synchronization tree @xmath15 . from proposition [ prop - gamma - hom ] ,
@xmath15 is obtained by contraction of the synchronization tree @xmath530 with respect to @xmath866 . from (
* theorem 3.5 ) ( cf .
remark [ rem : ht - caucal ] ) , we know that @xmath530 belongs to @xmath746 , and , as the contraction operation is a particular case of mso - interpretation , @xmath15 belongs to @xmath863 .
there is a @xmath13-regular synchronization tree which is not in @xmath861 .
consider the following @xmath13-regular recursion scheme : @xmath867 the synchronization tree @xmath16 defined by it has a single infinite branch @xmath868 ( with edges labelled a ) .
the out - degree of each vertex @xmath869 is @xmath870 since it is the source of @xmath871 edges labelled @xmath145 .
assume , towards a contradiction , that @xmath16 belongs to @xmath861 .
hence @xmath16 is isomorphic to the unfolding of some graph @xmath754 in @xmath764 from a vertex @xmath471 . by remark [ rem : rooted - graphs ]
, we can assume that all vertices are reachable from @xmath471 and hence that they all have finite out - degree .
this implies that @xmath754 contains a path @xmath872 along which the out - degree of the vertices grows exponentially ( i.e. the out - degree of @xmath291 is @xmath870 ) .
the following lemma shows that this is not possible in @xmath771 .
let @xmath212 be a graph in @xmath771 whose vertices have finite out - degree .
let @xmath873 be an infinite path in @xmath212 , possibly with repetitions .
then there exists a constant @xmath874 such that for all @xmath324 , the out - degree of @xmath869 is at most @xmath875 . by proposition [ prop : level - one ]
, we can assume without loss of generality that @xmath212 is a prefix - recognizable graph .
let @xmath876 be the relations involved in the definition of @xmath212 .
as the vertices of @xmath212 have finite out - degree , without loss of generality we may assume that all the @xmath877 are finite sets .
let @xmath376 denote the length of the longest word appearing in the @xmath877 , and let @xmath878 denote the number of words appearing in the @xmath877 .
then clearly @xmath879 for all @xmath650 , where for a word @xmath86 we denote its length by @xmath880 .
thus , introducing the notation @xmath881 , we have @xmath882 for all @xmath324 . as the out - degree of a vertex @xmath86 of @xmath212 is at most @xmath883 ,
we conclude that the out - degree of @xmath869 is at most @xmath875 for all @xmath324 with @xmath884 .
[ prop : delta - alg - in - graph3 ] all the @xmath13-algebraic synchronization trees are in @xmath864 and therefore have a decidable mso - theory .
let @xmath209 be an algebraic scheme over @xmath13 .
let @xmath16 be the @xmath13-term defined by @xmath209 , and @xmath15 be the synchronization tree defined by @xmath209 . by the mezei - wright theorem and proposition [ prop - tau ] ,
we have that @xmath885 .
more precisely , the tree @xmath15 is obtained by unfolding the graph @xmath470 ( as defined in section [ sec : morphism ] ) from the vertex @xmath222 and then applying a contraction with respect to @xmath886 .
the graph @xmath470 can be interpreted in the tree representing @xmath16 , which belongs to @xmath746 ( by theorem [ theorem : caucal ] ) , and hence @xmath470 belongs to @xmath863 .
the unfolding of @xmath470 from @xmath222 belongs to @xmath865 and its contraction @xmath15 to @xmath864 ( as contractions are particular cases of mso - interpretations ) .
there exists a @xmath13-algebraic synchronization tree which does not belong to @xmath865 . in this proof
, we consider @xmath887-labelled trees of a particular shape illustrated in figure [ fig : td0 ] .
these trees have a unique infinite branch whose edges are labelled @xmath12 .
in addition the vertex at depth @xmath5 along this branch has @xmath888 outgoing edges labelled @xmath198 , for some mapping @xmath889 . up to isomorphism , the tree is entirely characterized by the mapping @xmath376 and is denoted @xmath890 . in (
* theorem 4.5.3 ) , braud gives a necessary condition for a tree of this form - graph - combs which encompasses all trees @xmath890 . ] to belong to @xmath891 .
namely , if a tree @xmath890 , for @xmath889 belongs to @xmath891 then there exists a constant @xmath892 such that @xmath893 for all @xmath41 .
our proof goes as follows .
we first show that for the mapping @xmath894 , the tree @xmath895 , depicted in figure [ fig : td0 ] , is @xmath13-algebraic . towards a contradiction
, we assume that @xmath895 belongs to @xmath896 .
we then show there would exist a mapping @xmath897 satisfying @xmath898 , @xmath899 such that @xmath900 belongs to @xmath891 .
this statement contradicts braud s condition as there can not exist a constant @xmath202 such that @xmath901 , for all @xmath899 .
consider the @xmath13-algebraic scheme @xmath902 the tree defined by this scheme is isomorphic to the tree @xmath895 with @xmath903 for all @xmath41 .
= [ circle , draw = black , fill = black , inner sep = 0.25 mm , outer sep= 0.5 mm ] = [ inner sep = 0 mm , outer sep= 0 mm ] ( a0 ) at ( 0,0 ) ; ( a1 ) at ( 3,0 ) ; ( a2 ) at ( 6,0 ) ; ( b00 ) at ( -0.5,-1 ) ; ( b01 ) at ( 0.5,-1 ) ; ( b10 ) at ( 2,-1 ) ; ( b11 ) at ( 2.75,-1 ) ; ( b12 ) at ( 3.25,-1 ) ; ( b13 ) at ( 4,-1 ) ; ( b20 ) at ( 4.5,-1 ) ; ( b21 ) at ( 7.5,-1 ) ; ( a0 ) to[above ] node @xmath12 ( a1 ) ; ( a1 ) to[above ] node @xmath12 ( a2 ) ; ( a0 ) to[left ] node @xmath198 ( b00 ) ; ( a0 ) to[right ] node @xmath198 ( b01 ) ; ( a1 ) to[left ] node @xmath198 ( b10 ) ; ( a1 ) to[left ] node @xmath198 ( b11 ) ; ( a1 ) to[right ] node @xmath198 ( b12 ) ; ( a1 ) to[right ] node @xmath198 ( b13 ) ; ( a2 ) to[left ] node[yshift=1 mm ] @xmath198 ( b20 ) ; ( a2 ) to[right ] node[yshift=1 mm ] @xmath198 ( b21 ) ; ( a2 ) ( 9,0 ) ; ( a2 ) ( 5.5,-1 ) ; ( a2 ) ( 6.5,-1 ) ; ( b20 ) to node[below ] @xmath904 ( b21 ) ; assume now , towards a contradiction , that @xmath895 belongs to @xmath865 . consider the graph @xmath212 in @xmath863 from which @xmath16 can be obtained by unfolding from one of its roots @xmath230 . for all @xmath41 ,
we denote by @xmath905 the unique vertex in @xmath212 such that @xmath906 and by @xmath102 the set of vertices @xmath907 . as @xmath908 , we must have @xmath909 for @xmath910 and @xmath911 for all @xmath41 .
the set of vertices of @xmath212 is therefore equal to @xmath912 and its set of edges is : @xmath913 note that the sets @xmath102 are not necessarily pairwise disjoint and that in particular , @xmath212 is not necessarily a tree , as illustrated in figure [ fig : transform ] .
= [ circle , draw = black , fill = black , inner sep = 0.25 mm , outer sep= 0.5 mm ] = [ inner sep = 0 mm , outer sep= 0 mm ] ( a0 ) at ( 0,0 ) ; ( a1 ) at ( 2,0 ) ; ( a2 ) at ( 4,0 ) ; ( b00 ) at ( -0.5,-1 ) ; ( b01 ) at ( 0.5,-1 ) ; ( b10 ) at ( 1.5,-1 ) ; ( b11 ) at ( 2,-1 ) ; ( b12 ) at ( 2.5,-1 ) ; ( a0 ) to[above ] node @xmath12 ( a1 ) ; ( a1 ) to[above ] node @xmath12 ( a2 ) ; ( a0 ) to[left ] node @xmath198 ( b00 ) ; ( a0 ) to[right ] node @xmath198 ( b01 ) ; ( a1 ) to[above ] node @xmath198 ( b01 ) ; ( a1 ) to[left ] node @xmath198 ( b10 ) ; ( a1 ) to[right ] node[xshift=-1 mm ] @xmath198 ( b11 ) ; ( a1 ) to[right ] node @xmath198 ( b12 ) ; ( a2 ) to[above ] node[xshift=4mm , yshift=1 mm ] @xmath198 ( b01 ) ; ( a2 ) to[below ] node @xmath198 ( b12 ) ; ( a2 ) ( 5,0 ) ; ( a2 ) ( 3.5,-1 ) ; ( a2 ) ( 4,-1 ) ; ( a2 ) ( 4.5,-1 ) ; ( a0 ) at ( 0,0 ) ; ( a1 ) at ( 2,0 ) ; ( a2 ) at ( 4,0 ) ; ( b00 ) at ( -0.5,-1 ) ; ( b01 ) at ( 0.5,-1 ) ; ( b10 ) at ( 1.5,-1 ) ; ( b11 ) at ( 2,-1 ) ; ( b12 ) at ( 2.5,-1 ) ; ( a0 ) to[above ] node @xmath12 ( a1 ) ; ( a1 ) to[above ] node @xmath12 ( a2 ) ; ( a0 ) to[left ] node @xmath198 ( b00 ) ; ( a0 ) to[right ] node @xmath198 ( b01 ) ; ( a1 ) to[left ] node @xmath198 ( b10 ) ; ( a1 ) to[right ] node[xshift=-1 mm ] @xmath198 ( b11 ) ; ( a1 ) to[right ] node @xmath198 ( b12 ) ; ( a2 ) ( 5,0 ) ; ( a2 ) ( 3.5,-1 ) ; ( a2 ) ( 4,-1 ) ; ( a2 ) ( 4.5,-1 ) ; consider the mso - interpretation @xmath261 that erases the @xmath198-labelled edges @xmath89 whenever there exists a vertex @xmath156 such that @xmath914 and @xmath915 and keeps all other edges unchanged . by applying @xmath261 to @xmath212 ( as illustrated in figure [ fig :
transform ] , we obtain a tree @xmath916 with the same set of vertices as @xmath212 .
this tree has a unique infinite branch @xmath917 and , for all @xmath41 , the vertex @xmath905 has outgoing edges labelled @xmath198 to the vertices in @xmath918 the tree @xmath256 is hence isomorphic to @xmath900 where @xmath919 for all @xmath41 . to obtain a contradiction with ( * ? ? ?
* theorem 4.5.3 ) , we remark that for all @xmath899 , we have : @xmath920 & \geq & 2^{2^n } - n \cdot 2^{2^{n-1 } } \\ & \geq & 2^{2^{n-1 } } ( 2^{2^{n-1 } } - n ) \\ & \geq & 2^{2^{n-1}}. \\ \end{array}\ ] ] in this section , we prove theorem [ thm : carac - gamma - algebraic ] which states that @xmath10-algebraic synchronization trees are the contractions of trees in @xmath746 .
we start by showing that the synchronization trees in @xmath746 are @xmath10-algebraic .
[ prop : tree2-inc - alg ] synchronization trees in @xmath746 are @xmath10-algebraic .
let @xmath16 be a synchronization tree in @xmath746 .
by the first property of proposition [ prop : trees - in - trees2 ] , @xmath16 is the @xmath183-contraction of a deterministic tree @xmath921 labelled by @xmath224 . as @xmath10-algebraic synchronization trees are closed under contraction ( see proposition [ prop : contr - alg ] ) , it is enough to show that @xmath15 is @xmath10-algebraic .
we are going to show that a tree @xmath922 representing a @xmath352-term tree defining @xmath15 belongs to @xmath746 .
recall that @xmath352 is the signature @xmath923 allowing for sums of arbitrary arity that was introduced in remark [ rem : arbitrary - rank - sum ] on page .
thanks to theorem [ theorem : caucal ] , this implies that @xmath922 is defined by a @xmath352-algebraic recursion scheme . in turn this implies that @xmath15 is @xmath352-algebraic and hence @xmath10-algebraic ( see remark [ rem : arbitrary - rank - sum ] ) .
let @xmath924 be an arbitrary total order on @xmath74 .
the tree @xmath922 is obtained by applying a transduction @xmath925 to @xmath15 .
this transduction @xmath925 does the following : * whenever a vertex @xmath85 is the target of an edge labelled in @xmath74 and is not the source of an edge , then the transduction adds a new edge labelled @xmath17 from @xmath85 to a new vertex introduced by the transduction . * for every vertex @xmath86 with @xmath926 outgoing edges labelled @xmath927 , respectively , going to vertices @xmath928 , the transduction adds new vertices @xmath85 , @xmath929 , @xmath930 and edges @xmath931 , @xmath932 and @xmath933 .
if @xmath86 has an outgoing edge labelled by @xmath145 to a vertex @xmath85 then the transduction adds a vertex @xmath188 and two edges @xmath934 and @xmath935 . *
all the edges of the original structure are removed .
the tree @xmath922 obtained by applying @xmath925 to @xmath15 represents a term tree that defines @xmath15 .
as @xmath15 belongs to @xmath746 , it is ( up to isomorphism ) obtained by unfolding a graph @xmath936 from one of its roots @xmath230 .
furthermore it can be checked that the transduction @xmath925 commutes with the unfolding operation .
hence the tree @xmath922 is isomorphic to @xmath937 and therefore belongs to @xmath746 ( as @xmath938 belongs to @xmath771 ) .
[ thm : carac - gamma - algebraic ] the contractions of the synchronization trees in @xmath746 are the @xmath10-algebraic synchronization trees . by propositions [ prop : tree2-inc - alg ] and [ prop : contr - alg ]
, we have that each contraction of a synchronization tree in @xmath746 is @xmath10-algebraic . for the converse
, a @xmath10-algebraic synchronization tree is defined as the contraction with respect to @xmath939 of the @xmath10-term tree defined by an algebraic scheme over @xmath10 . from (
* theorem 3.5 ) , such a @xmath10-term tree belongs to @xmath861 .
moreover , it may be seen as a synchronization tree over the alphabet which contains , in addition to the letters in @xmath74 , the symbols @xmath427 and @xmath431 .
thanks to the second property of proposition [ prop : trees - in - trees2 ] , we have the following corollary .
every @xmath10-algebraic tree is bisimilar to a tree in @xmath746 .
call a synchronization tree * bounded * if there is a constant @xmath731 such that the outdegree of each vertex is at most @xmath731
. our aim in this section will be to offer a language - theoretic characterization of the expressive power of @xmath10-algebraic recursion schemes defining synchronization trees .
we shall do so by following courcelle
see , e.g. , @xcite and studying the branch languages of synchronization trees whose vertices have bounded outdegree .
more precisely , we assign a family of branch languages to each bounded synchronization tree over an alphabet @xmath74 and show that a bounded tree is @xmath10-algebraic if , and only if , the corresponding language family contains a deterministic context - free language ( dcfl ) . throughout this section
, we will call @xmath10-algebraic trees just algebraic trees , and similarly for regular trees .
suppose that @xmath73 is a bounded synchronization tree over the alphabet @xmath74 .
denote by @xmath731 the maximum of the outdegrees of the vertices of @xmath16 .
let @xmath183 denote the alphabet @xmath940 $ ] .
a * determinization of @xmath16 * is a tree @xmath941 over the alphabet @xmath183 which differs from @xmath16 only in the labelling as follows .
suppose that @xmath223 with outgoing edges @xmath942 labelled @xmath943 in @xmath16 .
then there is some permutation @xmath253 of the set @xmath944 $ ] such that the label of each @xmath945 in @xmath15 is @xmath946 .
consider a determinization @xmath15 of @xmath16 .
let @xmath223 and let @xmath947 denote the vertices on the unique path from the root to @xmath85 .
the * branch word * corresponding to @xmath85 in @xmath15 is the alternating word @xmath948 where @xmath949 denote the outdegrees of the vertices @xmath950 , and for each @xmath951 $ ] , @xmath952 is the label of the edge @xmath953 in @xmath15 .
the * branch language * @xmath954 corresponding to a determinization @xmath15 of @xmath16 consists of all branch words of @xmath15 .
finally , the family of branch languages corresponding to @xmath16 is : @xmath955 @xmath368 & { * + [ o][f=]{\makebox(4,4 ) { } } } { } \ar@<1ex>[d]^c \ar@<1ex>[r]^a & { * + [ o][f-]{\makebox(4,4 ) { } } } { } \ar@<1ex>[d]^c \ar@<1ex>[r]^a \ar@<1ex>[l]^b & { * + [ o][f-]{\makebox(4,4 ) { } } } { } \ar@<1ex>[d]^c \ar@<1ex>[r]^a \ar@<1ex>[l]^b & { * + [ o][f-]{\makebox(4,4 ) { } } } { } \ar@<1ex>[d]^c \ar@<1ex>[r]^a \ar@<1ex>[l]^b & \cdots \ar@<1ex>[l]^b \\ & { * + [ o][f-]{\makebox(4,4 ) { } } } { } \ar@<1ex>[u]^d \ar@<1ex>[r]^a \ar@<1ex>[d]^c & { * + [ o][f-]{\makebox(4,4 ) { } } } { } \ar@<1ex>[u]^d \ar@<1ex>[r]^a \ar@<1ex>[l]^b \ar@<1ex>[d]^c & { * + [ o][f-]{\makebox(4,4 ) { } } } { } \ar@<1ex>[u]^d \ar@<1ex>[r]^a \ar@<1ex>[l]^b \ar@<1ex>[d]^c & { * + [ o][f-]{\makebox(4,4 ) { } } } { } \ar@<1ex>[u]^d \ar@<1ex>[d]^c \ar@<1ex>[r]^a \ar@<1ex>[l]^b & \cdots \ar@<1ex>[l]^b \\ & { * + [ o][f-]{\makebox(4,4 ) { } } } { } \ar@<1ex>[u]^d \ar@<1ex>[r]^a \ar@<1ex>[d]^c & { * + [ o][f-]{\makebox(4,4 ) { } } } { } \ar@<1ex>[u]^d \ar@<1ex>[r]^a \ar@<1ex>[l]^b \ar@<1ex>[d]^c & { * + [ o][f-]{\makebox(4,4 ) { } } } { } \ar@<1ex>[u]^d \ar@<1ex>[r]^a \ar@<1ex>[l]^b \ar@<1ex>[d]^c & { * + [ o][f-]{\makebox(4,4 ) { } } } { } \ar@<1ex>[u]^d \ar@<1ex>[r]^a \ar@<1ex>[l]^b \ar@<1ex>[d]^c & \cdots \ar@<1ex>[l]^b \\ & \vdots \ar@<1ex>[u]^d & \vdots \ar@<1ex>[u]^d & \vdots \ar@<1ex>[u]^d & \vdots \ar@<1ex>[u]^d & \vdots } \ ] ] by way of example , consider the lts depicted in figure [ fig : bag ] .
this lts describes the behaviour of a bag over a four - letter alphabet , when we consider @xmath198 to stand for the output of an item that was input via @xmath12 , and @xmath376 to signal the output of an item that was input via @xmath202 .
the synchronization tree @xmath956 that is obtained by unfolding this lts from its start state is bounded .
in fact , the outdegree of each non - leaf node is three or four .
the branch words corresponding to the nodes of any determinization of the tree @xmath956 have the form @xmath957 where @xmath958 , @xmath959 , @xmath960 for each @xmath961 $ ] , each @xmath962 $ ] for @xmath951 $ ] and @xmath963 is a word with the property that , in any of its prefixes , the number of occurrences of the letter @xmath12 is greater than or equal to the number of occurrences of the letter @xmath198 , and the number of occurrences of the letter @xmath202 is greater than or equal to the number of occurrences of the letter @xmath376 .
moreover , for each @xmath951 $ ] , @xmath964 if and only if @xmath965 and @xmath966 .
( note that , when @xmath967 , the number of @xmath12 s in @xmath968 equals the number of @xmath198 s , and similarly for @xmath202 and @xmath376 . )
[ thm - dcfl ] a bounded synchronization tree @xmath16 is algebraic ( respectively , regular ) iff @xmath969 contains a dcfl ( respectively , regular language ) . in the proof , we make use of the following construction .
suppose that @xmath970 is a bounded synchronization tree over @xmath74 .
let @xmath731 be defined as above , and let @xmath20 be the signature containing the symbol @xmath971 of rank @xmath292 for each @xmath972 $ ] , the constant symbols @xmath17 and @xmath1 , and the letters of @xmath74 as symbols of rank @xmath1 .
we will sometimes write @xmath973 for @xmath17
. then each determinization @xmath15 of @xmath16 naturally corresponds to an ` alternating ' _ term tree _ @xmath974 in the initial continuous categorical @xmath20-algebra @xmath396 . as a partial function @xmath975
, the term tree @xmath974 is defined as follows . consider a vertex @xmath223 with corresponding branch word @xmath976 .
then @xmath974 is defined on both words @xmath977 and @xmath978 , and @xmath979 in addition , @xmath974 is defined on the empty word @xmath386 and @xmath980 , where @xmath981 is the outdegree of the root .
[ lem - a ] suppose that @xmath982 is bounded and has a determinization @xmath15 such that the @xmath20-term tree @xmath974 is algebraic ( regular , resp . )
. then @xmath16 is algebraic ( regular , resp . ) .
we can turn @xmath92 into a continuous categorical @xmath20-algebra by defining @xmath983 for each @xmath972 $ ] and @xmath984 , and similarly for morphisms between trees in @xmath92 . now up to natural isomorphism , there is a unique categorical @xmath20-algebra morphism @xmath985 . for any @xmath542 and for any determinization @xmath15 of @xmath16 , @xmath154 maps @xmath974 to a tree isomorphic to @xmath16 . since , by the mezei - wright theorem @xcite , @xmath986 preserves algebraic and regular objects , if @xmath974 is algebraic or regular , then so is @xmath16 .
[ lem - b ] suppose that @xmath542 is bounded and algebraic ( regular , resp . ) .
then @xmath16 has a determinization @xmath15 such that the @xmath20-term tree @xmath974 is algebraic ( regular , resp . ) .
suppose that @xmath16 is algebraic and bounded by @xmath731 .
then , by the mezei - wright theorem @xcite , there is an algebraic term tree @xmath987 such that @xmath988 is isomorphic to @xmath16 , where @xmath989 denotes the essentially unique continuous categorical @xmath10-algebra morphism @xmath528 .
we want to show that there is an alternating algebraic @xmath20-term tree @xmath990 such that @xmath991 is isomorphic to @xmath16 , where @xmath986 is the essentially unique continuous categorical @xmath20-algebra morphism @xmath992 .
since such a term tree @xmath993 is @xmath974 for some determinization @xmath15 of @xmath16 , this completes the proof .
we construct @xmath993 from @xmath994 in two steps .
first , we replace all maximal subterms all of whose vertices are labelled @xmath141 or @xmath17 by a single vertex labelled @xmath17 to obtain a @xmath10-term tree @xmath995 .
second , we consider vertices of @xmath995 labelled @xmath141 whose parent , if any , is labelled in @xmath74 . since @xmath16 is bounded by @xmath731 , the subterm rooted at such a vertex can be written as @xmath996 , where @xmath997 is a finite term all of whose vertices are labelled @xmath141 or @xmath17 or a variable in the set @xmath998 for some @xmath999 , whose frontier word is @xmath1000 , and each @xmath1001 is a @xmath10-term tree whose root is labelled in @xmath1002 .
we replace each such vertex by a vertex labelled @xmath1003 having @xmath5 outgoing edges labelled @xmath1004 connecting this vertex to the roots of @xmath1005 , respectively .
the first transformation is a monadic colouring @xcite ( a special case of monadic interpretation , which is also known as monadic marking @xcite ) , since there is a monadic second - order formula @xmath1006 characterizing those vertices @xmath171 of @xmath994 such that all vertices of the subterm rooted at @xmath171 are labelled @xmath141 or @xmath17 , but any other subterm containing @xmath171 has a vertex labelled in @xmath1007 : @xmath1008 ( here , @xmath1009 denotes that there is a path from @xmath171 to @xmath237 and @xmath1010 stands for @xmath1011 .
moreover , @xmath1012 denotes that @xmath1013 is labelled @xmath12 . )
thus , the first transformation gives a @xmath10-algebraic term tree , since algebraic term trees ( and in fact deterministic algebraic trees in the caucal s pushdown hierarchy ) are closed under monadic colourings ( * ? ? ? * proposition 1 ) .
in order to prove that the second transformation also gives an algebraic term tree , we argue on the level of graphs .
suppose that @xmath995 is the algebraic term tree obtained after the first step .
since @xmath995 is algebraic , it is the unfolding of a ( deterministic ) prefix recognizable graph @xmath212 from its root @xmath230 , see @xcite . without loss of generality
we may assume that every vertex of @xmath212 is accessible from @xmath230 .
our aim is to define a monadic transduction which , when applied to @xmath212 , produces a graph @xmath859 whose unfolding from vertex @xmath230 is @xmath993 . since , by proposition 1 in conjunction with lemma 2 in @xcite , prefix recognizable graphs are closed under monadic transductions , it follows that @xmath993 is algebraic .
we start by considering @xmath212 together with a disjoint copy of @xmath212 , whose vertices are ordered pairs @xmath1014 , where @xmath85 is vertex of @xmath212 .
the label of a vertex @xmath1014 is the label of @xmath85 in @xmath212 .
the edges are the edges of @xmath212 and an edge @xmath1015 labelled @xmath1016 for each vertex @xmath85 of @xmath212 .
edges in @xmath212 retain their label .
then we drop all vertices of the form @xmath1014 , where the label of @xmath85 is different from @xmath141 , using the formula @xmath1017 where the meaning of @xmath1018 is that there is an edge labelled @xmath1016 from @xmath237 to @xmath171 , which is satisfied by exactly those vertices @xmath1014 labelled @xmath141 .
moreover , we define new edges .
first of all , we keep all edges @xmath1019 of @xmath212 such that @xmath85 is labelled in @xmath74 , or @xmath85 is labelled @xmath141 but @xmath188 is not .
each such edge retains its label .
second , whenever @xmath85 and @xmath188 are both labelled @xmath141 in @xmath212 we introduce an edge @xmath1020 and an edge @xmath1021 whose label is the same as that of the edge @xmath1019 in @xmath212 . for this purpose
, we use the formulas in the free variables @xmath1022 , @xmath1023 where @xmath1024 .
last , for each edge @xmath1019 of @xmath212 such that @xmath85 is labelled @xmath141 but @xmath188 is not , we introduce an edge @xmath1025 whose label is that of the edge @xmath1019 .
this is done by utilizing the formula @xmath1026 where again @xmath1024 .
note that the unfolding of the graph constructed above from the vertex @xmath230 is @xmath995 .
next , consider any vertex @xmath85 labelled @xmath141 together with all the paths originating in @xmath85 leading to a vertex labelled in @xmath454 . since @xmath15 is bounded by @xmath731 , each such path is simple and the number of such paths is at most @xmath731
let @xmath1027 denote the ( not necessarily different ) end vertices of these paths , ordered ` lexicographically ' .
we relabel @xmath85 by @xmath1003 and introduce a new edge @xmath1028 labelled @xmath292 for each @xmath292 .
this is accomplished by using the following formulas .
let @xmath1029 denote a formula that says that the set of vertices @xmath8 forms a path from @xmath171 to @xmath237 , the label of each vertex in @xmath8 other than @xmath237 is different from @xmath141 , and the label of @xmath237 is in @xmath454 .
then for each @xmath5 , the formula @xmath1030 expresses that there are at least @xmath5 different paths from @xmath171 to some vertex @xmath237 labelled in @xmath472 , all of whose vertices different from @xmath237 are labelled @xmath141 . with the help of these formulas
we can also express that there are exactly @xmath5 such paths from @xmath171 .
finally , when @xmath1029 and @xmath1031 with @xmath1032 , we can express the fact that @xmath8 is lexicographically less than @xmath1033 by the formula @xmath1034 in addition to these new edges , we keep all edges originating from a vertex labelled in @xmath74 ( that are necessarily labelled @xmath1 ) .
all vertices of the form @xmath1014 become inaccessible from @xmath230 .
the unfolding of the new graph from vertex @xmath230 is almost an alternating term . in order to make it alternating , we have to add a new root labelled @xmath427 if @xmath230 is labelled in @xmath74 together with an edge to @xmath230 , and replace each edge @xmath1019 where both @xmath85 and @xmath188 are labelled in @xmath74 by new edges @xmath1035 and @xmath1036 , where @xmath86 is a new vertex labelled @xmath427 .
these edges are labelled @xmath1 .
the new graph is still obtained by monadic transduction , and its unfolding is the alternating term @xmath993 .
the same argument works in the regular case using the fact that regular terms are unfoldings of finite ( deterministic ) graphs , and that finite graphs are closed under monadic transduction .
_ proof of theorem [ thm - dcfl ] .
_ suppose that @xmath542 is bounded .
if @xmath16 is algebraic , then by lemma [ lem - b ] there is some determinization @xmath15 of @xmath16 such that @xmath974 is algebraic . by courcelle
s theorem , the branch language of @xmath974 is a dcfl .
but the branch language of @xmath974 is essentially @xmath954 .
suppose now that @xmath16 has a determinization @xmath15 such that @xmath954 is a dcfl .
then the branch language of @xmath974 is a dcfl , and thus by courcelle s theorem , @xmath974 is algebraic . the proof is completed by using lemma [ lem - a ] .
a similar reasoning applies in the regular case .
the language - theoretic characterization of the class of bounded algebraic synchronization trees offered in theorem [ thm - dcfl ] can be used to prove that certain trees are _ not _ algebraic .
[ prop : bagnotalg ] the synchronization tree @xmath956 associated with the bag over a binary alphabet depicted on figure [ fig : bag ] is not algebraic , even up to language equivalence .
recall that the branch words corresponding to the nodes of any determinization of the tree @xmath956 have the form @xmath957 where @xmath958 , @xmath959 , @xmath960 for each @xmath961 $ ] , each @xmath962 $ ] for @xmath951 $ ] and @xmath963 is a word with the property that , in any of its prefixes , the number of occurrences of the letter @xmath12 is greater than or equal to the number of occurrences of the letter @xmath198 , and the number of occurrences of the letter @xmath202 is greater than or equal to the number of occurrences of the letter @xmath376 .
moreover , @xmath1037 if and only if @xmath965 and @xmath966 .
the words accepted by that lts are those that in addition satisfy that the total number of occurrences of the letter @xmath12 in @xmath963 is equal to the number of occurrences of the letter @xmath198 , and the number of occurrences of the letter @xmath202 in @xmath963 is equal to the number of occurrences of the letter @xmath376 and which end in @xmath17 . if the language associated with any determinization of @xmath956 were context - free , then so would the language obtained by applying to each word in it the morphism that erases the letters @xmath1038 and renames each @xmath952 to @xmath1039 . however , that language is not context free .
therefore , theorem [ thm - dcfl ] yields that @xmath956 is not algebraic .
the above proposition is a strengthening of a classic result from the literature on process algebra proved by bergstra and klop in @xcite .
indeed , in theorem 4.1 in @xcite , bergstra and klop showed that the bag over a domain of values that contains at least two elements is not expressible in bpa .
moreover , by proposition [ prop : moreexpthanbpa ] , the collection of synchronization trees that are definable in bpa is strictly included in the set of @xmath10-algebraic ones ( theorem [ thm : delta2gamma ] ) .
therefore , proposition [ prop : bagnotalg ] is stronger than the above - mentioned inexpressibility result by bergstra and klop , and offers an alternative proof for it .
up to bisimilarity , we shall offer an even stronger statement in section [ sect : mso - bag ] ( see proposition [ prop : bag - bisim ] ) . as a final result , we can use theorem [ thm - dcfl ] to characterize the synchronization trees of bounded degree in @xmath746 .
the synchronization trees of bounded - degree in @xmath1040 are the @xmath10-algebraic synchronization trees of bounded - degree .
the direct inclusion is immediate . for the converse inclusion , we know by theorem [ thm - dcfl ] that a @xmath10-algebraic synchronization tree of bounded - degree @xmath16 has a determinization @xmath15 whose branch language is a deterministic context - free language .
in particular , this implies that @xmath15 belongs to @xmath746 . indeed from a deterministic pushdown automaton accepting the branch language of @xmath15
, we can construct a deterministic graph @xmath936 whose unfolding from some root @xmath230 is isomorphic to @xmath15 .
let @xmath1041 be the alphabet labelling @xmath212 .
consider the transduction @xmath925 defined as follows : * for every vertex @xmath86 , the transduction introduces new vertices @xmath928 ; * for every edge from @xmath86 to @xmath85 labelled by @xmath1042 , the transduction adds edges from each of the @xmath1043 , @xmath1044 $ ] , to @xmath1045 labelled by @xmath12 . unfolding
the graph @xmath938 from any of the vertices added by the transduction @xmath925 for the root @xmath230 gives a tree isomorphic to @xmath16 .
note that in the previous proof , @xmath925 can not simply rename the edges labelled @xmath1046 to @xmath12 .
consider for instance the case where @xmath212 consists of two vertices @xmath230 and @xmath156 and two edges @xmath1047 and @xmath1048 .
applying such a transduction would lead to a graph with only one edge .
in this section , we investigate the consequences of the decidability of monadic second - order logic for our synchronization trees . in this subsection we point out that the synchronization tree associated with the bag process depicted on figure [ fig : bag ] has an undecidable monadic theory ( even without the root being the source of an exit edge ) .
hence that tree is not in the caucal hierarchy and therefore , by proposition [ prop : delta - alg - in - graph3 ] , is not @xmath13-algebraic not even up to bisimilarity ( proposition [ prop : bag - bisim ] ) .
a similar result was obtained in ( * ? ? ?
* section 6.6.2 ) for a slightly richer structure . for completeness sake
, we give below a detailed proof of this result .
the synchronization tree @xmath1049 associated with the bag process has an undecidable mso - theory .
consider a @xmath2-counter machine whose program @xmath1050 is given as a sequence of instructions @xmath1051 where each @xmath1052 has one of the following forms : @xmath1053 where @xmath1013 is one of the counters @xmath1022 and @xmath1054 are integers between @xmath1 and @xmath5 . at any moment of time , the value of the counters @xmath171 and @xmath237 is described by an ordered pair @xmath1055 of non - negative integers .
the meaning of the above instructions is standard , where @xmath731 denotes the index of the instruction to be performed after execution of the given instruction , if the instruction is an increment or decrement , and @xmath1056 and @xmath1057 denote the indices of the instructions to be performed depending on the outcome of the test , if the instruction is of the last form . the machine with program @xmath1050 halts if a decrement instruction @xmath1058 is executed , but the current value of the counter @xmath1013 is @xmath17
. a well - known undecidable question for @xmath2-counter machines asks whether a @xmath2-counter machine started with @xmath1059 ever halts .
we encode the halting program for a counter machine with program @xmath1050 in monadic second order logic as follows .
let @xmath1060 be a set of variables associated with the instructions of @xmath1050 .
then , for each instruction @xmath1061 , we consider the formula @xmath1062 in the free first - order variable @xmath86 in the language with four binary predicates associated with the edge relations @xmath1063 , @xmath1064 . *
if @xmath1061 is of the form @xmath1065 , then @xmath1062 expresses that @xmath1066 holds for all @xmath85 such that @xmath1067 , where @xmath199 is @xmath12 if @xmath1068 and @xmath202 if @xmath1069 . *
if @xmath1061 is of the form @xmath1058 , then @xmath1062 expresses that there exists a @xmath85 with @xmath1067 , and @xmath1066 holds for all such @xmath85 , where @xmath199 is @xmath198 if @xmath1068 and @xmath376 if @xmath1069 . *
if @xmath1061 is of the form @xmath1070 , then @xmath1062 expresses that @xmath1071 and @xmath1072 hold for all @xmath1073 such that @xmath1074 and @xmath1075 , provided that there is no @xmath85 with @xmath1076 , and that @xmath1071 and @xmath1077 hold for all such @xmath1073 otherwise , where again @xmath199 is @xmath198 if @xmath1068 and @xmath376 if @xmath1069 .
now we assign to the machine with program @xmath1050 the formula @xmath1078\ ] ] where @xmath1079 asserts that the @xmath1080 are pairwise disjoint and jointly form an infinite path starting with the root , and @xmath1081 says that the root belongs to @xmath1082 .
then the machine does not halt iff the synchronization tree defined by the process on figure [ fig : bag ] is a model of @xmath1083 .
thomas showed in ( * ? ? ?
* theorem 10 ) that the monadic second - order theory of the infinite two - dimensional grid is undecidable .
however , we can not use that result to prove that the synchronization tree @xmath956 has an undecidable monadic second - order theory .
indeed , the unfolding of the infinite two - dimensional grid is the full binary tree , which has a decidable monadic second - order theory by rabin s celebrated tree theorem @xcite .
[ prop : bag - bisim ] the synchronization tree @xmath956 is not @xmath13-algebraic up to bisimilarity .
this statement follows from the more general remark that any synchronization tree @xmath16 that is bisimilar to a deterministic synchronization tree @xmath1084 having an undecidable mso - theory also has an undecidable mso - theory . given a formula @xmath239 with no free variables ,
consider the formula @xmath1085 where @xmath1086 is the formula @xmath239 in which all quantifications are relativized to @xmath8 and @xmath1087 states that if a node has a successor by an @xmath12-labelled edge then it has one and only one successor by an @xmath12-labelled edge which belongs to @xmath8 . if the formula @xmath1087 is satisfied on @xmath16 for some set of vertices @xmath1088 then @xmath16 restricted to the vertices in @xmath1088 is a deterministic tree isomorphic to @xmath1084 ( cf .
lemma [ lem : bisim - det - iso ] on page ) .
clearly @xmath1089 holds on @xmath16 if and only if @xmath239 holds on @xmath1084 .
this implies that the mso - theory of @xmath16 is undecidable . the argument used in this proof
can also be used to show that @xmath1049 does not belong to the caucal hierarchy up to bisimilarity .
it is well known that , for each bisimulation equivalence class @xmath1090 of synchronization trees in @xmath92 , there is a tree @xmath1091 such that for all @xmath1092 there is a _ surjective _
morphism @xmath1093 , and , moreover , the relation @xmath1094 is a bisimulation between @xmath16 and @xmath1095 .
furthermore , @xmath1095 is unique up to isomorphism . when @xmath1096 , we call @xmath1095 the _ minimization _ of @xmath16
. the minimization of a tree @xmath982 can be constructed in the following way .
we define an increasing sequence @xmath1097 of sets of vertices of @xmath16 , where @xmath1098 is a singleton set containing only the root of @xmath16 .
the construction will guarantee that , for each @xmath292 , the set @xmath1099 contains only vertices of depth at most @xmath292 , and whenever a vertex @xmath85 belongs to @xmath1099 and there is a path from a vertex @xmath86 to @xmath85 , then @xmath86 is also in @xmath1099 . given @xmath1099 and @xmath1100 of depth @xmath292 ,
consider the set @xmath1101 of successors of @xmath86 .
we may divide @xmath1101 into equivalence classes according to the bisimulation equivalence classes of the corresponding subtrees and the label of the edge coming from @xmath86 . to this end , we define @xmath1102 , for @xmath1103 , if the subtrees rooted at @xmath85 and @xmath188 are bisimilar and if @xmath211 and @xmath1104 for some @xmath143 . then we select a representative of each @xmath1105-equivalence class .
the set @xmath1106 consists of all vertices in @xmath1099 together with those vertices in @xmath1101 of depth @xmath1107 selected above , where @xmath86 ranges over the set of all vertices in @xmath1099 of depth @xmath292 .
finally , let @xmath1108 .
the subtree " of @xmath16 spanned by the vertices in @xmath75 is the minimization of @xmath16 .
it is known , see e.g. @xcite , that the minimization of a @xmath10-regular synchronization tree is @xmath10-regular .
in contrast with this result , we have : [ prop : quotientnotcaucal ] there exists a @xmath10-algebraic synchronization tree whose minimization does not have a decidable mso - theory , and hence does not belong to the caucal hierarchy and is neither a @xmath10-algebraic nor a @xmath13-algebraic synchronization tree .
let @xmath1109 .
consider the following @xmath10-algebraic scheme : @xmath1110 let @xmath16 be the synchronization tree defined the above scheme .
for a word @xmath1111 , we designate by @xmath1112 ( resp .
@xmath1113 ) the number of @xmath12 s ( resp . @xmath198
s ) in the longest suffix of @xmath86 that does _ not _ contain any occurrence of the letter @xmath202 ( resp .
@xmath376 ) .
intuitively , the tree @xmath16 consists of a full quaternary deterministic tree @xmath16 with edges labelled in @xmath1114 such that every vertex of @xmath16 is also the origin of two additional parallel disjoint branches .
since @xmath16 is deterministic , we may identify each vertex of @xmath16 with a word @xmath1115 .
the two additional branches at vertex @xmath86 of @xmath16 are such that their edge labels form the words @xmath1116 and @xmath1117 , respectively .
the minimization @xmath15 of @xmath16 is obtained by removing one of the two branches labelled @xmath1116 for all vertices @xmath86 of @xmath16 such that @xmath1118 .
the fact that @xmath15 has an undecidable mso - theory is based on a reduction from the halting problem for 2-counter machines with increment , reset and equality test @xcite .
the idea is , as usual , to define , for every such machine @xmath1119 , a closed mso - formula @xmath1120 such that @xmath1121 if and only if @xmath1119 does not halt .
when constructing the formula @xmath1122 , we associate to a vertex @xmath86 in @xmath1123 representing a possible history of the computation of @xmath1119 from its initial state , the integer @xmath1112 as the current value of the first counter , and the integer @xmath1113 as the current value of the second counter of the machine . assuming that the current values of the counters are given by vertex @xmath86
, we show how to simulate the various operations of the machine .
the increment of the first ( resp .
second ) counter is obtained by moving to vertex @xmath1124 ( resp .
@xmath1125 ) .
the reset of the first ( resp .
second ) counter is obtained by moving to @xmath1126 ( resp .
@xmath1127 ) .
finally , the test for equality between the two counters is performed by testing that vertex @xmath86 has only one outgoing edge labelled @xmath199 .
the details of the construction are similar to those in section [ sect : mso - bag ] . the above result yields that the collection of synchronization trees in the caucal hierarchy is _ not _ closed under minimization .
indeed , there is a @xmath10-algebraic tree whose minimization is not in the caucal hierarchy .
this leaves open the corresponding question for @xmath13-regular synchronization trees .
there are several questions that we leave open in this paper and that lead to interesting directions for future research .
* is there a strict expressiveness hierarchy for @xmath10- and @xmath13-algebraic recursion schemes that is induced by the maximum rank of the functor variables used in defining recursion schemes ? * is the minimal synchronization tree that is bisimilar to a @xmath13-regular synchronization tree also @xmath13-regular ? if not ,
is it in the caucal hierarchy ? * the @xmath10-algebraic scheme we use in the proof of proposition [ prop : quotientnotcaucal ] uses a binary functor variable . is the minimal synchronization tree that is bisimilar to a tree defined by a @xmath10-algebraic scheme involving only unary functor variables @xmath10-algebraic ?
r. milner .
an algebraic definition of simulation between programs . in proceedings
2nd joint conference on artificial intelligence , pages 481489 .
bcs , 1971 . also available as report no .
cs-205 , computer science department , stanford university .
w. thomas . constructing infinite graphs with a decidable mso - theory . in proceedings of the 28th international symposium on mathematical foundations of computer science , lncs 2747 , pp . 113124 , springer , 2003 . | we study algebraic synchronization trees , i.e. , initial solutions of algebraic recursion schemes over the continuous categorical algebra of synchronization trees .
in particular , we investigate the relative expressive power of algebraic recursion schemes over two signatures , which are based on those for basic ccs and basic process algebra , as a means for defining synchronization trees up to isomorphism as well as modulo bisimilarity and language equivalence .
the expressiveness of algebraic recursion schemes is also compared to that of the low levels in caucal s pushdown hierarchy .
[ multiblock footnote omitted ] |
the search for topological states of quantum matter has generated intensive interest in condensed matter physics @xcite .
recently , the quantum spin hall ( qsh ) state in two dimensions ( 2d ) and topological insulators in three dimensions ( 3d ) have been theoretically predicted and experimentally observed in a number of materials @xcite , and both of them are characterized by the @xmath2 topological indices @xcite .
soon afterwards , the concept of time - reversal - invariant ( tri ) topological superconductors has been proposed @xcite . closely related to qsh state and topological insulators ,
the @xmath0 and @xmath1 tri topological superconducting state has a full pairing gap in the bulk , and gapless majorana states at the edge and on the surface , respectively , which have potential applications in fault - tolerant topological quantum computation @xcite .
moreover , an emergent supersymmetry is naturally present in these systems as a consequence of the time - reversal symmetry @xcite .
great efforts have been made to search for topological superconductors , however , finding candidate materials for these new topological phases of matter is still challenging . a simple and general criterion has been proposed to test for tri topological superconductor based on the pairing amplitude on the fermi surface @xcite .
a @xmath0 tri superconductor is nontrivial if there are an
_ odd _ number of fermi surfaces with a _ negative _ pairing order parameter @xcite .
this physical criteria suggests to search for topological superconductors in nonconventional superconducting materials with inversion symmetry breaking @xcite and strong correlation @xcite .
recently , superconductivity has been realized in a doped topological insulator cu@xmath3bi@xmath4se@xmath5 @xcite .
such material has been proposed to be a @xmath1 topological superconductor , where a novel spin - triplet pairing with odd parity is favored by strong spin - orbit coupling ( soc ) based on a two - orbital model @xcite . however , the pairing symmetry in doped bi@xmath4se@xmath5 is still under active debates @xcite . on the other hand ,
the @xmath0 tri topological superconductor has not yet been discovered .
there are some theoretical discussions on possible tri topological superconductivity in noncentrosymmetric superconductors with the rashba spin splitting @xcite .
the key point here is for the spin split bands , one is paired into ( @xmath6 ) state , the other is paired into ( @xmath7 ) state .
however , to realize tri topological superconductor , the spin triplet @xmath8-wave pairing should be dominant over spin singlet @xmath9-wave pairing in the two spin split bands @xcite
. doped band insulators with strong soc may be good candidate materials in realizing an exotic pairing @xcite .
qsh effect has been realized in heterostructures @xcite , these systems have the advantage of great controllability of structure , doping , symmetry , and soc . in this paper
, we theoretically propose topological superconductivity in a doped qsh insulator .
we study the pairing symmetry of the newly predicted qsh insulator in decorated stanene films sn@xmath10
( @xmath11 -oh , -f , -cl , -br , and -i ) @xcite within a two - orbital model for its band structure .
when the interorbital attractive interaction is stronger than the intraorbital interaction , the resulting state realizes a topological superconductor .
we explicitly calculate the majorana edge spectrum in a tight - binding model .
finally , we discuss the robustness of topological superconductivity in doped sn@xmath10 films by comparing to @xmath1 superconductivity in bulk @xmath12-tin .
the organization of this paper is as follows .
after this introductory section , sec .
ii describes the effective two - orbital model for the superconductivity in a doped qsh insulator .
section iii presents the results on the pairing symmetry , phase diagram and edge state .
section iv presents discussion and possible experimental realization of topological superconductivity in tin film .
study of superconductivity in a doped qsh insulator requires the knowledge of its band structure and pairing mechanism .
the general results presented in this paper is generic for any doped qsh insulators . here
, we would like to start from a simple model describing the 2d qsh insulator sn@xmath10 for concreteness @xcite .
in fact , 3d @xmath12-sn was one of the first superconductors to be studied experimentally , with the critical transition temperature @xmath13 k. therefore , it is likely that doped sn@xmath10 film is also a superconductor . in the following ,
we assume that 2d sn@xmath10 is superconducting , and study under what condition it would also be a @xmath0 tri topological superconductor .
we leave the discussion on pairing mechanism to the end . a low - buckled geometry for 2d sni is shown in fig . [ fig1](a ) , where they have a stable @xmath14 configuration analogous to graphane .
the lattice symmetry is @xmath15 . as shown in fig .
[ fig1](b ) by first - principles calculations , the band structure of ag@xmath3sni is similar to its parent compound sni .
the low - energy bands of sni consist of an antibonding state of @xmath9 orbital and a bonding state of @xmath16 orbitals , labeled by @xmath17 and @xmath18 , respectively , which is similar as for hgte quantum well @xcite .
the effective hamiltonian describing these four bands near the @xmath19 point is given by the model of bernevig , hughes and zhang @xcite : @xmath20 here , @xmath21 and @xmath22 ( @xmath23 ) are pauli matrices acting spin and orbital , respectively . to the lowest order in @xmath24 , @xmath25 and @xmath26 account for the particle - hole asymmetry @xcite .
@xmath27 and @xmath28 guarantee that the system is in the inverted regime . the basis of eq .
( [ model ] ) is @xmath29 , @xmath30 , @xmath31 , @xmath32 , and the @xmath33 in the basis stand for the even and odd parity and @xmath34 , @xmath35 represent spin up and down states , respectively . with the chemical doping by ag adatom or electrical gating ,
the lattice symmetry is reduced to @xmath36 .
therefore , additional rashba terms will be added into effective hamiltonian due to inversion symmetry breaking @xcite . to the lowest order ,
the only possible term is @xmath37 where @xmath38 determines the strength of spin splitting . the effective model for ag@xmath3sni is given by @xmath39 .
the band structure is plotted in fig .
[ fig1](b ) , and the bands show small spin splitting which can be tuned to be large by gating . because of ag doping , the fermi energy @xmath40 lies in the conduction band approximately 0.31 ev above the band edge , which leads to a small fermi surface respecting full rotation symmetry around the @xmath41 axis .
[ fig1](c ) shows the linear relation between chemical potential and doping concentration , matches well with the parabolic band structure in 2d . as for the fermion pairing ,
we consider the following short - range density - density interactions , @xmath42 - 2vn_1(\mathbf{x})n_2(\mathbf{x}),\ ] ] where @xmath43 is the electron density in orbital @xmath44 .
@xmath45 represent @xmath46 and @xmath47 , respectively . @xmath48 and @xmath49 are effective intraorbital and interorbital interaction , respectively .
we assume that at least one of them is attractive , say due to phonons as in the case of superconductivity of 3d tin . the two - orbital @xmath48-@xmath49 model for 2d
ag@xmath3sni is @xmath50c_{\mathbf{k}}+\int d\mathbf{r}\mathcal{h}_{\text{int}}(\mathbf{r}).\end{aligned}\ ] ] in the following , we shall apply the criterion of ref . to investigate topological superconductivity in this noncentrosymmetric model .
to determine the superconducting phase diagram of the @xmath48-@xmath49 model , we construct the bogoliubov - de gennes ( bdg ) hamiltonian with mean - field approximation @xmath51\xi_{\mathbf{k}}.\ ] ] here @xmath52 are pauli matrices in nambu space and the basis @xmath53(@xmath54 ) .
the low energy physics with a small fermi surface has full rotation symmetry around the @xmath41 axis @xmath55 instead of 3-fold rotation symmetry @xmath56 of the point group @xmath36 in the lattice .
we classify all possible pairing potential @xmath57 according to time - reversal symmetry @xmath58 with @xmath59 complex conjugation , and @xmath60 with @xmath61 . in the weak coupling limit with purely short - range interaction ,
the mean - field pairing potential is @xmath62 independent . in table
[ table1 ] , only 6 forms can have nonzero values among the 16 possible products of @xmath63 and @xmath64 .
we find three different pairing symmetries with angular momentum @xmath65 under @xmath55 .
the form of the corresponding pairing order parameter @xmath66 , @xmath67 is shown explicitly @xmath68 @xmath69 is a spin - singlet , whereas @xmath70 and @xmath71 are interorbital spin - triplets .
the symmetry properties of @xmath66 are shown in table [ table1 ] .
[ table1 ] 3.3 in @cccc @xmath72 & matrix & @xmath55 & @xmath73 + @xmath69 & @xmath74 , @xmath75 & @xmath76 & @xmath77 + @xmath70 & @xmath78 & @xmath79 & @xmath77 + @xmath71 & @xmath80 & @xmath81 & @xmath77 + the excitation energy of quasiparticle are obtained by diagonalizing the bdg hamiltonian eq .
( [ bdg ] ) with fixing the pairing potential to each @xmath66 .
we find superconducting gap for @xmath70 has point nodes ( in the @xmath82-direction when one choose @xmath83 ) , and the others have full gap . to obtain the phase diagram
, we estimate the superconducting critical temperature @xmath84 by analyzing superconducting susceptibility for each pairing potentials . the standard pairing susceptibility @xmath85
is defined as @xmath86 $ ] , with @xmath87^{-1}$ ] the matsubara green function .
all other susceptibilities @xmath88 , @xmath89 , and @xmath90 can be obtained by replacing @xmath91 with their corresponding pairing potential @xmath92 , @xmath93 ( or @xmath94 ) , and @xmath95 ( or @xmath96 ) in table [ table1 ] .
a straightforward calculation shows that they can be expressed by @xmath85 , which contains the logarithmic divergence at the fermi surface .
the linearized gap equations for @xmath84 in each pairing channel are as @xmath97=0,\\ \delta_{2,3}:&\ \ v\chi_{2,3}(t_c)=1 . \end{aligned}\ ] ] using the band structure of @xmath98 , we can calculate the phase diagram numerically . in the limit of @xmath99 , we obtain the values of @xmath100 s analytically : @xmath101 , @xmath102 is the density of states .
@xmath103 , @xmath104 , @xmath105\chi_0 $ ] .
because @xmath106 , we find that @xmath70 always has a lower @xmath84 than @xmath71 . from the highest @xmath84
, only @xmath69 and @xmath71 appear in the phase diagram . by calculating their @xmath84 s from ( [ gapequation ] )
, we obtain the phase boundary @xmath107 fig .
[ fig2 ] shows the phase diagram as a function @xmath108 and @xmath109 , for positive ( attractive ) @xmath49 .
a significant part of the phase diagram is the @xmath71 phase , especially for the inversion symmetry breaking @xmath110 .
-@xmath49 model , showing highest @xmath84 phase as a function of @xmath109 and @xmath108 .
@xmath70 can not be leading instability in this phase diagram , for all the pairing states in @xmath111 are fully gapped .
solid ( blue ) and dashed ( red ) line are phase boundary for @xmath110 and @xmath112 , respectively.,width=316 ] next we consider the topological nature of the pairing state .
the present system belongs to the symmetry class @xmath113 in 2d @xcite , which is characterized by a @xmath2 topological invariant in contrast to @xmath114 in 3d .
since the system has @xmath73 symmetry without inversion symmetry , the criteria in ref . can be applied .
namely , a tri superconductor is a topological superconductor if ( 1 ) it has a full superconducting gap _ and _ ( 2 ) there are an odd number of fermi surfaces each of which encloses one tri point ( which satisfy @xmath115 up to a reciprocal lattice vector ) in the brillouin zone and has negative pairing .
the @xmath2 invariant is @xmath116^{m_j}.\end{aligned}\ ] ] here @xmath117 labels the fermi surface , @xmath118 is the number of the tri points enclosed by the @xmath117th fermi surface , @xmath119 $ ] denotes the sign of the paring amplitude of the @xmath117th fermi surface , @xmath120 is the eigenvectors of @xmath121 . here in our system , the @xmath71 pairing has opposite sign on the two fermi surfaces which gives rise to a topological superconductor phase , while @xmath69 has the same sign as shown in fig .
if we take the limit @xmath99 , the two fermi surfaces become degenerate , still the only odd - parity pairing @xmath71 phase is topological @xcite . as shown in fig .
[ fig2 ] , the topological superconductor phase @xmath71 is more favorable when inversion symmetry is breaking .
we confirm the system is exactly in the topological phase under such conditions . to obtain the topological protected gapless edge states ,
we solve the following tight - binding model describing continuous model eq .
( [ bdg ] ) in the low energy regime , @xmath122 , \nonumber\end{aligned}\ ] ] where @xmath123 denotes the nearest - neighbor site .
the hopping parameters @xmath124 , @xmath125 $ ] , and @xmath126 $ ] .
we consider the @xmath71 pairing state in the cylindrical geometry with periodic boundary condition in the @xmath127 direction and open one in the @xmath128 direction .
the energy spectrum of this model is shown in fig .
[ fig3](c ) .
one can see that there are helical majorana states crossing the bulk superconducting gap , where the right - going and left - going states are spin splitting ( very small ) , and localized at opposite edges .
therefore , nontrivial @xmath2 number in the bulk will lead to helical states at the edge . however , there are no edge states with @xmath69 pairing as in fig .
[ fig3](d ) , which are consistent with the previous study on bulk topological invariant .
topological number , fermi surface pairing amplitude and edge energy spectrum .
( a ) @xmath129 for @xmath71 pairing , and ( c ) edge spectrum shows that helical edge modes appears at each edge of the sample in the superconducting gap ; while ( b ) @xmath130 for @xmath69 pairing , and ( d ) no edge states .
all parameters are taken from ref .
finally , we discuss the robustness of topological superconductivity obtained from the two - orbital @xmath48-@xmath49 model and the possible pairing mechanism . as the phase diagram shows , the _ interorbital _ spin - triplet @xmath71 phase wins as long as the interorbital interaction exceeds over the intraorbital one ( @xmath131 ) .
this arises from the specific form of soc in the band structure , which favors @xmath71 pairing . also , inversion symmetry breaking and _ multi_-orbital system would result in more unconventional pairing .
therefore , doped qsh insulators with strong soc would offer better way to find topological superconductors .
the realistic value of @xmath48 and @xmath49 in tin films are difficult to estimate . nonetheless , the @xmath1 superconductivity in bulk @xmath12-sn is @xmath9-wave pairing from @xmath9 orbit via phonon - mediated mechanism .
it is likely that 2d sn@xmath10 is also a superconductor with a phonon driven pairing mechanism .
the bare value of attractive @xmath132 and @xmath133 are given by the electron - phonon coupling @xmath134 as @xmath135 .
such electron - phonon coupling strength @xmath134 in 2d can be modulated and could even be larger than that in 3d @xcite . with 2d sn@xmath10 film on different insulating substrates such as cdte or inas , the strain from substrate can cause expansion and shrinkage of interlayer spacing , therefore the phonon spectrum can be greatly modulated .
the topological property persists in sn@xmath10 with lattice constant mismatch from @xmath136 to @xmath137 @xcite .
thus , one could maximize @xmath134 without changing the topological properties of sn@xmath10 .
in reality , @xmath132 is usually larger than @xmath133 , and results in a largest energy gain by forming the @xmath9-wave pairing .
however , the effective interaction given by @xmath48 and @xmath49 should include effects of the coulomb interaction and other possible renormalizations .
the coulomb repulsion @xmath138 and @xmath139 renormalizes the bare value of @xmath132 and @xmath133 , respectively .
@xmath140 due to smaller interorbital wavefunction overlap , and weaker coulomb screening in 2d makes @xmath138 larger than that in 3d .
the effective interaction parameters are given by @xmath141 , and @xmath142 .
therefore the stronger intraorbital repulsion would lead to @xmath143 .
in summary , we have studied topological superconductivity in a doped qsh insulator and propose 2d doped sn@xmath10 as a potential candidate .
a wealth of qsh insulating materials could lead to the discovery of the tri topological superconductor , which supports the existence of majorana edge states and majorana zero energy modes in vortex cores .
we hope the theoretical work here could aid the search for topological superconductor phases in real materials .
this work is supported by the us department of energy , office of basic energy sciences , division of materials sciences and engineering , under contract no .
de - ac02 - 76sf00515 and the defense advanced research projects agency microsystems technology office , mesodynamic architecture program ( meso ) through the contract no .
n66001 - 11 - 1 - 4105 .
33ifxundefined [ 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 link:\doibase 10.1063/1.3293411 [ * * , ( ) ] link:\doibase 10.1103/revmodphys.82.3045 [ * * , ( ) ] link:\doibase 10.1103/revmodphys.83.1057 [ * * , ( ) ] link:\doibase 10.1038/nature08916 [ * * , ( ) ] @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) link:\doibase 10.1038/nphys1274 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.95.146802 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.98.106803 [ * * , ( ) ] link:\doibase 10.1103/physrevb.78.195424 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.102.187001 [ * * , ( ) ] link:\doibase 10.1103/physrevb.78.195125 [ * * , ( ) ] @noop _ _ , aip conf .
1134 ( aip , new york , 2009 ) , p. 22 .
link:\doibase 10.1103/revmodphys.80.1083 [ * * , ( ) ] link:\doibase 10.1103/physrevb.81.134508 [ * * , ( ) ] link:\doibase 10.1103/physrevb.34.8190 [ * * , ( ) ] link:\doibase
10.1103/physrevlett.104.057001 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.105.097001 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.107.217001 [ * * , ( ) ] link:\doibase 10.1103/physrevb.86.064517 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.108.057001 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.110.117001 [ * * , ( ) ] link:\doibase 10.1103/physrevb.79.060505 [ * * , ( ) ] link:\doibase 10.1103/physrevb.79.094504 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.108.147003 [ * * , ( ) ] @noop * * , ( ) link:\doibase 10.1103/physrevlett.111.136804 [ * * , ( ) ] @noop @noop * * , ( ) link:\doibase 10.1103/physrevb.79.214526 [ * * , ( ) ] link:\doibase 10.1103/physrevb.81.220504 [ * * , ( ) ] link:\doibase 10.1126/science.1105130 [ * * , ( ) ] | time - reversal - invariant topological superconductors have a full paring gap in the bulk and gapless majorana states at the edge or on the surface . here
, we theoretically propose topological superconductivity in a doped quantum spin hall insulator .
we study the pairing symmetry of a doped two - dimensional tin film within a two - orbital model , and find that a novel spin - triplet pairing is favored when the interorbital attractive interaction is stronger than the intraorbital interaction .
we propose that a doped tin film is a good candidate for a @xmath0 topological superconductor .
edge channels are studied in a tight - binding model numerically .
finally , we discuss the robustness of topological superconductivity in two - dimensional tin films by comparing to @xmath1 superconductivity in bulk tin . |
for many decades , statisticians have made attempts to prepare the bayesian omelette without breaking the bayesian eggs ; that is , to obtain probabilistic likelihood - based inferences without relying on informative prior distributions .
a recent example is murray aitkin s recent book , _ statistical inference _ , which is the culmination of a long research program on the topic of integrated evidence , exemplified by the discussion paper of @xcite .
the book , subtitled _ an integrated bayesian / likelihood approach _ , proposes handling statistical hypothesis testing and model selection via comparisons of posterior distributions of likelihood functions under the competing models or via the posterior distribution of the likelihood ratios corresponding to those models .
( the essence of the proposal is detailed in section [ small ? ] . ) instead of comparing bayes factors or performing posterior predictive checks ( comparing observed data to posterior replicated pseudo - datasets ) , _ statistical inference _ recommends a fusion between likelihood and bayesian paradigms that allows for the perpetuation of noninformative priors in testing settings where standard bayesian practice prohibits their usage @xcite or requires an extended decision - theoretic framework @xcite .
while we appreciate the considerable effort made by aitkin to place his theory within a bayesian framework , we remain unconvinced of the said coherence , for reasons exposed in this note .
l0.22 from our bayesian perspective , and for several distinct reasons detailed in the present note , integrated bayesian / likelihood inference can not fit within the philosophy of bayesian inference . aitkin
s commendable attempt at creating a framework that incorporate the use of arbitrary noninformative priors in model choice procedures is thus incoherent in this bayesian respect . when using improper priors lead to meaningless bayesian procedures for posterior model comparison , we see this as a sign that the bayesian model will not work for the problem at hand . rather than trying at all cost to keep the offending model and define marginal posterior probabilities by fiat ( whether by bic , dic , intrinsic bayes factors , or posterior likelihoods ) , we prefer to follow the full logic of bayesian inference and recognize that , when one s bayesian approach leads to a dead end , one must change either one s methodologies or one s beliefs ( or both ) .
bayesians , both subjective and objective , have long recognized the need for tuning , expanding , or otherwise altering a model in light of its predictions ( see , for example , @xcite and @xcite ) , and we view undefined bayes factors as an example where otherwise useful methods are being extended beyond their applicability .
to try to work around such problems without altering the prior distribution is , we believe , an abandonment of bayesian principles and , more importantly , an abandoned opportunity for model improvement .
the criticisms found in the current review are therefore not limited to aitkin s book ; they also apply to previous patches such as the deviance information criterion ( dic ) of @xcite ( which also uses a posterior " expectation of the log - likelihood ) and the pseudo - posteriors of @xcite(which make an extensive use of the data in their product of predictives ) .
unlike the author , who has felt the call to construct a partly new @xcite if tentatively unifying foundation for statistical inference , we have the luxury of feeling that we already live in a comfortable ( even if not flawless ) inferential house .
thus , we come to aitkin s book not with a perceived need to rebuild but rather with a view toward strengthening the potential shakiness of the pillars that support our own inferences .
a key question when looking at any method for probabilistic inference that is not fully bayesian is : for the applied problems that interest us , does the proposed new approach achieve better performances than our existing methods ?
our answer , to which we arrive after careful thought , is no . as an evaluation of the ideas found in _
statistical inference _ , the criticisms found in this review are inherently limited .
we do not claim here that aitkin s approach is wrong _ per se _ merely that it does not fit within our inferential methodology , namely bayesian statistics , despite using bayesian tools .
we acknowledge that statistical methods do not , and most likely never will , form a seamless logical structure
. it may thus very well be that the approach of comparing posterior distributions of likelihoods could be useful for some actual applications , and perhaps aitkin s book will inspire future researchers to demonstrate this .
_ statistical inference
_ begins with a crisp review of frequentist , likelihood and bayesian approaches to inference and then proceeds to the main issue : introducing the `` integrated bayes / likelihood approach '' , first described in chapter 2 .
much of the remaining methodological material appears in chapters 4 ( `` unified analysis of finite populations '' ) and 7 ( `` goodness of fit and model diagnostics '' ) .
the remaining chapters apply aitkin s principles to various examples .
the present article discusses the basic ideas in _ statistical inference _ , then consider the relevance of aitkin s methodology within the bayesian paradigm .
this quite small change to standard bayesian analysis allows a very general approach to a wide range of apparently different inference problems ; a particular advantage of the approach is that it can use the same noninformative priors . " _ statistical inference _ , page xiii the quite small change " advocated by _
statistical inference _ consists in envisioning the likelihood function @xmath0 as a generic function of the parameter @xmath1 that can be processed a posteriori ( that is , with a distribution induced by the posterior @xmath2 ) , hence allowing for ( posterior ) cdf , mean , variance and quantiles .
in particular , the central tool for aitkin s model fit is the posterior cdf " of the likelihood , @xmath3 as argued by the author ( chapter 2 , page 21 ) , this small change " in perspective has several appealing features : * the approach is general and allows to resolve the difficulties with the bayesian processing of point null hypotheses , being defined solely by the bayesian model associated with @xmath0 ; * the approach allows for the use of generic noninformative and improper priors , again by being relative to a single model ; * the approach handles more naturally the
vexed question of model fit " , still for the same reason ; * the approach is `` simple . '' as noted above , the setting is quite similar to spiegelhalter et al.s ( @xcite ) dic in that the deviance @xmath4 is a renaming of the likelihood and is considered a posteriori " both in @xmath5 $ ] and in @xmath6 , where @xmath7 is a bayesian estimator of @xmath1 , since @xmath8 the discussion of @xcite made this point clear , see in particular @xcite , even though the authors disagreed .
@xcite make a similarly ambiguous proposal that also relates to @xcite by its usage of cross - validation quantities .
we however dispute both the appropriateness and the magnitude of the change advocated in _ statistical inference _ and show below why , in our opinion , this shift in paradigm constitutes a new branch of statistical inference , differing from bayesian analysis on many points .
first , using priors and posteriors is no guarantee that inference is bayesian @xcite .
empirical bayes techniques are witnesses of this @xcite .
aitkin s key departure from bayesian principles means that his procedure has to be validated on its own , rather than benefiting from the coherence inherent to bayesian procedures .
the practical advantage of the likelihood / bayesian approach may be convenience , but the drawback is that the method pushes both the user and the statistician _ away _ from progress in model building . within a model , even while giving meaningless ( or at least not universally accepted ) values for marginal likelihoods that are needed for bayesian model comparison .
it does when interest shifts from @xmath9 to @xmath10 that the bayesian must set aside most of noninformative @xmath11 and , perhaps reluctantly , set up an informative model .
see , e.g. , @xcite and @xcite for some current perspectives on bayesian model choice using noninformative priors . ]
we envision bayesian data analysis as comprising three steps : ( 1 ) model building , ( 2 ) inference , and ( 3 ) model checking . in particular , we view steps ( 2 ) and ( 3 ) as separate .
inference works well , with many exciting developments still in the coming , handling complex models , leading to an unlimited range of applications , and a partial integration with classical approaches ( as in the empirical bayes work of @xcite , or more recently the similarities between hierarchical bayes and frequentist false discovery rates discussed by @xcite ) , causal inference , machine learning , and other aims and methods of statistical inference . even in the face of all this progress on inference ,
bayesian model checking remains a bit of an anomaly , with the three leading bayesian approaches being bayes factors , posterior predictive checks , and comparisons of models based on prediction error and other loss - based measures .
( decision - theoretic analyses as in @xcite , while intellectually convincing , have not gained the same amount of popularity . )
unfortunately , as aitkin points out , none of these model checking methods works completely smoothly : bayes factors depend on aspects of a model that are untestable and are commonly assigned arbitrarily ; posterior predictive checks are , in general , conservative " in the sense of producing @xmath12-values whose probability distributions are concentrated near @xmath13 ; and prediction error measures ( which include cross - validation and dic ) require the user to divide data into test and validation sets , lest they use the data twice ( a point discussed immediately below ) . the setting is even bleaker when trying to incorporate noninformative priors @xcite and new proposals are clearly of interest . a persistent criticism of the posterior likelihood approach ( ... )
has been based on the claim that these approaches are ` using the data twice , ' or are violating temporal coherence . " _ statistical inference _
, page 48 using the data twice " is not our main reservation about the method if only because this is a rather vague concept .
obviously , one could criticize the use of the posterior expectation " of the likelihod as being the ratio of the marginal of the twice replicated data over the marginal of the original data , @xmath14 = \int l(\theta , x ) \pi(\theta|x)\,\text{d}\theta = \dfrac{m(x , x)}{m(x)}\,,\ ] ] similar to @xcite ( a criticism clearly expressed in the discussion therein ) .
however , a more fundamental issue is that the posterior " distribution of the likelihood function can not be justified from a bayesian perspective . _ statistical inference _ stays away from decision - theory ( as stated on page xiv ) so there is no derivation based on a loss function or such .
our primary difficulty with the integrated likelihood idea ( and dic as well ) is ( a ) that the likelihood function does not exist a priori and ( b ) that it requires a joint distribution to be properly defined in the case of model comparison .
the case for ( a ) is arguable , as aitkin would presumably contest that there exists a joint distribution on the likelihood , even though the case of an improper prior stands out ( see below ) .
we still see the concept of a posterior probability that the likelihood ratio is larger than @xmath15 as meaningless . the case for ( b )
is more clear - cut in that when considering two models , hence a likelihood ratio , a bayesian analysis does require a joint distribution on the two sets of parameters to reach a decision , even though in the end only one set will be used .
as detailed below in section [ prodpost ] , this point is related with the introduction of pseudo - priors by @xcite who needed arbitrary defined prior distributions on the parameters that do not exist . in the specific case of an improper prior , aitkin s
approach can not be validated in a probability setting for the reason that there is no joint probability on @xmath16 .
obviously , one could always advance that the whole issue is irrelevant since improper priors do not stand within probability theory .
however , improper priors do stand within the bayesian framework , as demonstrated for instance by @xcite and it is easy to give those priors an exact meaning .
when the data are made of @xmath17 iid observations @xmath18 from @xmath19 and an improper prior @xmath20 is used on @xmath1 , we can consider a _ training sample _
@xcite @xmath21 , with @xmath22 such that @xmath23 if we construct a probability distribution on @xmath1 by @xmath24 the posterior distribution associated with this distribution and the remainder of the sample @xmath25 is given by @xmath26 this distribution is independent from the choice of the training sample ; it only depends on the likelihood of the whole data @xmath27 and it therefore leads to a non - ambiguous posterior distribution on @xmath1 .
however , as is well known , this construction does not lead to produce a joint distribution on @xmath28 , which would be required to give a meaning to aitkin s integrated likelihood .
therefore , his approach can not cover the case of improper priors within a probabilistic framework and thus fails to solve the very difficulty with noninformative priors it aimed at solving .
this is further illustrated by the use of haldane s prior in chapter 4 of _ statistical inference _ , despite it not allowing for empty cells in a contingency table @xcite .
`` the @xmath12-value is equal to the posterior probability that the likelihood ratio , for null hypothesis to alternative , is greater than 1 ( ... ) the posterior probability is @xmath12 that the posterior probability of @xmath29 is greater than 0.5 . '' _ statistical inference _
, pages 4243 those two equivalent statements show that it is difficult to give a bayesian interpretation to aitkin s method , since the two posterior probabilities " quoted above are incompatible . indeed ,
a fundamental bayesian property is that the posterior probability of an event related with the parameters of the model is not a random quantity but a number . to consider the
posterior probability of the posterior probability " means we are exiting the bayesian domain , both from logical and philosophical viewpoints . in chapter 2 , aitkin exposes his ( foundational ) reasons for choosing this new approach by integrated bayes / likelihood .
his criticism of bayes factors is based on several points we feel useful to reproduce here : 1 .
[ i ] have we really eliminated the uncertainty about the model parameters by integration ?
the integrated likelihood ( ... ) is the expected value of the likelihood .
but what of the prior variance of the likelihood ? "
( page 47 ) .
[ ii ] any expectation with respect to the prior implies that the data has not yet been observed ( ... ) so the integrated likelihood `` is the joint distribution of random variables drawn by a two - stage process .
( ... ) the marginal distribution of these random variables is not the same as the distribution of @xmath30 ( ... ) and does not bear on the question of the value of @xmath1 in that population '' ( page 47 ) .
[ iii ] we can not use an improper prior to compute the integrated likelihood .
this eliminate the usual improper noninformative priors widely used in posterior inference . "
( page 47 ) .
[ iv ] any parameters in the priors ( ... ) will affect the value of the integrated likelihood and this effect does not disappear with increasing sample size "
( page 47 ) .
[ v ] the bayes factor is equal to the posterior mean of the likelihood ratio between the models " _ [ meaning under the full model posterior ] _
( page 48 ) . 6 .
[ vi ] `` the bayes factor diverges as the prior becomes diffuse .
( ... ) this property of the bayes factor has been known since the lindley / bartlett paradox of 1957 '' ( page 48 ) .
the representation [ i ] of the integrated " ( or marginal ) likelihood as an expectation under the prior @xmath31\ ] ] is unassailable and is for instance used as a starting point for motivating the nested sampling method @xcite .
this does not imply that the extension to the variance or to any other moment stated in [ i ] has a similar meaning , nor that the move to the expectation under the posterior is valid within the bayesian paradigm . while the difficulty [ iii ] with improper priors is real , and while the impact of the prior modelling [ iv ] may have a lingering effect , the other points can be easily rejected on the ground that the posterior distribution of the likelihood is meaningless within a bayesian perspective .
this criticism is anticipated by aitkin who protests on pages 48 - 49 that , given point [ v ] , the posterior distribution must be meaningful , " since the posterior mean is meaningful " , but the interpretation of the bayes factor as a posterior mean " is only an interpretation of an existing integral ( in the specific case of nested models ) , it does not give any validation to the analysis .
( the marginal likelihood may similarly be interpreted as a prior mean , despite depending on the observation @xmath32 , as in the nested sampling perspective .
more generaly , bridge sampling techniques also exploit those multiple representations of a ratio of integrals , @xcite . )
one could just as well take [ ii ] above as an argument _ against _ the integrated likelihood / bayes perspective .
in the case of unrelated models to be compared , the fundamental theoretical argument against using posterior distributions of the likelihoods and of related terms is that the approach leads to parallel and separate simulations from the posteriors under each model . _ statistical inference
_ recommends that models be compared via the distribution of the likelihood ratio values , @xmath33 where the @xmath34 s and @xmath35 s are drawn from the respective posteriors .
this choice is similar to scott s ( @xcite ) and to congdon s ( @xcite ) mistaken solutions exposed in @xcite , in that mcmc simulations are run for each model separately and the resulting samples are then gathered together to produce either the posterior expectation ( in scott s , 2002 , case ) or the posterior distribution ( for the current paper ) of @xmath36 which do not correspond to genuine bayesian solutions ( see @xcite ) .
again , this is not as much because the dataset @xmath32 is used repeatedly in this process ( since reversible mcmc produces as well separate samples from the different posteriors ) as the fundamental lack of a common joint distribution that is needed in the bayesian framework .
this means , e.g. , that the integrated likelihood / bayes technology is producing samples from the product of the posteriors ( a product that clearly is not defined in a bayesian framework ) instead of using pseudo - priors as in @xcite , i.e. of considering a joint posterior on @xmath37 , which is [ proportional to ] @xmath38 this makes a difference in the outcome , as illustrated in figure [ fig : sxot ] , which compares the distribution of the likelihood ratio under the true posterior and under the product of posteriors , when assessing the fit of a poisson model against the fit of a binomial model with @xmath39 trials , for the observation @xmath40 .
the joint simulation produces a much more supportive argument in favor of the binomial model , when compared with the product of the posteriors .
( again , this is inherently the flaw found in the reasoning leading to scott s , 2002 , and congdon s , 2006 , methods for approximating bayes factors . ) _ comparison of the distribution of the likelihood ratio under the correct joint posterior and under the product of the model - based posteriors , when assessing a poisson model against a binomial with @xmath39 trials , for @xmath40 .
the joint simulation produces a much more supportive argument in favor of the negative binomial model , when compared with the product of the posteriors .
_ ] although we do not advocate its use , a bayesian version of aitkin s proposal can be constructed based on the following loss function that evaluates the estimation of the model index @xmath41 based on the values of the parameters under both models and on the observation @xmath32 : @xmath42 here @xmath43 means that model @xmath41 is chosen , and @xmath44 denotes the likelihood under model @xmath41 . under this loss , the bayes ( optimal )
solution is @xmath45 > 1/2\\ 2 & \text{otherwise , } \end{cases}\ ] ] which depends on the _ joint _ posterior distribution on @xmath37 , thus differs from aitkin s solution .
we have @xmath46 = & \pi(\mathcal m_1| x ) \int_{\theta_2 } \mbox{pr}^{\pi_1}\left [ l^1(\theta_1 ) > l^2(\theta_2)| x , \theta_2 \right ] \,\text{d}\pi_2(\theta_2)\\ & + \pi(\mathcal m_2| x ) \int_{\theta_1 } \mbox{pr}^{\pi_2}\left [ l^1(\theta_1 ) > l^2(\theta_2)| x , \theta_1\right ] \,\text{d}\pi_1(\theta_1)\,,\end{aligned}\ ] ] where @xmath47 and @xmath48 denote the log - likelihoods and where the probabilities within the integrals are computed under @xmath49 and @xmath50 , respectively .
( pseudo - priors as in @xcite could be used instead of the true priors , a requirement when at least one of those priors is improper . ) an asymptotic evaluation of the above procedure is possible : consider a sample of size @xmath17 , @xmath27 .
if @xmath51 is the true " model , then @xmath52 and we have @xmath53 & = \mbox{pr } \left [ -\mathcal x^2_{p_1 } > l^2(\theta_2)- l^2(\hat{\theta_1 } ) \right ] + o_p(1/\sqrt{n } ) \\ & = f_{p_1}\left [ l^1(\hat{\theta_1 } ) - l^2(\theta_2 ) \right ] + o_p(1/\sqrt{n})\,,\end{aligned}\ ] ] with obvious notations for the corresponding log - likelihoods , @xmath54 the dimension of @xmath55 , @xmath56 the maximum likelihood estimator of @xmath57 , and @xmath58 a chi - square random variable with @xmath54 degrees of freedom . note
also that , since @xmath59 , @xmath60 where @xmath61 denotes the kullback
leibler divergence and @xmath62 denotes the _ projection _ of the true model on @xmath63 : @xmath64 , we have @xmath65 = 1 + o_p(1)\,.\ ] ] by symmetry , the same asymptotic consistency occurs under model @xmath66 . on the opposite
, aitkin s approach leads ( at least in regular models ) to the approximation @xmath67,\ ] ] where the @xmath68 and @xmath69 random variables are independent , hence producing quite a different result that depends on the asymptotic behavior of the likelihood ratio .
note that for both approaches to be equivalent one would need a pseudo - prior for @xmath66 ( resp .
@xmath51 if @xmath70 were _ true _ ) as tight around the maximum likelihood as the posterior @xmath71 , which would be equivalent to some kind of empirical bayes type of procedure .
furthermore , in the case of embedded models , @xmath66 and @xmath72 , aitkin s approach can be given a probabilistic interpretation .
to this effect , we write the parameter under @xmath51 as @xmath73 , @xmath74 being a fixed known quantity , and under @xmath75 as @xmath76 , so that comparing @xmath51 with @xmath66 corresponds to testing the null hypothesis @xmath77 .
aitkin does not impose a positive prior probability on @xmath51 , since his prior only bears on @xmath66 ( in a spirit close to the savage - dickey representation , see @xcite ) .
his approach is therefore similar to the inversion of a confidence region into a testing procedure ( or vice - versa ) . under the model @xmath78 , denoting by @xmath79 the log - likelihood of the bigger model , @xmath80 & \approx & \mbox{pr}\left [ \mathcal x^2_{p_2-p_1 } > - l(\hat{\theta}_1(\psi_0 ) , \psi_0)+ l(\hat{\theta}_1,\hat{\psi})\right ] \\ & \approx & 1 - f_{p_2-p_1 } [ - l(\hat{\theta}_1(\psi_0 ) , \psi_0)+ l(\hat{\theta}_1,\hat{\psi } ) ] , \end{aligned}\ ] ] which is the approximate @xmath12-value associated with the likelihood ratio test .
therefore , the aim of this approach seems to be , at least for embedded models where the bernstein
von mises theorem holds for the posterior distribution , to construct a _
bayesian _ procedure reproducing the @xmath12-value associated with the likelihood ratio test . from a frequentist point of view it is of interest to see that the posterior probability of the likelihood ratio being greater than one is approximately a @xmath12-value , at least in cases when the bernstein - von mises theorem holds , e.g. for embedded models and proper priors .
this @xmath12-value can then be given a finite - sample meaning ( under the above restrictions ) , however it seems more interesting from a frequentist perspective than from a bayesian one . from a bayesian decision - theoretic viewpoint , this is even more dubious , since the loss function ( [ loss ] ) is difficult to interpret and to justify .
without a specific alternative , the best we can do is to make posterior probability statements about @xmath81 and transfer these to the posterior distribution of the likelihood ratio ( .. ) there can not be strong evidence in favor of a point null hypothesis against a general alternative hypothesis . " _
statistical inference _ , pages 4244 we further note that , once _ statistical inference _ has set the principle of using the posterior distribution of the likelihood ratio ( or rather of the divergence difference since this is at least symmetric in both hypotheses ) , there is a whole range of outputs available including confidence intervals on the difference , for checking whether or not they contain zero . from our ( bayesian ) perspective , this solution ( a ) is not bayesian for reasons exposed above
, ( b ) is not parameterization invariant , and ( c ) relies once again on an arbitrary confidence level .
we have focused in this review on aitkin s proposals rather than on his characterizations of other statistical methods . in a few places , however , we believe that there have been some unfortunate confusions from his part . on page 22
, aitkin describes bayesian posterior distributions as `` formally a measure of personal uncertainty about the model parameter , '' a statement that we believe holds generally only under a definition of `` personal '' that is so broad as to be meaningless . as we have discussed elsewhere
( gelman , 2008 ) , bayesian probabilities can be viewed as `` subjective '' or `` personal '' but this is not necessary . or , to put it another way ,
if you want to label my posterior distribution as `` personal '' because it is based on my personal choice of prior distribution , you should also label inferences from the proportional hazards model as `` personal '' because it is based on the user s choice of the parameterization of cox ( 1972 ) ; you should also label any linear regression ( classical or otherwise ) as `` personal '' as based on the individual s choice of predictors and assumptions of additivity , linearity , variance function , and error distribution ; and so on for all but the very simplest models in existence .
in a nearly century - long tradition in statistics , any probability model is sharply divided into `` likelihood '' ( which is considered to be objective and , in textbook presentations , is often simply given as part of the mathematical specification of the problem ) and `` prior '' ( a dangerously subjective entity to which the statistical researcher is encouraged to pour all of his or her pent - up skepticism ) .
this may be a tradition but it has no logical basis . if writers such as aitkin wish to consider their likelihoods as objective and consider their priors as subjective , that is their privilege .
but we would prefer them to restrain themselves when characterizing the models as others .
it would be polite to either tentatively accept the objectivity of others models or , contrariwise , to gallantly affirm the subjectivity of one s own choices .
aitkin also mischaracterizes hierarchical models , writing `` it is important not to interpret the prior as in some sense a
_ model for nature _ [ italics in the original ] that nature has used a random process to draw a parameter value from a higher distribution of parameter values '' on the contrary , that is exactly how we interpret the prior distribution in the ideal case .
admittedly , we do not generally approach this ideal ( except in settings such as genetics where the population distribution of parameters has a clear sampling distribution ) , just as in practice the error terms in our regression models do not capture the true distribution of errors . despite these imperfections
, we believe that it can often be helpful to interpret the prior as a model for the parameter - generation process and to improve this model where appropriate .
_ statistical inference _ points out several important facts that are individually known well ( but perhaps not well enough ! ) , but by putting them all in one place it foregrounds the difficulty or impossibility of putting all the different approaches to model checking in one place . we all know that the @xmath12-value is in no way the posterior probability of a null hypothesis being true ; in addition , bayes factors as generally practiced correspond to no actual probability model . also , it is well - known that the so - called harmonic mean approach to calculating bayes factors is inherently unstable , to the extent that in the situations where it does work , " it works by implicitly integrating over a space different from that of its nominal model . yes , we all know these things , but as is often the case with scientific anomalies , they are associated with such a high level of discomfort that many researchers tend to forget the problems or try to finesse them .
it is refreshing to see the anomalies laid out so clearly . at some points , however , aitkin disappoints .
for example , at the end of section 7.2 , he writes : `` in the remaining sections of this chapter , we first consider the posterior predictive @xmath12-value and point out difficulties with the posterior predictive distribution which closely parallel those of bayes factors . ''
he follows up with a section entitled `` the posterior predictive distribution , '' which concludes with an example that he writes `` should be a matter of _ serious _ concern [ emphasis in original ] to those using posterior predictive distributions for predictive probability statements . ''
what is this example of serious concern ?
it is an imaginary problem in which he observes 1 success in 10 independent trials and then is asked to compute the probability of getting at most 2 successes in 20 more trials from the same process .
_ statistical inference
_ assumes a uniform prior distribution on the success probability and yields a predictive probability or 0.447 , which , to him , `` looks a vastly optimistic and unsound statement . '' here , we think aitkin should take bayes a bit more seriously . if you think this predictive probability is unsound , there should be some aspect of the prior distribution or the likelihood that is unsound as well .
this is what good ( @xcite ) called `` the device of imaginary results . ''
we suggest that , rather than abandoning highly effective methods based on predictive distributions , aitkin should look more carefully at his predictive distributions and either alter his model to fit his intuitions , alter his intuitions to fit his model , or do a bit of both .
this is the value of inferential coherence as an ideal .
several of the examples in _ statistical inference _ represent solutions to problems that seem to us to be artificial or conventional tasks with no clear analogy to applied work .
they are artificial and are expressed in terms of a survey of 100 individuals expressing support ( yes / no ) for the president , before and after a presidential address ( ... ) the question of interest is whether there has been a change in support between the surveys ( ... ) .
we want to assess the evidence for the hypothesis of equality @xmath82 against the alternative hypothesis @xmath83 of a change . " _ statistical inference _
, page 147 based on our experience in public opinion research , this is not a real question .
support for any political position is always changing .
the real question is how much the support has changed , or perhaps how this change is distributed across the population .
a defender of aitkin ( and of classical hypothesis testing ) might respond at this point that , yes , everybody knows that changes are never exactly zero and that we should take a more `` grown - up '' view of the null hypothesis , not that the change is zero but that it is nearly zero .
unfortunately , the metaphorical interpretation of hypothesis tests has problems similar to the theological doctrines of the unitarian church .
once you have abandoned literal belief in the bible , the question soon arises : why follow it at all ?
similarly , once one recognizes the inappropriateness of the point null hypothesis , it makes more sense not to try to rehabilitate it or treat it as treasured metaphor but rather to attack our statistical problems directly , in this case by performing inference on the change in opinion in the population . to be clear
: we are not denying the value of hypothesis testing . in this example
, we find it completely reasonable to ask whether observed changes are statistically significant , i.e. whether the data are consistent with a null hypothesis of zero change .
what we do not find reasonable is the statement that `` the question of interest is whether there has been a change in support . ''
_ ( a ) hypothetical graph of presidential approval with discrete jumps ; ( b ) presidential approval series ( for george w. bush ) showing movement at many different time scales .
if the approval series looked like the graph on the left , then aitkin s question of interest " of whether there has been a change in support between the surveys " would be completely reasonable . in the context of actual public opinion data , the question does not make sense ; instead , we prefer to think of presidential approval as a continuously - varying process . _ , title="fig : " ] _ ( a ) hypothetical graph of presidential approval with discrete jumps ; ( b ) presidential approval series ( for george w. bush ) showing movement at many different time scales .
if the approval series looked like the graph on the left , then aitkin s question of interest " of whether there has been a change in support between the surveys " would be completely reasonable . in the context of actual public opinion data , the question does not make sense ; instead , we prefer to think of presidential approval as a continuously - varying process . _ , title="fig : " ] all this is application - specific .
suppose public opinion was observed to really be flat , punctuated by occasional changes , as in the left graph in figure [ fig : president ] .
in that case , aitkin s question of `` whether there has been a change '' would be well - defined and appropriate , in that we could interpret the null hypothesis of no change as some minimal level of baseline variation . real public opinion
, however , does not look like baseline noise plus jumps , but rather shows continuous movement on many time scales at once , as can be seen from the right graph in figure [ fig : president ] , which shows actual presidential approval data . in this example
, we do not see aitkin s question as at all reasonable .
any attempt to work with a null hypothesis of opinion stability will be inherently arbitrary .
it would make much more sense to model opinion as a continuously - varying process .
the statistical problem here is not merely that the null hypothesis of zero change is nonsensical ; it is that the null is in no sense a reasonable approximation to any interesting model .
the sociological problem is that , from @xcite onward , many bayesians have felt the need to mimic the classical null - hypothesis testing framework , even where it makes no sense .
aitkin is unfortunately no exception , taking a straightforward statistical question estimating a time trend in opinion and re - expressing it as an abstracted hypothesis testing problem that pulls the analyst away from any interesting political questions .
`` the posterior has a non - integrable spike at zero .
this is equivalent to assigning zero prior probability to these unobserved values . '' _ statistical inference _
, page 98 a skeptical ( or even not so skeptical ) reader might at this point ask , why did we bother to write a detailed review of a somewhat obscure statistical method that we do not even like ?
our motivation surely was not to protect the world from a dangerous idea ; if anything , we suspect our review will interest some readers who otherwise would not have heard about the approach ( as previously illustrated by @xcite ) . in 1970 , a book such as _ statistical inference _
could have had a large influence in statistics .
as aitkin notes in his preface , there was a resurgence of interest in the foundations of statistics around that time , with lindley , dempster , barnard , and others writing about the intersections between classical and bayesian inference ( going beyond the long - understood results of asymptotic equivalence ) and researchers such as akaike and mallows beginning to integrate model - based and predictive approaches to inference .
a glance at the influential text of cox and hinkley ( 1974 ) reveals that theoretical statistics at that time was focused on inference from independent data from specified sampling distributions ( possibly after discarding information , as in rank - based tests ) , and `` likelihood '' was central to all these discussions .
forty years on , a book on likelihood inference is more of a niche item .
partly this is simply part of the growth of the field with the proliferation of books , journals , and online publications , it is much more difficult for any single book to gain prominence .
more than that , though , we think statistical theory has moved away from iid analysis , toward more complex , structured problems .
that said , the foundational problems that _ statistical inference _
discusses are indeed important and they have not yet been resolved . as models get larger , the problem of `` nuisance parameters '' is revealed to be not a mere nuisance but rather a central fact in all methods of statistical inference . as noted above
, aitkin makes valuable points known , but not well - enough known about the difficulties of bayes factors , pure likelihood , and other superficially attractive approaches to model comparison .
we believe it is a natural continuation of this work to point out the problems of the integrated likelihood approach as well . for now , we recommend model expansion , bayes factors where reasonable , cross - validation , and predictive model checking based on graphics rather than @xmath12-values .
we recognize that each of these approaches has loose ends .
but , as practical idealists , we consider inferential challenges to be opportunities for model improvement with the bayesian realm rather than motivations for a new theory of noninformative priors that takes us in uncharted territories . | for many decades , statisticians have made attempts to prepare the bayesian omelette without breaking the bayesian eggs ; that is , to obtain probabilistic likelihood - based inferences without relying on informative prior distributions .
a recent example is murray aitkin s recent book , _ statistical inference _ , which presents an approach to statistical hypothesis testing based on comparisons of posterior distributions of likelihoods under competing models .
aitkin develops and illustrates his method using some simple examples of inference from iid data and two - way tests of independence .
we analyze in this note some consequences of the inferential paradigm adopted therein , discussing why the approach is incompatible with a bayesian perspective and why we do not find it relevant for applied work .
* keywords : * foundations , likelihood , bayesian , bayes factor , model choice , testing of hypotheses , improper priors , coherence . |
high energy heavy - ion collisions enable the study of strongly interacting matter under extreme conditions . at sufficiently high collision energies quantum - chromodynamics ( qcd )
predicts that hot and dense deconfined matter , commonly referred to as the quark - gluon plasma ( qgp ) , is formed .
the experiment alice @xcite at the cern large hadron collider ( lhc ) @xcite , allows the study of the qcd matter in a new energy domain .
high @xmath2 partons produced in the initial stage of the collisions , have been identified as a valuable probe of the medium .
they are only observed indirectly , as a collimated jet of hadrons originating from the hadronization of the partonic shower .
comparing the properties of the jet fragmentation in proton - proton and heavy - ion collisions will reveal the modifications induced by the medium on the hard scattered partons .
ideally , one needs to know the 4-momentum of the parton when it has been produced in the hard scattering and after it has been modified by the medium .
this can be achieved by selecting particular hard processes in which there is a photon in the final state . since the photon does not interact with the medium , its 4-momentum is not modified and thus provides a measure of the hard scattered parton emitted back - to - back with the photon . measuring the hadrons
opposite to the photon is thus a promising way to measure the jet fragmentation and imbalance between photon and hadrons to quantify the modifications due to the medium . to establish a reference measurement for heavy - ion collisions ,
proton - proton collision data collected with alice in 2010 have been analyzed with the ultimate goal to construct the direct photon - charged hadron correlations .
minimum bias data have been collected in pp collisions at center of mass @xmath3tev .
the present results have been obtained by analyzing about 160 million events .
the preliminary result is presented together with inclusive photon - charged hadrons correlation and @xmath0-charged hadrons spectra all showing the characteristic di - jet pattern from where the momentum imbalance @xmath1 is extracted .
the experimental technique consists in tagging events with a leading trigger and measuring the distribution of hadrons associated to this leading trigger from the same event .
such a measurement requires an excellent photon and @xmath0 identification and the measurement of charged and neutral hadrons with good @xmath4 resolution . in alice , the electromagnetic calorimeters , phos
( @xmath5 and @xmath6 = 100@xmath7 ) and emcal ( @xmath8 and @xmath9 = 100@xmath7 ) @xcite , are capable to measure photons with high efficiency and resolution . in the calorimeters , electromagnetic particles
are detected as clusters of cells in the calorimeters .
roughly we have identified @xmath0 candidate as a pair of clusters with invariant mass around the @xmath0 mass between 110 and 160 mev / c@xmath10 , and single clusters are identified as inclusive photon candidates .
no particle identification has been applied yet so that the single cluster sample may contain a sizable fraction of charged particles which develop a shower in the calorimeters or high-@xmath4 merged @xmath0 cluster which can not be reconstructed by invariant mass .
the central tracking system ( its and tpc ) , covering the pseudo - rapidity @xmath11 and the full azimuth , is used for charged track measurements , contributes to the direct photon identification by applying the isolation technique .
three different trigger particles have been selected for the correlation measurements : ( i ) the charged trigger is chosen as the track with highest transverse momentum among all the tracks from the same event , ( ii ) the photon cluster trigger is defined as the calorimeter cluster with highest energy and no charged track from the same event has momentum larger than photon cluster , ( iii ) the @xmath0 trigger is selected as the cluster pair within the appropriate invariant mass range and with the highest transverse momentum in the event .
the azimuthal correlation between the trigger particles ( charged particle , single cluster ) and charged hadrons are shown in fig .
[ fig : clusterch ] . the near side ( @xmath12 ) and away side ( @xmath13 ) peaks
are clearly observed .
the correlation with cluster triggers shows larger di - jet peaks reflecting the fact that the neutral trigger selection enhances the probability that the trigger is the leading particle of the jet fragmentation compared to the less restrictive charged trigger selection .
the azimuthal correlations from inclusive photon clusters and @xmath0 triggers show quite similar shapes ( fig .
[ fig : photonpi0 ] ) , indicating that most of the inclusive photon clusters are @xmath14 decay photons .
azimuthal correlation distributions for inclusive cluster trigger and @xmath0 triggers on the trigger particles with @xmath15 gev / c in pp collisions at @xmath16 = 7 tev . ]
azimuthal correlation distributions for inclusive cluster trigger and @xmath0 triggers on the trigger particles with @xmath15 gev / c in pp collisions at @xmath16 = 7 tev . ] by selecting isolated triggers , i.e. the trigger satisfies : the sum of the transverse momentum of the hadrons inside a cone with size @xmath17 around the trigger candidate carries less than 10 % of the trigger s transverse momentum , we can enrich the sample with direct photons or single hadron jets .
the near side peak is suppressed by construction , whereas the away side peak remains and a slight difference is due to the imperfect isolation parameters used in the analysis ( fig .
[ fig : isoemctrigdphi ] ) indicating the existence of di - jet events with one of the jet being a hard fragmenting jet or eventually a direct photon .
because of the hadronization , we do not have direct access to the parton kinematics and therefore can measure neither the fragmentation function nor the magnitude of partonic transverse momentum @xmath1 which modifies the ideal @xmath18 kinematics .
however , the isolated photon/@xmath0 triggered correlation could be used to extract the partonic level kinematics to the extend that the leading order kinematics dominates , as suggested by the phenix analysis @xcite : @xmath19 where @xmath20 is the trigger fragmentation variable and @xmath21 is the ratio between away and near side hard scattered partons , @xmath22 is similar to @xmath23 but at the hadronic level , and @xmath24 is the projection of trigger particle deviates from the parton before fragmentation ( see detail in @xcite ) .
the values of @xmath25 are determined by measuring the width of the away side peak @xmath26 , using the fitting function described in @xcite .
the fitted away side peak width shows in fig .
[ fig : pout ] , the width is weakly depend on the trigger @xmath2 .
the isolated trigger represents the hard scattered parton direction approximately ( @xmath27 ) , therefore , @xmath28 1 and @xmath290 .
the @xmath25 values we have measured for @xmath30gev / c and @xmath31gev / c is consistent with another measurement obtained from charged di - hardon correlations in alice .
the measured value agrees the extrapolated value at lhc energies from available worldwide data @xcite . however this preliminary analysis does not allow to draw any conclusion other than these results indicate the expected behavior .
exciting physics will certainly come with the final analysis of large statistics within well - calibrated detectors and all efficiency corrections .
00 k. aamodt et al 2008 , _ jinst _ * 3 * s08002 .
l. evans and p. bryant ( editors ) 2008 , _ jinst _ * 3 * s08001 .
alice collaboration 1999 , http://alice.web.cern.ch / alice / tdr/. a. adare et al .
, 2006 _ phys .
d _ * 74 * 072002 .
a. adare et al . , 2010 _
d _ * 82 * 072001 .
y. x. mao et al . , 2008 , _
j. c _ * 57 * 613 ; _ qm2009 poster_. | comparing the measurements of the hadronic final state from partonic showers in proton - proton and heavy - ion collisions will reveal the modifications generated by the medium on partons produced in hard scatterings .
this can be achieved by selecting the hard processes in which there is a direct photon in the final state .
the experimental technique consists in tagging events with a well identified high energy direct photon and in measuring the azimuthal angle correlation with charged hadrons . to establish a reference measurement for heavy - ion collisions ,
proton - proton collision data collected with alice have been analyzed .
preliminary results are presented together with photon and @xmath0-charged hadrons correlations showing the characteristic di - jet pattern from where the partonic momentum @xmath1 is extracted .
triggers , azimuthal correlation , isolation cut , @xmath1 21.10.hw , 25.75.gz , 21.10.hw , 12.38.mh |
the general principles in non - living systems play significant roles in living systems .
how do the principles in gravity theory or in quantum mechanism impact on our understanding of life ?
an organism can not keep active without the supply of energy due to the first law in thermodynamics . and it can not live long without the supply of minus entropy due to the second law in thermodynamics .
but it seems that there are no direct effects of the relativity principle or the uncertainty principle on life .
we found that the holographic principle , which is likely only one of several independent conceptual advances needed for progress in quantum gravity @xcite@xcite@xcite@xcite@xcite , profoundly constraints the forms of life and substantially impacts on the evolution of life .
the holographic principle states that there is a precise , general and surprisingly strong limit on the information content of spacetime regions .
the number of quantum states in a spatial region is bounded from above by the surface of the region measured in the unit of four - fold planck areas .
this entropy bound is a strong constraint on any theory about our universe .
if this principle is true , field theory or string theory , where there are infinite degrees of freedom , can not be the ultimate theory . and if this principle can be applied to the phenomenon of life , the degrees of freedom in a living system will also be constrained . from this point of view
, the principles in relativity or in quantum theory constrain life in an alternative way .
the holographic principle indicates that there is a strict relationship between the information storage capacity of the space and the complexity of any organism wherein .
such a basic idea can be illustrated by a simple example .
whatever a living system with @xmath0 degrees of freedom is , we can conclude that it can never exist in a universe with a horizon area less than @xmath1 , where @xmath2 is the planck length . in this paper , we estimated the immense degrees of freedom for living systems according to the theory of gene regulatory networks and boolean networks @xcite @xcite .
we found a contradiction between the possible degrees of freedom of living systems and the maximum information storage capacity in the observed universe .
then we reconciled this contradiction in terms of the causality between the possible sequences of macromolecules for the actual living systems , which is equivalent to the existence of language of genes .
we propose evidences of language of genes and we can explain the outline of protein length distributions by a linguistic mechanism of generation of protein sequences .
we can also explain the leaps in the evolution of biological complexity according to the entropy bound .
information properly bridges biology and physics @xcite@xcite@xcite , which gives deep insights into the nature of life . with the development of genetics , we know that the gene regulatory networks play significant roles in development and evolution of life @xcite . based on the theory of self - organization ,
kauffman proposed a general theory of boolean networks to describe the gene regulatory networks , where the interactions between genes can represented by boolean operations between the nodes of the network @xcite .
thus , the degrees of freedom of a living system can be estimated by the number of states of the corresponding boolean network .
proteins are the elementary units in the activities of life .
so a living organism can be represented by a dynamical system of all the proteins in its body .
we denote the set @xmath3 as all possible protein sequences with a cutoff of protein length @xmath4 .
proteins are chains concatenated by @xmath5 amino acids .
so there are @xmath6 elements in the set @xmath3 .
we define a boolean network @xmath7 as the boolean network whose nodes are elements of @xmath3 ( fig .
according to the definition of boolean networks , there are two states for each node of a boolean network : `` on '' or `` off '' @xcite .
a state of @xmath8 represent that some nodes are `` on '' while the others are `` off '' .
so a proteome can be represented by a state of @xmath8 , where only the nodes corresponding to protein sequences in the proteome are `` on '' .
the state space @xmath9 consists of all possible states of @xmath7 whose number is @xmath10 an actual species can be represented by a point in @xmath9 .
the evolution of a species can be illustrated by a trajectory in @xmath9 ( fig .
2b ) . as a preliminary consideration ,
the degrees of freedom of a living system can be estimated by the logarithm of number of states @xmath11 which we will reconsider later on . according to the holographic principle
, we can calculate that the information in the observed universe is about @xcite @xmath12 this value is too large for non - living systems .
for example , the information of black body cosmic background photons is about @xmath13 bits , which may be the largest degrees of freedom for possible non - living systems .
but it is still much less than @xmath14 .
the remaining information storage capacity in our universe has not been wasted however for there being living systems .
the degrees of freedom for living systems are so immense that may exceed the maximum information storage capacity in the observed universe .
the structure of chains of genetic macromolecules essentially provides immense degrees of freedom for living systems , because the number of possible protein sequences can be as large as @xmath15 . for a living system
, the degrees of freedom may be equivalent to that of the observed universe if the protein length is about @xmath16 amino acids .
interestingly , the most frequent protein length for the life on our planet is about @xmath17 .
the immense degrees of freedom of living systems originate from the great number of possible sequences in @xmath7 .
most of the degrees of freedom come from the states of @xmath7 in which about half the nodes are `` on '' . on the other hand , the degrees of freedom can also come from the states in which only a minority of nodes are `` on '' .
our living systems belong to the latter case , where there are only thousands of proteins in actual proteomes .
[ fig1 ] the estimate of immense degrees of freedom of a living system in the above , however , seriously contradicts the holographic principle if we consider the actual life around us .
the average protein length in a proteome ranges about from @xmath18 amino acids to @xmath19 amino acids , and a certain number of proteins are longer than thousands of amino acids . according to the preliminary estimate
, the degrees of freedoms for the actual living systems on our planet will be much larger than the maximum degrees of freedoms in the observed universe @xmath14 .
we have to reconcile the contradiction between the preliminary estimate of degrees of freedom of living systems and the conclusion of the holographic principle .
if the holographic principle is not invalid , we must find ways to shrink the preliminary estimate of the degrees of freedom of living systems .
we introduce the causality between the states of @xmath7 to reveal the additional constraint on the degrees of freedom by the entropy bound . at the beginning of the evolution on the planet
, the first living system may be denoted as an inertial state @xmath20 .
when the degrees of freedom of @xmath7 is greater than @xmath14 , not all the states of @xmath7 can have causal relationship with @xmath20 unless the holographic principle is untrue .
we define the set @xmath21 as all the states that have causal relationship with @xmath20 , which has @xmath22 states and is only a proper subset of @xmath9 ( fig .
the nodes of @xmath21 constitute @xmath23 , which is a subset of @xmath3 .
an actual living system at present corresponds to a dynamic system evolving only in the state space @xmath21 and a meaningful protein sequence in biology must belong to @xmath23 .
the degrees of freedom of a living system , therefore , can be defined by the number of states in @xmath24 : @xmath25 where @xmath22 is much less than @xmath26 and @xmath27 can be rightly less than @xmath14 .
[ fig1 ] the causality provides a physical explanation to distinguish a part of sequences @xmath23 from all possible sequences @xmath28 .
not all the amino acid chains or base chains are meaningful in biology . according to the theory of formal language ,
a language is defined by a subset of all the sequences concatenated by letters in a given alphabet @xcite .
the choice of a subset @xmath23 from @xmath3 is a natural way to define a formal language ( fig .
2 ) . the protein or dna languages originate in the constraint on the degrees of freedom of life by the entropy bound .
the alphabet of protein language consists of @xmath5 amino acids , and the alphabet of the language of genes consists of @xmath29 bases .
the arrangement of the letters in the sequences should be determined by some grammars . although there are various entropy bounds , there is no difference for the requirement of finite degrees of freedom in life and the requirement of the language of genes for all the theories .
to some extent , the language of genes is a consequence of the principles in quantum gravity .
the phenomenon of life is constrained strictly by the entropy bound .
the requirement of the order of sequences by the grammars can not be explained in the context of classical physics because the degrees of freedom of life can be infinite .
the ability of speaking for human beings is determined by genes . that we can communicate with each other instinctively
can be attributed to our common genes .
the human language can be viewed as a transformation of cell language @xcite .
the information storage capacity of a natural language can also be estimated by the similar calculation in the above .
for instance , we estimated that there are up to @xmath30 bits of information can be written in a language with @xmath31 letters and the length of words in the language is @xmath32 , which is much less than the protein length . in this sense , the natural language is simpler than the language of genes .
the value @xmath33 is much less than the information in the observed universe @xmath14 .
so the description of the universe by natural language is always a simplified version of the actually complex world .
interestingly , there were not rare cases to reach the same goal by different routes in the history of natural sciences , such as , riemannian geometry and general relativity , or the theory of bundles and gauge theories .
such encounters may come from that all the descriptions in different subjects have a common ultimate theory of all the information in the universe , although we can not understand all the details of the world by only one subject .
[ fig3 ] several attempts have been made over the past three decades to combine linguistic theory with biology @xcite @xcite .
the distribution of the number of occurrences of protein domains in a genome can be a good fit of the power - law distribution known as zipf s law in linguistics , and we can distinguish between the protein linguistics and the language of genes according to the theory of formal language @xcite .
so the experimental observations support the existence of languages in the sequences of macromolecules . on one hand ,
they are required by the holographic principle . on the other hand , they are consequences of the evolution of life at the molecular level @xcite@xcite@xcite@xcite .
the alphabets of amino acids or bases formed at the beginning of life . and genetic code developed and fixed in the early stage of evolution .
all these factors can determine whether a sequence is permitted in a life , which is equivalent to the role of grammars at the molecular level .
we found a strong evidence of the underlying mechanism in the organization of amino acids in protein sequences by studying the correlations between protein length distributions , which indicates the languages in the protein sequences .
the protein length distribution corresponds to a vector @xmath34 where there are @xmath35 proteins with length @xmath36 in the complete proteome of a species and @xmath37 is the cutoff of protein length .
our data of the protein length distributions are obtained from the data of @xmath38 complete proteomes in the database predictions for entire proteomes @xcite .
the discrete fourier transformation of the protein length distribution is : @xmath39 let @xmath40 denotes the frequency of the highest peak @xmath41 in the discrete fourier transformation of the protein length distribution for a species .
we found that there is an interesting relationship between the frequency @xmath40 and the average protein length @xmath42 of species .
the distribution of species in @xmath43 plane shows an regular pattern : the species in the three domains ( archaebacteria , eubacteria and eukaryotes ) gathered in three rainbow - like arches respectively ( fig 3 ) .
this pattern strongly indicates the intrinsic correlation among the protein length distributions , which can never achieve if the protein length distributions are stochastic .
the periodic - like fluctuations in the protein length distribution @xcite may also originate in the underlying mechanism of generation of protein sequences .
we propose a model to reveal the underlying mechanism in the protein sequences according to tree adjoining grammar @xcite . in the model ,
protein sequences can be generated by tree adjoining operations , i.e. , substituting the initial tree or auxiliary trees into to each other by identifying the inner nodes ( fig .
4a ) @xcite .
there is only one variant @xmath44 in the model , which is the probability of substitutions in the adjoining operations and denotes different species .
a certain number of proteins can be generated when @xmath44 is fixed , hence we obtain a protein length distribution by the model ( fig .
the properties of protein length distributions can be explained by the simulation . the outline and the fluctuations of the simulated protein length distribution agree with the actual protein length distributions in principle .
we show that there is a close relationship between the protein length distributions and grammar rules .
the fluctuations in the distributions are determined by the grammar rules .
the same grammar rule corresponds to the same distribution . if changing grammar rules , we obtain different outlines and fluctuations of distribution .
this result suggests that the fluctuations in actual protein length distributions are intrinsic properties of certain species and may infer the underlying mechanism on the order of protein sequences .
the evolution of complexity of life is not a linear course of increment @xcite@xcite .
the entropy bound can also explain the leaps in the evolution of biological complexity .
consequently we can outline the macroevolution of life .
the gene regulatory networks are accelerating networks @xcite@xcite . according to this theory
, the evolution of complexity of any accelerating networks has to be slowed down and will stop at an upper limit of complexity . hence there must be upper limits of complexity in both of the evolution of biological complexity for prokaryotes and eukaryotes , where the entropy bound is a natural upper limit .
the whole evolution of biological complexity can be , therefore , divided into three steps : the evolution of unicellular life , the evolution of multicellular life and the evolution of society of human beings .
the cambrian explosion divided the first two steps .
and we found that the evolution of multicellular life has reached its upper limit because the maximum non - coding dna content is near to @xmath45 at present .
the civilization of human beings appeared , which can be taken as an alternative form of biological complexity .
the entire evolution of biological complexity should be governed by a universal mechanism of evolution . the universal language of genes in species may harmonize the evolution of life in the biosphere .
croft , l. j. , lercher , m. j. , gagen , m. j. & mattick , j. s. is prokaryotic complexity limited by accelerated growth in regulatory overhead .
_ arxiv preprint archive _ [ online ] , http://arxiv.org/abs/q-bio.mn/0311021 ( 2003 ) . | we show that the holographic principle in quantum gravity imposes a strong constraint on life .
the degrees of freedom of an organism can be estimated according to the theory of boolean networks , which is constrained by the entropy bound .
hence we can explain the languages in protein sequences or in dna sequences .
the overall evolution of biological complexity can be illustrated . and
some general properties of protein length distributions can be explained by a linguistic mechanism . |
single particle quantum devices work essentially in a quantum mechanical way by using the whole features of quantum states , especially the phases of quantum states .
a typical example is the single electron transistor ( set ) @xcite and its photon analogue @xcite , which plays a central role in quantum manipulation and quantum measurement .
a set is a mesoscopic system that allows confined electrons to tunnel to the metallic leads .
it turns on and off again every time one electron is added to the isolated region kastner9293 . unlike the conventional transistors , which can be understood by using classical concepts , the set is a quantum mechanical one in substance .
it can be utilized for measuring the quantum effects in josephson junction superconductor circuit and nano - mechanics resonators shnirman9801,schwab03,korotkov05 . in this article
, we will pay attention to another single particle quantum device , which can be understood as the quantum analogue of the conventional diode device . in practice ,
diode devices are indispensable for building various electronic circuits in the traditional electronics .
accordingly , once a coherent diode device has been implemented , a single - particle quantum circuit may be further realized in the level of single quantum state by using of the two basic elements , single particle transistor and diode .
compared with the traditional electrocircuits based on the density distribution of electrons , the new quantum circuit makes full use of the quantum properties including quantum phases in information processing . in this sense
, we can also regard it as a necessary element in quantum information science and technology .
a major research effort in recent years is to seek a simplest single particle system with diode features @xcite , which is characterized by the asymmetric performance of transmission coefficients along the opposite directions . in the present investigation
, we find that a three - site tight - binding ring system threaded by an aharonov - bohm ( ab ) magnetic flux has analogy to a perfect diode in a wide range of parameters .
in addition , our studies are not restricted to the charged particle case . for a neutral particle system ,
an effective magnetic flux is induced by applying a globe rotation @xcite , which can also break the time reversal symmetry and lead to the asymmetric transmissions .
it is shown that the difference of the two output currents is linearly proportional to the rotational angular velocity approximately , which behaves as a new kind of extended sagnac gyroscope @xcite .
decades ago , bttiker _
et al . _ have found that the conductance of a multi - terminal sample is asymmetric in the presence of an ab flux buttiker86,azbel81,buttiker85 .
realization of such mechanism in quantum device requires its size being smaller than the phase - breaking length . thus seeking a minimized system with asymmetric feature is attractive in practice .
a discrete system may be a good candidate to accomplish such a task .
it is of both theoretical and practical importance to study this theory in a discrete system .
moreover , we show practical applications of the quantum interference effect as single electron diode and single particle gyroscope . in the general discussion part
, it is interesting to find that some special configurations may protect the transmission coefficients from symmetry breaking even if the ab magnetic flux is imposed .
this paper is organized as follows . in sec .
ii , the general @xmath0-site ring - shaped model with attaching leads is presented , and an analytical method for calculating the transmission coefficients is introduced . in sec .
iii , we study the asymmetric transmission behavior of the @xmath1-site system by using the exact solution . in sec .
iv , a single particle gyroscope and its physical realization in optical lattice is proposed . in sec .
v , we give a general discussion about the condition of asymmetric transmission by making use of green s function approach , which also confirms the results obtained above .
conclusions are summarized at the end of the paper . in the appendix , a proof is given to show the equivalence between the bethe ansatz and the green s function method .
arc chains @xmath2},a^{[2]},\dots , a^{[m]}\}$ ] and @xmath3 attaching leads @xmath4},b^{[2]},\dots , b^{[m]}\}$ ] threaded by a magnetic flux .
the input and possible output currents are marked by wavy arrows .
( b ) schematic illustration of the three - site chiral coherent scattering system with diode features.,width=8 ] the central system we concern is described by an @xmath0-site tight - binding ring threaded by a magnetic flux shown in fig .
[ model](a ) .
the hamiltonian reads @xmath5here , @xmath6 ( @xmath7 ) is the fermion creation ( annihilation ) operator at the @xmath8th site .
the external magnetic field does not exert force on the bloch electrons , but makes hopping integral @xmath9 between sites @xmath8 and @xmath10 pick up an ab phase factor @xmath11 in the peierls approximation @xcite , where @xmath12@xmath13 is the flux quanta and @xmath14 is the total magnetic flux ; @xmath15 is the chemical potential of the central system . in the same figure ,
@xmath3 half - infinite tight - binding leads are attached to @xmath3 sites on the ring , and the corresponding hamiltonian are@xmath16 } + t\sum_{i}b_{i}^{\dag \left [ l\right ] } b_{i+1}^{\left [ l% \right ] } + \mathrm{h.c.}\right ) .\]]without loss of generality , we let @xmath17 lead as the input lead , while all the other indexed @xmath18 leads are output ones
. the spin degree of freedom is omitted for notational brevity here since the model does not contain interaction involving spin . in order to investigate the scattering problem through such a system in the framework of bethe ansatz @xcite more efficiently ,
we regard the @xmath0-site ring as @xmath3 arc tight - binding chains connected by the head and tail . as illustrated in fig .
[ model](a ) , the length of the @xmath19th arc chain @xmath20 } $ ] is @xmath21 , and the @xmath19th lead @xmath22 } $ ] is joined with the @xmath23th site of the @xmath19th arc chain . the total number of sites on the ring satisfies @xmath24 .
then the hamiltonian @xmath25 is rearranged with @xmath26}}a_{j}^{\dag \left [ l\right ] } a_{j+1}^{\left [ l\right ] } + e^{i\phi _ { n_{l-1}}^{[l-1]}}a_{n_{l-1}}^{\dag \left [ l-1\right ] } a_{1}^{\left [ l% \right ] } + \mathrm{h.c.}\right ] , \notag \\
h_{l } & = & -\sum_{l=1}^{m}\left ( ga_{1}^{\dag \left [ l\right ] } b_{1}^{\left [ l% \right ] } + t\sum_{i}b_{i}^{\dag \left [ l\right ] } b_{i+1}^{\left [ l\right ] } + % \mathrm{h.c.}\right ) .\end{aligned}\]]the scattering wave function is supposed to be@xmath27 } \left ( j\right ) & = & a_{1}\left ( l\right ) e^{i\left ( qj-\sum_{m=1}^{j-1}\phi _ { m}^{[l]}\right ) } + a_{2}\left ( l\right ) e^{-i\left ( qj+\sum_{m=1}^{j-1}\phi _ { m}^{[l]}\right ) } , \notag \\ \psi _ { b}^{\left [ l\right ] } \left ( j\right ) & = & b\left ( l\right ) e^{ikj}+\delta _ { l , l_{in}}e^{-ikj},\end{aligned}\]]where @xmath28 actually defines the reflection coefficient@xmath29while @xmath30 for @xmath18 give the transmission coefficient from the @xmath31th lead to the @xmath19th lead , @xmath32the coefficients @xmath33 , @xmath34 and @xmath35 should satisfy the connecting conditions and the schrdinger equation@xmath36 } \left ( n_{l-1}+1\right ) = \psi _ { a}^{\left [ l% \right ] } \left ( 1\right ) , \notag \\ & & -g\psi _ { b}^{\left [ l\right ] } ( 1)-j\left [ \!e^{i\phi _ { 1}^{\left [ l\right ] } } \psi _ { a}^{\left [ l\right ] } ( 2)+e^{-i\phi _ { n_{l-1}}^{\left [ l-1\right ] } } \psi _ { a}^{\left [ l-1\right ] } ( n_{l-1})\right ] \notag \\ & & = ( e-\omega ) \psi _ { a}^{\left [ l\right ] } ( 1 ) , \notag \\ & & -t\psi _ { b}^{\left [ l\right ] } \left ( 2\right ) -g\psi _ { a}^{\left [ l\right ] } \left ( 1\right ) = e\psi _ { b}^{\left [ l\right ] } \left ( 1\right ) .
\label{ba}\end{aligned}\]]the relation between @xmath37 and @xmath38 is also given by the schrdinger equation,@xmath39}}\psi _ { a}^{\left [ l\right ] } \!\left ( j+1\right ) + e^{-i\phi _ { j-1}^{[l]}}\psi _ { a}^{\left [ l\right ] } \!\left ( j-1\right ) \right ] \!\!=\left ( \!e-\omega \right ) \psi _ { a}^{\left [ l\right ] } \!\left ( j\right ) , \notag \\ & & -t\left [ \psi _ { b}^{\left [ l\right ] } \!\left ( j+1\right ) + \psi _ { b}^{\left [ l\right ] } \!\left ( j-1\right ) \right ] \!=\!e\psi _ { b}^{\left [ l\right ] } \!\left ( j\right ) , \end{aligned}\]]for @xmath40 , i.e.,@xmath41then the @xmath42 coefficients @xmath43 are fully determined by solving the above @xmath42 independent equations in eq .
( [ ba ] ) .
it shows that the bethe ansatz method can provide the exact wave function of the scattering state . *
* * * for small size system , the explicit wave function allows us to clarify the mechanism of the asymmetric transmission from the viewpoint of interference .
hereafter , we mainly consider the simplest case of @xmath44 and @xmath45 shown in fig . [ model](b ) .
the bethe ansatz approach gives the exact solution of transmission and reflection coefficients as @xmath46 where@xmath47 we notice that in general , the transmission coefficients @xmath48 and @xmath49 from one lead to the other two are not identical , i.e. , the current flow across two arbitrary leads is unidirectional . on the other hand ,
the clockwise coefficients @xmath50and @xmath51 , defined as @xmath48 , have the same value , so do the anti - clockwise coefficients @xmath52 defined as @xmath49 .
namely , the transmission coefficients have dextrorotary and levorotatory characteristics .
we will discuss the general asymmetric scattering problem in detail in sec .
v. we will see that this kind of feature is reasonable for a non - bipartite system with broken time - reversal symmetry induced by the external magnetic flux .
( black solid lines ) and anti - clockwise direction @xmath49 ( red dashed lines ) . (
a ) @xmath48 and @xmath49 with respect to @xmath53 in the case of @xmath54 ( upper panel ) and @xmath55 ( lower panel ) , where @xmath56 , @xmath57 , and @xmath58 .
it shows that the system behaves like a perfect diode at @xmath59 ( upper panel ) and within a wide range @xmath60 $ ] ( lower panel ) .
( b ) transmission spectrum in the case of @xmath61 and @xmath58 with @xmath62 ( upper panel ) and @xmath63 ( lower panel ) .
the perfect diode energy can be adjusted by the magnetic flux @xmath64 .
( c ) plots of @xmath48 and @xmath49 at @xmath65 as a function of @xmath66 with parameters @xmath67 , @xmath57 and @xmath58 .
the transmission coefficients of two directions coincide in the limit of both @xmath68 and @xmath69 , but diverge dramatically at the critical point @xmath54 .
( d ) when @xmath56 , @xmath70 , and @xmath59 , the diode feature is optimal at the resonant point @xmath71.,width=8 ] now we concentrate on some special cases of the @xmath1-site ring to exemplify this feature .
\(1 ) @xmath56 , @xmath72 , and @xmath73 , the bloch electrons are injected at @xmath59 with momentum @xmath74 .
( [ tr ] ) and ( [ tr1 ] ) becomes @xmath75it means that the current can only flow in the clockwise direction with no reflection .
thus the system is a perfect diode if we regard the @xmath23th lead as the source and @xmath76th lead as the drain .
the profiles of the corresponding transmission spectra , @xmath48 and @xmath49 as functions of @xmath77 for @xmath54and @xmath78 are plotted in fig .
[ transmission](a ) .
one can see that , when @xmath55 and @xmath56 , the diode is perfect even if the input energy @xmath79 is shifted within the region @xmath80 $ ] .
this shows the great tolerant as a perfect quantum device .
\(2 ) @xmath81 and @xmath73 , the input energy is determined by @xmath37 as @xmath82 , @xmath83therefore , for a given @xmath77 , we can get a perfect diode device by tuning @xmath84 .
the transmission spectrum of the @xmath85 and @xmath86 cases are shown in [ transmission](b ) to give a visual impression .
\(3 ) @xmath67 , @xmath72 and @xmath73 , we plot @xmath48 and @xmath49 at @xmath87 with respect to @xmath88 in fig .
[ transmission](c ) .
the point of @xmath54 denotes the diode case . beyond that critical point , the difference between @xmath89 and @xmath49 gets smaller . in
the limit of @xmath68 , @xmath48 and @xmath49 converged to @xmath90 .
\(4 ) @xmath56 , @xmath72 , and @xmath59 . the dependence of @xmath48 and @xmath91 on the frequency detuning @xmath15 is shown in fig .
transmission(d ) .
it is apparent that the resonant case is advantageous to forming a perfect diode .
a diode system based on the above mechanism is different from the classical ones in semiconductor electronics .
the conventional diode such as p - n junction is made up based on the different density distribution of electrons in the p - type and n - type materials .
when the external voltage is absent and the equilibrium is reached , there is a difference of chemical potentials between the two materials .
consequently , current will flow readily in the forward biased direction since the applied voltage decreases the barrier ; but not in the reverse biased direction because the barrier is raised .
by contrast , the single particle diode device presented here works in the region with no chemical potential difference .
it makes full use of the pure quantum interference phenomenon induced by the aharonov - bohm effect .
so far we have seen that the threaded magnetic flux plays an important role in controlling the amount and direction of currents .
it can be applicable to the more extended system . actually ,
if the system is rotated , an effective magnetic field will be induced in the rotating frame of references .
therefore , the difference between the currents of the two output leads can reflect the rotational angular velocity of the system . that is the basic idea of making up a gyroscope by using the above mentioned asymmetric scattering system .
the analysis below can be applied to a neutral - particle system by simply choosing @xmath92 . for a rotating system with angular frequency @xmath93
, an additional term @xmath94 should be added on the hamiltonian @xcite in the non - inertial frame , where @xmath95 is a constant depends on the geometry of the central system . in the meantime
, the hamiltonian of leads @xmath96 remains the same since the leads are along the radial directions .
then the hamiltonian of the central system becomes @xmath97 or@xmath98 , \]]where @xmath99 , @xmath100 .
the total effective magnetic flux @xmath101depends on the angular frequency @xmath93 . for simplicity ,
we only focus on the case of @xmath59 injection . if the total effective magnetic flux is absent , i.e. , @xmath102 , the transmission and reflection are symmetric due to the fact @xmath103 with @xmath104 \left [ g^{4}+t^{4}\left ( \omega + j\right ) ^{2}\right ] } .\]]then we have @xmath105 , which characterizes the asymmetrical feature . for small effective magnetic flux
@xmath106 , we [email protected] .
, the current difference @xmath108 is a linear function of @xmath109 in the vicinity of @xmath102 . for a typical case @xmath110 , we plot @xmath108 as a function of @xmath111 in fig .
[ gyroscope](a ) . in the resonant case @xmath73
, it shows that this gyroscope have the advantage of good linear response within a wide range of @xmath112 . in practice , in the case of extremely large rotation frequency @xmath93
, a compensate magnetic flux can be added to ensure it works in the linear region .
for some off - resonant cases , e.g. , @xmath113 , the ratio of @xmath108 to @xmath109 at @xmath102 is larger than the one of the resonant case .
the larger ratio implies a higher sensitivity , so it is good for measurement .
depends linearly on the total effective magnetic flux @xmath111 around @xmath102 approximately . here , @xmath114 , @xmath59 , the frequency detuning @xmath115
is chosen as @xmath116 ( red dashed line ) , @xmath58 ( black solid line ) , and @xmath117 ( blue dot line ) . (
b ) intensity distribution of the three - site ring - shaped optical lattice generated by two laguerre - gauss laser beams.,width=8 ] as a result , the instantaneous value @xmath118 can be obtained by measuring @xmath119 directly .
the cumulative rotation angle is just an integration of the angular velocity over time,@xmath120therefore , such a system can measure the angular velocity and orientation precisely as a gyroscope . actually , the most famous modern gyroscope is the laser gyroscope based on sagnac effect @xcite ( two waves propagating along the opposite directions though a closed rotating ring will induce a relative phase difference , and the position of the interference fringes depends on the angular velocity ) .
our proposal is an extention of the sagnac gyroscope since the interference effect is also used .
however , the difference of the two output currents is measured here instead of observing the interference pattern directly , which may enhance the sensitivity of the results . our asymmetric coherent scattering system may be realized by trapping atoms in a ring - shaped optical lattice . the optical lattices with periodic boundary condition have been produced by the superposition of two laguerre - gauss ( lg ) laser modes theoretically and experimentally amico05,arnold07 . the lg beam with frequency @xmath15 , wave vector @xmath37 , and amplitude @xmath121 propagating along the @xmath122 axis
has the form@xmath123where @xmath124 and @xmath125 is the beam waist .
@xmath126 is the generalized laguerre polynomial . to generate a 1d optical lattice ring with @xmath0 traps
, parameters are chosen as @xmath127 , @xmath128 , @xmath129 , and @xmath130 .
the intensity distribution of the @xmath44 case is shown in fig .
[ gyroscope](b ) .
in this section , we generally consider in which case the asymmetric transmission can happen for a general multi - terminal tight - binding system . for an arbitrary single - particle tight - binding model with attaching leads
, we will derive the condition of the asymmetric transmission @xmath131 between lead @xmath132 and @xmath38 by using the green s function formalism introduced in ref . @xcite and references therein .
we first briefly review green s function method for multi - terminal system .
the retarded green s function of the system as a function of input energy @xmath133
is@xmath134where @xmath135 is the hamiltonian of the central system , and @xmath136 as the total self - energy denotes the contribution of the half - infinite leads , @xmath137 . on the single particle basis @xmath138 , @xmath139 , @xmath140 , @xmath141 , both @xmath135 and @xmath142
are @xmath143 matrix .
especially , each @xmath142 has only one non - zero matrix element at the cross point of row @xmath132 and column @xmath132 , @xmath144which is because the @xmath132th lead is only attached to the @xmath132th site of the central system via @xmath145 } + \mathrm{h.c.})$ ] .
then the advanced green s function @xmath146 and @xmath147 matrix @xmath148 $ ] are obtained sequentially . the transmission spectrum from lead @xmath132 to lead @xmath38 ( @xmath149 )
is given by @xmath150 .
\label{greenfunction}\]]applying it to the ring system mentioned in sec .
ii - iv , one can see that the green s function formalism gives the same exact solution as the one form bethe ansatz method .
the equivalence of the two approaches is proved in the appendix .
in general , the green s functions @xmath151 ( eq . ( [ gr ] ) ) and @xmath152 are complex . from eq .
( [ greenfunction ] ) , the transmission coefficient @xmath153 is @xmath154 \notag \\ & = & ( \gamma _ { q})_{qq}\mathrm{re}(g^{r})_{qp}(\gamma _ { p})_{pp}\mathrm{re}% ( g^{r\dag } ) _ { pq } \notag \\ & & -(\gamma
_ { q})_{qq}\mathrm{im}(g^{r})_{qp}(\gamma _ { p})_{pp}\mathrm{im}% ( g^{r\dag } ) _ { pq } \notag \\ & = & ( \gamma _ { p})_{pp}(\gamma _ { q})_{qq}\left [ \mathrm{re}(g^{r})_{qp}^{2}+% \mathrm{im}(g^{r})_{qp}^{2}\right ] \notag \\ & = & ( \gamma _ { p})_{pp}(\gamma _ { q})_{qq}\left\vert ( g^{r})_{qp}\right\vert ^{2 } , \label{tpq}\end{aligned}\]]where we have use the identity @xmath155together with the property that the @xmath156 matrix has only one non - zero matrix element . similarly , for the inverse transport,@xmath157therefore , once @xmath158we have a symmetric scattering @xmath159 . when this condition is not satisfied , i.e. , @xmath160 , an asymmetric scattering is obtained .
( gr ) tells us that the condition is not only related to the intrinsic symmetry of the central hamiltonian @xmath161 , but also depends on the features of the leads .
for example , if we only attach two leads on the same @xmath1-site central ring investigated in sec .
iii , the asymmetric transmission phenomena will disappear .
this result is in agreement with the viewpoint in ref . buttiker85
that the simplest example must involve at least three channels .
since @xmath135 is hermitian and @xmath136 is diagonal , the term@xmath162can be written as a matrix as@xmath163the diagonal matrix elements are noted by @xmath164 , while the off - diagonal ones are denoted by @xmath165 .
the inverse matrix of @xmath166 has the form @xmath167where @xmath168 is the adjugate matrix of @xmath166 , which is the transpose of the matrix of cofactors .
since the determinant@xmath169 is a constant , our problem is reduced to the condition of the identity @xmath170 . since a matrix and its transpose
have the same determinant , we notice that @xmath171 and @xmath172 have the form@xmath173here , @xmath174 is a polynomial function constructed by @xmath164 and @xmath165 . generally speaking ,
if all of the @xmath164 and @xmath165 are non - zero random complex numbers , @xmath175 is probably not equal to @xmath176 .
however , under some specific conditions , the asymmetric transmission will disappear .
now we focus on the resonant systems ( there is no chemical potential difference between the leads and the central lattice ) and take some typical cases as an illustration .
\(1 ) obviously , when there is no external magnetic flux breaks the time reversal symmetry , we have all the @xmath177 , then @xmath178 , the condition eq .
( [ condition ] ) is automatically satisfied .
\(2 ) when all the @xmath164 are real , i.e. , @xmath179 , @xmath180 , the condition eq .
( [ condition ] ) is also satisfied . because of eq .
( selfenergy ) and ( [ ekq ] ) , this case requires that the incident momentum @xmath37 is either @xmath181 or @xmath182 , which corresponds to injecting an electron at the top or bottom of the bloch energy band .
thus it is not very easy to be realized in practice .
\(3 ) a realizable but not trivial case : the symmetric scattering can be obtained even though the magnetic field is present if the geometry of the lattice is bipartite and electrons are injected at @xmath59 .
a bipartite lattice is a lattice that can be divided into a and b sublattices , such that a site in a is connected only with sites in b and vice versa . at this time
, the function @xmath174 is a polynomial contains only either even order of @xmath164 or odd order of @xmath164 . in other words , if one term of @xmath174 is a product of even number of @xmath164 and some @xmath165 , no term of @xmath174 contains the product of odd number of @xmath164 , thus @xmath174 is called even with respect to @xmath164 .
the situation is similar when @xmath174 is odd with respect to @xmath164 .
the momentum for @xmath59 is @xmath183 , then all the @xmath164 are pure imaginary , @xmath184 . therefore , @xmath185 if @xmath174 is even with respect to @xmath164 , while @xmath186 if @xmath174 is odd with respect to @xmath164 . for both cases
, we have @xmath187 which represents a symmetric scattering .
\(4 ) for the bipartite central lattice , @xmath188 equals to @xmath189 even though the input energy @xmath190 if one of the following conditions is satisfied .
( i ) @xmath0 is odd , there is no lead connected to the sublattice with @xmath191 sites , or there are no more than two leads connected to the sublattice with @xmath192 sites .
( ii ) @xmath0 is even , there are no more than one lead connected to either sublattice a or sublattice b.
in this paper , we take the three - site tight - binding ring as an explicit example to study the scattering problem of some coherent nanostructure with or without time reversal symmetry by virtue of exact solutions .
it is discovered that , induced by the threaded or effective magnetic flux , the asymmetric transmission can happen in such a system . with this interesting property , on the one hand ,
such a system behaves as a quantum diode for a charged particle due to the pure quantum interference effect . on the other hand ,
it can also be served as a gyroscope for either a charged or neutral particle when a global rotation providing an effective magnetic flux .
we have shown that the difference of the output currents is linear proportional to the rotational angular velocity approximately .
the observable effects presented in this paper are hopeful to be realized experimentally in the quantum dot system and the ultra - cold atom optical lattice system .
we also try to generalize the results from the three - site system to the general ones . according to the green s function approach @xcite
, it is demonstrated that the intrinsic symmetries of the whole system , including the geometry of lattice and leads , determine the symmetry of scattering coefficients together .
protected by the bipartite configuration , a symmetry scattering may be obtained even if the time reversal symmetry is broken for the central system .
this work is supported by nsfc no .
10474104 , 60433050 , 10874091 and no . 10704023 , nfrpc no .
2006cb921205 and 2005cb724508 .
in this appendix , we prove that the bethe ansatz method and green s function method are equivalent for determining the transmission coefficients .
the hamiltonian of a general central system with @xmath0 sites reads@xmath193we suppose that there is a set of @xmath8 defined as @xmath194lead@xmath195 , where @xmath196 .
if @xmath197lead@xmath198 , a half infinite tight - binding lead is attached to the @xmath8th site .
the total hamiltonian of leads is@xmath199}+t\sum_{j^{\prime } = 1}^{\infty } b_{j^{\prime } } ^{\dag \lbrack j]}b_{j^{\prime } + 1}^{[j]}+\mathrm{h.c.}\right ) .\]]according to the bethe ansatz method , the scattering state is supposed to be@xmath200}\left ( j^{\prime } \right ) b_{j^{\prime } } ^{\dag
\lbrack j]}\left\vert 0\right\rangle , \]]where @xmath201}\left ( j^{\prime } \right ) = b\left ( j\right ) e^{ikj^{\prime } } + \delta _ { j , p}e^{-ikj^{\prime } } .
\label{awaveb}\]]the @xmath132th lead has been chosen as the input lead . from the schrdinger equation@xmath202@xmath203 and @xmath204}\left ( j\right )
$ ] should satisfy@xmath205}\left ( 1\right ) = e\psi _ { a}\left ( j\right ) \label{ascheq}\]]for any @xmath8 , and @xmath206}\left ( 2\right ) & = & e\psi _ { b}^{[j]}\left ( 1\right ) , \notag \\ -t\left [ \psi _ { b}^{[j]}\left ( j^{\prime } + 1\right ) + \psi _ { b}^{[j]}\left ( j^{\prime } -1\right ) \right ] & = & e\psi _ { b}^{[j]}\left ( j^{\prime } \right ) \label{apsib}\end{aligned}\]]for @xmath197lead@xmath198 .
( [ awaveb ] ) and ( [ apsib ] ) give the dispersion relation@xmath207together with the equation for @xmath197lead@xmath198 , @xmath208for future convenience , we define a column vector @xmath209 , @xmath210eq .
( [ ascheq ] ) can be simplified as@xmath211 x_{j^{\prime } } \notag \\ & = & \left ( h_{c}\right ) _ { j , p}-\delta _ { j , p}\left ( e+\frac{g^{2}}{t}% e^{-ik}\right ) , \label{alineqs}\end{aligned}\]]which is a system of linear equations in the variables @xmath212 . in the matrix form ,
( [ alineqs ] ) becomes@xmath213where@xmath214the matrix elements of @xmath166 , @xmath215 , and @xmath136 are @xmath216where @xmath217 .
the solution of @xmath212 can be given by the cramer s rule @xcite .
without loss of generality , we focus on the solution of @xmath218 with @xmath219lead@xmath198 but @xmath220,@xmath221here , the matrix elements of @xmath222 are@xmath223the determinant of @xmath222 equals to the one of another matrix @xmath224 , where@xmath225and @xmath226is a real number .
so that@xmath227@xmath228 is the @xmath229th algebraic cofactor of @xmath166 , which is actually @xmath230 times the determinant of the submatrix obtained by removing @xmath132th row and @xmath38th column from @xmath166 .
therefore , the transmission coefficient from lead @xmath132 to lead @xmath38 is@xmath231the definition of the inverse of a matrix tells us that eq .
( [ ascheq ] ) is actually@xmath232 on the other hand , from green s function method eq .
( [ tpq ] ) , @xmath233where @xmath234 . then eq .
( atpqg ) and ( [ atpqb ] ) are same , the bethe ansatz method is equivalent to the green s function method for determining the transmission coefficients .
moreover , the bethe ansatz method can also gives the exact wave function . | we study the single particle scattering process in a coherent multi - site system consisting of a tight - binding ring threaded by an aharonov - bohm flux and several attaching leads .
the asymmetric behavior of scattering matrix is discovered analytically in the framework of both bethe ansatz and green s function formalism .
it is found that , under certain conditions , a three - site electronic system can behave analogous to a perfect semiconductor diode where current flows only in one direction .
the general result is also valid for a neutral particle system since the effective magnetic flux may be implemented by a globe rotation .
this observation means that the three - site system can serve as an orientation measuring gyroscope due to the approximate linear dependence of the current difference of two output leads on the rotational angular velocity . |
the idea of tipping points has captured the public s attention from topics as diverse as segregation , marketing , rioting , and global warming .
robustness considerations have extended beyond engineering and ecology to political regimes , computer algorithms , and decision procedures . analyzing and exploiting path dependence plays a significant role in technology spread , institutional design , legal theory and the evolution of culture .
however these concepts have not been generally and formally defined and , as a result , the terms uses across these various applications are hardly consistent . at times
a tipping point refers to a threshold beyond which the system s outcome is known .
other times ` tipping point ' is used to describe an event that suffices to achieve a particular outcome , or an aspect of such an event , or the time of such an event .
another use of tipping points is to label the conditions to which the system is most sensitive . the idea is frequently tied up with processes such as positive feedback , externalities , sustainable operation , perturbation , etc . robustness and path dependence share
also in this preponderance of senses and this paper aims to elucidate the distinctions among these and other uses of the terms . to accomplish this conceptual analysis this paper puts forth formal definitions for each concept ( with an implied algorithm ) to measure properties of system dynamics .
the analysis utilizes markov model representations of systems and so definitions of the foundational concepts of system dynamics ( equilibrium , basin of attraction , support , etc . )
are first provided
. then various tipping point - related concepts are described , defined , and illustrated with a simplified graphical example .
this treatment is then repeated for robustness - related concepts and then again for a variety of path sensitivities in system dynamics .
an additional section identifies projects for future work in considerable detail . through presentations and conversations
the techniques presented here have garnered considerable interested from within academics and from government and industry .
the next step is clearly to apply these measures to existing data and models to refine the measures and contribute to science .
a planned extension provides a methodology ( and software ) to ( automatically ) generate the state transition representation from observational and model - generated data .
this software tool is necessary for many of the intended and most useful applications of these measures .
there is an additional and unexpected potential conceptual benefit to the robustness formalization presented here for the philosophical study of dispositional properties .
dispositions such as fragile , soluble , and malleable have long resisted necessary and sufficient conditions to distinguish them from categorical properties ( like red , liquid , and square ) .
the formal definition of one dispositional property ( robust ) may shed some conceptual light on how to proceed .
other research thrusts that extend the conceptual and methodological benefits of what follows are also presented .
this paper applies methods from ergodic theory , network theory and graph theory to whole systems encoded as markov models to find and measure tipping points , robustness , and path sensitivity in system dynamics .
the fusion of these techniques to this purpose is novel , but certainly there is nothing unprecedented about analyzing systems to find these properties or modeling dynamics with markov models .
but since this paper represents the first marriage of these two realms , the story thus far must be told as two separate threads .
much previous work in finding and measuring properties of system dynamics has focused on explanation - the answering of the ` why ' question .
not surprisingly since these papers , books , and discussions were couched in scientific contexts where a particular phenomenon ( or class of phenomena ) required explanation .
each such jaunt into explaining tipping points , robustness , or path dependence was accompanied by a custom - suited methodology capable of generating and detecting that property in the model provided ( to answer the ` how ' question ) .
these models achieved varying levels of generality , but each was limited by the desire to explain the property in a particular model or context .
this is limiting because in order to explain how a process generates ( say ) path dependent behavior one has to model that process explicitly .
the current work is one of pure methodology rather than a purported model of any particular system or causal apparatus .
it is meant to be completely abstract and general and therefore capable of measuring these system properties in any system . because it does not model any generating process it can not address the ` why ' or ` how ' questions .
it is not meant to .
this paper answers the ` whether ' and ` how much ' questions .
these questions are also asked in previous work , but results could not be compared between models because the methodology was model - specific .
a general methodology provides a framework through which all modelers ( and some data analysts ) can determine whether and how much of each of these properties of system dynamics obtains @xmath0 and compare results across models regardless of the generating mechanisms .
the ability to compare measures across systems is achieved through a focus on scale - free measures - measures that do not depend on the size of the system being analyzed .
this framework allows scientists to focus on making appropriate models of their subject - matter by eliminating the burden of figuring out how to measure these properties for their model .
here i review some previous work that includes the similar measured properties to highlight where this methodology might prove useful .
the term ` tipping point ' was first coined by morton grodzins in 1957 @xcite to describe the threshold level of non - white occupants that a white neighborhood could have before `` white flight '' occurred .
the term continued to be used in this context through the work of eleanor wolf @xcite and thomas schelling @xcite who also extended the concept to other similar social phenomena . though these researchers had a specific usage with narrow focus , the idea of a critical parameter value past which aggregate behavior is recognizably different spread across disciplines where its meaning and application varied considerably .
malcolm gladwell s pop sociology book _ the tipping point _
@xcite has played a significant part in bringing the term to the public s awareness .
the notion of tipping point most frequently used by gladwell is an event that makes something unusual ( such as hush puppy shoes ) become popular .
more precisely this is a critical value for producing a phase transition for percolation in certain heterogeneous social network structures .
this form of tipping point behavior also appears in the work of mark granovetter @xcite and peyton young @xcite for the propagation of rioting behavior and technology respectively .
this version of tipping will play only a minor role in what follows , however the fact that the expression has made it into the everyman s conceptual vocabulary boosts the importance of establishing rigorous scientific definitions to disambiguate loose usage .
a recent trend in reports of climate change is to refer to a hypothesized tipping point in global warming and ice cap and glacial melting .
james e. hansen has claimed that `` earth is approaching a tipping point that can be tilted , and only slightly at best , in its favor if global warming can be limited to less than one degree celsius . ''
@xcite this usage reflects hansen s belief that `` humans now control the global climate , for better or worse . ''
gabrielle walker states , `` a tipping point usually means the moment at which internal dynamics start to propel a change previously driven by external forces .
@xcite '' it is unclear whether walker s and hansen s comments are compatible ; the conceptual ambiguity of the terms may be making them talk past each other .
but even if their usage is meaningful within their fields , they fail as general characteristics . identifying tipping points ( as a property of system dynamics ) should not depend on whether humans are in control of system behavior or what is driving the dynamics ( even if explaining why those are the dynamics does ) .
but not all heretofore definitions of the term ` tipping point ' have been loose or subject - matter specific .
it is often deployed as a semi - technical term in equation - based models of various sorts .
for example , it can refer to an unstable manifold in a differential equation model , the set of boundary parameters for comparative statistics @xcite , or inflection points in the behavior of functional models .
each of these uses of ` tipping points ' conforms to our intuitive sense of the term s meaning and at some slightly higher level of abstraction these tipping behaviors are the same - and match the definitions provided in this paper . but not all models can be faithfully represented as systems of equations and this limits the usefulness of equation - dependent definitions .
one set of tests we can perform on the compatibility of current analysis is to generate markov models based on the existing differential equation and comparative static models and then determine whether the definitions provided here identify the same states as tipping points .
such a project is left for future work .
robustness considerations are already a common analysis path for researchers in many fields : ecology , engineering , evolutionary biology , logistics , computer science , decision theory , and even statistics .
models in these fields are often developed specifically to enhance system robustness , avoid system failures , mitigate vulnerabilities , and otherwise cope with variations in an unpredictable environment .
these previous analyses provide some understanding of what features make certain systems persist and others fail , but there is little in the area of general theory .
one hope is that the present construction of general measures of robustness - related concepts will inform and facilitate the construction of general theories of what systemic features produce these properties of system dynamics .
if it can reveal that robust configurations and dynamics in these varied fields can be captured by a single measure , then we will have taken the first step towards a unified theory of robustness .
understanding how social systems can be both simultaneously flexible and strong has garnered increasing interest recently . in an upcoming book
jenna bednar investigates how institutional design can affect the robustness of a federalist governing body @xcite .
`` by explicitly acknowledging the context dependence of institutional performance , we can understand how safeguards intersect for a robust system : strong , flexible , and able to recover from internal errors . ''
bednar has identified the properties that make institutional system robust ( compliance , resilience , and adaptation ) in a way that is somewhat specific to the subject matter . that is beneficial and to be expected for explaining and improving the robustness of political institutions .
such an analysis stands to gain from the conceptual refinements derived from formal measures of multiple robustness - related features of system dynamics .
bednar s work especially underlies the thought that understanding many systems of interest requires more than traditional equilibria analysis ; the dynamic nature of dissipative structures ( see example [ dissipativestructure ] ) requires new notions of stability , resilience , and robustness that i hope to help inform through the provided measures .
thomas sargent has made extensive use of principles from robust control theory in his analysis of monetary policy and pricing ( and other topics ) .
the sense of ` robustness ' used in robust control theory is a gap between modeled levels and actual levels of parameters .
it is used to formalize misinformation , uncertainty , and lack of confidence in agents knowledge and , more generally , to facilitate high levels of performance despite errors and in known less - than - ideal conditions .
this sense of robustness applies across the decision theoretic sciences and planning literature ( e.g. the work of rob lempert and company @xcite ) . however , not all robustness analyses are to cope with uncertainty . in genetics
the term robustness refers to a species consistency of phenotype through changes in the genotype .
robustness can be considered at two levels : 1 ) through how much mutation is a member of a species viable and 2 ) how much genetic variation is required to transform a species physical characteristics .
the first level takes genetic profiles of organisms and determines which can survive to reproductive age and which can not ( or are sterile ) .
the number of genotypic variations that remain viable is a measure of the species robustness @xmath0 according to that usage . on the evolutionary time scale
we wish to understand how incremental genetic drift is responsible for large phenotypic variations over time .
walter fontana has demonstrated that a network of neutral mutations ( ones that do not affect fitness ) can sufficiently explain the observed punctuated equilibria ( see example [ punctuatedequilibria ] ) in species evolution @xcite . though fitness may remain neutral through some genetic variation , the connection between fitness change and phenotype change is strong . a model that tracks fitness through genetic variations
could then approximately measure how robust each stage in the evolutionary progression is .
it is clear that these two concepts of robustness are distinct ; and they are both distinct from control and decision theories usage as well .
we can add robustness measures from statistics and computer science to the variety of senses that ` robustness ' can take . in computer science an algorithm , procedure , measure , or
process is robust if small changes ( errors , abnormalities , variations , or adjustments ) have a proportionally small affect on the algorithm , procedure , measure , or process .
the time complexity of two algorithms may change in different ways .
algorithm a may require one step per input ( @xmath1 ) and algorithm b may require one step per two to the power of the input size ( @xmath2 ) ; in this case algorithm a is more robust to changes in input size .
statistical robustness is either when an estimator performs sufficiently well despite the assumptions required by that estimator being violated or when ( like in computer science ) a measure changes little compared to changes in the input . for example , the median is a more robust measure than the mean because to alter the median a data point has to cross the median point , whereas any input value change will change the mean s value . and there are more variations in information theory , data security , engineering , law , ecology , and just about every field has their own version of robustness .
they all share certain high - level conceptual commonalities , but differ in their details and criterion for application .
the definitions below produce necessary and sufficient conditions for the application of several robustness - related concepts .
there are many different ways in which systems can cope with variation are each has its own definition
. this level of refinement ( combined with a very inclusive markov modeling technique ) may be able to bring discussions of robustness in the different fields to a single table and foster inter - discipline research .
the level of interest in explaining path dependent processes has risen in recent years .
this is in part due to an overall appreciation for the importance of such dynamics in complex systems across domains .
this is also partly due to a natural need to explain observed path dependent phenomena such as convention lock in ( e.g. qwerty keyboards ) , climate change , and political instability . and
another part is due to an increased prevalence in models wherein path dependence could potentially be formally measured .
each technique to measure path dependence requires a definition to characterize path dependence in a manner measurable by the formal machinery presented . because previous work focused on explaining path dependence through demonstrating sufficient mechanisms to generate it , and this paper s goals are to provide a general system - level definition and way to measure it
, the current work will only barely touch on previous research .
yet insofar as the definitions should be compatible it is worth taking a look at previous , recent formal definitions of path dependence in the literature . according to james mahoney @xcite
there are three basic characteristics of path dependence in the social science literature .
the first type of path dependence appreciates sensitivity to events that take place in the early stages of a sequence of events .
secondly , there are some historical events early in the sequence that are not explained by prior events . finally , the sequence of events exhibit some kind of `` inertia '' culminating in an equilibrium - type outcome .
though later analyses ( including this one ) deny these characteristics in favor of other ones , this work does demonstrate some popular thoughts in the formal literature on path dependent processes .
paul pierson presents one definition that contrasts with mahoney s . in pierson
s work path dependence describes how early historical events act to select among multiple possible equilibria .
this equilibria selection process , however , is due to exogenous shocks to the system , a characterization which does not seem necessary for a definition of path dependence .
it does ring true that if only one equilibria - like outcome is possible then system ought to be characterized as * path independent * ( or at least that the outcome be characterized as path - independent ) .
kollman & jackson have a variant of page s definition of path dependence ( see below ) - one that applies only to specifically parameterized dynamical system models and requires `` very specific and stringent conditions in order for there to be path dependence @xcite . ''
path dependence is revealed when a certain time - varying autoregressive parameter is ( or converges to ) one as the system s dynamics progress through specified shocks .
one of their results is that `` several steps ought to be taken prior to proposing that some process is path dependent .
such a proposal should be based on rigorous analysis of the process at issue .
we can not offer a set of computational tools that can be used ` off the shelf . '
there is no substitute for theoretical modeling appropriate to the system under study as a first step.@xcite '' indeed , building a data - generating model is a necessary first - step for measuring path dependence in most cases .
however , if data satisfies certain properties ( described in the next section ) then the methodology presented in this paper can be used `` off the shelf '' to measure the degree of many forms of path dependence ( see _ automatically generating the markov model _ in the future work section for more details ) . if one is primarily interested in finding out whether and how much path dependence a process produces , however , the model need not be specified with the details required by their methodology
such model - level tinkering is only necessary to provide an explanation of whence the path dependence .
but if we accept that there are very many mechanisms that can generate path - dependence and that this feature of the mechanisms will be revealed through the data they produce , then we can ignore the mechanisms specifics when measuring their degrees of path dependence .
this is the approach taken here .
the work of scott page in identifying the types and causes of path dependence @xcite is closest in flavor to the work presented here .
though again an example of mechanism identification , the motivation identified in his introduction applies equally to this paper and is worth including here . _
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ attempts to extend what is meant by path dependence reflect a need for a finer unpacking of historical causality .
we need to differentiate between types of path dependence .
the way to do that is with a formal framework .
an obvious advantage of having such a framework is that we can conduct empirical analyses and discern whether the evidence supported or refuted a claim of the extent and scope of the sway of the past .
that said , empirical testing of a framework of causality is far from the only reason for constructing a framework for modeling historical forces .
formal models discipline thicker , descriptive accounts @xcite . by boiling down causes and effects to their spare fundamentals ,
they enable us to understand the how s and why s ; they tell us where to look and where not to look for evidence .
they also help us to identify conditions that are necessary and/or sufficient for past choices and outcomes to influence the present .
@xcite _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ the differences that page recognizes in the underlying forces driving law making , pest control , or technology choice are points well taken .
insofar as the different forces generate differentiable system dynamics the techniques of this paper will be able to identify and precisely measure how much and in what ways past states influence the future .
the properties defined through the formal framework presented below is intended to guide scientists towards characteristics of the original system that merit closer examination .
markov modeling has a long history in mathematics , engineering , and in applications to fields as diverse as condensed matter physics , genomics , sociology , and marketing .
techniques to use markovian processes to uncover information about system dynamics fall within the field of ergodic theory . several properties and techniques from standard markov modeling
are employed below . in their abstract form
one can compute features such as the equilibrium distribution , expected number of steps between two states ( with standard deviation ) , reversibility , and periodicity .
these features gain added meaning when interpreted for the system being modeled , but this paper utilizes them as part of defining ( and creating algorithms to uncover ) interesting system dynamic properties .
computer scientists have long been analyzing networks in the form of actual communication networks as well as various abstractions from these problems .
they have invented several useful measures and exceptionally well - crafted algorithms to calculate connectedness , load - bearing properties , path switching , transmission speed , and packet splitting and fusion to name a few .
my analysis borrows heavily from this work in terms of algorithms , though each has been repurposed to the abstract markov model system representation .
computer science is also the home of finite state machines : mathematical objects that share their states & transitions structure with markov models ( though state machines are frequently not probabilistic ) .
the states of a finite state machine represent the internal states of some agent and the transitions represent the behavior rules by which agents change their states .
few of the techniques invented to analyze finite state machines will apply to this research because few are adapted to purely probabilistic transitions .
hardware engineers and their physicist partners have worked out several interesting measures for circuit design problems .
multiple paths , variable resistance , flow injection , capacitance and many other characteristics of electronic circuits have analogs in the markov models presented below .
though these are only partially explored at this stage , future work will look deeply at borrowing techniques from circuit research . and
finally graph theory offers a few useful measures for our purposes , and moreover provides a wealth of definitions for graph structure and node relationships .
structural properties will play a larger role in future work addressing changes of resolution and in establishing equivalence classes of system dynamics .
also , many of the features that graph theory identifies have been given alternative definitions that underlie the probabilistic nature of the markov model analysis .
few explicit references to previous work in these methodological subjects appear below because the methods used generally fall in the category of common knowledge .
when a specific algorithm or specialized technique is used , a reference is provided .
the measures defined here are meant to stand on their own as improvements in our conceptual understanding of the included features of system dynamics . by differentiating and formally defining these properties of processes we gain both a common vocabulary with which to discuss our models and a detailed typography of behavior to include and detect in system models .
many of the applications i have in mind are to include these measures in constructive models across multiple disciplines where the models are iterated with multiple initial settings and/or have stochastic parameters . these include game theoretic models , network models , physical models , and the whole gamut of models
which may be considered agent - based .
certain static data sets , the sorts collected by surveys , are also analyzable via this methodology .
the data must satisfy certain criteria to be thus analyzable - basic properties it must have before even considering the statistical issues involved with particular applications .
there must be 1 .
data across time ( because we re measuring properties of dynamics ) 2 .
repeated system states ( so the markov model is nt deterministic ) 3 . known ( or known to be fixed ) time between observations to have repeated system states from collected data we typically will need data from multiple independent trials . in some cases independence will not be true but will be a useful / necessary approximation .
for example , voter data from each state or county are not truly independent , but each state or county could constitute a separate trial . in many ways these considerations parallel issues already present in statistical analysis , though the motivation for these requirements are quite different .
other features of typical statistical analysis will also show up in this technique s application ( e.g. correlation , covariance , fixed effects , kurtosis ) , but they will not be highlighted except where doing so enhances the discussion . in some cases
the data may be recoded or otherwise translated into system states in such a way to as to highlight those features of the system we wish to track and uncover the tippiness , robustness and path sensitivities of . in some cases this will be a resolution choice , in others this will be a shift to recording markov states as rate changes , and in others it may be converting fixed - time dynamics into event - drive dynamics .
it will take expertise to determine which , if any , conversion is necessary and a great deal of trial and error to develop that expertise . eventually standards will be uncovered as people build proficiency in this methodology .
a final consideration has nothing to do with the structure and format of the data . even if the data is amenable to the analyses presented below
, the data may not be observations from a system for which we think robustness or path sensitivity apply .
for example , even if we could run the robustness analysis on voter data , it is not clear what the result would be telling us .
similarly , given any set of real numbers we can calculate the mean value , but there is no useful interpretation of that mean value for some sets of real numbers ( e.g. telephone numbers ) .
path sensitivity , however , is something we expect to uncover in poll data and so the amenability of the voter data to the markov modeling enables this analysis .
the modeling and analysis techniques need to be both possible _ and _ appropriate , but only a human can determine appropriateness .
one promising application of the measures and methods defined in this paper is in developing self - managing large - scale systems - so - called _
autonomic systems_. autonomic systems apply top - down measures of their internal mechanisms to adapt to changes in resource needs and availability .
some computer systems , such as internet routers , satellites , autopilots , and explorer robots , use autonomic management programs .
another major application could be logistics management : the routing of parcels , fuel , food , luggage , etc . to minimize service breakage and costs
. with formal measures of robustness in hand ( especially in combination with understanding of path sensitive trajectories ) these systems could guide themselves to maintain their function and cope with environmental perturbations using levers uncovered through the analyses presented below . because my research extends into each of these research arms the potential gain for having these measures constitutes a significant personal motivation to develop them .
at its most basic a markov model is a collection of states and a set of transition probabilities between pairs of states .
there are several ways to represent a markov model , but the standard techniques are to 1 ) model the states as vertices ( nodes ) and the transition probabilities as weighted edges of a graph or 2 ) use such a graph s corresponding adjacency matrix or edge list .
different applications of markov modeling take different system features as the states , but the nodes in the markov models used here represent a complete description of a state of the system ( see below ) .
the transition probabilities represent either observed system dynamics or theoretically posited state changes . given that states and transitions are defined this way it is clear that the set of states and their transition probabilities are constant for the markov models utilized in this paper . before beginning the breakdown of the aforementioned phenomena into their various categories ,
a general typology of state spaces will be helpful .
a system state is a complete set of instantiations of the aspects of the system ( values for variables , existence for agents , etc . ) . throughout
we will analyze systems with a finite ( but possibly arbitrarily large ) number of states each with a finite number of aspects .
insofar as some parameters may take on unbounded values ( e.g. a continuum of real values ) bounding the number of the parameters ( i.e. dimensionality of the parameter space ) does not ensure a bounded or discrete state space .
the analysis that follows is limited to a finite , discrete state space achieved by binning continuous parameters .
[ aspects ] a state in the markov model is a complete specification of the @xmath3 aspects of one configuration of the system .
@xmath4 where @xmath5 is the value of aspect @xmath6 in state @xmath7 [ samestate ] two states are represented as one state of the markov model if all the aspects of the two states are identically valued .
@xmath8 an obvious ( but still useful ) corollary results directly from the truth values of biconditional equivalence .
[ difference ] a difference in any aspect marks a different state of the system .
@xmath9 if our system is an iterated strategic - form game played by six players @xmath10 each with four possible actions @xmath11 then each state of the system has six aspects and each aspect takes on one of four values .
that is @xmath12 and a particular state @xmath13 might be @xmath14 .
there are @xmath15 combinations of four actions for six players , but the markov model may not include all of them .
recall that the model is expected to be built from either collected data or a theoretical model so some combinations of aspect values may be unobserved or theoretically impossible or irrelevant . to analyze schelling s segregation model
@xcite we have several options for how we capture the states of the system .
consider an @xmath16 grid with 50 agents of two types .
we could choose to track the @xmath17 and @xmath18 coordinates of each agent in the model which would generate states with 100 aspects ( two for each agent ) .
we could instead track whether each agent is happy with its neighborhood so there would be 50 binary _ \{yes , no } _ aspects to each state . alternatively the aspects could represent what is in each of the 64 grid spaces with values from _ \{empty , agent type1 , agent type2}_. note that not all @xmath19 combinations are possible because there are fixed numbers of each type of agent and two grid spaces change every transition ( agent moving ) . these and other specifications could be combined with each other and/or measures of the configuration ( e.g. how clustered the agents are ) .
the choice of what to count as the aspects of the states will determine what the measures defined below can reveal .
a set of @xmath20 states is demarked with boldface type : @xmath21 .
the set of all the states in the markov model is @xmath22 which has size @xmath23 thus @xmath24 is also the number of nodes in the graph representation .
the state of the system at time @xmath25 ( denoted @xmath26 ) changes to @xmath27 in discrete , homogenous time intervals .
state transitions are probabilistic and specified by the system s transition diagram or matrix ( see figure [ markovexample ] ) .
we write the probability of transitioning from state @xmath28 to state @xmath29 as @xmath30 .
it will later be useful to denote the set of transitions @xmath31 and the size of this set as @xmath32 .
m n p2 mm c & p p p p 1 & 2 & 3 & 4 + + [ cols= " > " , ] & .5 & 0 & 0 & .5 + 0 & 0 & 1 & 0 + 0 & .9 & .1 & 0 + + & = + + [ o][f]@=15 mm 1 @(l , u)[]^.25 _
.25 ^.5 & 2 ^.5 ^.5 + 3 @(l , d)[]_1 & 4 ^.1^.9 + ( a ) transition matrix & ( b ) transition diagram + following the standard definition from probability theory : [ probability ] the sum of a state s exit probabilities must equal one .
@xmath33 the entry in the transition matrix at row @xmath7 and column @xmath34 represents @xmath35 and so each row must sum to 1 .
the probability of a state change equals the probability that each of the aspects of the state changes .
@xmath36 this theorem , which relates state transitions back to changes in their constituent aspects , follows directly from definition [ samestate ] of state sameness and applies to self transitions ( i.e. @xmath37 ) as well .
it is important to note that the state change probability is not the sum of the aspect change probabilities .
each single or multiple aspect change produces a distinct , independent state change with its own probability .
this property ( and others related to aspect changes ) will be useful in the discussion of levers below .
to use a markov diagram to represent system dynamics we will need to define various types of system behaviors in terms of system states , sets of states , and state transitions . as a preliminary to the common features of system behavior
i will present definitions of some structural features that will be utilized . in graph theory a path ( of length @xmath38 )
is typically defined as a set of vertices and edges satisfying the schema @xmath39 where the edge @xmath40 links the vertex @xmath41 to @xmath42 @xcite .
self - transitions , which represent both a lack of change and a change too small to count as a state change , are an important feature of markov modeling and hence both nodes and edges may be repeated along paths .
so ` path ' as it is used here is the broader notion sometimes called a ` walk ' in the graph theory literature .
since there are neither multi - edges nor hyper - edges in a markov diagram the set of vertices ( or the set of edges ) alone is sufficient to uniquely specify a path as long as successively repeated vertices ( or edges ) are interpreted as self - transitions .
a path in a markov model could be defined as a set satisfying the same schema used in graph theory , but we will use a slightly different definition to make the probabilistic aspects explicit .
[ path ] a _ path _ is an ordered collection of states and transitions such that from each state there exists a positive probability to transition to the successor state within the collection .
a path from @xmath28 to @xmath29 denoted @xmath43 or @xmath44 is the set of states @xmath45 such that * @xmath46 * @xmath47
* @xmath48 this definition establishes necessary and sufficient conditions for @xmath49 to be a path , but does not provide a schema for specifying a particular path . to specify intermediate states ( _ markers _ )
for the system to pass through we can write @xmath50 to denote a path from @xmath28 to @xmath51 that passes through ( at least ) @xmath29 .
such a path is merely the conjunction of the two subpaths @xmath44 and @xmath52 .
any number of markers can be thus specified .
the order of the states specified must be satisfied by the path taken , but it does preclude other states from being visited between the marked states . to specify a long sequence of path markers this paper uses the notation @xmath53 . to completely specify each state along a path we adopt the notation @xmath54 for short sequences and @xmath55 for long ones . [ exactpath]an exact path ( @xmath56 ) satisfies definition [ path ] of a path above .
@xmath57 where for each subpath the @xmath58 in item ( ii ) of definition [ path ] equals 1 .
though the term ` transition ' appeared above in the general description of markov models , it was not defined precisely .
[ transition ] the _ transition _ from @xmath28 to @xmath29 is @xmath59 .
[ length]the _ length _ of a path is the number of transitions taken between the first and last states .
@xmath60 this ( possibly overly complicated ) formal definition of length simply uses features of the definition of path above , but it is equivalent to the number of edges traversed along the path .
a path built from a set of states is at least as long as the number of states in the set .
@xmath61 this theorem follows from the fact that states may be revisited along the path .
the definition of length suffices for completely specified paths , but the length of merely marked paths can take a range of values depending on the exact sequence of nodes visited .
[ cycle ] a _ cycle _ is a path that starts and ends with the same state .
@xmath62 a cycle of length one is a self - transition .
this observation follows directly from the definition of a transition and theorem [ exactpath ] .
[ elementarypath ] an _ elementary path _ is a path that visits each state within the path exactly once from this definition it is clear that an elementary path is a path with no cycles - including no self - transitions .
a restriction to elementary paths is particularly helpful for ascertaining certain features ( e.g. path existence ) because simple algorithms exist for them and because of the following property .
the length of an elementary path equals the number of states in the path .
though this fact obviously follows from the definition of path with the exclusion of cycles , providing a formal definition is trivial and obvious and will be omitted . graph and network theorists have developed a great many algorithms for finding paths , calculated their lengths , and measuring properties germane to their application in those fields .
some of those will come up later in measuring properties of system dynamics , but the definitions and theorems presented above will suffice to move forward in examining our markov models .
this subsection provides definitions and sketches of some algorithms for common structural properties of markov models used to represent system dynamics .
many of these features have existing definitions in terms of matrix operations or limiting distributions , but this paper will present alternative definitions in many cases .
the definitions will make focused use of the finitude of the markov models , the resolution of the states , and the granularity of the probability measurements .
my motivation for the alternative definitions is to facilitate clear intuitions about a system s processes and how to measure them precisely .
[ equilibrium ] a system state that always transitions to itself is called an _ equilibrium _ or _
stable state_. , stability refers to a tendency to self - transition .
hence an equilibrium is equivalent to a * fully * stable state . ]
an equilibrium @xmath40 is a state @xmath28 such that @xmath64 in some cases a set of states plays a role similar to that of an equilibrium .
[ orbit ] an _ orbit _ is a set of states such that if the system enters that set it will always revisit every member of the set and the system can never leave that set .
@xmath45 is an orbit if @xmath65 and @xmath66 [ oscillator ] an _ oscillator _ is an orbit that is also a cycle .
equilibria can not be proper subsets of an orbit assume @xmath67 .
from definition [ orbit ] @xmath68 such that @xmath69 .
since @xmath40 is in @xmath70 definition [ orbit ] implies @xmath71 .
definition [ equilibrium ] implies that @xmath72 which further implies that @xmath73 . for all @xmath74 , @xmath75 .
therefore by contradiction with definition [ orbit ] @xmath76 . given the definitions of equilibrium and orbit above it is clear that an equilibrium is an orbit of just one state . [ attractor ] an _ attractor _ ( denoted @xmath77 ) is either an equilibrium state or an orbit of the system .
the attractor states in @xmath78 are those where @xmath79 as @xmath80 . or in a more computable formulation , given a desired degree of significance for the probability measures @xmath81 such that @xmath82 if and only if @xmath28 is an attractor state .
although it becomes a little more complicated if one wants to isolate the individual equilibria and orbits .
techniques ( such as this one ) for measuring properties ( whether mathematical or computational ) will be presented in an appendix in future versions and are occasionally referred to in this text or in footnotes . ]
the choice of a resolution determines whether an orbit appears as an equilibrium or _
vice versa_. in cases where attractor avoidance is the aim of the model ( see tipping and robustness below ) we can collapse orbits into a single attractor state without loss of information . for this reason i will use ` @xmath77 ' as if it were a single state except in cases where it being an orbit affects the analysis . in later sections we will encounter the idea of a dissipative structure ( see example [ dissipativestructure ] ) for which equilibria analysis is inappropriate .
it is not the case that these systems fail to have attractors , it s just that the goal of such systems is to remain in continual flux and avoid equilibria and other `` point attractors '' ( i.e. attractors that incorporate a small percentage of the total number of states ) .
given the definitions established above every markov model must have some set of states satisfying the conditions for being an attractor .
[ minattractor ] every system has at least one attractor .
assume that there exists a system with no attractors .
if @xmath45 is not an orbit then by definition [ orbit ] for some @xmath84 it must be the case that either 1 ) there is some time in the future after which @xmath28 does not get visited or 2 ) some state outside @xmath45 gets visited .
let @xmath20 be @xmath85
. for any @xmath20 , case ( 1 ) implies that any orbit must be smaller than @xmath20 and case ( 2 ) implies that there must be at @xmath86 states in the system .
for @xmath87 it is not possible for the orbit to be smaller so any orbit must contain at least @xmath88 states . by induction on @xmath20
it must be the case that if there are no orbits of size @xmath89 ( @xmath90 is the whole system ) then any orbit must be of size @xmath91 which is impossible .
that contradicts the assumption that there exists a system with no attractors , so every system must have at least one attractor .
[ basin ] those states from which the system will eventually move into a specific attractor are said to be in that attractor s _ basin of attraction_. the basin of @xmath77 or @xmath92 is a set of states @xmath45 such that .
see previous footnote for more details . ]
@xmath93 some systems may spend a great deal of time in a basin of attraction before reaching the attractor located within it thus making system behavior in the basin similar to an orbit itself ( also note for the discussion of robust sets below ) . in such cases
it is sometimes helpful to utilize the following property to describe and make inferences about system behavior .
once in a basin of attraction the system can never leave it . @xmath94 [ support ] the _ support _ of a state ( also known as its _ in - component _ ) is the set of states which have a path to it .
the support of @xmath28 or @xmath95 is the set of states such that @xmath96 we can expand this definition to the support of a set of states @xmath45 as the union of the supports of the members of @xmath45 .
some facts relating these features of system dynamics are clear from the above definitions .
an attractor is a subset of its basin of attraction , and an attractor s basin of attraction is a subset of its support .
@xmath97 the equilibrium may be the only member of either its basin or support , but if it is the only member of its support then it is disconnected from the rest of the graph ( e.g. @xmath98 in figure [ equilbasin ] ) .
the _ indeterminate states _ of a system , ones that are not members of a basin of attraction , convey a wealth information about the system s dynamics and its future states . recall that the whole system may be a single orbit and there may be no indeterminate states .
but if there are multiple attractors , then the indeterminate states are the ones in multiple attractors supports .
[ overlap ] the _ overlap _ of a collection of states ( whether attractors or not ) is the set of states in all of their in - components ( i.e. the intersection of supports ) .
the overlap of @xmath99 , written @xmath100 , is the set of states in @xmath101 basins of attraction can only contain one equilibrium state or orbit and hence the basins for two different attractors can not overlap .
supports of different collections of attractors may or may not have sets of overlapping states .
the overlap states are of interest because these are the states with positive probabilities for ending up in each of the attractors for which the supports are overlapping .
if no attractors supports overlap then every attractor s support is just its basin of attraction and the system has a _
deterministic outcome_. models referred to as dynamical systems models typically use sets of differential equations and are typically fully deterministic because the map from @xmath17 to @xmath102 is always a function in the strict sense . because the dynamics are produced from these functions , every initial condition is mapped to a particular equilibrium state .
certain parameterized systems allow changes in their state maps that can change the number , location , and/or `` strength '' of equilibria .
but for any given value of those parameters , each initial value still has one possible outcome .
many systems can be usefully modeled with such fully deterministic systems , but in this paper we are mostly concerned with the non - deterministic parts of systems because that is where the critical points occur ( see next section ) .
[ outdegree ] a state s _ out - degree _ is the number of distinct successor states ( states that may be immediate transitioned into ) .
the out - degree @xmath103 of state @xmath28 equals @xmath104 @xmath51 will be used to denote a neighboring state and @xmath105 the set of neighboring states .
[ indegree ] the number of states that can transition into a state is its _ in - degree_. @xmath106 the in - degree measure will be rarely used in what follows ( except for algorithms utilizing a reversed markov model subgraph ) and will not need its own symbol . because of this asymmetry the term ` degree ' will refer to a state s out - degree unless otherwise noted .
[ reach ] the _ reach _ of a state ( also called its _ out - component _ ) is the set of states that the system may enter by following some sequence of transitions ; i.e. _ all possible _ future states given an initial state .
the reach of @xmath28 or @xmath107 is the set of @xmath29s such that @xmath108 [ decreasingreach ] every successor state s reach is less than or equal to the intial state s reach .
@xmath109 [ reachpath ] @xmath29 is in the reach of @xmath28 if and only if there exists at least one path from @xmath28 to @xmath29 .
@xmath110 this theorem , which can be used as an alternate definition of reach , follows from the definitions of path and reach .
an obvious corollary of the definitions of basin and reach is that if the reach of a state @xmath28 includes only one attractor then @xmath28 must be in that attractor s basin of attraction @xmath111 an attractor state s reach is just the states within the attractor ; an equilibrium s reach is itself . for all transitions @xmath59 it must be the case that
@xmath112 from the definition of reach since @xmath113 satisfies @xmath114 .
furthermore every @xmath115 in the reach of @xmath29 is also in the reach of @xmath28 because for whatever @xmath6 makes it true that @xmath116 , @xmath117 makes it true for @xmath28 . to achieve the inequality it suffices for there to be at least one state in @xmath107 not in @xmath118 .
it is _ prima facie _ obvious that it is possible for @xmath119 such that @xmath120 and @xmath121 .
theorem [ decreasingreach ] generalizes to all paths ( which is just a sequence of transitions ) so that reach never increases as the systems transitions along any path .
this property relies on the fact that the transition structure is fixed for the markov models used in this paper ; future work will relax this requirement and the theorem does not necessarily hold for models with changing dynamics .
a _ strongly connected component _ of a directed network is a set of vertices such that there is a path from every vertex in the set to every vertex in the set ( including itself ) .
we will find the same concept useful , but this paper adopts a different name for it .
[ core ] a _ core _ of a set is a subset wherein every member of the subset is in the reach of every member of the subset .
the core of some set @xmath45 is written @xmath122 and is a subset satisfying the condition @xmath123 some sets will have multiple cores the set of @xmath45 s cores can be called @xmath45 s _
mantle_. [ reachcycle]every cycle within @xmath45 is ( at least part of ) a core and every state in a core is in at least one cycle .
[ reachcore]every state in a core has the same reach .
the states on the boundary of a set are a useful set of states to identify .
[ perimeter ] the _ perimeter _ of a set , @xmath124 , is a collection of those states in the set that may transition to states outside the set .
that is , @xmath45 such that @xmath125 in keeping with the core and mantle analogy , the perimeter states of the mantle of @xmath45 will be referred to as @xmath45 s _ crust_. perimeter states themselves , without further specification , describe one commonly deployed ( though weak ) concept of tipping points , although we will see in the next section that specifying different base sets produces different types of tips . in the next section
we apply the above definitions in different combinations and different contexts to identify various system behaviors .
most extant systems analysis focuses on equilibria , but a lot of interesting behavior happens away from equilibrium . in the indeterminate states of a system
we can not know precisely which states the system will reach or which state it will be in at a given time in the future , we only know probability distributions over the future states . but by understanding a system s behavior we might know whether some particular change facilitates a specific outcome or path through the dynamic - this may be helpful information .
discerning these sorts of facts about system dynamics can increases one s information about the system ( in both the technical and colloquial senses ) .
considerations such as this one are the building blocks of the formal theory of tipping points , robustness , and related phenomena immediately to follow .
using markov models and the states and sets defined above as a springboard , this section defines and briefly describes several terms related to the concept ( or more to the point , concepts ) of tipping points .
critical phenomena and tipping points of various kinds share the defining feature that ( for whatever reason ) behavior is different before and after some transition .
behavior in this analysis is just the properties of systems dynamics .
there are , of course , many ways in which the properties of system dynamics can differ and each way is a different kind of tipping phenomenon . recalling that state changes occur if and only if there is a change in some aspect of the initial state ( see corollary [ difference ] ) , our analysis of tipping phenomena starts with state aspects .
[ levers ] the _ levers _ of a state are the aspects of a state such that a change in those aspects is sufficient to change the system s state .
the levers of @xmath28 , denoted @xmath126 , is @xmath127 so given a state in the markov model , the levers are those aspects of the state that are different in any neighboring state and each such @xmath128 is a distinct lever . in some cases
neighboring states will differ by more than one aspect . in those cases
the respective element of the set @xmath126 will be a list of all the aspects that need to change as one lever .
[ leverpoint ] a _ lever point _ is a transition resulting from a change in a particular aspect ( or set of aspects ) .
an aspect s lever points is the collection of transitions that a change in that aspect ( or those aspects ) alone generates .
the lever points of @xmath128 is the set of transitions created by @xmath129 levers and lever points work complementarily : for levers we pick a state and find the aspects that change and for lever points we pick the aspect and find the state changes it produces .
it is occasionally helpful to refer to the aspect change(s ) that generate a specific transition .
[ leverset ] @xmath130 symbolizes the _ lever set _ of @xmath59 : the aspect or aspects that differ between @xmath28 and @xmath29 . in some applications we will be interested in how many aspects change for a transition .
[ magnitude ] the _ magnitude _ of the lever set of a specific transitions is @xmath131 .
though levers as they are defined here do not depend on the ability to control that aspect , the choice of ` lever ' for this concept is motivated by the realization that in some models control of some aspects is available .
one may be performing a tipping points analysis precisely because one is choosing levers to bring about one state versus another ( or agents within the model may be choosing ) .
[ policychange]imagine a model wherein each aspect is a variable representing some part of a policy ( e.g. amount of money spent on each line item ) .
each aspect change has an associated cost ( legal , bureaucratic , time , etc . ) .
the modeler may be trying to determine the lowest cost , feasible route from the current policy to some desired policy ; or perhaps to determine how far policy can be changed on a specific budget .
the cumulative magnitudes of lever sets along a path may adequately approximate such a cost measure . in general the sum of the magnitudes along
a path is a rough measure of how difficult it is for the system to behave that way .
techniques from circuit design applied to the markov model may be gainfully applied to such models . in some contexts
we may wish to know how much change an aspect is responsible for across the system s dynamics .
[ strength]the _ strength _ of a lever is the sum of the probabilities of all transitions that result from changing that lever .
the strength of @xmath128 is equal to @xmath132 this measure is not scale - free since the sum depends on the number of transitions in the markov model , but it is useful for comparing levers within a system .
the strength measure could be used , for example , to determine which aspects to control to maximize ( or minimize ) one s ability to manipulate the system .
it could also be associated with a cost of letting that aspect vary over time .
we will revisit levers below in other forms as they apply to other measures of system dynamics . in some cases
we are interested not just in which aspects change through a transition but also in the * * values * * of levers at transitions .
[ threshold ] a _ threshold _ or _ threshold point _ is a particular value for a lever such that if the value of the aspect crosses the threshold value it generates a transition .
so @xmath17 is a threshold value of @xmath59 if @xmath133 this definition can be applied _
mutatis mutandis _ for a set of values @xmath134 for a lever set which can be distinguished by the name _ threshold line _ when appropriate . if there are multiple states with transitions crossing the same threshold value then knowing that information refines our understanding of the lever s role in system dynamics . thus determining the threshold value for one transition is merely a means to the end of determining the strength of the levers with that threshold .
[ thresholdstrength ] the _ threshold strength _ of @xmath17 is the strength of the levers for which @xmath17 is the threshold value : @xmath135 if a particular value for a particular aspect plays a large role in system dynamics then crossing that threshold is another oft - used version of `` tipping point '' behavior .
these definitions of threshold and threshold strength only require that the end state s value be different from the start state s value . in common usage , however , thresholds establish different and separate boundary values for ascending and descending values .
if a threshold only affects system dynamics in one direction then we can determine that from the markov model using the following definitions .
an _ upper bound threshold _ of @xmath59 is a value @xmath17 such that @xmath136 a _ lower bound threshold _ of @xmath59 is a value @xmath17 such that @xmath137 the threshold strength measure can be adapted to these ascending and descending definitions in the obvious ways .
sets satisfying these definitions can tell us how frequently crossing that threshold in that direction acts as a lever .
if there are multiple states with transitions crossing the same threshold value then knowing that information refines our understanding of the lever s role in system dynamics .
this general definition admits examples from many different kinds of systems and can even apply to parts of systems ( such as agents ) . in granovetter s model of riot
spreading @xcite we can talk of each agent having its own threshold - the number of rioting agents necessary to make each agent join the riot .
this is just the same threshold definition applied to a lever set where the levers happen to be the same feature of each agent . in granovetter
s model the threshold value is the same in both directions . we can also talk of thresholds in the properties of the system dynamics that track how the system transitions through states . instead of being a value for an aspect within the model
, it would be a value for one of the measures defined in this paper
. the following might be the case for some system : once the energy level of the current state drops below three the system is at most three transitions away from being in an attractor . though having an energy level is not part of the system that the markov model represents we can associate this property of system dynamics with each state of the system as if it were one of its possible levers
then we can explore the relations of specific values of this property of system dynamics to its other dynamics .
relating values of properties of system dynamics back to aspects within the model will also provide useful information in many cases .
in some system analyses the property of interest is what is available for the future @xmath0 in the most general terms .
if one does not know much a system s dynamics then even knowing how many states could potentially be transitioned to provides an informational benefit .
the measures below become increasingly refined and detailed , but we start with some simple measures that may suffice for some applications .
[ stretch ] a state s is the number of states in its reach .
so the stretch of @xmath28 equals @xmath138 [ criticalbehavior ] a system dynamic ( i.e. a particular state transition ) is considered _ critical behavior _ if and only if it produces a decrease in stretch ; that is , critical behavior is any @xmath59 such that @xmath139 in addition to identifying the transitions that limit the system s future states , we can also measure how critical the transition is .
subtracting the end state s stretch from the start state s stretch provides such a measure , but it is not scale - free and so can not be readily compared across different systems .
we can normalize the stretch difference with the size of the system to which it is being applied to produce a percentage measure .
[ stretchgap ] the _ stretch - gap _ of a transition is the change in the percent of the total number of states that can be reached .
this quantity equals @xmath140 because this measure includes the total number of states in the system it clearly is not scale - free either .
however despite this limitation it does provide information about the system s future and is an intuitive way to compare transitions within the same system - even at different resolutions .
as example [ dropexample ] below demonstrates the stretch - gap reports how much of the system s state space is cut off by each transition and this information could be used , for example , to manipulate system dynamics to prolong system longevity .
we also have an alternative , fully scale - free , measure of the drop in reach across a transition . [ criticality ] a * transition s * _ criticality _ is one minus the ratio of the start and end states stretch .
the criticality of @xmath59 equals @xmath141 recall from theorem [ decreasingreach ] that from any initial starting point , as the system transitions through its states the sizes of the states reach are monotonically decreasing . as a result of that theorem
we have the following corollary regarding the range of values for the ratio of reaches . a transition s criticality will be between zero and one .
let @xmath142 and @xmath143 for transition @xmath59 . from theorem [ decreasingreach ] @xmath144 .
@xmath145 produces @xmath146 which yields a criticality of zero . for @xmath147
we can decrease @xmath148 or increase @xmath149 to find the other bound , but since @xmath148 and @xmath149 are natural numbers increasing @xmath149 is the better approach . using a well - known mathematical fact suffices for finding the other bound : @xmath150 .
transitions within a cycle ( which includes self - transitions ) always have zero criticality and zero stretch - gap .
this observation follows from observations [ reachcycle ] and [ reachcore ] and the definition of a core .
the concept of criticality agrees with this measure insofar as any transition that has no affect on what states may be visited in the future should not be a critical transition .
[ dropexample ] the system represented as figure [ criticalityexample ] has thirty - three states in total .
each one is color - coded by its stretch .
compare the patterns in color to the attractors , basins , and support in figure [ equilbasin ] and the overlap in figure [ overlapdiagram ] .
stretch alone , though a rather simplistic measure , works decently to partition the system dynamics into regions of similar behavior .
stretch - gap performs well as a discriminator that groups states into these regions in a way similar to spectral analysis for detecting community structure in networks @xcite . to wit
, stretch drops between zero and four states within any basin or overlap , but drops four to eighteen states crossing a boundary ( in this example ) .
this is not completely reliable , of course , but for many systems this simple technique may provide all the information required ; and it may be the best one can do with available data .
@xmath151 has a stretch of 21 , @xmath152 has a stretch of 6 , and @xmath153 has a stretch of 2 .
there are 33 states in this system so the stretch - gaps of @xmath154 and @xmath155 are 45.45% and 12.12% respectively .
that means that 45.45% fewer of the system s states can be reached after the @xmath155 transition .
we can also use this to determine the stretch - gap of @xmath156 as 57.58% regardless of the particular path taken .
only the start and end states stretches are necessary to calculate this , but the result is always equal to the sum of the stretch gaps of each transition taken . let s compare these figures to the criticality of the same transitions .
the criticality of @xmath154 is 0.714 and the criticality of @xmath155 is 0.667 .
that means that the system only has 71.5% of the possible future states in state @xmath152 as it did in @xmath151 .
a composite measure is also possible for the criticality of @xmath156 .
it can be calculated just using the start and end states stretches , using the standard percentage of a percentage of a percentage @xmath0 calculation .
so the criticality of @xmath157 these measures above are intended to be just rough measures useful in certain limited contexts and when information about the system is limited .
for starters , these measures only consider only the structure of the markov models , not the probabilities .
also , they apply to transitions rather than states .
[ statecriticality ] the _ criticality _ of a * state * is the probabilistically weighted sum of the criticality of all the transitions from that state .
so to find the criticality of @xmath28 we calculate @xmath158 where by convention that is the sum over @xmath28 s neighbors . for @xmath29 that are not neighbors
. this convention will be used throughout - including cases where limiting an operation to neighbors matters . ]
because by definition [ probability ] the sum of the probabilities sum to one , state criticality will also be a scale - free measure with values between zero and one .
all these criticality measures quantify the constriction of future possibilities on a state - by - state basis which is useful if we want to keep our options open " . as we will see later
that is sometimes exactly what we want to measure , but sometimes we will want to measure system dynamics with reference to some particular features and that is what the following definitions for tipping points allow us to measure . as a refinement of levers from definition [ levers ] we can apply the lever concept to critical states to identify another feature of system dynamics .
[ criticallever ] a state has a _
critical lever _ if a change in that aspect ( or those aspects ) of the state will reduce the reach .
this merely combines the concept of a lever with the concept of critical behavior ( definition [ criticalbehavior ] ) .
by looking more deeply at the aspects driving the state changes and calculating the magnitude and strength of different critical levers we can gain a better understanding of how microfeatures generate the macrobehavior of the model . as mentioned in the section introduction
, the common feature of the measures in this section is that some states or transitions mark a shift in the properties of a system s dynamics . for the criticality measures above the difference
was the number of reachable states .
the following measures generalize to any sets distinguished by a chosen characteristic .
given states exhaustively compartmentalized by the property ( or properties ) of interest the following techniques can find where shifts occur and measure their magnitude . for some models we are interested in the achievement of a particular state ( e.g. an equilibrium ) or a particular system behavior ( e.g. a path linking two states ) .
we denote the particular state ( or set ) of interest as the _ reference state _ ( or _ reference set _ ) .
below we will see examples of specific reference states ( e.g. attractors and functional states ) but first the general case .
there are many ways in which behavior may change with respect to a reference state or set ( e.g. probability of reaching it , probability of returning to it , or probability of visiting an intermediate state ) : each property may partition the states into different equivalence classes ( groups with the same value of the property ) .
it is the movement between equivalence classes that counts as tipping behavior .
a _ tipping point _ is a state which is in the perimeter of an equivalence class for some property .
recall from definition [ perimeter ] that perimeter states are those from which the system s dynamics can leave the specified set .
because the sets here are determined by the properties of system behavior leaving a set implies a change in that behavioral property - and that is a tip .
this definition does not preclude that the system could tip back into a previously visited set : that possibility depends on what property is establishing the equivalence classes .
[ climatechange1 ] a climate change model that relates the co@xmath159 content of the atmosphere to global temperature may have states that are grouped together according to a shared property of those states ( e.g. sea level , precipitation , glacial coverage ) . due to feedback mechanisms in the system it is likely the case that these qualitative features change in punctuated equilibria ( see also example [ punctuatedequilibria ] ) thus producing equivalence classes for some states of the system . _
ex hypothesi _ people can manipulate the level of co@xmath159 to higher or lower values .
the values at which the property shifts happen may differ for the increasing and decreasing directions , but the point is that co@xmath159 levels could raise temperatures to the point where glaciers disappear and then later lower past the point where glaciers will form again .
for some systems behavior can tip out of a equivalence class and then later tip back in .
phase transitions in condensed matter physics are another example of reversible tipping behavior .
so while some have posited that tipping points are points of no return for system behavior , that turns out to be true only for certain systems and is not properly part of the definition .
dynamics of staying , leaving , returning , and avoiding a specified set of states will be covered in the section below on robustness . here
we continue with ways to quantify changes in what is possible for system dynamics for different states and transitions .
these measures apply for any reference state or reference set , but for convenience and intuition pumping the following presentation will adopt the notation of attractor ( @xmath77 ) for a reference state and set and @xmath78 for a collection of reference states . recall from definition [ attractor ] that attractors may be equilibria or orbits and both possibilities were symbolized with @xmath77 and treated as singular .
that convention will be continued here .
@xmath78 is a collection of independent reference states and sets each of which satisfies a property while @xmath77 may be a set of states that collectively satisfies a particular dynamical property ( such as an orbit as a whole satisfies the dynamical property of an equilibrium ) .
[ energylevel ] the _ energy level _ of a state is the number of reference states within its reach . we write this as @xmath160 and it equals @xmath161 energy level quantities partition the system s states into equivalence classes .
[ energyplateau ] the equivalence class mapping created by states energy levels is called the system s _ energy plateaus_. each energy plateau is a set @xmath162 .
[ energyprecipice ] the change in energy across a transition is called an _ energy precipice _ or _
energy drop_. we can measure the magnitude of an energy precipice in the obvious way : @xmath163 [ energydrop ] an energy precipice is never negative : @xmath164 . from definition [ energylevel ]
the energy level of any state @xmath28 is the number of attractors in @xmath107 . by theorem [ decreasingreach ] for any @xmath165 .
this decreasing reach property implies @xmath166 since the end state of a transition always has a lower or equal energy level , @xmath167 is always greater than or equal to zero .
we now can measure the degree to which a state is likely to be the site of a tip ( in a similar fashion to definition [ statecriticality ] of state criticality above ) .
[ tippiness ] the _ tippiness _ of a state is the probabilistically weighted proportional drops in energy of its immediate successors : @xmath168 [ tippinessrange ] tippiness ranges from zero to one .
the lower bound occurs when all neighbors can reach the same number of reference states : @xmath169 .
denotes the degree of a state and @xmath51 is a neighboring state . ] in this case tippiness equals @xmath170 by theorem [ probability ] . by theorem [ energydrop ] @xmath171 .
when @xmath172 the upper bound occurs when a state has every attractor and only attractors as neighbors .
again by theorem [ probability ] @xmath173
. attractors by definition [ attractor ] have an energy level of one and the energy level of @xmath28 in this case is @xmath174 so @xmath28 s tippiness is @xmath175 .
thus the upper bound of @xmath28 s tippiness goes to 1 as @xmath176 .
note that tippiness uses the ratio of energies rather than the difference ; this makes tippiness a dimensionless metric and thus comparable across any state or system . sometimes one will be more interested in minimizing the magnitude of energy drops , or avoiding states with the highest expected magnitude of energy drops , which have obvious formulations given the above definitions . as another refinement of levers from definition [ levers ]
we can apply the lever concept to tipping points to relate tips to the individual aspect changes that drive them .
[ tippinglever ] a state has a _ tipping lever _ with respect to some specified set of states if a change in that aspect ( or those aspects ) of the state will take the system out of that set of states .
this merely combines the concept of a lever with the general concept of tipping behavior .
so while every state accept equilibria has levers , only perimeter states have critical levers . identifying the tipping levers of certain sets of states is precisely what we d like a `` tipping point '' analysis to reveal because it is just the aspects of a tipping point that actually change when the system tips out of a set of states .
these ideas are further refined in the analysis of robustness and related concepts below .
this section will use the markov model framework provided above to establish formal definitions of several related concepts : robust , sustainable , resilient , recoverable , stable , and static ; as well as their counterparts : susceptible , vulnerable , fragile , and collapsible . as before , it is unlikely that any mathematically precise definition will maintain all the nuances of the full concept sharing the same name .
furthermore , existing definitions and formal treatments of the same concepts or that use the same terms risk distracting from or confusing the thrusts of this research .
what is more important than the terms used is that the definition is useful and we can easily refer to it , though some care has been spent on finding the closest word to the provided definition . adding to the ambiguous and often synonymous usage of these terms are the conceptual questions that arise in considering their dispositional nature .
dispositional properties are philosophically troubling for many reasons but the philosophical troubles will not interfere with their definition and ascription here ( mostly these stem from the question of whether their subjunctive conditional status distinguishes them from categorical properties ) .
i mention this here because i want to hint that in addition to the obvious and direct application of this methodology to improve the performance capabilities of systems , it may also produce insights into the nature of dispositional properties in general .
this analysis reveals how these particular dispositional properties are behavioral in nature and emerge from the microbehavior of the system components .
the philosophical issues will be addressed in separate work ( see future work subsection below ) , but it may be interesting to the reader to consider how to apply the following methodology to investigate other dispositional properties such as soluble , malleable , affordable , and differentiate them from other types of properties with the much sought after necessary and sufficient conditions .
the measures in this first set are conceptually simple with intuitive mathematical definitions and straightforward algorithms .
they nevertheless identify important features of system dynamics and act as building blocks for more sophisticated measures .
[ statestability ] a * state s * _ stability _ is how likely that state is to self - transition . @xmath28
s stability is @xmath177 while this may seems a trivial property , it is consistent with a useful distinction from system dynamics : the difference between stable and unstable equilibria . due to the resolution of the markov model s states
, an attractor state will include a neighborhood of aspect values around the equilibrium point values .
thus exit behavior from the attractor node includes the response to small perturbations to ( or variations around ) the equilibrium point values .
stable states will tend to stay within this neighborhood and this is reflected in a high self transition probability value . since values that are nearby an unstable equilibrium but not exactly on the equilibrium point values will tend to move away from the equilibrium values , we would see this reflected in low self - transition probabilities .
these results exactly match attributions of stability and instability in the markov model via definition [ statestability ] we can extend stability to apply to sets of states in the obvious way .
[ setstability ] the _ stability _ of a * set * is the probability that the system will not transition out of the set given that the system starts within the set .
we calculate this as the average of the individual states exit probabilities , so set stability is @xmath178 this is a crude measure because is does nt properly reflect the probability of staying within the set over time , it only looks one time period ahead . a more sophisticated notion of staying within a set of states
is presented by definition [ sustainability ] below of sustainability , but in some cases ( discussed in that subsection ) the two measures generate the same value . the word ` static ' is often used to indicate a lack of dynamics in a system , and that is the sense attached to the following formal definition . recalling that self - transitions can be interpreted as a lack of transition it aggregates the lack of transitions among states .
[ static ] the degree to which a set is _
static _ is the average of the states stability values : @xmath179 this definition , though simple , captures how likely a system is to be in the same state for consecutive time steps in a way that is comparable across sets and systems with different numbers of states .
what this definition fails to capture is that sets with equilibria will spend an infinite amount of time in them whereas sets lacking equilibria will continue to transition for eternity ( even if that s within an attractor ) ; and yet because this measure uses average stability it is easy to construct cases where an equilibrating systems has a lower static level on the given definition .
static and stable set measurements are similar in their calculation but distinct in their sense .
set stability is a measurement of lack of change , but it is a lack of change out of a set ( though it ignores dynamics that stay with in the set ) .
it is therefore only applicable if the set chosen is smaller than the whole system .
staticness can apply to the whole system and is useful for comparing systems overall level of dynamism . if @xmath180 ( i.e. a set with one state )
then the static measurement equals the stability measurement .
this theorem clearly follows from the fact that the sum of transitions staying within a set equals the sum of self - transitions for a set of one state .
also in this case set stability naturally equals state stability because the set is a state . for any set , the set stability measure is always greater than or equal to the level of staticness . for any given @xmath84 , the set stability measure is @xmath181 the probability of transferring to another state in @xmath45
, @xmath182 is greater than or equal to zero .
if that equals zero then @xmath28 s contribution to set stability becomes @xmath183 which is equal to @xmath28 s contribution to the static measure . if @xmath184 for any @xmath185 then @xmath45 s set stability is greater than its degree of being static .
on the other end of the spectrum as stability and staticness are measures of how likely the system is to change states . because the above measures are defined in terms of probabilities , most simple measures of the presence of dynamics can be calculated as one minus the appropriate measure above .
there is one additional simple measure to present here ; it is a rough measure of how predictable state changes are .
the _ turbulence _ of a set is the average percentage of states that its states can transition into .
we can calculate @xmath45 s turbulence with the average ratio of each state s degree to the number of states in @xmath45 : @xmath186 this measure ranges from zero to one where the zero case occurs as @xmath187 and a turbulence of one means that the set is fully connected ( including self - transitions for each state ) .
the idea is that when each state has only a few possible transitions then there are far fewer possible paths through the system dynamics .
if each state can transition into many others then , like with the common usage of ` turbulent ' , there is a great deal of uncertainty regarding the path that a series of transitions will take . plotting the degree distribution would reveal a set s ( which might be the whole system or just a portion ) turbulence profile .
if one were to find something like a power - law distribution ( where a few states have many transitions and most have just a few transitions ) the high - degree states would seem to satisfy yet another concept of tipping behavior . combining turbulence profiling with ( for example ) the identification of perimeter states could be used to classify systems by the dynamical properties ( see future work section for more details ) .
though the turbulence measure may provide sufficient information in many systems , it fails to differentiate the effects of high and low probability transitions .
transition weights clearly play a role in determine how confident one can be that a particular trajectory will be taken rather than another .
for example , if all but one of a state s transitions have very small probabilities associated with them then the set should be considerable less turbulent than if all the transitions are equally probable .
[ weightedturbulence ] as a refinement of turbulence , _ weighted turbulence _ of the state @xmath28 equals zero if @xmath188 and for @xmath189 can be calculated as @xmath190 because by definition [ probability ] the sum of exit probabilities sum to one , the average of the exit probabilities is @xmath191 regardless of the number and their individual weights .
the innermost component of this calculation , therefore , finds the difference between each transition and the average weight and squares it .
squaring has the dual effect of producing an absolute value and intensifying differences ; the intensification is not crucial and is merely adopted by convention . because turbulence is maximal when each weight is equal , we subtract the differences from one to calculate each state s turbulence .
case is that the definition produces a value of one instead of zero .
this is an artifact of the fact that if there is only one edge then all the edges have a weight equal to the average weight - and that is the case that produces maximal turbulence for all other @xmath192s . ]
to determine the weighted turbulence of a set we simply average each included state s weighted turbulence in common parlance something is sustainable if it can perpetually maintain its operation , function , or existence .
it is often used in connection to environmental considerations such as whether humans are using up resources faster than they can be replenished or to the ecological question of whether population dynamics will drive any species to extinction .
political institutions , academic reading groups , pools of workers , and any other system that undergoes inflows and outflows of its parts and might collapse or fail is a potential subject for sustainability considerations . roughly speaking , for this analysis a set of states is sustainable if the system can stay within that set of states .
there are multiple ways to calculate a measure of this sort and each reports a slightly different concept of sustainability .
as a crude approximation to the long - term sustainability we can find the cumulative sum of the @xmath193 power of the set stability measure from definition [ setstability ] up to some sufficiently large @xmath58 ( see observation [ bigenough ] below ) .
we could call such a measure _ naive sustainability _ and identify conditions for its appropriate application , but instead we will move on to a more sophisticated measure .
the previous measure is crude because within a set there may be ( for example ) heavily weighted cycles such that if the system starts in one of the cycle - states it is very likely to go around the cycle for a long period . to properly account for this , while still remaining agnostic over
which state of @xmath45 the system starts in , we calculate a refined sustainability measure .
[ sustainability ] the _ sustainability _ of @xmath45 is the average cumulative long - term probability density of future states that remain in the set starting from each state in the set .
power up to @xmath194 ( see observation [ bigenough ] ) .
this is done each time to clear out the probability mass outside the set so that it ca nt return .
the result is the sum of the resulting vectors for the states in @xmath45 each step .
usually algorithms are been banished to the appendix , but understanding this calculation is likely to make understanding definition [ susceptible ] of susceptibility much clearer . ]
[ bigenough ] if the chosen set does not contain an attractor then this calculation is unproblematic because some probability density `` escapes '' the set each iteration and there exists some time @xmath58 after which the remaining probability density in each state of @xmath45 is less than any arbitrarily chosen minimum resolution .
sustainability measurements are therefore only appropriate for sets that do not include attractors . if an application to sets including attractors were deemed useful then we could separate out the basin(s ) of the attractor(s ) from the other states and apply the sustainability measure above to the remaining states of the set .
it is still not clear how to recombine the two subsets into a single measure or how to cope with epistemic barriers to knowing whether a set contains an attractor before running the analysis and thus whether this would be necessary . because sustainability is a cumulative measure the calculation will not produce a probability .
but because the cumulative sum is divided by the size of @xmath45 it is normalized and comparable across differently sized sets . what we uncover through this process is a measure of the expectation over time that the system will stay in the set given that the system starts somewhere in it .
if @xmath45 is chosen to be an energy plateau then all exit transitions are one - way .
if there are no cycles in @xmath45 then the probability mass will quickly dissipate from within @xmath45 yielding a very low sustainability measure .
if there is at least one cycle then there is a chance that the system will stay within the set indefinitely , but since that cycle can not be an attractor the system will leave the set in expectation .
the stronger the weights of the transitions among the cycle states the greater the sustainability measure .
the energy plateau application is especially helpful for the `` keep our options open '' mindset and where each equilibria is a different form of system failure ( e.g. for dissipative structures ) .
it can even be useful to calculate the sustainability of a basin of attraction ( excluding the attractor itself ) - which is also an energy plateau .
the system may exhibit interesting and long - lived behavior within a basin of attraction that may reveal much more about the processes affecting a system than just which attractor it is likely to end up in .
time to equilibria may be on the galactic time - scale and knowing that will alter our interpretation of the system s characteristics . the term ` susceptible ' is typically followed by ` to ' and an indication of what the thing is susceptible to .
i preserve that usage with the measure presented here .
we will talk of sets being susceptible and sets can be defined according to different properties for different applications .
states are what sets are susceptible to .
this may sound odd , but the probability of transitioning out of a set depends on which state within the set the system is currently in .
so the characteristic represented by staying within a set risks being lost to a degree contingent upon the state . [ susceptible ] the degree to which @xmath45 is susceptible to @xmath28 is how much more ( or less ) likely it is to transition out of @xmath45 conditional on it being in a particular state @xmath28 of @xmath45 compared to the sustainability of @xmath45 overall .
@xmath196 given this definition we can see that a positive susceptibility means a lower probability to stay within @xmath45 .
we can also determine and measure the set s susceptibility to the lever points of an aspect .
recall from definition [ leverpoint ] that the lever points of an aspect are all those transitions that result from a change in that aspect .
we can calculate the susceptibility of @xmath45 to a collection of lever points as @xmath197 sustainable / susceptibility analysis can be used to help systems maintain a performance level .
we can take the set to be a contiguous collection of states that count as functional in some system : such as all the configurations of an airplane that the autopilot can manage .
for the airplane system some state changes will be exogenous perturbations due to environmental factors ( wind , rain , pressure , lightening , passenger movement , etc . ) , others will be endogenous control adjustments by either the pilot or the autopilot , and some will be a mix .
first one would calculate the sustainability of the whole set of autopilot capable states
. then one would calculate how susceptibility that set is to each state ( or smaller collection of states ) . using this information the autopilot and/or pilot could select actions that minimize susceptibility across the states visited and this means maximizing the probability of staying within the set of autopilot capable states .
this example can be generalized to any case where maintaining functionality is the modeler s goal .
stability , staticness , and sustainability are different ways to measure a system s dynamics tendency not to leave a state or set ; we now turn to measures of returning to a state or set once it has been left .
[ resilience ] a state s _ resilience _ is the cumulative probability of returning to a state given that the system starts in that state .
the resilience of @xmath28 equals @xmath198 it is the sum of the individual probabilities of returning in @xmath199 time steps . because the sum of exit probabilities of every state equals one and the probability of traversing a path is the product of the states along the path this cumulative sum is always less than or equal to one and is a true probability measure .
[ fragility ] a state s _ fragility _ is a measure of how likely it is that the system will never return to that state .
this is just one minus the resilience of that state .
so equilibria have zero fragility and states with no return paths have a fragility of one .
measuring the degree of fragility requires the same calculation as measuring the resilience , but finding out whether a state is ever revisited is much easier because we can utilize our definition of a state s reach .
[ brittle ] a state is _ brittle _ if and only if it has a fragility value of one ( i.e. a resilience value of zero ) .
brittle states are the ones such that @xmath200 except for the brittle states which have a specific formal significance , the choice of whether to use a resilience or fragility measure will depend on which feature the user would like to highlight ( glass half - full or half - empty )
. we can also define the resilience and fragility of a set in an analogous way .
[ setresilience ] _ set resilience _ is the probability that the system will return to a set if the initial state of a sequence is within the state .
@xmath201 though the definition is exactly parallel to the single - state case , the algorithm to calculate this probability is considerably more difficult .
a few facts about entering and leaving sets will help refine our understanding .
* transitions exit @xmath45 through the perimeter states @xmath202 of @xmath45 .
* transitions enter @xmath45 through a set of entry points @xmath203 of @xmath45 .
* we can refine the definition of set resilience to @xmath204 .
so to calculate set resilience we need first to find all the paths from each element in @xmath202 to each element in @xmath203 . in the worst case this can be done in @xmath205 time via a breadth - first search .
set fragility is one minus set resilience .
[ resilienceiszero ] if @xmath45 is an energy plateau then the resilience of @xmath45 is zero .
by definition [ energyplateau ] an energy plateau contains all the states in the system with the same number of attractors in their reach .
any transition out of such a set would be to a state with a different energy level and by theorem [ energydrop ] it must be a lower energy level . also by theorem [ energydrop ]
no transition can be to a higher energy level .
hence if a system transitions out of an energy plateau then it can never transition back into it .
if the system can not transition back into the set @xmath45 then by definition [ setresilience ] @xmath45 s resilience is zero .
the susceptibility measure determines how sustainability changes depending on the specific starting state . we may wish to have a similar measure for resilience that reports how likely the system dynamics are to return to a state given that it exits via a particular transition .
[ recoverable ] a transition out of the set is _ recoverable _ to the degree that the system will return to the set after the transition .
@xmath45 is recoverable from @xmath59 to the degree calculated by @xmath206 note that leaving via a particular transition is the same as exiting due to a particular lever change .
thus we can uncover the recoverability of a set of lever points ( from definition [ leverpoint ] ) for a particular aspect as the average of the recoverability of each transition in it . also note that there may be multiple paths from @xmath28 back into each @xmath203 of @xmath45 .
each path leading from @xmath45 back into @xmath45 can be called a _ recovery path_. continuing with the autopilot example ,
imagine that there are many known points of failure for maintaining autopilot control .
each of these is a transition out of the set @xmath45 via a known lever change .
but not all failures are equally as problematic . by calculating the recoverability of each of the failure transitions
they can be ranked by their seriousness .
such a ranking can guide both the pilot in adjusting to the failure and the autopilot to avoid it in the first place .
again , the autopilot example can be generalized to the maintenance of any system : political regimes , sports clubs , ecosystems , viable crop production , etc .
sustainability measures the likelihood of a system s dynamic s staying in a certain set given that it starts within that set and resilience measures how likely it is to return to the set if the dynamics leave the set , but these measures do not include the case where the system s state starts outside the set and then enters it . when the set of interest is an energy plateau resilience is always zero ( as shown by theorem [ resilienceiszero ] ) but the set may still receive probability mass from parts of the system with higher energy levels . and in cases where a non - equilibrium analysis is appropriate we might be comparing different subsets within an energy plateau ( e.g. the relative probability mass of two cores within the mantle of an energy plateau - see example [ punctuatedequilibria ] below ) . in this final subsection
the above measures will culminate in the most inclusive measures of system robustness . before defining the measure that allows for inflow , we first define a measure that combines the features of sustainability and reliability . [ reliable ]
the _ reliability _ of a set is the average cumulative long - term probability density over the states in the set given that the system starts within that set .
@xmath207 this measure combines the concepts of sustainability and resilience , but it is not just the sum of those two measures . reliability starts the flow in the set and calculates the probability of being in each state on each consecutive time step . it does restrict the probability mass summation to the specified set , but it tracks probability mass throughout the system . the reason that this is nt merely a sum of resilience and sustainability is because when combining those two it was not possible to track probability mass that leaves the set , cycles back into the set , and then circulates within the set ( and maybe even repeats this process ) . with reliability
we can reincorporate probability flow that leaves and then re - enters the set .
a characteristic captured by the chosen set is reliable if it can be maintained or , if lost , can be regained .
if @xmath45 is an energy plateau then @xmath45 s reliability equals @xmath45 s sustainability .
this theorem does not follow directly from theorem [ resilienceiszero ] because there is no direct link between resilience and reliability , but the reasoning is the same . because there can not be any paths leading out of an energy plateau back into it , all the probability mass that @xmath45 gets for the reliability measure is from the initial distribution . leaving mass
never returns so that produces an equivalent measure as not counting the returning mass : this is the sustainability measure .
hence the sustainability values in figure [ sustainabilityfigure ] are also those energy plateaus reliability values . finally we add to the reliability measure the possibility that the system did not start in the set , but transitions into it .
[ robust ] the _ robustness _ of a set is the average cumulative long - term probability density over the states in the set given that the system may start at any state .
robust characteristics not only have high retaining power and recoverability , they also draw the system in from states outside the characteristic set . sets with high robustness values are sets that the system s dynamics tends towards .
that description makes robust sets sound a lot like attractors ; and this is as we would expect .
attractors will typically have high robustness measures on account of their perfect sustainability and the fact that typically several states will lead into them . ) or a proper subset of an attractor . ]
the attractor - like behavior related to robust sets provides interesting and useful insights into many systems dynamics .
[ punctuatedequilibria ] sets that behave like ( and are defined as ) equilibria in other modeling techniques may be revealed to be highly robust sets under the current analysis .
the phenomena of _ punctuated equilibria _ describes a system that spends long periods of time in characteristic patterns with interspersed and short - lived periods of rapid change . in the markov model representation
we might see a mantle of an energy plateau with multiple highly robust cores .
these cores could have relatively short transition paths among them .
each is a different cohesive pattern with larger probabilities of staying in than going out .
but because these cores are not attractors the system will eventually transition out of them and into the next core .
[ dissipativestructure ] one of the foci of complex systems science is the study of the self - maintaining ( or _ autopoietic _ ) nature of dissipative structures .
dissipative structures are those where a continual flow of energy , matter , or other resource is necessary to maintain system structure and performance .
biological systems are like this , constantly changing and adapting to maintain functionality , and so are many other complex systems .
these are systems where there are no equilibria or all equilibria are states to be avoided so that the energy level of the system remains mostly constant .
some set(s ) of states are preferred to others for exogenous reasons ( functionality , performance , diversity , longevity , or other utility measures ) and the goal is to maximize time spent in the desired states .
the goal might also be to maintain some characteristic feature of transient system behavior .
the current techniques offer new measures of behavior for non - equilibrium analysis .
these can be used to embed an existing equilibria model into a larger context and/or to push down the level of analysis to see what is happening inside an `` equilibrium '' state . using a definition parallel to that of susceptibility
, the following measure calculates how much more ( or less ) likely the system is to be in the set conditional on the dynamics starting in a particular state ( not necessarily in that set ) .
[ vulnerable ] a set s _ vulnerability _ is the difference in the average long - term probability density over the states in the set compared to the density generated by starting in @xmath28 @xmath209
this section s status is somewhere between completed work and future work - which i suppose makes it work in progress .
it has not been prepared to the degree of rigor of the previous sections , but there is sufficient material here to provide a strong indicator of how a markov model representation can be used to uncover a variety of measures related to path sensitivity .
the algorithms for the presented measures ( and more ) have been prepared , but the mathematical and conceptual work needs more smoothing and filling in .
it is offered in its present form for review , evaluation , and feedback purposes .
path sensitivity is not a single , well understood and properly defined concept . for starters ,
different features of system dynamics can have the path sensitivity property : outcomes can be path sensitive , processes can be path sensitive , measures can be path sensitive , and paths can be path sensitive .
previous work has focused on explaining why a particular system exhibits path sensitivity or how a particular mechanism can generate it , but they have not provided general , causally agnostic measures applicable to any system s dynamics .
what follows is a collection of distinctions and formal definitions for several different forms of path sensitivity .
and like the above measures of tipping- and robustness - related concepts , these measures apply to markov models representing either observational or model - generated data .
one type of path sensitivity results when a transition excludes some states from any possible future of the system s dynamics .
such an occurrence is closely related to the measures of criticality and tipping behavior presented in definitions [ criticality ] and [ tippiness ] respectively .
[ weakpathpreclusion ] any reduction in the size of the reach across a transition is instance of _ weak path preclusion_. the degree of path preclusion of @xmath59 is the criticality measure of @xmath59 . such a definition is likely to be useful in limited number of models but it is conceptually intuitive that merely excluding a set of states from the future of a system s dynamics is a relevant form of path sensitivity . in applications geared towards preserving , tracking , or monitoring some characteristic of the system or its behavior
we need a concept that accounts for the preclusion of that characteristic .
[ strongpathpreclusion ] when there is a reduction in the number of reference states or sets ( e.g. attractors or specific equivalence classes ) that can be reached then this is tipping behavior that exhibits _ strong path preclusion_. the strong path preclusion of @xmath59 is measured by the tippiness of @xmath59 .
recall from example [ climatechange1 ] that not being able to return to a state or set was * not * properly part of the definition of tipping behavior .
whether a tip is path preclusive or not marks an important feature of the transition .
in many cases we will be more interested in whether a transition is path preclusive than whether it is tipping out of some characteristic .
it is this dynamic and its importance for understanding system dynamics that gives tippiness its relevance as a measure .
some systems have only one long - term outcome ( a singular attractor ) and some have none ( when the whole system is an orbit ) .
in either case we might care less about where the system goes than how it gets there , i.e. which intermediate states the system realizes between two anchor states .
we might care about the exact path our dynamics takes through the system s states because some states are preferred to others ( for exogenous reasons ) or because they have different dynamical properties ( e.g. susceptibility or vulnerability measures ) .
two different tips may include the same reference states but force the system dynamics to take different paths to get there .
[ trajectoryforcing]_trajectory forcing _ is when a particular transition sends the dynamics down a specified sequence of states .
the _ force _ of an exact path from @xmath28 to @xmath29 can be measured as the product of the probabilities of all the transitions required to stay that course : @xmath210 given this general definition it is clear that forcing is relative to the specific set of connected states . if , for example , one wanted to maximize the probability of reaching a specific equilibrium then calculating the force of each path from the tips of the current core to that equilibrium would provide the necessary guidance ( see figure [ trajectoryforcingfigure ] ) .
to @xmath211 the red path from @xmath212 to @xmath213 in the figure has a force of @xmath214 and the green path has a force of @xmath215 .
force is not the same thing as the probability of reaching @xmath213 given @xmath216 and @xmath217 respectively ; that is calculated by the tippiness of those two transitions with only @xmath213 as a reference state .
these force measurements are the probabilities of following those exact paths to get between those two states . for non - exact paths force
can be calculated as the sum of the probabilities of the paths that visit each of the specified markers .
the term `` markovian '' means memoryless in the related fields of mathematics where it refers to any processes wherein future probabilities of events do not depend on past occurrences ( e.g. poisson processes ) .
the markov model representation may therefore seem an unlikely tool for uncovering dependencies in paths of system dynamics .
the conditional probability assignments to the transitions imply that the current state is sufficient for knowing the probability distributions of future states . but
another way to interpret this markov model structure is that it encapsulates the probability distribution of future states * if * all one knows is the current state .
it can also be used to track correlations in the specific paths followed .
[ pathdependence ] a state s exit transitions are _ path dependent _ if and only if the distribution of their probabilities changes conditional on previous states . the degree to which @xmath28 s transitions are path dependent on a set of historical sets @xmath218 equals @xmath219 this definition is very general and is meant to capture all the different ways that probability distributions could change due to different types of historical sets ( see page @xcite for several types of path dependence ) .
this therefore admits to a refinement for each type of historical set dependence : exact path leading to @xmath28 , unordered collection of states preceding @xmath28 , existence of a path @xmath220 for different @xmath221 , stability of @xmath218 , length of @xmath218 , and many others .
but insofar as each of these conditions can alter the transition probability distribution , their degree of path dependence can be measure in the same way . ]
consider the situation depicted in figure [ pathdependencefigure ] ; both states @xmath222 and @xmath151 transition into state @xmath152 which may transition to either @xmath153 or @xmath223 ( edges @xmath17 and @xmath18 respectively ) .
the path dependence of @xmath152 s transitions is revealed through the markov model when there are significant correlations in ( say ) @xmath224 and @xmath225 .
so even though it may be the case that @xmath226 , i.e. @xmath227 , it may not be the case that @xmath228 or that @xmath229 .
assume that the system dynamics enter @xmath152 equally often from both @xmath222 and @xmath151 .
given the system at @xmath152 let s assume @xmath230 .
analyzing the individual time series of data may reveal that @xmath231 and @xmath232 .
so @xmath152 s path dependence on @xmath233 .
@xmath152 s path dependence on @xmath234 . considering the conditional probabilities
@xmath231 and @xmath232 the relative value of these figures matches intuition .
this paper is very much a work in progress and so each definition , measure , theorem , algorithm and example is a subject for future work .
in addition to bolstering the current contents i have several planned avenues for expansion and spin - off projects , some of which are outlined in this section .
good methodology exists as a facilitator to good science , so the first and perhaps most important extension of this project is to apply these measures to models within substantive research projects . over the past year of development
i have given a few presentations and have engaged in many conversations about these techniques and their potential to reveal interesting features of system dynamics .
the potential collaborative projects that resulted from these discussions ( outlined below ) are in addition to planned applications to personal research in the evolution of culture and morality , institutional design , biological contagion control , supply chain management , ecological robustness , resource sustainability , and various philosophical implications of seeing multiple systems dynamics as instances of the same underlying phenomena . within the university of michigan
several parties have expressed interest in this methodology .
qing tian ( working with dan brown ) plans to use vulnerability analysis to identify the social and physical levers of the well - being of people in the poyang lake area of china where flooding frequently disrupts economic and social activity .
abe gong in the department of political science and public policy wants to incorporate these tools for research into far - from - equilibrium dynamics in organizational change .
dominick wright ( working with scott atran ) want to find points of susceptibility to terrorist cell activity ; intelligently disrupting sustainable operation of terrorist networks might offer low - impact methods to benefit national security .
warren whatley of the department of economics and the center for african america studies would like to apply the path dependency measures to his data and econometric model of the 18@xmath235 century british slave trade in america to identify any lasting effects on african economic and political stability .
chris chapman of the multimedia development team at the university of michigan medical school wants to uncover user behavior patterns in educational technology in order to improve retention and pinpoint weak links to improve educational efficacy .
several other individuals have expressed interest in the technique , but the above selection suffices to demonstrate the potential benefits this methodology has for substantive research fields across multiple disciplines .
this methodology has also garnered interest from the private sector .
state farm insurance has expressed interest in building a model to better understand the effects of word - of - mouth spread in insurance company choices that exploits tipping behavior and path dependence .
lockheed martin is pushing to develop software to read in data from existing simulations and engineering tools to build the markov model and identify the features of system dynamics defined herein .
palantir technologies is considering incorporating these analysis capabilities into their social network and financial market analysis software platforms . because the methods are designed to be as general possible , algorithms implementing them could be constructed as code libraries for popular scientific programming languages ( java , c , c++ , python , matlab , mathematica , etc . ) and made open source for popular consumption and widespread use .
i have started to pursue government grant funding to develop this software package . through each of these applications and others that follow i expectt to uncover exceptions , caveats , refinements , and alternatives to the work as it stands now .
i also expect to discover many more features of system dynamics that can be uncovered via the markov model representation .
this paper presents a first attempt at capturing the above - defined properties ; through applied research i will be able to validate these measures usefulness and improve them where necessary .
as mentioned in the text , i have sketched a methodology to generate the markov model representation from data sets ( whether from a database or collected from a generative model ) .
because actual data sets and parameter spaces are typically quite large , a tool to automatically create the markov transition diagram is necessary to perform the above measurements .
part of designing this tool will be specifying precisely what the restrictions and requirements of a data set are .
another part includes providing the algorithms to use available data to generate a system s states and transitions properly weighted . because the methodology presented in this paper is essentially a statistical technique ( more on this below ) some time
will be spent to demonstrate that the assumptions made are the minimal assumptions and that the structure generated is the most justified result from the observable sample .
one major goal of complex systems research is to identify common underlying mathematical properties in a myriad of seemingly very different phenomena .
the markov modeling technique allows us to create a common representation of almost any system s dynamics .
differences in the definitions of system states , however , will still mask many of the similarities .
that difficulty notwithstanding we can make great gains by identifying network _ motifs _
( repeated patterns in the graph structure ) and establishing cross - disciplinary equivalence classes of system behavior .
achieving this goal will require solving issues with the choice of system resolution and `` playing with '' the resolution to find the matching patterns . though this may sound suspicious , changing the resolution is nothing more than altering the level of organization to which we are applying the properties .
as long as we are consistent in our application of these techniques then we may be able to discover similarities in many complex systems dynamics .
there are two potential non - trivial objections to the above - given probabilistic accounts of properties of system dynamics .
the first is that probabilistic definitions are inadequate because we aim to understand these features as properties that systems possess rather than dynamics they _ might _ have . to answer this question
we may first need a better grasp of dispositional versus categorical properties more generally ( see below ) .
but it may be the case the other definitions cashed out purely in terms of structural properties of markov models may be what some people would find more intuitive .
it could also be that the definitions these potential objectors are seeking can not be formulated within markov models at all .
as long as the above definitions reveal useful distinctions and patterns of system behavior the project was a success , but still better ( or at least different and also useful ) measures may be available if build from a different formal foundation .
i will , naturally , continue to pursue other and hugely different measures of system dynamics .
the other objection to the probabilistic definitions provided is that a person may insist that for many of these concepts the definition is incomplete without the causal explanation for how it comes about
. like all other statistics - like approaches ( see `` metastatistics '' below ) these measures may be realized by many different micro - level dynamics .
some of those dynamics may not seem proper candidates for robustness or tipping behavior even if the data they generate reveals it as such from this analysis .
but if this were to happen then i would consider the project a huge success
. this would be similar to discovering scale - free degree distributions in many different networks from disparate research fields .
finding that common property urged researchers to pursue more deeply the phenomena and they eventually uncovered several different mechanisms by which scale - free network may be created .
our understanding of each of those systems greatly increased because we had a common yardstick with which to measure them .
the probabilistic measures presented here are not intended to replace or make unnecessary the deeper scientific analysis - they are supposed to foster it . while working on the formal definitions of robustness - related properties i realized that these are all dispositional properties ; and dispositional properties constitute a long - standing philosophical problem .
that connection immediately made me wonder if my mathematical formalism might shed some new light on how to differentiate dispositional properties from categorical ones .
dispositional properties are philosophically troubling for many reasons , but primarily these stem from the question of whether their subjunctive conditional status distinguishes them from categorical properties
. it may therefore be interesting to consider how to apply my methodology to investigate other dispositional properties such as soluble , malleable , affordable , and differentiate them from other properties with accepted necessary and sufficient conditions . if in addition to the obvious and direct application of this methodology to improve the performance capabilities of systems my research also produces insights into the nature of dispositional properties in general that would be unanticipated but certainly welcome news .
let s look at some details of the problem .
a big part of the problem is that any property can be given a subjunctive conditional description , but a property would be dispositional if and only if such a definition is the only possible one .
color properties are known as primary properties and should be excellent candidates of clear - cut categorical properties . yet being red is dispositional in the sense that nothing seems red in the dark . whether or not redness ( or conductive , or triangular , or @xmath0 ) is dispositional in the same way that fragility is has not been solved .
my mathematical analysis reveals how particular dispositional properties ( robustness - related ones ) are behavioral in nature and emerge from the microbehavior of the system components .
what i hope is that this can be expanded to develop necessary and sufficient conditions for properties to be dispositional in different ways @xmath0 or at least a step in a helpful direction .
i have said elsewhere in this paper that i consider the methodology presented here to be similar in kind to statistics .
it starts with data ( perhaps generated from simulations of a model ) , fits a model ( a markov model ) to the data , and then purports to describe the real system with measures over that model ( my definitions ) .
statistics as we usually see it takes a different kind of model ( some form of distribution or estimator ) , but its purpose and general method of attack are very similar
. and this procedure is clearly different from other sorts of models in that neither standard statistics nor my methodology can explain the phenomena being analyzed . statistics ( and my methodology ) can produce evidence that some generative theory - driven model does explain the observed data , but the theory behind the generative model is what is doing the explaining .
standard statistics and my methodology are certainly not unique in their abilities to measure but not explain phenomena .
much of complex network analysis can be seen in this light as well .
the network representation facilitates the calculation of measures on the generating data but not because the links identified in the network representation are in the actual system s features .
classifier systems , bayes nets , hidden markov models , and neural nets are all further examples where the formal representation can permit measures and produce predictions without mirroring the structure and dynamics of the underlying behavior - generating system . seeing all these different techniques under the same metastatistical light may allow us to 1 ) bridge gaps among these techniques , 2 ) identify broader guidelines for the proper application and interpretation of these techniques , and 3 ) find new statistics - like techniques with desired features .
young , h. peyton .
`` the diffusion of innovations in social networks '' _ the economy as a complex evolving system _ , vol .
iii , lawrence e. blume and steven n. durlauf , eds .
( oxford university press , 2003 ) . | this paper draws distinctions among various concepts related to tipping points , robustness , path dependence , and other properties of system dynamics . for each concept a formal definition
is provided that utilizes markov model representations of systems .
we start with the basic features of markov models and definitions of the foundational concepts of system dynamics
. then various tipping point - related concepts are described , defined , and illustrated with a simplified graphical example in the form of a stylized state transition diagram .
the tipping point definitions are then used as a springboard to describe , formally define , and illustrate many distinct concepts collectively referred to as `` robustness '' .
the final definitional section explores concepts of path sensitivity and how they can be revealed in markov models .
the definitions provided are presented using probability theory ; in addition , each measure has an associated algorithm using matrix operations ( excluded from current draft ) .
finally an extensive future work section indicates many directions this research can branch into and which methodological , conceptual , and practical benefits can be realized through this suite of techniques . |
the particular morphology and dimensions of nanowires ( nws ) have opened in the last years many perspectives in applications ranging from electronics , optoelectronics to energy harvesting and storage@xcite .
the successful incorporation of dopants in the nanowires is key for the implementation of such devices .
doping of nanowires has been a challenging task , due to the particular growth mechanisms@xcite . in this work
we report on the electrical properties of c - doped gaas nanowires obtained by the ga - assisted method@xcite .
as c is not soluble in ga , its incorporation in the nanowire core through the droplet remains difficult .
our strategy is then to grow an intrinsic nw core and to grow a doped shell around it .
we study the doping and surface effects in nws resistivity , field effect mobility as well as the minority carrier diffusion length .
the nanowires consisted of an undoped gaas core of about 70 nm in diameter and a c - doped shell .
the gaas nanowires were grown on a ( 111 ) si substrate under conditions that maximize the yield of vertical nanowires , as described elsewhere@xcite : a rotation of 7 rpm , a flux of ga equivalent to a planar growth rate of @xmath6 @xmath7 , a substrate temperature of @xmath8 @xmath9 and a v / iii beam equivalent pressure ( bep ) ratio of 50 .
subsequently to the gaas nanowire synthesis , growth conditions were radically changed for the obtaining of a @xmath10 nm gaas shell .
the ga flux was stopped for about 10 min while the substrate temperature was lowered to @xmath11 @xmath9 and the v / iii bep ratio was increased to @xmath12@xcite .
a p - type doping was achieved by adding a flux of carbon during the growth of the shell . to test the effect of c doping
, we grew a series of samples with different c flux given by the current applied to the cell , as reported in table 1 .
equivalent samples have been fabricated with the addition of a capping layer consisting of 15 nm al@xmath13ga@xmath14as and 6 nm of intrinsic gaas .
+ to give an accurate estimation the resistivity of the nanowires , multiple contacts were fabricated@xcite .
multiple four point measurements have been performed on around 100 samples .
the realization of a large number of devices has been automatized thanks to our auto - contacting software @xcite .
the electrical contacts consisted of pd / ti / au ( @xmath15 nm ) metal layer directly in contact with the doped shell . for the capped nanowires ,
the al@xmath13ga@xmath14as shell was removed using a solution of citric acid and hydrogen peroxide ( 2:1 ) leading an etching rate of @xmath16 @xmath17 .
bottom gate transistors , with 200 nm thick sio@xmath18 layer used as a gate dielectric , were realized in order to extract the nanowire majority carrier mobility .
surface states play an extremely important role in the opto - electronic properties of gaas nanowires . in ambient conditions
there is a formation of a thin oxide layer on the nw surface resulting in a pinning of the fermi level . in gaas
the pinning occurs near the middle of the band gap , producing a depletion region .
the existence of this region reduces the electrically active part of the nws@xcite .
we can avoid this problem by passivating the gaas with a wider band gap semiconductor , such as algaas@xcite . in this study
we study how a passivation layer affects the electronic properties of gaas nws with different doping concentrations .
+ 7 types of samples have been analyzed and the shell doping concentration , the resistance and the mobility are reported in table 1 . .doping shell concentration ,
resistance per unit length and mobility of the nanowires analyzed .
`` p '' corresponds to the passivated nws and `` u '' to the unpassivated ones . [
cols="<,^,^,^ " , ] [ tab1 ] the conductivity @xmath19 is obtained from 4 point measurements and the mobility @xmath20 is derived from bottom gate field effect transistor characteristics : @xmath21 where @xmath22 and @xmath23 are respectively the gate and the source - drain voltages , @xmath24 is the distance between source and drain and @xmath25 is the device capacitance . for a structure formed by a nanowire gated through the substrate , it is found@xcite : @xmath26 where @xmath27 is the dielectric constant of insulating layer
, @xmath28 is the nw radius and @xmath29 the dielectric thickness . in the case of passivated nws
we consider the capping layer as a capacitance in series with the bottom gate dielectric . for
200 nm sio@xmath18 thickness with @xmath30 the typical capacitance value of our devices is around @xmath31 pf .
typical output curves of our devices are shown in fig .
+ we assume that all acceptors are ionized at room temperature , meaning that the holes concentration can be directly linked to the c concentration . as there is no depletion region within the gaas doped shell of the passivated nws ( 1c , 2c , 3c and 4c )
, we can consider that the electrically active part corresponds to the entire doped shell .
we can therefore introduce the conductivity of the doped shell @xmath32 calculated on the doped cross section area and directly extract the doping density @xmath33 when nws are not passivated , one needs to consider the fermi level pinning at the nw surface and the existence of a depletion layer . solving the poisson equation and using the boundary conditions of a vanishing electric field at the surface we can calculate the depletion region width . for the doping concentration analyzed , the surface potential @xmath34 at the gaas surface is found to be around @xmath35 ev@xcite .
the depletion region @xmath36 follows@xcite : @xmath37 where @xmath38 is the gaas dielectric constant and @xmath39 the doping concentration in the shell . @xmath39 and @xmath36 can be deduced from the electrical measurements using the following system : @xmath40 where @xmath41 is the resistance per unit length , @xmath20 is the carrier mobility and @xmath42 is the radius of the undoped nanowire core . + .
the inset shows the nw bottom - gate transistor configuration .
b ) mobility vs shell doping concentration for the 7 types of nws analyzed .
1u , 2u and 3u are unpassivated nws while 1p , 2p , 3p and 4p are passivated .
c ) nw resistance per length in function of the doping concentration .
the black and red stripes highlight the region of minimum resistance per length for unpassivated and passivated nws.,width=6 ] results on the mobility and resistance per length obtained for all different types of nanowries are shown in fig [ fig1]b and [ fig1]c . the mobility measured increases inversely with the doping concentration for both passivated and unpassivated nws .
this behavior is commonly observed in semiconductors and is related to the scattering of carriers by the doping impurities .
interestingly , we observe a reduction of mobility with the increase in carrier concentration , which results in the saturation of the resistivity for doping concentrations higher than @xmath43 @xmath4 .
the saturation values correspond to @xmath44 k@xmath45/@xmath20 m and @xmath46 k@xmath45/@xmath20 m respectively for the passivated and unpassivated nws . + a priori one would expect a higher mobility for passivated nanowires with respect to the unpassivated . in fig .
[ fig1]b we observe that this is only true for the doping concentrations higher than @xmath3 @xmath4 . in order to shed some light to this issue , we have simulated the band profile across the nanowire with @xmath47 software ( copyright ( c ) 2012 , nextnano gmbh , germany ) and calculated the carrier distribution inside the nws .
figure 2 shows the profile of the hole concentration for both passivated and unpassivated nws and for two doping concentrations : @xmath48 @xmath4 and @xmath49 @xmath4 ( fig .
[ fig2 ] ) . in the unpassivated case
the fermi level pinning was used . and @xmath49 @xmath4 ) .
the distribution holes profile is represented in the case of nws capp ( blue lines ) and nws uncap ( red lines ) . in the unpassivated nws
, the depletion region has the effect of keeping the holes far from the surface .
the numerical analysis has been performed with @xmath50 . ] as shown in fig .
[ fig2 ] holes are mainly distributed in the doped shell , but a small portion is also found in the core . in the case of unpassivated wires , the surface depletion leads to a decrease in the amount of carriers in the shell .
this is especially notable for the low doping concentration . from this
we conclude that two mechanisms may contribute to the observation that the highest mobility is for low doped unpassivated nws .
first , the depletion region in unpassivated and low doped nws ensures that transport of carriers occurs at a certain distance from the surface and reduces the surface - related scattering .
second , a higher proportion of the transport occurs through the undoped core , which exhibits highest mobility due to the absence of dopants .
+ finally , we used electron beam induced current ( ebic ) to study the mean free path of minority carriers .
the semiconductor - metal interface at the contacts produces a schottky barrier .
when exciting electron - hole pairs with light or an electron beam , the potential gradient produced by the schottky barrier attracts minority carriers ( electrons ) into the contacts , which is measured by an electrical current .
the current exponentially decays with the distance from the contact @xcite : @xmath51 where @xmath52 is the electron mean free path . in order to understand what region of the nanowire
is excited with the electron beam , we simulate the trajectory of electrons from the beam with casino @xcite , shown in ( fig . [ fig3]a ) .
we simulate the trajectories of 200 incident electrons with an energy of @xmath53 kev and a probe current of @xmath54 pa , corresponding to the experimental conditions used . as it can be seen in fig .
[ fig3]a , the excitation is limited to the doped shell of the nanowire .
kev and a probe current of @xmath55 pa .
b ) experimental current decay as a function of distance to the contact for 3 nws samples ( 2u , 2p and 4p ) .
the minority carrier mean free path extracted increase up to @xmath5 nm for passivated nws with a doping concentration of @xmath56 @xmath4 . ]
figure [ fig3]b shows the exponential decay of the current for an unpassivated and two passivated nws with hole concentrations respectively of @xmath57 and @xmath49 @xmath58 .
the lowest mean free path is @xmath59 nm for the unpassivated nanowire .
the passivated nanowires exhibit longer mean free paths for minority carriers , up to @xmath5 nm .
despite the majority carrier ( holes ) mobility is lower for passivated nws , a longer diffusion path is found for the minority carriers ( electrons ) .
this is explained again by the fermi level pinning at the surface and its respective electric field , which attracts the beam - induced electrons towards the surface .
consequently , the electron main free path in unpassivated nws is reduced by the larger defect and trap densities at the surface .
in conclusion , we have measured the carrier concentration , mobility and minority diffusion length for core - shell gaas nanowires were the shell is c - doped at different concentrations .
the effect of the surface states is evaluated by comparing the values for passivated and unpassivated nws . at low doping concentrations
the highest mobility is observed for unpassivated nanowries , while at higher doping concentrations the passivated nanowires offer the best characteristics . the highest mobility ( 30 @xmath60/(v@xmath1s ) )
was obtained for unpassivated nws with a doping concentration of @xmath61 @xmath60/(v@xmath1s ) .
finally , we have measured the minority carrier diffusion lengths up to @xmath5 nm for passivated nws with a doping concentration of @xmath56 @xmath60/(v@xmath1s ) .
the authors thank financial support from : the swiss national science foundation under grant no .
2000021- 121758/1 and 129775/1 ; nccr qsit ; the european research council under grant `` upcon '' .
j. wallentin , n. anttu , d. asoli , m. huffman , i. aberg , m. magnusson , g. siefer , p. fuss - kailuweit , f. dimroth , b. witzigmann , h. xu , l. samuelson , k. deppert , and m. borgstrominp nanowire array solar cells achieving 13.8% efficiency by exceeding the ray optics limit , science * 339*(6123 ) , 10571060 ( 2013 ) .
krogstrup , h.i .
jrgensen , m. heiss , o. demichel , j.v .
holm , m. aagesen , j. nygard , a.f .
i morral , and a.f .
morralsingle - nanowire solar cells beyond the shockley queisser limit , nature photonics * 7*(4 ) , 306310 ( 2013 ) . c. colombo , d. spirkoska , m. frimmer , g. abstreiter , and a. fontcubertai morralga - assisted catalyst - free growth mechanism of gaas nanowires by molecular beam epitaxy , phys .
* 77*(apr ) , 155326 ( 2008 ) . e. russo - averchi , m. heiss , l. michelet , p. krogstrup , j. nygard , c. magen , j. ramon morante , e. uccelli , j. arbiol , and a. fontcubertai morralsuppression of three dimensional twinning for a 100% yield of vertical gaas nanowires on silicon , nanoscale * 4*(jan ) , 14861490 ( 2012 ) .
j. dufouleur , c. colombo , t. garma , b. ketterer , e. uccelli , m. nicotra , and a. fontcubertai morralp - doping mechanisms in catalyst - free gallium arsenide nanowires , nano letters * 10*(5 ) , 17341740 ( 2010 ) .
p. blanc , m. heiss , c. colombo , a.d .
mallorqui , t.. safaei , p. krogstrup , j. nygard , and a. f. i morralelectrical contacts to single nws : a scalable method allowing multiple devices an a chip . application to a single nw radial p - i - n junction , int .
j. nanotechnology , ( accepted ) .
g. landgren , r. ludeke , y. jugnet , j.f .
morar , and f.j .
himpselthe oxidation of gaas(110 ) : a reevaluation , journal of vacuum science technology b : microelectronics and nanometer structures * 2*(3 ) , 351 358 ( 1984 ) .
d. drouin , a.r .
couture , d. joly , x. tastet , v. aimez , and r. gauvincasino v2.42 : a fast and easy - to - use modeling tool for scanning electron microscopy and microanalysis users , scanning * 29*(3 ) , 92101 ( 2007 ) . | the resistivity and the mobility of carbon doped gaas nanowires have been studied for different doping concentrations .
surface effects have been evaluated by comparing upassivated with passivated nanowires .
we directly see the influence of the surface : the pinning of the fermi level and the consequent existence of a depletion region lead to an increase of the mobility up to 30 @xmath0v@xmath1s@xmath2 for doping concentrations lower than @xmath3 @xmath4 .
electron beam induced current measurements show that the minority carrier diffusion path can be as high as @xmath5 nm for passivated nanowires . |
the 122-family iron - based superconductors , such as bafe@xmath8(as@xmath9p@xmath10)@xmath8 and @xmath92 are widely described in the literature @xcite by a three - band model that is composed of one hole pocket located at the center of the brillouin zone @xmath18 and two electron pockets centered at two specific momenta @xmath19 and @xmath20 .
recently , this model was used by fernandes _
_ to study the nature of quantum phase transitions in the superconducting dome of some iron pnictides @xcite .
after integrating out the fermionic degrees of freedom @xcite and including the kinetic terms , one can obtain an effective lagrangian density @xmath93 where the parameters @xmath27 , @xmath28 , @xmath29 , @xmath30 , @xmath31 , and @xmath32 , are defined in ref .
our rg analysis starts from this effective field theory . in order to examine the effect of ordering competition on the superfluid density
, we introduce a small external gauge potential @xmath51 that couples to the superconducting order parameter @xcite and thus have an additional term @xmath94 where the lorentz gauge @xmath95 is utilized and the parameter @xmath53 is a new coupling constant . in the superconducting dome ,
the order parameter @xmath45 acquires a finite vacuum expectation value , i.e. , @xmath96 the quantum fluctuation of @xmath45 around its mean value is believed to play an important role @xcite and needs to be seriously taken into account .
it is convenient to introduce two new fields @xmath47 and @xmath48 by defining @xmath97 with @xmath98 . in many field - theoretic treatments of ordering competition , specially in the context of condensed matter systems , the quantum fluctuation of order parameter in the ordered phase
is usually omitted .
however , more careful analysis @xcite have showed that this approximation is not appropriate and that the order parameter fluctuation around its mean value can be significant . in order to entirely reveal the physical effects of ordering competition
, we will consider @xmath47 and @xmath48 as quantum field operators and study their interactions with the magnetic order parameters @xmath67 and @xmath68 .
substituting the re - parameterized field operator eq .
( [ eq_parametrization ] ) into the total lagrangian density , we get a new effective model @xmath99 where a number of effective parameters are defined on the basis of the fundamental parameters @xmath27 , @xmath28 , @xmath29 , @xmath30 , @xmath31 , @xmath32 , and @xmath53 , as given by eq .
( 5 ) in the main part of the paper .
our rg analysis will be based on this effective model , assuming that the coupling constants take small values .
we now proceed to treat the effective theory by performing a detailed rg analysis @xcite . to this end , we first make the following scaling transformations @xmath100 where @xmath101 and @xmath58 is a running length scale . under these transformations ,
the field operators @xmath67 , @xmath68 , @xmath47 , and @xmath48 should be re - scaled as @xmath102 for simplicity , we first consider the sdw qcp with @xmath42 , where the magnetic order parameters @xmath67 and @xmath68 both have vanishing mean values and their quantum fluctuations are critical .
analogous to the scheme presented in ref .
@xcite , we have derived the following rg equations for the seven fundamental parameters @xmath104 .
\\ \frac{du_m}{dl } = u_m+\frac{1}{2\pi^2}\left\{(8u_mg_m-17u^2_m-11g^2_m)\left(\frac{4a_s\lambda } { u_s}+1\right)-\frac{3\lambda^2}{8}(4a_s+1)\right.\\ \hspace{0.77cm}\left.+\frac{16a_s\lambda^2}{u_s}(2u_m - g_m ) \left(1+\frac{4a_s\lambda}{u_s}+2a_s\right)-\frac{6a_s\lambda^3}{u_s } \left[1 + 2\left(\frac{a_s\lambda}{u_s}+2a_s\right)\right]\right\ } , \\ \frac{du_s}{dl } = u_s+\frac{1}{\pi^2}\left\{2\lambda^2\left[4\left(a_m-\frac{a_s\lambda}{u_s}\right)-1\right ] -\frac{9u^2_s}{4}(4a_s+1)+54a_su^2_s(1 + 6a_s)\right.\\ \hspace{0.77cm}\left.-\frac{4\lambda^2_{\delta a}}{3 } \left(\frac{4a_s\lambda_{\delta a}}{u_s}+1\right ) + \frac{32a_s\lambda^3}{u_s } \left(1+\frac{6a_s\lambda}{u_s}\right ) + \frac{11072a_s\lambda^3_{\delta a}}{35u_s } \left(1+\frac{6a_s\lambda_{\delta a}}{u_s}\right)\right\ } , \\
\frac{dg_m}{dl } = g_m+\frac{1}{2\pi^2}\left\{3(u_m^2 + 3g^2_m-8u_mg_m)\left(\frac{4a_s\lambda}{u_s}+1\right ) + \frac{\lambda^2}{8}(4a_s+1)\right.\\ \hspace{0.77cm}\left.+\frac{16a_s\lambda^2}{u_s}(2g_m - u_m)\left(1+\frac{4a_s\lambda}{u_s } + 2a_s\right)-\frac{2a_s\lambda^3}{u_s}\left[1 + 2\left(\frac{a_s\lambda } { u_s}+2a_s\right)\right]\right\ } , \\
\frac{d\lambda}{dl } = \lambda + \frac{1}{\pi^2 } \left[\frac{4a_s\lambda^3}{u_s}\left(1+\frac{4a_s\lambda}{u_s}+2a_s\right ) -\frac{3u_s\lambda}{8}(4a_s+1)+9a_su_s\lambda(1 + 6a_s)-2\lambda^2 \left(\frac{2a_s\lambda}{u_s}+2a_s+1\right)\right.\\ \hspace{0.77cm}\left.-\lambda(2u_m - g_m)\left(\frac{4a_s\lambda}{u_s}+1\right ) + \frac{16a_s\lambda^2}{u_s}(2u_m - g_m)\left(1+\frac{6a_s\lambda}{u_s}\right ) + 6a_s\lambda^2\left(1+\frac{2a_s\lambda}{u_s}+4\alpha_s\right)\right ] , \\
\frac{d\lambda_{\delta a}}{dl } = \lambda_{\delta a}+\frac{1}{3\pi^2 } \left\{27a_su_s\lambda_{\delta a}(1 + 6a_s)-12\lambda^2_{\delta a } \left(1 + 2a_s+\frac{2a_s\lambda_{\delta a}}{u_s}\right)-\frac{9u_s\lambda_{\delta a}}{8}(4a_s+1)\right.\\ \hspace{0.77cm}\left.+\frac{64a_s\lambda^3_{\delta a}}{u_s } \left(1 + 2a_s+\frac{4a_s\lambda_{\delta a}}{u_s}\right)+36a_s\lambda^2_{\delta a } \left[1 + 2a_s\left(2+\frac{\lambda_{\delta a}}{u_s}\right)\right]\right\}. \end{array}\right.\end{aligned}\ ] ] when the system moves away from the sdw qcp and goes to a lower doping concentration , the stripe - sdw order parameters @xmath67 and @xmath68 also develop nonzero mean values , i.e. , @xmath105 to include the quantum fluctuation of @xmath67 and @xmath68 around their mean values , we introduce two new fields @xmath106 and @xmath107 : @xmath108 with @xmath109 . now the problem becomes more complicated than the case of sdw qcp . after lengthy but straightforward calculations ,
we obtain a set of rg equations : @xmath110+\frac{32a_s\lambda^2_{\delta a}}{3u_s } \bigl(1+\frac{4\lambda_{\delta a}a_s}{u_s}\bigr ) + \lambda\bigl[1 + 2\bigl(\frac{a_s\lambda } { u_s}+a_m\bigr)\bigr]\\ \hspace{0.77cm}+\frac{3u_s}{4}(1 + 2a_s)+\frac{8a_sa^2_m\lambda^2}{u_s } \bigl[1 + 8\bigl(\frac{a_s\lambda}{u_s}+a_m\bigr)\bigr ] + \lambda_{\delta a}\bigl(1+\frac{2\lambda_{\delta
a}a_s}{u_s}\bigr ) + 4a^2_m\lambda\bigl[1 + 6\bigl(\frac{a_s\lambda}{u_s}+a_m\bigr)\bigr]\bigr\},\\ \frac{da_m}{dl } = 2\bigl(\frac{a_s\lambda}{u_s } + a_m\bigr)+\frac{1}{4\pi^2}\bigl\{12a_m\bigl[a_m(g_m+5u_m)-6u_m\bigr ] \bigl[1 + 4\bigl(\frac{a_s\lambda}{u_s}+a_m\bigr)\bigr]-72a^3_mu_m \bigl[1 + 8\bigl(\frac{a_s\lambda}{u_s}+a_m\bigr)\bigr]\\ \hspace{0.77cm}-\frac{4a_s\lambda^2}{u_s }
\bigl[1 + 2\bigl(\frac{a_s\lambda}{u_s}+a_m\bigr)+2a_s\bigr ] -2(2u_m - g_m)\bigl[1 + 2\bigl(\frac{a_s\lambda}{u_s}+a_m\bigr)\bigr]-\frac{\lambda}{2}(1 + 2a_s)\\ \hspace{0.77cm}-4a^2_m(g_m-17u_m)\bigl[1 + 6\bigl(\frac{a_s\lambda}{u_s } + a_m\bigr)\bigr]\bigr\ } + \left(\frac{\lambda}{u_s}\frac{da_s}{dl } + \frac{a_s}{u_s}\frac{d\lambda}{dl } - \frac{a_s\lambda}{u_s^2}\frac{du_s}{dl}\right),\\ \frac{du_m}{dl } = u_m + \frac{1}{4\pi^2 } \bigl\{576a_m u_m(2u_m - g_m ) \left[1 + 6\bigl(\frac{a_s\lambda}{u_s}+a_m\bigr)\bigr ] - 16a^2_m\bigl(130u^2_m - 94u_m g_m + g^2_m\bigr ) \bigl[1 + 8\bigl(\frac{a_s\lambda}{u_s}+a_m\bigr)\bigr]\right.\\ \hspace{0.77cm}+\frac{32a_s\lambda^2}{u_s}(2u_m - g_m)\bigl[(1 + 4a^2_m)(1 + 2a_s)+4\bigl(\frac{a_s\lambda}{u_s } + a_m\bigr)(1 + 8a^2_m)\bigr ] + \frac{16a_s\lambda^3}{u_s}\left[1 + 2\left(\frac{a_s\lambda}{u_s}+2a_s\right)\right]\\ \hspace{0.77cm}\left.-(37u^2_m+25g^2_m-10u_mg_m ) \bigl[1 + 4\bigl(\frac{a_s\lambda}{u_s}+a_m\bigr)\bigr ] + 4608a^3_mu_m(2u_m - g_m)\bigl[1 + 10\bigl(\frac{a_s\lambda}{u_s}+a_m\bigr)\bigr]\right\},\\ \frac{dg_m}{dl } = g_m + \frac{1}{2\pi^2}\bigl\{\frac{32a_s\lambda^2}{u_s}(2g_m - u_m)\bigl[(1 + 4a^2_m)(1 + 2a_s ) + 4\bigl(\frac{a_s\lambda}{u_s}+a_m\bigr)(1 + 8a^2_m)\bigr]\\ \hspace{0.77cm}-4a^2_m\bigl[(g_m+73u_m)(7g_m-5u_m)+9(u_m - g_m)^2\bigr ] \bigl[1 + 8\bigl(\frac{a_s\lambda}{u_s}+a_m\bigr)\bigr]\\ \hspace{0.77cm}-\bigl[12(u_m - g_m)(2g_m - u_m)+6(u_m+g_m)^2-\frac{\lambda^2}{4}\bigr ] \bigl[4\bigl(\frac{a_s\lambda}{u_s}+a_m\bigr)+1\bigr]\\ \hspace{0.77cm}+576a_mu_m(2g_m - u_m)\bigl[6\bigl(\frac{a_s\lambda}{u_s}+a_m\bigr)+1\bigr]\bigr\},\\ \frac{du_s}{dl } = u_s+\frac{4}{\pi^2}\left\{\frac{27a_su^2_s}{2}(1 + 6a_s ) -\frac{144a_sa_mu_m\lambda^3}{u_s}\bigl[1 + 10\bigl(\frac{a_s\lambda}{u_s}+a_m\bigr)\bigr ] -\frac{\lambda^2}{2}\bigl[4\bigl(\frac{a_s\lambda}{u_s}+a_m\bigr)+1\bigr]\right.\\ \hspace{0.77cm}-\frac{9u^2_s}{16}(4a_s+1 ) - \frac{4\lambda^2_{\delta a}}{3 } \bigl(\frac{4\lambda_{\delta a}a_s}{u_s}+1\bigr ) -a^2_m\lambda^2\bigl[1 + 8\bigl(\frac{a_s\lambda}{u_s}+a_m\bigr)\bigr]\\ \hspace{0.77cm}\left.+\frac{8a_s\lambda^3}{u_s } \left(1+\frac{6a_s\lambda}{u_s}\right)+\frac{2768a_s\lambda^3_{\delta a}}{35u_s } \left(1+\frac{6a_s\lambda_{\delta a}}{u_s}\right)\right\},\\ \frac{d\lambda}{dl } = \lambda+\frac{\lambda}{\pi^2}\left\{9a_su_s(6a_s+1 ) + 72a_mu_m\bigl[6\bigl(\frac{a_s\lambda}{u_s}+a_m\bigr)+1\bigr ] -(2u_m - g_m)\bigl[4\bigl(\frac{a_s\lambda}{u_s}+a_m\bigr)+1\bigr]\right.\\ \hspace{0.77cm}-\frac{3u_s}{8}(4a_s+1)+\frac{96a_sa^2_m\lambda^2}{u_s } \bigl(\frac{a_s\lambda}{u_s}+a_m\bigr)-2\lambda\bigl[1-\frac{2a_s\lambda}{u_s}(1 + 4a^2_m)\bigr ] \bigl[2a_s+2\bigl(\frac{a_s\lambda}{u_s}+a_m\bigr)+1\bigr]\\ \hspace{0.77cm}-4a^2_m(20u_m - g_m)\bigl[1 + 8\bigl(\frac{a_s\lambda}{u_s}+a_m\bigr)\bigr]+576a^3_mu_m \bigl(1 + 10\bigl(\frac{a_s\lambda}{u_s}+a_m\bigr)\bigr)\\ \hspace{0.77cm}\left.+\frac{16a_s\lambda}{u_s}(2u_m - g_m)\left(1+\frac{6a_s\lambda}{u_s}\right ) + 6a_s\lambda\left[1 + 2a_s\left(\frac{\lambda}{u_s}+2\right)\right]\right\},\\ \frac{d\lambda_{\delta a}}{dl } = \lambda_{\delta a}+ \frac{\lambda_{\delta
a}}{3\pi^2}\left\{27a_su_s(1 + 6a_s ) -\frac{9u_s}{8}(4a_s+1)+\frac{64a_s\lambda^2_{\delta a}}{u_s } \bigl[1 + 2a_s\bigl(1+\frac{2\lambda_{\delta a}}{u_s}\bigr)\bigr]\right.\\ \hspace{0.77cm}\left.-12\lambda_{\delta a}\bigl[1 + 2a_s\bigl(1+\frac{\lambda_{\delta a } } { u_s}\bigr)\bigr]+36a_s\lambda_{\delta a}\left[1 + 2a_s\left(2+\frac{\lambda_{\delta a } } { u_s}\right)\right]\right\}. \end{array}\right.\end{aligned}\ ] ] these equations are used to analyze the impact of ordering competition on the global phase diagram in the main context of the paper . | we consider the impact of the competition among superconductivity , spin density wave , and nematic order in iron pnictides , and show that the ordering competition substantially reshapes the global phase diagram .
we perform a detailed renormalization group analysis of an effective field theory of iron pnictides and derive the flow equations of all the physical parameters . using these results , we find that superconductivity can be strongly suppressed by the ordering competition , and also extract the @xmath0-dependence of superfluid density .
moreover , the phase transitions may become first order .
interestingly , our rg analysis reveal that the nematic order exists only in an intermediate temperature region @xmath1 , but is destroyed at @xmath2 by thermal fluctuation and at @xmath3 by ordering competition .
this anomalous existence of nematic order leads to a back - bending of the nematic transition line on the phase diagram , consistent with the observed reentrance of tetragonal structure at low temperatures .
a modified phase diagram is obtained based on the rg results .
competition between distinct long - range orders is a common phenomenon that we frequently meet when studying a number of unconventional superconductors , including cuprates @xcite , heavy fermion compounds @xcite , and iron pnictides @xcite .
although ordering competition is a general concept and occurs in various patterns , the most frequently studied is that superconductivity compete and coexist with antiferrogmagnetism or nematic order @xcite .
when these orders coexist in a bulk superconductor , one expects that a well - defined quantum critical point ( qcp ) exists somewhere in the superconducting ( sc ) dome .
an important question is how to probe the widely predicted qcp in realistic experiments .
recently , we studied the physical effects of the competition between superconductivity and nematic order in a @xmath4-wave cuprate superconductor @xcite , and found that the superfluid density @xmath5 is suppressed at the nematic qcp significantly . according to our analysis ,
the suppression of @xmath5 is indeed caused by two scenarios .
first , the ordering competition reduces the charge condensate .
second , the gapless nodal quasiparticles couple strongly to the critical fluctuation of nematic order , which excites more normal quasiparticles out of the condensate .
we further showed that the suppression effect is significant solely at the qcp @xcite , thus @xmath5 should exhibit a deep valley at this point . based on these results , we proposed @xcite that the nematic qcp can be probed by measuring london penetration depth @xmath6 , which satisfies @xmath7
. clearly , the deep valley of @xmath5 corresponds to a sharp peak of @xmath6 .
thus far , no experimental evidence for the suppression of superfluid density has been reported in cuprates .
it is interesting that hashimoto _
et al .
_ have measured the penetration depth @xmath6 in an iron pnictide bafe@xmath8(as@xmath9p@xmath10)@xmath8 and observed a sharp peak of @xmath6 @xcite , which was claimed to signal the existence of a qcp of certain competing order beneath the sc dome @xcite .
this observation has stimulated considerable theoretical interest @xcite on the properties of the proposed qcp in bafe@xmath8(as@xmath9p@xmath10)@xmath8 and its relationship with the observed peak of @xmath6 .
however , the above finding seems to be at odds with the fact that there are two transition lines going across the superconducting line @xmath11 , namely a nematic transition line @xmath12 and a spin density wave ( sdw ) transition line @xmath13 .
this important issue was addressed by fernandes _
et al . _
@xcite within an effective theory that consists of sc , sdw , and nematic order parameters . the theoretical analysis of ref . @xcite demonstrated that the sdw and nematic transition lines penetrate separately into the sc dome , but merge at certain temperature , giving rise to a single qcp , which is schematically shown in fig .
[ fig_coexist ] . in spite of the interesting progress , our understanding of the physical effects of ordering competition is still quite limited , and more research effort is required . in particular
, it is necessary to investigate how the global phase diagram is influenced by ordering competition , which can help us to clarify many important issues about the nature of quantum phase transitions in iron pnictides . in this paper
, we study the global phase diagram of some iron pnictides , such as bafe@xmath8(as@xmath9p@xmath10)@xmath8 and ba(fe@xmath9co@xmath10)@xmath8as@xmath8 , by carefully investigating the impact of the competition among superconductivity , sdw order and nematic order . in order to examine the role played by the quantum fluctuation of various order parameters
, we will perform an extensive renormalization group ( rg ) analysis @xcite and obtain the rg equations for all the physical parameters that are introduced to describe the system .
we also extract the @xmath0-dependence of the superfluid density @xmath14 from the solutions of rg equations , and find that the superconductivity may be drastically suppressed by the ordering competition .
in addition , the rg results clearly show that the phase transitions become first order .
moreover , we have paid special attention to the fate of the transition line of nematic order in the sc dome .
interestingly , our rg analysis have discovered that the nematic order can only exist in an intermediate temperature region @xmath15 , but is destroyed at @xmath2 by thermal fluctuation and at @xmath3 by ordering competition .
such a phenomenon is found to occur in a wide region of the sc .
this result indicates that the nematic transition line bends back towards lower values of @xmath16 , which is shown in the schematic phase diagram fig .
[ fig_coexist ] .
we notice that nandi _
et al . _
@xcite have observed a reentrance of the tetragonal structure at low @xmath0 in the sc dome .
this observation is phenomenologically analogous to our theoretical result about the fate of nematic order .
it is known @xcite that many of the physical properties of bafe@xmath8(as@xmath9p@xmath10)@xmath8 and @xmath17 can be described by a three - band model that composed of one hole pocket located at the center of the brillouin zone @xmath18 and two electron pockets centered at two specific momenta @xmath19 and @xmath20 .
the hamiltonian is usually written as @xcite @xmath21 where @xmath22 represents two sorts of interactions : @xmath23 here , @xmath24 denotes the pair hoping interaction and @xmath25 the density - density interaction , which are responsible for the sdw order and superconductivity respectively @xcite .
we mention here that the @xmath24 terms describe the magnetic hund s coupling interactions @xcite , which are known to be important in multi - band electronic systems @xcite . as demonstrated in ref .
@xcite , the fermionic degrees of freedom can be fully integrated out , leading to @xmath26 where the parameters @xmath27 , @xmath28 , @xmath29 , @xmath30 , @xmath31 , and @xmath32 are defined in @xcite .
this model will be our starting point .
the transition lines for the sdw and sc orders are obtained by taking @xmath33 and @xmath34 , respectively . @xmath35
represent the sdw order parameters that generate long - range magnetic order for @xmath36 and @xmath37 respectively @xcite .
for @xmath38-wave superconductor , a universal sc gap is introduced such that @xmath39 @xcite .
an ising - type nematic order is induced by the magnetic order @xcite , and represented by the @xmath40 .
fernandes _ et al .
_ @xcite have recently studied the nature of quantum phase transitions in iron pnictides within the above effective theory and argued that the sdw and nematic orders merge at certain point in the sc dome . in this paper
, we will make a systematic rg analysis .
our aim is two fold .
first , we would like to examine the impact of the quantum fluctuations of sc and magnetic order parameters on the fate of phase transitions and the global phase diagram of the system , since these quantum fluctuations are known to play a vital role in systems that exhibit competing orders @xcite .
second , we attempt to extract the @xmath0-dependence of the superfluid density @xmath14 from the rg solutions . as aforementioned , @xmath14 can be suppressed by two different scenarios : the competitive interaction between distinct orders , and the coupling between fermionic quasiparticles and competing order . while the latter scenario has been studied in some recent references @xcite , the former is rarely considered in the literature @xcite . as will be shown below
, the influence of ordering competition on @xmath14 is prominent and can be efficiently obtained from our rg results . to simplify consideration , we first concentrate on the sdw qcp , corresponding to @xmath41 on fig .
[ fig_coexist ] , at which @xmath42 and the magnetic order parameters have vanishing mean values , i.e. , @xmath43 . in the sc dome ,
the sc order parameter develops a nonzero mean value , i.e. , @xmath44 . to study the quantum fluctuation of @xmath45 around @xmath46 , we introduce two new fields @xmath47 and @xmath48 , and then decompose @xmath45 as @xmath49 with @xmath50
moreover , to compute the superfluid density , it is convenient to couple @xmath45 to a gauge potential @xmath51 . recall that an important property of sc state is the happening of anderson - higgs mechanism , which leads to the meissner effect .
this mechanism needs to be properly included in the rg analysis .
we now introduce gauge potential @xmath51 to the effective theory via the standard minimal coupling , i.e. , @xmath52 , where @xmath53 is the coupling between @xmath51 and @xmath45 .
straightforward calculations gives rise to the following effective model @xmath54 where we have defined a number of new effective parameters from @xmath27 , @xmath28 , @xmath29 , @xmath30 , @xmath31 , @xmath32 , and @xmath53 : @xmath55 as a consequence of the anderson - higgs mechanism , the massless goldstone boson induced by gauge symmetry breaking is swallowed by the gauge boson @xmath51 , which then acquires an effective mass term @xmath56 .
the superfluid density @xmath5 is proportional to the gauge boson mass , i.e. , @xmath57 @xcite . in the following
, we will extract the @xmath0-dependence of @xmath14 by computing the @xmath58-dependence of @xmath59 , where @xmath58 is a running length scale .
however , @xmath60 is related intimately to other parameters , so we need to solve all the rg equations self - consistently . as shown in eq .
( [ eq_para_trans ] ) , all the new effective parameters are given by the seven original parameters .
our rg calculations are performed to the one - loop level in powers of the coupling constants .
analogous to the analysis of ref .
@xcite , we have derived the rg equations for all the coupling parameters ( see supplemental material @xcite for the details ) . to solve the equations , we assign the initial values of parameters as @xmath61 , and @xmath62 .
sc and sdw orders are supposed to coexist , so @xmath29 , @xmath30 , @xmath31 , and @xmath32 satisfy the constraint @xmath63 @xcite .
the main conclusion is independent of these assumptions .
there are three main results , to be explained one by one below .
first of all , we consider the fate of the associated phase transitions . after solving the rg equations
, we show the @xmath58-dependence of various parameters in fig .
[ fig_runaway_gm_t ] .
the quadratic coupling parameters @xmath30 , @xmath29 , and @xmath32 all flow to infinity eventually as @xmath64 , thus the system does not approach any stable fixed point in the low energy region . as discussed in previous works
@xcite , this results is usually regarded as a signature of an instability towards first - order transitions since a second - order transition is always associated with the presence of a stable infrared fixed point @xcite .
however , we can see from fig .
[ fig_runaway_gm_t ] that the parameters @xmath30 , @xmath29 , and @xmath32 increase with growing @xmath58 very slowly for small values of @xmath58 .
their magnitudes become large only when @xmath58 grows beyond certain threshold . therefore , before running to large values , the rg results of the @xmath58-dependence of these parameters are still reliable and give us very useful information about the physical properties of the system @xcite .
secondly , we address the impact of the ordering competition on the nematic transition line .
a number of experiments have observed an interesting transition from an orthorhombic structure to a tetragonal structure at low @xmath0 in the sc dome of @xmath17 @xcite and @xmath65 @xcite .
it turns out that the nematic order exists in an intermediate range of @xmath0 and disappears once @xmath0 is lower than certain value .
remarkably , this unusual behavior can be naturally obtained in our rg analysis . to demonstrate this ,
let us consider the property of parameter @xmath31 , whose sign determines whether the nematic order is present .
if @xmath66 , only one of the two order parameters @xmath67 and @xmath68 develops a finite mean value @xcite due to tetragonal symmetry breaking . in this case , the nematic order is present . on the other hand
, we have @xmath69 if @xmath70 , which indicates that the nematic order is absent @xcite .
therefore , to judge whether the nematic order exists at certain @xmath0 , we need to compute the @xmath0-dependence of @xmath31 from the rg results . the detailed @xmath58-dependence of @xmath31 is depicted in fig .
[ fig_runaway_gm_t ] .
as @xmath58 grows , @xmath31 first increases steadily , but decreases rapidly for large values of @xmath58 and becomes negative at certain critical value @xmath71 .
notice that @xmath71 is also the length scale at which @xmath30 , @xmath29 , and @xmath32 diverge .
we can translate the @xmath58-dependence of @xmath31 to a @xmath0-dependence by using the transformation @xcite @xmath72 .
the inset of fig .
[ fig_runaway_gm_t ] clearly shows that the positive @xmath31 becomes negative as @xmath0 decreases immediately below temperature @xmath73 , which means the nematic order is entirely suppressed at @xmath3 .
therefore , the nematic order can only exist in the intermediate range between @xmath74 and its transition temperature @xmath75 .
it is destroyed by the thermal fluctuation at @xmath76 , and by the strong competition among superconductivity , sdw , and nematic order at @xmath77 .
since @xmath75 is supposed to be always higher than @xmath13 @xcite , this behavior gives rise to a back - bending effect of the nematic transition line on the phase diagram shown in fig .
[ fig_coexist ] , which is consistent with the observed reentrance of tetragonal structure at low temperatures @xcite . to acquire a better knowledge of the phase diagram
, we also wish to know how the nematic transition line varies with @xmath0 as we move away from the sdw qcp . in the doping region @xmath78 , the sdw order parameters @xmath67 and @xmath68
develop finite mean values .
we here only present the main results ( see supplemental material @xcite for more details ) .
fig .
[ fig_g_m_ii ] shows the @xmath0-dependence of @xmath31 for different values of bare parameter @xmath79 . for the chosen values of @xmath79
, @xmath31 always first grows with decreasing @xmath0 and then becomes negative once @xmath0 is below certain threshold , which means the nematic order is suppressed at low @xmath0 .
in addition , the temperature scale at which @xmath31 changes sign increases as @xmath79 grows . in principle , the quantity @xmath80 , given by the sc transition temperature , also varies as @xmath79 grows .
however , despite this complexity , one can conclude from our rg results that the nematic transition line can not intersect with the horizontal axis of fig .
[ fig_coexist ] , but should instead merge somewhere with the sc transition line .
this property leads to a considerable modification of the global phase diagram , as visualized in fig .
[ fig_coexist ] . according to our results ,
the reentrance of tetragonal phase occur in a wide range of doping @xmath16 . finally , we turn to the @xmath0-dependence of superfluid density @xmath5 . within the effective model given by eq .
( [ eq_effective_l ] ) , the superfluid density @xmath81 . to obtain the @xmath0-dependent @xmath82
, we also utilize the transformation @xmath83 , where a suitable choice of @xmath80 is the sc temperature @xmath84 . at first glance ,
the rg results seem to suggest that @xmath85 diverges rapidly at @xmath74 .
however , because the transitions become first order , the @xmath0-dependence of @xmath14 is reliable only at @xmath86 . here
, we choose the value @xmath87 with @xmath88 being a little higher than @xmath74 to re - scale @xmath14 , and then show @xmath89 in fig .
[ fig_denisty_l ] , where the initial values are the same as those adopted in fig .
[ fig_runaway_gm_t ] .
an obvious conclusion is that the ordering competition leads to a strong @xmath0-dependence of @xmath5 , which decreases very rapidly as @xmath0 grows . for finite @xmath79 ,
the behaviors of @xmath14 are depicted in fig .
[ fig_density_ii ] . as @xmath79 increases ,
the suppression of @xmath14 becomes weaker .
since @xmath11 sensitively depends on the superfluid density , we can infer that @xmath11 should be suppressed to some extent in the region @xmath78 and that this effect is most significant at @xmath41 , as shown in fig .
[ fig_coexist ] .
hashimoto _ et al . _
@xcite has measured the superfluid density @xmath14 in a number of superconductors , including iron pnictides and heavy fermion compounds , and claimed to unveil a universal @xmath90 behavior over a wide range of temperatures .
this behavior was argued @xcite to be caused by the coupling between magnetic fluctuation and fermionic excitations .
our analysis has clearly showed that ordering competition alone can not account for the @xmath90 behavior . as depicted in the inset of fig .
[ fig_denisty_l ] , only in a very restricted region of @xmath0 could we extract an approximate @xmath91 behavior . on the other hand , we can see from the main panel of fig .
[ fig_denisty_l ] that , ordering competition does make a significant contribution to @xmath14 and hence can not be simply neglected .
bearing these two points in mind , we believe that both ordering competition and fermionic excitations need to be properly incorporated in a more refined model of @xmath14 . in summary , we have studied the impact of the strong competition among superconductivity , sdw order , and nematic order on the global phase diagram of iron pnictides by performing a systematic rg analysis within an effective field theory .
the main results are summarized in the schematic phase diagram presented in fig .
[ fig_coexist ] .
j.w .
is supported by the china postdoctoral science foundation under grants 2015t80655 and 2014m560510 , the national natural science foundation of china under grant 11504360 and the fundamental research funds for the central universities ( p. r. china ) under grant wk2030040074 .
g.z.l .
is supported by the national natural science foundation of china under grant 11274286 .
10 p. a. lee , n. nagaosa , and x .-
g .
wen , rev .
mod .
phys . * 78 * , 17 ( 2006 ) .
h. v. lhneysen , a. rosch , m. vojta , and p. wlfle , rev .
mod .
phys . * 79 * , 1015 ( 2007 ) .
o. stockert , s. kirchner , f. steglich , and q. si , j. phys .
. jpn . *
81 * , 011001 ( 2012 ) .
y. kamihara , t. watanabe , m. hirano , h. hosono , j. am .
chem .
soc . * 130 * , 3296 ( 2008 ) .
x. h. chen , t. wu , g. wu , r. h. liu , h. chen , and d. f. fang , nature ( london ) * 453 * , 761 ( 2008 ) .
g. f. chen , z. li , d. wu , g. li , w. z. hu , j. dong , p. zheng , j. l. luo , and n. l. wang , phys .
rev .
lett .
* 100 * , 247002 ( 2008 ) .
m. rotter , m. tegel , and d. johrendt , phys .
rev .
lett .
* 101 * , 107006 ( 2008 ) .
i. r. fisher , l. degiorgi , and z. x. shen , rep .
. phys . * 74 * , 124506 ( 2011 ) .
p. j. hirschfeld , m. m. korshunov , and i. i. mazin , rep .
. phys . * 74 * , 124508 ( 2011 ) .
a. v. chubukov , annu .
rev . condens .
matter phys .
* 3 * , 57 ( 2012 ) .
m. vojta , adv .
phys . * 58 * , 699 ( 2009 ) ; e. fradkin _
et al . _ ,
annu .
rev . condens .
matter phys .
* 1 * , 153 ( 2010 ) .
g .- z .
liu , j .-
r .
wang , and j. wang , phys .
rev .
b * 85 * , 174525 ( 2012 ) .
k. hashimoto , k. cho , t. shibauchi , s. kasahara , y. mizukami , r. katsumata , y. tsuruhara , t. terashima , h. ikeda , m.a .
tanatar , h. kitano , n. salovich , r.w .
giannetta , p.walmsley , a. carrington , r. prozorov , and y. matsuda , science * 336 * , 1554 ( 2012 ) .
a. levchenko , m. g. vavilov , m. khodas , and a.v .
chubukov , phys .
rev .
lett . * 111 * , 177003 ( 2013 ) .
d. chowdhury , b. swingle , e. berg , and s. sachdev , phys .
rev .
lett . * 111 * , 157004 ( 2013 ) .
r. m. fernandes , s. maiti , p. wlfle , and a. v. chubukov , phys .
rev .
lett . * 111 * , 057001 ( 2013 ) .
k. g. wilson , rev .
mod .
phys .
* 47 * , 773 ( 1975 ) .
r. shankar , rev .
mod .
phys . * 66 * , 129 ( 1994 ) . e. g. moon and s. sachdev , phys .
rev .
b * 82 * , 104516 ( 2010 ) .
r. m. fernandes , a. v. chubukov , and j. schmalian , nat .
phys .
* 10 * , 97 ( 2014 ) .
s. nandi , m. g. kim , a. kreyssig , r. m. fernandes , d. k. pratt , a. thaler , n. ni , s. l. budko , p. c. canfield , j. schmalian , r. j. mcqueeney , and a. i. goldman , phys .
rev .
lett . * 104 * , 057006 ( 2010 ) .
s. maiti and a. v. chubukov , phys .
rev .
b * 82 * , 214515 ( 2010 ) .
a. v. chubukov , d. v. efremov , and i. eremin , phys .
rev .
b * 78 * , 134512 ( 2008 ) f. wang , h. zhai , y. ran , a. vishwanath , d .- h
. lee phys .
rev .
lett . *
102 * , 047005 ( 2009 ) .
l. de medici , phys .
rev .
b * 83 * , 205112 ( 2011 ) .
t. schickling , f. gebhard , j. bnemannand , l. boeri , o. k. andersen , and w. webe , phys .
rev .
lett . * 108 * , 036406 ( 2012 ) .
l. de medici , g. giovannetti , and m. capone , phys .
rev .
lett . * 112 * , 177001 ( 2014 ) .
r. m. fernandes , a. v. chubukov , j. knolle , i. eremin , and j. schmalian , phys .
rev .
b * 85 * , 024534 ( 2012 ) .
j .- h .
she , j. zaanen , a. r. bishop , and a. v. balatsky , phys .
rev .
b * 82 * , 165128 ( 2010 ) .
z. nussinov , i. vekhter and a. v. balatsky , phys .
rev .
b * 79 * , 165122 ( 2009 ) ; a. j. millis , phys . rev .
b * 81 * , 035117 ( 2010 ) .
j. wang and g .- z .
liu , phys .
rev .
d * 90 * , 125015 ( 2014 ) .
t. nomoto and h. ikeda , phys .
rev .
lett . *
111 * , 167001 ( 2013 ) .
see supplemental material at http : .... for more detailed analytical treatment of the effective theory and for the expression of the coupled rg equations of all the effective parameters .
b. i. halperin , t. c. lubensky , and s .- k .
ma , phys .
rev .
lett . * 32 * , 292 ( 1974 ) .
r. m. fernandes and j. schmalian , phys .
rev .
b * 82 * , 014521 ( 2010 ) .
e. domany , d. mukamel , and m. e. fisher , phys .
rev .
b * 15 * , 5432 ( 1977 ) ; j. h. chen , t. c. lubensky , and d. r. nelson , phys .
rev .
b * 17 * , 4274 ( 1978 ) ; j. rudnick , phys .
rev .
b * 18 * , 1406 ( 1978 ) ; h. h. iacobson and d. j. amit , ann .
phys . * 133 * , 57 ( 1981 ) .
j. wang , a. eberlein , and w. metzner , phys .
rev .
b * 89 * , 121116(r ) ( 2014 ) .
d. k. pratt , w. tian , a. kreyssig , j. l. zarestky , s. nandi , n. ni , s. l. budko , p. c. canfield , a. i. goldman , and r. j. mcqueeney , phys .
rev .
lett . * 103 * , 087001 ( 2009 ) .
a. d. christianson , m. d. lumsden , s. e. nagler , g. j. mac - dougall , m. a. mcguire , a. s. sefat , r. jin , b. c. sales , and d. mandrus , phys .
rev .
lett . * 103 * , 087002 ( 2009 ) .
a. kreyssig , m. g. kim , s. nandi , d. k. pratt , w. tian , j. l. zarestky , n. ni , a. thaler , s. l. budko , p. c. canfield , r. j. mcqueeney , and a. i. goldman , phys .
rev .
b * 81 * , 134512 ( 2010 ) .
r. m. fernandes , s. a. kivelson , and e. berg , arxiv:1504.03656 .
j .- h .
she , m. j. lawler , and e .- a .
kim , phys .
rev .
b * 92 * , 035112 ( 2015 ) .
k. hashimoto , y. mizukami , r. katsumata , h. shishido , m. yamashita , h. ikeda , y. matsuda , j. a. schlueter , j. d. fletcher , a. carrington , d. gnida , d. kaczorowski , and t. shibauchi , pnas * 110 * , 3293 ( 2013 ) . |
rich information on neutrino properties , oscillations , the supernova mechanism , and very dense matter is contained in the neutrinos from core - collapse supernovae@xcite .
existing detectors such as super - kamiokande@xcite should accurately measure the @xmath5 component of the supernova signal .
however the very interesting @xmath1 , @xmath2 , @xmath6 , and @xmath7 ( collectively @xmath8 ) components may be detected without direct energy information and or in the presence of significant backgrounds from other neutrino induced reactions .
therefore , additional @xmath8 detectors could be very useful .
perhaps the ultimate " supernova detector involves neutrino - nucleus elastic scattering @xcite . the count rate in such a detector could be very high because the coherent elastic cross section is large and all six neutrino components ( @xmath9,@xmath5 , and the four @xmath8 ) contribute to the signal .
in particular , the detector is sensitive to @xmath8 , which are expected to have a high energy and large cross section .
elastic scattering detectors can have yields of a few or more @xmath8 events _ per ton _ for a supernova at 10 kpc ( @xmath10 m ) .
this is an increase by a factor of 20 or more over existing light - water detector yields of hundreds of @xmath5 and tens of @xmath8 events _ per kiloton_. furthermore , the energy of nuclear recoils provides direct information on the @xmath8 spectrum .
existing detectors measure @xmath8 via neutral - current inelastic reactions on oxygen@xcite , deuterium @xcite , or carbon @xcite . here
the observed energy deposition does not depend on the neutrino energy as long as it is above threshold .
perhaps neutrino - proton elastic scattering@xcite can be detected in kamland@xcite .
this is similar to neutrino - nucleus elastic scattering but has a smaller cross section .
alternatively , it may be possible to detect @xmath8 using inelastic excitations of pb .
proposals include using lead perchlorate , as suggested by elliott@xcite , omnis @xcite and land @xcite . here
some information on @xmath8 energies may be obtained by measuring the ratio of single- to two - neutron knockout .
however , the inelastic pb cross sections are somewhat uncertain . in contrast , neutrino - nucleus elastic cross sections can be calculated accurately with very little theoretical uncertainty .
the @xmath8 spectrum depends on how neutrinos thermalize with matter in a supernova , and is somewhat uncertain .
keil , raffelt , and janka have studied the effects of nn bremsstrahlung , pair annihilation , and nucleon recoil on the @xmath8 spectrum @xcite .
these effects can be measured with an elastic - scattering detector .
obtaining direct information on @xmath8 energies may be very important because the difference in energies for @xmath8 compared to @xmath9 or @xmath5 is the primary lever arm for observing neutrino oscillations .
for example , @xmath11 oscillations could lead to high - energy @xmath9 .
however , deducing the oscillation probability may depend crucially on knowing how hot the @xmath8 were to begin with .
neutrino - nucleus elastic scattering itself is flavor blind " .
therefore , the signal should be independent of neutrino oscillations ( among active species ) .
thus elastic scattering may provide a baseline with which to characterize the supernova source .
comparing this information to other flavor - dependent information and theoretical simulations may provide the best evidence of oscillations .
the kinetic energy of the recoiling nuclei is low , typically below 100 kev .
it is difficult to detect such low - energy events in the presence of radioactive backgrounds .
furthermore , scintillation signals from the nuclear recoils may be reduced by quenching because of the very high ionization density .
however , recent progress in designing detectors for low - energy solar neutrinos suggests that detection may be feasible
. in general , backgrounds for solar neutrinos with a low count - rate signal may be more severe than those for a supernova , where all of the events are concentrated in a few - second interval . cryogenic low energy astrophysics with noble gases ( clean )
is a proposed detector for low - energy solar neutrinos based on scintillation in an ultrapure cryogenic liquid@xcite .
this will detect electrons from neutrino - electron scattering at energies comparable to the recoil energy of nuclei from supernova neutrinos . in this paper
, we discuss the utility of clean for supernovae detection via neutrino - nucleus elastic scattering .
there is considerable interest in the direct detection of weakly interacting massive particles ( wimps ) .
these are expected to produce recoiling nuclei with a spectrum somewhat similar@xcite to that of supernova neutrino - nucleus elastic scattering .
again , backgrounds for wimp detection may be larger than for supernovae because of the low wimp count rates .
present wimp detectors , for example cdms@xcite , have small target masses .
however , future detectors may be larger .
it is important to search for neutrinoless double - beta decay , as this can distinguish dirac from majorana neutrinos .
existing @xmath12ge experiments use multikilogram masses @xcite .
the next - generation experiments such as majorana @xcite or genius @xcite may employ up to a ton of ge . the need for good energy resolution , to tell neutrinoless from two - neutrino double - beta decay , often aids in the detection of low - energy recoils .
we calculate that the largest double - beta decay experiments may soon be sensitive to galactic supernovae via elastic scattering .
finally , micropattern gas detectors @xcite may have a threshold low enough to detect nuclear recoils .
this may allow the observation of neutrino - nucleus elastic scattering using reactor antineutrinos .
thus the technical requirements for detecting low - energy solar neutrinos , wimps , double - beta decay , and supernovae via nuclear - elastic scattering may be similar .
one detector or approach for low - threshold , low - background , large mass measurements may have applications in multiple areas including supernova detection via neutrino - nucleus scattering . in section
ii , we fold elastic scattering cross sections with a model supernova neutrino spectrum to produce recoil spectra and yields .
we consider a range of noble - gas targets from @xmath3he to @xmath13xe along with @xmath14c , @xmath15si , @xmath12ge , @xmath16cd , @xmath17te , and @xmath4pb .
we also discuss extrapolating yields to nearby isotopes .
section iii focuses on the liquid - ne - based clean detector , which appears to be very promising .
a monte carlo simulation of backgrounds in clean is presented and compared to the expected supernova signal .
we discuss the large signal - to - background ratio , choice of fiducial volume , and possible detector thresholds .
we conclude in section iv .
the neutrino - nucleus elastic - scattering cross section @xmath18 is @xcite , @xmath19 for a neutrino of energy @xmath20 scattering at angle @xmath21 .
the fermi constant is @xmath22 .
this coherent cross section depends on the square of the weak charge @xmath23 @xmath24 of a nucleus with @xmath25 neutrons and @xmath26 protons .
the weak mixing angle is sin@xmath27 @xcite .
we assume a spin - zero target .
finally , the ground - state elastic form factor @xmath28 at momentum transfer @xmath29 , @xmath30 is @xmath31.\ ] ] here @xmath32 is the neutron density and @xmath33 is the proton density .
the form factor is normalized @xmath34 .
we neglect a small correction from the single - nucleon form factors .
the inclusion of @xmath28 is crucial for heavier targets .
however , we evaluate it at relatively small @xmath35 so the exact form of the densities is not important .
the proton density is often well constrained by measured charge densities .
for simplicity we use theoretical densities from simple relativistic - mean - field calculations using the successful nl3 effective interaction @xcite .
these calculations assume spherical ground states and do not include pairing corrections .
the use of other densities is not expected to change our results significantly .
we now consider a simple standard model " for the supernova - neutrino spectra , see for example @xcite .
this model is close to what others have used . a total energy of @xmath36 ergs ( 1 erg @xmath37 j )
is assumed to be radiated in neutrinos from a supernova at a distance @xmath38 of 10 kpc ( @xmath10 m ) . for simplicity
we use boltzmann spectra at temperatures of @xmath39 , 5 and 8 mev for the @xmath9 , @xmath5 and @xmath8 components respectively . here
@xmath40 is the boltzmann constant , which we set to one in the rest of the paper .
the use of fermi dirac spectra ( at zero chemical potential ) should give similar results .
however , neutrino - nucleus elastic scattering is sensitive to the high - energy tails in the spectra .
therefore non thermal spectra could modify our results somewhat and should be investigated in future work .
we assume equal partition in energy among the @xmath9 @xmath5 and the four @xmath8 components .
therefore this standard supernova radiates a total of @xmath41 , @xmath42 and @xmath43 neutrinos .
the time integral of the neutrino flux at earth @xmath44 for a neutrino of energy @xmath20 is @xmath45 for @xmath46 , or @xmath8 . microscopic simulations of supernovae suggest that equal partition of energy may be good only to @xmath47 25 % .
furthermore , there is important uncertainty in the @xmath8 spectrum , with estimates of @xmath48 ranging from @xmath49 to 8 mev .
it is an important goal of elastic - scattering detectors to measure @xmath48 .
therefore , the predictions of our supernova spectrum have significant uncertainties
. nevertheless , this simple model should provide order - of - magnitude estimates and may allow easy comparisons to other calculations . the yield of recoiling nuclei with energy @xmath50 and mass @xmath51 is @xmath52 where @xmath53 is the total number of target atoms . for our boltzmann spectra
this integral is simple : @xmath54 with @xmath55 . for large recoil energy @xmath50 , @xmath56 is proportional to @xmath57 for light nuclei the high - energy tail continues to hundreds of kev and is produced by the scattering of very - high - energy @xmath8 .
however , for heavier nuclei the tail is sharply reduced by the nuclear form factor .
we consider first noble - liquid detectors from @xmath3he to @xmath13xe and then a range of other detectors in order of increasing mass number @xmath58 from @xmath14c to @xmath4pb .
the yield @xmath56 from eq .
( [ yield ] ) is shown in figure 1 for detectors made of @xmath3he , @xmath59ne , @xmath60ar , @xmath61kr , and @xmath13xe .
we do nt mean to imply that detectors would be feasible with all of these liquids .
however , we show these nuclei to illustrate how the yield depends on @xmath58 for a broad range of @xmath58 .
the spectra in figure 1 are peaked at low recoil energy @xmath50 .
increasing @xmath58 raises the cross section because coherent scattering is proportional to @xmath62 .
thus at low energies @xmath56 increases significantly with @xmath58 .
however , as @xmath58 increases the spectrum is strongly shifted to lower energies by the form factor and the large target mass .
the energy integral of @xmath56 , or total yield , is given in table i in events per ton of detector .
also listed are events above a threshold of 5 , 10 , 25 , or 50 kev .
finally , the average recoil energy of the nuclei is given .
this average is influenced by a small number of events at high energies , while the spectrum is peaked at low energies .
the optimal choice of @xmath58 may involve a trade - off between cross section , which favors high @xmath58 , and recoil energy , which favors small @xmath58 .
this choice may depend on the attainable threshold .
furthermore , the choice of target material depends on a host of other practical considerations including the presence of possible backgrounds from radioactive isotopes .
although @xmath3he has a relatively small cross section it has high recoil energies .
we find a total yield in table i of 0.85 events per ton .
helium - based solar - neutrino detectors include heron @xcite and tpc @xcite ( or hellaz @xcite ) .
perhaps nuclei near @xmath63 give a reasonable balance between cross section and recoil energy .
however , ar may have backgrounds from radioactive @xmath64ar and @xmath65ar , while kr may have backgrounds from @xmath66kr .
xenon is being used in several dark - matter , double - beta decay and solar - neutrino detectors including xmass @xcite , xenon @xcite and zeplin@xcite .
the total yield is very large , 31 events per ton .
however , because @xmath13xe is heavy , there is a strong premium on obtaining a low threshold .
background from the @xmath67 double - beta decay of @xmath68xe should not be a problem for a supernova detector . above a threshold of 25
kev the yield is a relatively slow function of @xmath58 .
therefore one may have considerable freedom in the choice of target material .
we consider the clean liquid - ne solar - neutrino detector @xcite at some length .
the total yield of 3.99 events per ton is dominated by 3.08 @xmath8 events with only 0.38 @xmath9 and 0.53 @xmath5 events .
one is very sensitive to the @xmath8 spectrum .
for example , if the @xmath8 temperature is not the expected 8 mev but instead is near that for @xmath5 @xmath69 mev , the spectrum in figure 1 is greatly changed .
this verifies that the recoil spectrum contains direct information on the @xmath8 energies . a monte carlo simulation of backgrounds in clean is presented in section iii .
we now consider a number of other targets .
detectors based on organic scintillator such as borexino and kamland @xcite have a yield of 2.50 events per ton of @xmath14c , see table i. with any carbon - based detector there will be backgrounds from @xmath70c .
borexino should have a ratio @xmath70c/ @xmath14c of the order @xmath71 @xcite . at a concentration of @xmath71
there will be about 2 @xmath70c decays per ton during the 10 seconds of a supernova neutrino burst .
this is comparable to the number of elastic recoils .
however , the @xmath70c background should have a different spectrum and can be well measured at other times .
therefore , this background may not prevent the use of carbon as a supernova detector , even if it does prevent the observation of pp solar neutrinos .
quenching may be a serious problem for organic scintillator .
the amount of light produced from recoiling c ions may be much less than that for recoiling electrons @xcite .
an organic scintillator will also have events from neutrino - proton elastic scattering @xcite .
we estimate a yield of about 0.33 events per ton of ch@xmath72 . because of the light proton mass these events will have a larger recoil energy .
the proton elastic - scattering cross section has a theoretical uncertainty of 10 to 20 % from possible strange quark contributions to the nucleon s axial current and spin@xcite
. it would be very useful to have better laboratory measurements of neutrino - proton elastic scattering .
in contrast , strange quarks are not expected to make significant contributions for supernova neutrino - nucleus elastic scattering .
indeed , there is almost no theoretical uncertainty in the neutrino - nucleus elastic cross sections .
silicon detectors such as those described in @xcite have a yield of 5.5 events per ton and the recoil spectrum is shown in figure 2 , while @xmath12ge detectors have a yield of 18.6 events per ton and an average recoil energy of 9.5 kev .
the double - beta decay experiment majorana @xcite is proposed to have a 500 kg mass , while genius @xcite has a proposed mass of one ton .
these detectors should have very low backgrounds and low thresholds .
therefore they should be sensitive to a supernova at 10 kpc .
it is remarkable that such small target masses can yield statistics for our own galaxy comparable to those for the historic imb and kamiokande signals from sn1987a in the large magellanic cloud .
finally , @xmath16cd , @xmath17te and @xmath4pb yields are listed in table i. the yield and spectrum for @xmath17te is very close to @xmath13xe ( see figures 1 and 2 ) since they both have 78 neutrons .
the heavy nucleus @xmath4pb has a very large yield of 47.5 events per ton .
however , the average recoil energy is only 2.6 kev .
backgrounds and the need for a very low threshold may make a pb detector very difficult to build .
the nuclei in figures 1 and 2 display a range of recoil spectra . for point nuclei
there would be a single universal spectral shape , with the recoil energy decreasing and the yield increasing with increasing @xmath58 .
however , the different nuclear form factors modify the spectra for heavy nuclei .
finally , we provide a simple formula to extrapolate the yields in table i to nearby isotopes .
if one ignores small changes in the form factors of nearby nuclei , then the yield will be approximately proportional to the square of an effective weak charge , @xmath73 , @xmath74 with @xmath75 from eq .
( [ qw ] ) , and @xmath76 for odd @xmath58 nuclei and @xmath77 for even @xmath58 nuclei .
this factor takes into account the axial current of the last nucleon with @xmath78 .
because this term adds in quadrature with @xmath79 , it makes a very small contribution , except for very light systems such as the proton .
for example , using eq .
( [ qeff ] ) we find that @xmath80ne and @xmath81ne have cross sections respectively 1.29 and 1.48 times that of @xmath59ne .
natural ne is 0.3 % @xmath80ne and 8.8 % @xmath81ne so this will lead to a slight increase in yield over that for pure @xmath59ne .
clean , a detector concept based on liquid ne , was originally proposed for the detection of low - energy solar neutrinos .
it will also have high sensitivity to weakly interacting massive particles ( wimps ) .
liquid neon has a high scintillation yield , has no long - lived radioactive isotopes , and can be easily purified using cold traps .
in addition , neon is inexpensive , dense , and transparent to its own scintillation light , making it practical for use in a large self - shielding apparatus .
here we consider a clean detector in which a stainless steel tank holds 200 tons of liquid neon , half of which is exposed to a wavelength shifter to convert the ultraviolet light to the visible . inside the tank and suspended in the liquid neon are several thousand photomultipliers .
a diagram of the proposed clean design is shown in figure 3 .
clean will also be sensitive to supernova neutrinos , detected through neutrino - nuclear scattering . in this case
the entire active neon mass inside the wavelength shifter can be used , with a possible modest fiducial volume cut to reduce radioactive backgrounds . of prime importance for determining the sensitivity of clean to neutrino - nuclear scattering
is the determination of the light yield of liquid neon for nuclear recoils . because the density of excitation in scintillator is typically higher for nuclear recoils than for electron recoils , the chemically excited species are more likely to interact , increasing the likelihood that energy will be lost through mechanisms that do not produce light .
this quality is often expressed as a quenching factor " , the ratio of light emitted for a nuclear recoil to the light emitted from an electron recoil , per unit deposition energy .
while the quenching factor for liquid neon has not yet been measured , we expect it to be similar in magnitude to the quenching factor measured for liquid xenon .
recent ( and widely disagreeing ) measurements of the liquid xenon quenching factor@xcite are 22 % and 43 % .
the amount of quenching for liquid neon should be less than that for liquid xenon , as the density ( 1.2 ) of liquid neon is less than that of liquid xenon , while the scintillation mechanism in liquid neon is qualitatively similar .
for the following simulations , we assume that the quenching factor for liquid neon is 0.25 . clearly the scintillation quenching factor in liquid neon would have to be accurately measured in order to properly interpret any supernova data .
as for all neutrino detectors , a prime design consideration in clean is the reduction of radioactive backgrounds . we expect that any radioactive species suspended in the liquid neon will be removed by passing the neon through charcoal or similar adsorbant ; however there will remain a high rate of gamma rays entering the liquid neon after being emitted by the surrounding photomultipliers , photomultiplier support structure , wavelength shifter , and stainless steel tank containing the liquid neon .
figure 4 shows the expected supernova neutrino recoil spectrum for a clean detector with 100 tons of active liquid neon , assuming 3750 scintillation photons per mev , 100 % efficiency for the wavelength shifter , photomultiplier coverage of 75 % , and a photomultiplier quantum efficiency of 15 % .
also shown is the expected radioactive background for 10 seconds of observing time , assuming that the combined gamma and x - ray emission is dominated by the photomultiplier glass and the wavelength shifter substrate .
the simulation assumes photomultipliers 20 cm in diameter , each with mass 650 g , with 30 ng per g of u and th and 60 @xmath82 g per g of k in the glass .
the wavelength shifter is assumed to be evaporated on quartz wafers of 1 mm thickness , with a u and th concentration of 1 ng per g. from @xmath83 seconds of simulated data , we find a background of 62 @xmath84 8 events in 10 seconds within the energy range of 0 to 200 kev . here
the uncertainty corresponds to the @xmath84 1-sigma interval for a particular 10-second observation .
since liquid neon has no long - lived radioactive isotopes that would limit its practical threshold , the clean detector could conceivably trigger on as few as two photoelectrons .
the accidental coincidence rate in clean will be low , as the photomultiplier dark count rate will be suppressed at liquid - neon temperature ( 27 k ) .
thus we expect that the full 100 tons of liquid neon could be viewed with a threshold of 2 photoelectrons , equivalent to a recoil energy of about 5 kev .
the expected 62 background events is smaller than the expected supernova signal of 330 ne elastic events above 5 kev in 100 tons .
the radioactive backgrounds , while small in comparison to the supernova signal , can be lowered further through position resolution , since most gamma rays will deposit their energy in the outer edges of the liquid neon sphere .
recently , monte carlo simulations have shown that the location of ionizing radiation events in clean can be determined by analyzing the pattern of photomultiplier hits .
these simulations are described in detail in an upcoming publication@xcite . here
we show only some results relevant to supernova neutrino detection .
we find that a mild fiducial radius cut , leaving a mass of 70 tons , virtually eliminates gamma - ray background for the purposes of supernova neutrino detection ( only 2.6 @xmath84 1.6 events in the fiducial volume in 10 seconds ) .
these results are shown in figure 5 . in the case of a fiducial volume cut ,
detecting a few more photoelectrons will improve the convergence of the position resolution algorithm .
currently we can analyze events that produce as few as 8 photoelectrons ( recoil energy of 21 kev ) if position cuts are applied . in general , as the position cut is increased or the number of photoelectrons is reduced , algorithm convergence should be carefully checked .
the expected 2.6 background events are much smaller than the expected supernova signal of 140 ne events above 21 kev in 70 tons .
the signal - to - noise ratio reported here , though already quite large , might be significantly improved by the development of photomultipliers with lower inherent radioactivity .
such photomultipliers are under investigation by the clean and xenon collaborations .
in addition , plans are being made to measure the quenching factor for nuclear recoils in liquid neon , as this measurement is important for the determination of the sensitivity of clean to wimp particles , as well as for supernova neutrinos .
detectors with reduced radioactive backgrounds may be able to study supernovae via neutrino - nucleus elastic scattering .
this could provide important information on the flux and spectrum of @xmath8 ( @xmath1 and @xmath2 ) .
elastic - scattering detectors could see a few or more events per ton for a supernova at 10 kpc .
this is 20 or more times the number of events per ton of existing water detectors .
the clean experiment , based on the detection of scintillation in liquid ne , is a prime example of a detector that would be sensitive to the elastic scattering of supernova neutrinos .
its active mass of 100 tons may yield almost 400 @xmath8 events .
in addition , many other detectors , including the largest dark - matter and double - beta decay experiments , may also be sensitive to supernova neutrinos via elastic scattering .
observation of neutrino - nuclear elastic scattering will complement supernova signals from other detectors .
water detectors such as sno or super - k will detect @xmath8 without @xmath8 spectral information .
this energy information could be important for neutrino oscillations .
kamland may be able to measure @xmath85 elastic scattering .
this contains energy information just like @xmath86 scattering .
the small cross section for @xmath85 scattering may be compensated by a large detector mass .
this may yield only slightly smaller statistics than clean .
we strongly encourage development of detectors based on both @xmath85 and @xmath86 elastic scattering since they may have different backgrounds , thresholds , and systematic errors . furthermore ,
the very large @xmath86 elastic cross sections may allow even larger statistics in future detectors .
we thank john beacom , steve elliott , and harry miley for useful discussions and acknowledge financial support from doe grant de - fg02 - 87er40365 and nsf grant 0226142 .
we acknowledge the hospitality of the aspen center for physics , where some of this work was done .
see for example , j. f. beacom , hep - ph/9909231 .
super - kamiokande collaboration , astro - ph/0007003 .
a. drukier and l. stodolsky , phys . rev .
* d30 * ( 1984 ) 2295 .
blas cabrera , lawrence m. krauss , and frank wilczek , phys .
* 55 * ( 1985 ) 25 .
k. langanke , p. vogel and e. kolbe , phys .
* 76 * ( 1996 ) 2629
. c. j. virtue , astro - ph/0103324 . v. barger , d. marfatia and b. p. wood , hep - ph/0011251 .
john f. beacom , will m. farr , petr vogel , phys.rev .
* d66 * ( 2002 ) 033001 .
s. r. elliot , phys .
rev . * c62 * ( 2000 ) 065802
. j. engel , g. c. mclaughlin , and c. volpe , hep - ph/0209267 .
d. b. cline _ et al .
_ phys . rev .
* d50 * ( 1994 ) 720 ; p. f. smith , astropart . phys .
* 8 * ( 1997 ) 27 ; j. j. zach _ et al .
a484 * ( 2002 ) 194 . c. k. hargrove _ et al .
_ , astropart . phys . * 5 * ( 1996 ) 183 .
mathias th .
keil , georg g. raffelt and hans - thomas janka , astro - ph/0208035 .
d. n. mckinsey and j. m. doyle , journal of low temperature physics * 118 * ( 2000 ) 153 .
j. d. lewin and p. f. smith , astroparticle phys .
* 6 * ( 1996 ) 87 .
a. abusaidi _
et al . _ ,
* 84 * ( 2000 ) 5699 .
l. baudis _
et al . _ ,
* 83 * ( 1999 ) 41 ; c. e. aalseth , _ et al .
_ , ( igex collaboration ) , yad . phys . * 63 * ( 2000 ) 1299 .
l. debraeckeleer , talk at workshop on the next generation u.s .
underground science facility , wipp , june 12 - 14 , 2000 , carlsbad , new mexico , usa ; c. e. aalseth _ et al .
_ , in proc . of taup2001 , gran sasso , italy , september 2001 , ed .
a. bettini _ et al . _
see also http://majorana.pnl.gov .
h. v. klapdor - kleingrothaus , l. baudis , g. heusser , b. majorovits , h. paes , hep - ph/9910205 .
p. barbeau , j. i. collar , j. miyamoto , and i. shipsey , hep - ex/0212034 .
d. z. freedman , d. l. tubbs , and d. n. schramm , annu .
* 27 * ( 1977 ) 167 .
sld collaboration , k. abe _
et al . _ ,
* 86 * ( 2001 ) 1162 .
g. a. lalazissis , j. knig and p. ring , phys . rev . * c55 * ( 1997 ) 540 .
k. takahashi , m. watanabe , k. sato and t. totani , phys .
* d64 * ( 2001 ) 093004 ; t. totani , k. sato , h. e. dalhed and j. r. wilson , astrophys .
j. * 496 * ( 1998 ) 216 .
r. e. lanou , h. j. maris , and g. m. seidel , phys .
* 58 * ( 1987 ) 2498 ; s. r. bandler et al .
, j. low temp .
phys . * 93 * ( 1993 ) 785 .
g. bonvicini , d. naples and v. paolone , nim a * 491 * ( 2002 ) 402 ; g. bonvicini and a. schreiner , nim a * 493 * ( 2002 ) 90 .
a. de bellefon for the hellaz collaboration , nucl .
b ( proc . suppl . )
* 70 * ( 1999 ) 386 .
m. nakahata , http://www.mpi-hd.mpg.de/nubis/www_lownu2002/transparency/nakahata_lownu2002.pdf .
e. aprile _ et al .
_ , astro - ph/0207670 ; y. suzuki , astro - ph/0008296 .
david b. cline , hanguo wang , y. seo , astro - ph/0108147 .
borexino collaboration , g. alimonti _
et al . _ ,
* 16 * ( 2002 ) 205 . also at hep - ex/0012030 .
l. a. ahrens _ et al .
_ , phys . rev .
* d35 * ( 1987 ) 785 .
et al . _ , nim a * 449 * ( 2000 ) 147 .
et al . _ , epjc , to be published .
k.j.coakley and d. n. mckinsey , to be published .
lllllll + target & y & y @xmath87 & y @xmath87 & y @xmath87 & y @xmath87 & @xmath88 + & & 5 kev & 10 kev & 25 kev & 50 kev & ( kev ) + @xmath3he & 0.85 & 0.82 & 0.79 & 0.72 & 0.62 & 240 + @xmath14c & 2.5 & 2.2 & 2.0 & 1.6 & 1.1 & 83 + @xmath59ne & 4.0 & 3.3 & 2.9 & 2.0 & 1.2 & 46 + @xmath15si & 5.5 & 4.2 & 3.4 & 2.1 & 1.1 & 31 + @xmath60ar & 9.4 & 6.6 & 5.0 & 2.5 & 0.99 & 21 + @xmath12ge & 18.6 & 9.6 & 5.8 & 1.7 & 0.30 & 9.5 + @xmath61kr & 19.8 & 9.5 & 5.5 & 1.4 & 0.20 & 8.4 + @xmath16cd & 26.3 & 9.7 & 4.6 & 0.70 & 0.041 & 5.7 + @xmath17te & 31.8 & 10.1 & 4.3 & 0.47 & 0.014 & 4.8 + @xmath13xe & 31.1 & 9.8 & 4.1 & 0.43 & 0.012 & 4.8 + @xmath4pb & 47.5 & 7.3 & 1.7 & 0.022 & 0.001 & 2.6 + [ table1 ] yield in events per ton for a supernova at 10 kpc assuming different target materials . also listed is the number of events above thresholds of 5 , 10 , 25 or 50 kev . finally the average recoil energy @xmath88 is given . | development of large mass detectors for low - energy neutrinos and dark matter may allow supernova detection via neutrino - nucleus elastic scattering .
an elastic - scattering detector could observe a few , or more , events per ton for a galactic supernova at 10 kpc ( @xmath0 m ) .
this large yield , a factor of at least 20 greater than that for existing light - water detectors , arises because of the very large coherent cross section and the sensitivity to all flavors of neutrinos and antineutrinos .
an elastic scattering detector can provide important information on the flux and spectrum of @xmath1 and @xmath2 from supernovae .
we consider many detectors and a range of target materials from @xmath3he to @xmath4pb .
monte carlo simulations of low - energy backgrounds are presented for the liquid - neon - based cryogenic low energy astrophysics with noble gases ( clean ) detector .
the simulated background is much smaller than the expected signal from a galactic supernova . |
by mixing fermion species of unequal particle number , recent experiments have revealed many aspects of the exciting physics of imbalanced fermion gases @xcite . in these experiments ,
a trap is loaded with an ultracold spin mixture consisting of two lithium hyperfine states .
the existence of an easily - accessible feshbach resonance allows the tuning of the interspecies interactions with great accuracy so that the entire bec - bcs crossover with number imbalance may be explored @xcite .
the _ bcs regime _ involves a small and negative s - wave scattering length , with the fermions forming weakly bound cooper pairs . on the other hand , in the _ bec regime _ , where the scattering length is positive , the particles form tightly bound molecular pairs forming a condensate @xcite . in between
lies the _ unitary regime _ , characterized by a diverging scattering length ( signifying the appearance of a bound state ) , where particles are strongly interacting .
the divergence of the scattering length implies that the system is universal , in the sense that only a single length scale characterizes it .
much work has been done in an attempt to understand the ground state properties of the imbalanced clouds ; however , despite theoretical advances @xcite , some problems remain unresolved .
the mit data @xcite are in good agreement with monte carlo ( mc ) simulations @xcite and renormalization group ( rg ) predictions @xcite , which point to the existence of a _ three - shell _ structure consisting of a superfluid ( sf ) core phase , a partially polarized and a fully - polarized normal ( n ) phase . from these experiments ,
the homogeneous phase diagram has been determined and is in agreement with both the mc and rg results . on the other hand ,
the results of the rice group s experiments @xcite are are less well understood .
for example , they find that superfluidity does not break down at high polarizations , which is incompatible with the results of the mit experiments and thus also in contradiction to theoretical predictions of a critical polarization , above which superfluidity breaks down . since both the mit phase diagram and the mc and rg predictions were obtained under the assumption that the local density approximation ( lda ) holds , and since the rice experiments involve smaller traps , it is likely that the answer to the riddle lies in the breakdown of lda ( in other words , finite - size effects ) ; indeed , taking as an ansatz a _ two - shell _ structure , the rice data can be recovered by using a n - sf interface tension as a fitting parameter @xcite . for our purposes , it is sufficient to say that the mit experiments are well - described by theory , and furnish direct evidence that a n - sf interface exists , separating an unpolarized sf from a partially polarized n. in this article we argue that the n - sf interface may block the thermal equilibration process between the n and sf phases ; one consequence of this will be to cause a temperature imbalance to appear across the interface .
incorporating such temperature difference in the existing models may provide the key to a better understanding of the experiments .
we present a detailed study of the scattering off the n - sf interface within the hartree - fock - bogoliubov - de gennes ( hfbdg ) formalism , followed by a calculation of the heat conductivity across the interface .
we find that this conductivity drops exponentially fast with increasing @xmath0 with @xmath1 the sf gap and @xmath2 the temperature .
our model predicts the conductivity to be vanishingly small for the experiments at unitarity .
a summary of our results was published earlier in ref . .
the article is structured as follows .
we start off in section [ sec_bulk ] by discussing the bulk physics of imbalanced fermion gases .
this is done using a mean - field hamiltonian which incorporates both the gap formation and the hartree - fock terms .
we present analytical results for the equations of state which enable us to write down the condition for coexistence between the n and the sf phases .
then , in section [ sec_intscatt ] , we discuss in detail how particles scatter on a n - sf interface according to the hartree - fock - bogoliubov - de gennes formalism .
we find a rich variety of interfacial scattering processes , depending on the energy and momentum of the incident particle .
the implications of this scattering are outlined in section [ sec_transort ] , where we obtain the thermal conductivity exactly and with a good analytical approximation . finally in section [ sec_timescales ]
we calculate that for the experiments at lowest temperatures , the typical timescale associated with the equilibration of a temperature difference across the interface , is of order seconds and therefore comparable with the trap lifetime .
fermion mixtures in the bec - bcs crossover are the subject of a very active research field .
although some theoretical models give accurate results for the bec and/or bcs regimes , no model yet allows an analytical , quantitative description across the entire crossover . in the following , we use the
so - called _ one - channel model _ at zero temperature ; this is a straightforward generalization of the bcs model to arbitrarily strong coupling and is widely used in the literature , as it is known to embody most qualitative features of the bec - bcs crossover for the currently experimentally relevant feshbach resonances .
interactions between fermions and the bosonic bound states may be explicitly incorporated within a two - channel model @xcite .
we start off by deriving the hartree - fock - bogoliubov - de gennes ( hfbdg ) equations .
we then focus on the physics of an interface separating two semi - infinite phases : a sf and a polarized n. we write down the equations of state in these phases , and use them to find the condition for two - phase coexistence . in this section ,
we extend the work reported in ref . by including the hf term into the bdg equations and treating the case of unequal species masses .
the system under consideration consists of two fermionic species @xmath3 and @xmath4 with masses @xmath5 and chemical potentials @xmath6 for @xmath7 and which we assume to be trapped by the external potential @xmath8 .
the intraspecies interactions are via p - wave scattering which can be safely ignored at low temperatures . on the other hand ,
the interactions between the two spin species are taken to be contact interactions characterized by the coupling constant @xmath9 with @xmath10 and @xmath11 the s - wave scattering length . within a generalized hf approximation
, one may work with the following hamiltonian @xcite : @xmath12,\nonumber\end{aligned}\ ] ] where @xmath13 , the gap @xmath14 and the hf terms @xmath15 and @xmath16 .
as usual , one diagonalizes the hamiltonian by a bogoliubov transformation i.e. by writing the operators @xmath17 in terms of fermionic quasiparticle operators @xmath18 : @xmath19,\\ \widehat{\psi}_\downarrow(\textbf{r},t)&=\sum_{\ell } e^{-ie_{\ell } t/\hslash } \left[u_{\ell , \downarrow}(\textbf{r})\hat{c}_{\ell , \downarrow}-v_{\ell,\downarrow}(\textbf{r})\hat{c}^\dag_{\ell,\uparrow } \right],\end{aligned}\ ] ] where @xmath20 and @xmath21 are the quasiparticle wave functions , @xmath22 and @xmath23 . then , if under this transformation the hamiltonian is diagonalized , the relations @xmath24=-e_\ell \hat{c}_{\ell , i}$ ] and @xmath25=e_n \hat{c}^{\dag}_{\ell , i}$ ] must hold such that one straightforwardly finds the four _ hartree - fock - bogoliubov - de gennes _ ( hfbdg ) , or _
blonder - tinkham - klapwijk _ , equations @xcite : [ bdg ] @xmath26 the wave functions @xmath27 and @xmath28 which appear in the first two equations are not coupled to the wave functions @xmath29 and @xmath30 of the last two equations so that , apart from normalization , they can be treated separately .
in the rest of this section , we will be concerned with the study of a homogeneous system of two fermion species at fixed chemical potentials @xmath31 and @xmath32 and with no external potential @xmath33 .
furthermore we treat both the case of a n and a sf phase . at zero temperature and for negative scattering length
, there always exists a ( meta-)stable state with @xmath34 and for which both species densities are equal ; we call this state the _ bcs state _ or _
sf phase_. for this state , we define now the _ effective chemical potential imbalance _ @xmath35 and the _ average effective chemical potential _ @xmath36 as follows : @xmath37 without loss of generality , we further assume that @xmath38 .
note that in the sf phase , both spin species have equal densities which implies equal hf potentials @xmath39 for the spin species .
one can then find the ground state energy of the _ homogeneous _ sf system as a function of @xmath31 , @xmath32 and @xmath11 .
solving the hfbdg eqs . , and minimizing the ground state energy with respect to the variables @xmath1 and @xmath40 , the gap and number equations at zero temperature are : @xmath41,\\ u_{_s}&=-\frac{g}{2}\int \frac{\text{d}^3\mathbf{k}}{(2\pi)^3}\left[1-\frac{\varepsilon_{_\mathbf{k}}-\mu_{_s } } { \sqrt{(\varepsilon_{_\mathbf{k}}-\mu_{_s})^2+\delta^2}}\right],\end{aligned}\ ] ] where @xmath42 .
note that the above equations are coupled by the term @xmath40 which appears in the definition of @xmath36 .
the ground state grand potential per unit volume in terms of @xmath1 and @xmath36 is : @xmath43.\end{aligned}\ ] ] we are mostly interested in the bcs side of the bec - bcs crossover where the effective chemical potentials @xmath44 and @xmath36 can be taken as positive . performing the integrals over the three - dimensional wave vectors
leads to the analytical expressions for the gap , the hf potential and the grand potential per unit volume @xcite : [ eqstate ] @xmath45 ^ 2/\eta,\label{eqstate2}\\ u_{_s}%&=\sqrt{2m_{_+}}a_{12}\hslash^{-1}(\mu_{_s})^{3/2}\left[p_{_{1/2}}(\eta)-\eta^{-1}p_{_{3/2}}(\eta)\right].\nonumber\\ & = \mu_{_s}\left[1-\eta^{-1}p_{_{3/2}}(\eta)/p_{_{1/2}}(\eta)\right],\label{eqstate3}\\ \omega_{_s } % & = -\frac{\sqrt{2}(m_{_+})^{3/2 } ( \mu_{_s})^{5/2}v}{20\pi\hslash^3(-\eta)^{1/2}}\label{eqstate1}\\ % & \times\left[(\eta^{-2}-5)p_{_{1/2}}(\eta)+4\eta^{-1}p_{_{3/2}}(\eta)\right],\nonumber\\ & = -\frac{(m_{_+})^{3/2}\epsilon_{_a}^{5/2}\eta^2 } { 80\pi\hslash^3|p_{_{1/2}}(\eta)|^5}\nonumber\\ & \quad\times\left[(\eta^{-2}-5)p_{_{1/2}}(\eta)+4\eta^{-1}p_{_{3/2}}(\eta)\right].\label{eqstate1b}\end{aligned}\ ] ] with @xmath46 the legendre function , @xmath47[zeradef ] and we introduced the energy scale which is set by the scattering length @xmath48[scattdef ] .
expression allows to write the grand potential in terms of @xmath49 ; indeed , for each value thereof , one can find @xmath50 by solving the relation : @xmath51 ^ 4,\end{aligned}\ ] ] such that the grand potential is fully determined by @xmath49 only . for the _ normal state _
, one finds the following expressions for the grand potential per unit volume of the ground state and for the hf potentials : [ grootpot ] @xmath52\nonumber,\\ \frac{u_\uparrow}{\epsilon_{_a}}&=-\frac{4\sqrt{2}}{3\pi}\left(\frac{m_\downarrow}{m_{_+}}\right)^{3/2}\left(\frac{\mu_\downarrow - u_\downarrow}{\epsilon_{_a}}\right)^{3/2}\label{normeqstate2},\\ \frac{u_\downarrow}{\epsilon_{_a}}&=-\frac{4\sqrt{2}}{3\pi}\left(\frac{m_\uparrow}{m_{_+}}\right)^{3/2}\left(\frac{\mu_\uparrow - u_\uparrow}{\epsilon_{_a}}\right)^{3/2}\label{normeqstate3}.\end{aligned}\ ] ] at fixed chemical potentials , the attractive interactions induce an increased normal state density and a decreased ground - state grand potential .
eqs . and
are coupled in a non - trivial manner and can be solved numerically to find @xmath53 and @xmath54 . before writing down the n - sf coexistence condition
, we explain how a variation of the chemical potential imbalance between the spin species may change the ground state . a small imbalance ( or polarization )
gives rise to a sf state while a large polarization leads to a n state .
intuitively this is expected since it is known from bcs theory that pair formation is energetically favorable for particles with opposite momenta . by shifting the fermi energies ,
an energy cost must be paid which is associated with the matching of the fermi levels so as to enable pair formation ; when this energy cost is too high , the n state prevails . in a _ trapped _ imbalanced fermion system , both the n and the sf phase may be present , bounded by an interface . using a sufficiently large number of particles ,
a local density approximation is justified there and amounts to using an effective local chemical potential @xmath55 with @xmath56 the chemical potential at the trap center and @xmath57 @xcite . since in experiments
both spin species feel the same trapping potential , the imbalance @xmath58 is constant throughout the trap . on the other hand , the average chemical potential @xmath59/2
$ ] varies : it is large at the center of the trap and decreases towards the trap boundary .
the ground state energy of the sf increases with increasing value of the gap ; as seen from eqs .
, the gap value is proportional to the average chemical potential @xmath36 and independent of @xmath35 .
therefore , if present , the sf phase will be at the trap center where @xmath36 is maximal .
moving radially outwards , two - phase coexistence of the n and sf phase occurs at the position where there is a balance between the energy gained by creating , on the one hand , a gap in the sf , and , on the other hand , a density difference in the n phase . to establish now the condition for coexistence between the n and the sf phase , recall that we have a system containing two fermion species @xmath60 with masses @xmath5 and chemical potentials @xmath6 for which interactions are characterized by a negative scattering length @xmath11 .
we aim to find the coexistence condition as a function of the parameters @xmath11 , @xmath31 and @xmath32 .
the sf state is fully described by the parameters @xmath1 and @xmath40 , which can be extracted from eqs . , and .
the normal state , on the other hand , is characterized by @xmath53 and @xmath54 which are solved for using eqs . and .
the coexistence condition obtained by equating the grand potentials of eqs . and is : @xmath61=\left(\frac{m_\uparrow}{m_{_+}}\right)^{3/2}\left(\frac{\mu_\uparrow - u_\uparrow}{\epsilon_{_a}}\right)^{5/2 } + \left(\frac{m_\downarrow}{m_{_+}}\right)^{3/2}\left(\frac{\mu_\downarrow - u_\downarrow}{\epsilon_{_a}}\right)^{5/2}.\end{aligned}\ ] ] by condition one can find the surface of coexistence in the space of the parameters ( @xmath62 ) .
for one specific case it is easy to relate the gap value to the chemical potentials at coexistence ; this occurs in case of equal masses @xmath63 without incorporation of the hf term .
the energy scale @xmath64 can then be eliminated and by a numerical analysis , one finds to a good approximation the clogston relation , originally derived for superconductors in a magnetic field @xcite : @xmath65 the clogston limit is exact in the bcs limit , when @xmath66 .
lastly , it is common to write the scattering length @xmath11 in units of @xmath67 , the latter defined by the relations @xmath68^{2/3}/(2m_{_+})$ ] ; we find that in the sf : @xmath69^{1/3}.\nonumber\end{aligned}\ ] ]
in this section , we explain in full detail how particles scatter from a n - sf interface for imbalanced fermion gases . by taking a simple model for the interface , we describe the different scattering regimes and we give the transmission coefficients for scattering across the interface ; the latter will be used for calculating transport quantities in the next section . to model the n - sf interface , we start with a geometry wherein the @xmath70 plane separates the n phase at @xmath71 from the sf phase at @xmath72 or : @xmath73 with @xmath74 the heaviside function .
this approximation was also used by andreev @xcite for studying the normal - superconductor interface .
a particle of energy @xmath75 incident on the interface may scatter to states which have the same energy and these are deduced from the spectra in the sf and in the n phase . in the n phase ,
we have for species @xmath76 : @xmath77 we depict with full lines this spectrum in the left panel of fig .
[ fig1 ] where one recognizes two approximately linear energy branches @xmath3 and @xmath4 near the ( unequal ) fermi surfaces . in the sf on the other hand
, one finds using the hfbdg equations , an @xmath78 and a @xmath79 spectrum : @xmath80 where @xmath81 .
these gapped spectra are depicted in the right panel of fig .
the @xmath78 branch is composed of particles of phase @xmath3 and holes of phase @xmath4 and vice versa for the @xmath79 spectrum .
one can find that the minimal values attained by the @xmath78 and @xmath79 spectra are @xmath82 where @xmath83/2 $ ] with @xmath84 , whence the lowest spectrum touches the @xmath85 axis or becomes _ ungapped _ under the condition that @xmath86 .
we assume a gapped spectrum .
since the gap function is constant in both the n and the sf phase , we can decompose the quasiparticle wave functions @xmath87 and @xmath88 , associated with the @xmath78-branch , into their fourier components : @xmath89.\nonumber\end{aligned}\ ] ] here @xmath90 and @xmath91 are complex scattering amplitudes and the sub- and superscripts @xmath92 and @xmath93 denote normal , superfluid , particle - like and hole - like , respectively .
we split the vectors @xmath94 into its components parallel to the wall @xmath95 and its @xmath96-component @xmath97 , @xmath98 , @xmath99 and @xmath100 perpendicular to the interface . at fixed energy ,
there exist eight quasiparticle states , four of which in the n and four in the sf phase .
these states are represented by dots in fig .
[ fig1 ] : filled dots are particle - like states and empty dots are hole - like states ( the latter have opposite group and phase velocity ) . by use of the spectra of eqs . and
, one relates the wavenumbers @xmath99 , @xmath100 and @xmath101 to the wavenumber @xmath102 of a particle in the n phase : [ wavevectors ] @xmath103 where we defined : [ additional ] @xmath104 and @xmath84 .
let us now use these equations to try to understand the physics of the scattering processes at the interface .
consider for example a particle incident on the interface from the n side with wavenumber @xmath102 ( along the @xmath96-axis ) and energy @xmath105 ; this is represented in fig . [ fig1 ] by a curly arrow .
whatever processes occur at the interface , energy must be conserved ; but out of the seven remaining states in the n phase with the same energy , only four have a group velocity directed away from the interface .
this implies that @xmath106 .
the particle can now undergo either specular reflection to the state marked in fig .
[ fig1 ] by @xmath107 , andreev reflection to the state @xmath101 or it can penetrate the sf as a particle - like ( to @xmath108 ) or as a hole - like quasiparticle ( to @xmath109 ) . note that andreev reflection is also of importance in other ultracold systems such as in fermionic traps without population imbalance @xcite and in bosonic traps near a n - sf interface in the absence @xcite or presence @xcite of a lattice .
the scattering amplitudes depend on the energy @xmath105 of the incident particle , represented by the long dashed line in fig .
consider for instance decreasing the energy of the incident particle to a value smaller than the lowest value of the @xmath78-branch ; since no sf states are available , transmission to the sf is then not possible .
mathematically this is marked by the wavenumber of one state in the @xmath78-branch becoming imaginary , indicating the presence of an evanescent wave in the sf .
furthermore , we can also raise the long dashed line of fig .
[ fig1 ] , that is , increase the energy @xmath105 , to a value above the local maximum at @xmath110 of the spectrum of species @xmath4 in the n phase ; this will inhibit andreev reflection .
the crossover of either of the @xmath111 and @xmath108 and the wave vector @xmath112 from real to imaginary ( or vice versa ) signifies a change in the scattering mechanism . defining @xmath113[chipdef ] ,
such transitions are encountered when ( see eq . ):
@xmath114 the curves @xmath115-@xmath116 separate the possible scattering regimes and are shown in figs .
[ fig2a ] and [ fig2b ] where the latter depicts a general ( gerenic ) diagram and the former assumes equal species masses and zero hf potentials . to better understand the various scattering regimes , we switch to working with the energy @xmath117 of eq . instead of @xmath105 .
recall that @xmath117 takes values in the interval @xmath118 where @xmath119/2 $ ] ; clearly @xmath120 in case @xmath121 and @xmath122 .
if the energy of the particle incident on the interface is small enough or @xmath123 , there is insufficient energy to excite quasiparticles inside the sf ; the particle is then fully reflected as a superposition of a particle and a hole .
this regime is labelled i in figs .
[ fig2a ] and [ fig2b ] .
the situation where @xmath124 i.e. regime ii is physically forbidden since the wave vector of the incident particle ( i.e. @xmath112 ) is imaginary .
suppose now that a particle is incoming with energy @xmath117 above the threshold value @xmath125 .
although a particle in regime iii appears to have sufficient energy for transmission , in undergoes complete reflection .
particles in regime iv may be transmitted and reflected but all targeted states are particle - like . in both regimes
iv and v , there are no reflected holes ; that is , andreev reflection does not occur . however , in regime v , both particle - like and hole - like excitations are present in the sf .
finally , in regime vi , both particle - like and hole - like excitations , or both andreev and normal reflection , are allowed .
the blocking of scattering regimes can be understood as being caused by a too large angle of incidence at the interface , similar to total internal reflection . while the critical angle is roughly constant for optical total internal reflection , here the critical angles of incidence depend on the energy of the incoming particle .
generally , a particle of species @xmath3 has an angle of incidence @xmath126 which satisfies @xmath127 , or in terms of the energy : @xmath128 the critical angles for transmission , hole - like transmission and andreev reflection can then be obtained as a function of the energy @xmath105 by substitution of @xmath129 from eq . into eq . . to summarize , the lightly - shaded regimes in figs .
[ fig2a ] and [ fig2b ] describe normal particles of species @xmath3 which undergo complete reflection , while the rightmost , heavily - shaded regime is unphysical .
_ only incident particles in the unshaded regimes may excite quasiparticles inside the sf . _ if we consider now holes of spin species @xmath4 which are incident on the interface from the n phase , they can only be transmitted in a regime analogous to vi in the diagram .
furthermore , one can find now also the scattering diagram , applied for particles of spin species @xmath4 and holes of spin species @xmath3 incident on the interface and scattering to the @xmath79 branch .
the scattering diagram is then qualitatively similar to that of fig .
[ fig2a ] and the curves @xmath115-@xmath116 are given by eqs .
after performing the transformations @xmath130 , @xmath131 , @xmath132 and @xmath133 . at this stage
, we can discuss the influence of the interaction parameter @xmath134 , the hf potentials and the species mass asymmetry on the diagram of fig .
[ fig2a ] as they may strongly affect the transport properties which will be calculated in the following section . _ the interaction regime _ first of all , the @xmath129-@xmath117 diagrams of the deep bcs interaction regime and the unitary interaction regime have a different topology . assume for now that @xmath122 and that @xmath135 .
in the deep bcs regime , the relation @xmath136 implies that line @xmath137 in fig .
[ fig2a ] is shifted horizontally to much higher values of @xmath129 so that regime vi becomes by far the most important scattering regime .
therefore quasiparticle reflections occur almost only via the andreev mechanism .
it is known that this involves transport of particles , but no transport of energy across the interface .
in contrast , at unitarity where @xmath138 ( since @xmath139 as derived from eq . in case @xmath122 and @xmath140 ) , phase vi does not even exist .
this means that quasiparticles with energy @xmath117 above the threshold @xmath1 can not undergo andreev reflection but only normal ( specular ) reflection , by which neither particles nor energy are carried across the interface . _
the hf potentials _ assume again that @xmath122 ; the hf potentials @xmath141 , @xmath53 and @xmath54 are seen to be present in the expressions which describe the curves @xmath115-@xmath116 of fig .
[ fig2a ] .
however , it follows from these equations that _ the hf potentials induce only horizontal shifts of the curves @xmath115-@xmath116 but do not affect the upper and the lower bounds of regime i_. whereas line @xmath137 is shifted to the right , the curves @xmath115 , @xmath142 and @xmath116 are always shifted to the left .
this can be deduced from the fact that @xmath143 and @xmath144 since the interactions are attractive and the density of the sf exceeds the density of the n phase .
_ the species mass asymmetry _ in the presence of a species mass asymmetry @xmath145 , regime i may become either larger or smaller .
this is due to a shift of its lower bound @xmath146/2 $ ] ( obtained by taking @xmath147 in eq . ) and a shift of the upper bound @xmath148 .
it is seen that both shifts have the same sign but can either increase or decrease the surface area of regime i. also , the species mass asymmetry induces a shift of line @xmath116 as is seen from eq . .
in order to study the transport properties in the next section , we first seek to relate the scattering amplitudes @xmath90 and @xmath91 of eqs . to the transmission coefficients .
the amplitudes are determined by matching the quasiparticle wave functions of eq . and their derivatives at the interface @xmath70 @xcite .
in its most general form , this leads to the equations : @xmath149 where we defined @xmath150[varsigmadef ] .
as argued before , depending on the various scattering regimes , one must set some scattering amplitudes to zero .
one finds : @xmath151 we give the solutions for the scattering amplitudes in expressions - in the appendix . from the hfbdg eqs .
it readily follows that the quasiparticle density @xmath152 and the quasiparticle currents @xmath153 -\frac{i } { 2m_\downarrow}\left[v_\downarrow\boldsymbol{\nabla } v_\downarrow^*-v_\downarrow^*\boldsymbol{\nabla } v_\downarrow\right],\end{aligned}\ ] ] satisfy the `` continuity equation '' @xmath154 .
note , however , that they do not express conservation of particles or mass . the same expressions are valid for the @xmath79 branch after the transformation @xmath155 .
we define now the _ transmission coefficient _ @xmath156[transdef ] of an incident particle of energy @xmath117 and momentum @xmath94 as the ratio of the transmitted to the incoming current along the @xmath96-axis .
as argued before , the transmission coefficients vanish in regimes i , ii and iii of fig .
[ fig2a ] ; for regimes iv , v and vi ( and for holes ) we give their analytical forms in expression in the appendix . for energies
@xmath117 slightly above the transmission threshold @xmath157 ( for regimes v and vi ) one can analytically find the following behavior @xmath158 this is similar to the case of particles scattering from a hf potential of height @xmath125 @xcite .
moreover , in the bcs regime wherein , by the andreev approximation , one can assume @xmath159 , the transmission coefficients reduce to : @xmath160 apart from the species mass asymmetry @xmath161 , this is the expression which was obtained by andreev in ref . @xcite .
thermally excited particles incident on the surface will be reflected unless they have an energy of at least @xmath162 . at low enough temperatures ,
not many particles will have enough energy to penetrate .
this leads to a significant decrease of heat conductivity through the interface , a phenomenon well - known for superconductors @xcite .
we will now argue that the same blocking of energy transport occurs for imbalanced fermion gases , and discuss the influence of the interaction regime on this phenomenon .
heat transport across the interface can be studied using the transmission coefficients calculated in last section . using the relation : @xmath163
one can write the particle - like _ heat flux _ through the interface and to the @xmath78 branch as : @xmath164 where @xmath165[fermidef ] is the fermi - dirac distribution and the integration is performed over all regimes of the @xmath129-@xmath117 plane . for the heat flux
@xmath166 carried by _ hole _
excitations , the integration is only over regime vi and by use of @xmath167 , we can obtain the same expression as eq .
but with @xmath168 instead of @xmath169 and with @xmath170 instead of @xmath171 , as given in . by the same approach ,
the contributions of the @xmath79-branch are then found by performing the transformations @xmath172 , @xmath131 , @xmath132 and @xmath133 . in equilibrium ,
quasiparticles are incident on both sides of the interface such that the n - sf flux is balanced by an equal sf - n flux .
suppose , however , that the temperature on the n side is @xmath173 higher than on the sf side ( and @xmath174 ) . a net heat flow @xmath175 then results from the scattering to the @xmath78-channel ( @xmath176 ) and to the @xmath79-channel ( @xmath177 ) . for small temperature differences ,
we may write @xmath178 with @xmath179 the _ heat conductivity _ caused by scattering to the @xmath78 channel . within the kapitza approach , we have : @xmath180 in the deep bcs regime , where the andreev approximation may be used , one may obtain an analytical expression for @xmath181 . taking @xmath159 and assuming the temperature to satisfy @xmath182 effects regime vi of figs .
[ fig2a ] and [ fig2b ] to be the sole important scattering regime ; in other words , all reflections are due to andreev reflection .
we obtain for the heat flux to the @xmath78 channel ( see eq . ):
@xmath183(k_{_b}t)^{3/2}}{\sqrt{\delta\sqrt{\widetilde{m}}}}.\end{aligned}\ ] ] for the @xmath78-channel the heat conductivity of eq . in the bcs regime
is then obtained : @xmath184}{\sqrt{k_{_b}t\delta\sqrt{\widetilde{m}}}},\nonumber\end{aligned}\ ] ] which amounts to andreev s result when @xmath121 and @xmath122 @xcite .
the heat flux and heat conductivity of the @xmath79-channel is again found after the transformation @xmath185 , @xmath186 , @xmath187 and @xmath132 .
the energy carried by the @xmath79 branch is a factor @xmath188 lower than that of the @xmath78 branch and due to coexistence @xmath189 and since @xmath190 , it can be neglected .
. demonstrates that @xmath181 decays exponentially fast with decreasing temperature .
this can be understood as follows . for low enough temperature i.e. @xmath191 , only incident particles ( and holes ) with energies slightly above the threshold @xmath192 can contribute to the heat conductivity ; they have , however , a very low statistical weight : @xmath193 it is this weight factor which also appears in expression .
the question now is : what will happen to this exponential decay if we tune the interactions from the bcs to the unitary regime ? in order to answer this , we have numerically calculated the ratio of @xmath194 to the conductivity in the n phase @xmath195 , and show this in fig .
[ fig3 ] as a function of @xmath196 and we assumed @xmath122 and @xmath140 .
_ we find that @xmath197 decreases drastically below @xmath198 and is independent of the interaction parameter @xmath199_. also in fig .
[ fig3 ] we plot the heat conductivity of eq . ; it is seen to give a good fit for all scattering regimes is eliminated using the coexistence condition eq . .
the reason for this good agreement is that the threshold line @xmath200 of the diagram of fig .
[ fig2a ] is unaffected by changing the scattering regime .
furthermore , relation which describes the transmission coefficient slightly above the threshold line guarantees that @xmath171 displays the same behavior in regimes v and vi . as argued before , the main effect of varying @xmath201 from large and negative ( deep bcs ) to zero ( unitarity ) , is to shift the curves @xmath115-@xmath116 in diagram of fig .
[ fig2a ] .
both the strong exponential decay in @xmath202 and the square root dependence of @xmath171 on @xmath203 are unaffected by the variation of @xmath201 and the introduction of the hartree potentials , hence the very similar behavior of @xmath197 for all scattering regimes under study . on the other hand ,
what is different between the scattering regimes is the value of @xmath1 ; it is small in the sense that @xmath204 in the bcs limit and is comparable with @xmath205 at unitarity .
the introduction of the hartree potentials has two implications on the heat resistivity .
first of all , the scattering diagram will change by horizontal shifts of lines @xmath115-@xmath116 ; as argued before , this does not affect the steep exponential decay of resistivity .
secondly , the coexistence condition will change , that is , the relation between @xmath1 and @xmath93 will change .
this , on the other hand may strongly affect the exponential decay as is seen from eq . .
at unitarity , the relation between @xmath1 and @xmath93 is known from monte carlo ( mc ) @xcite simulations and a superfluid local density approximation ( slda ) @xcite . from these
we adapt the values @xmath206 , @xmath207 , @xmath208 @xcite and finally @xmath209 @xcite . in fig .
[ fig3bis ] , we compare the andreev approximation with values of the one - channel model ( full line ) and of the mc and slda results ( dashed line ) .
we see that the one - channel model predicts the sharp decrease of @xmath181 to be at higher temperatures than the mc and slda results .
we find that the interface conductivity at the lowest experimental temperatures is very low in the sense that the resulting characteristic equilibration time is of order of seconds i.e. the trap lifetime .
we explain now how we estimate the characteristic equilibration time @xmath210 .
consider again a temperature bias @xmath173 across the interface .
we are interested in the resulting total heat flux @xmath211 , induced by the excess heat @xmath212 in the n phase . by definition ,
the heat flux @xmath213 is the amount of heat transferred per unit time per unit surface area .
one can therefore estimate @xmath210 by the relation @xmath214 with @xmath215 the interface area .
moreover , at position @xmath216 in the degenerate n phase @xmath76 , one has @xmath217 $ ] with @xmath218 the particle density and @xmath7 @xcite . combining this with @xmath219
, one finds @xmath220 assume now an isotropic trap of @xmath221 particles at polarization @xmath222 with radial trapping frequency @xmath223hz such that the local density approximation is valid @xcite . for this configuration
, we calculate that the characteristic time of relaxation @xmath210 is of order seconds when @xmath224 .
this means that according to our estimates , the relaxation time is of the order of the lifetime of the system at the lowest attainable experimental temperature @xmath225 such that in these experiments , the n and sf phases may not be equilibrated .
at least two interesting experimental consequences of the low thermal conductivity for ultracold gases are different to what is the case for a sc .
first of all , in a superconductor , the heat conductivity has a lattice component which dominates the electron component at low temperatures ; the lack of such a component in the system under study makes the decrease of the conductivity more significant @xcite .
secondly , at very low temperatures , a n - sf temperature gradient will entail a pressure difference @xmath226 across the interface , and therefore the physical movement of the interface ; this is impossible for normal - superconductor junctions .
taking @xmath227 and an infinite thermal resistivity , one finds @xcite : @xmath228 where @xmath229 is the @xmath230 pressure of species @xmath76 .
depending on the time scales involved , mechanical equilibrium might be reached before thermal equilibrium .
the step function model for the gap ( see eq . ) is adequate for tackling the problem at hand , that is for estimating the heat conductivity across the interface .
first of all , this model should gives qualitatively correct results as long as the quasiparticle wavelengths are of the same order or less as the length over which the gap varies .
indeed , this is clearly valid in the bcs limit , when the healing length for the gap , further denoted as @xmath231 , is of order @xmath232 and @xmath233 .
moreover , near unitarity , the step - function model is also justified since the length of variation of the gap is smaller than the quasiparticle wavelength @xcite .
therefore , the step function of eq
. serves as a good approximation for all regimes under consideration .
second , the step - model for the gap is justified whenever the interface width must be small as compared to the variation length of the trapping potential .
this is also realized in experiments .
we argue now that , even if we would take into account the exact interface profile , our results would hardly be affected . by solving the bogoliubov - de - gennes equations in a self - consistent manner , the interface profile for the gap could be determined @xcite . at zero temperature ,
the concomitant interface width is of order @xmath234 . as was shown in ref .
, such smooth interface gives rise to a zero transmission coefficients for particles with energy below the gap , and above the gap , smaller transmission coefficients than predicted by our step - function model .
this simply means that particles incoming to the interface with energy below the gap can not cross over to the superfluid phase and therefore they can not carry energy across the interface . hence our conclusions of a blocking of energy transport across the interface remains valid , independent of the interface width .
note also that , due to the decreased transmission coefficient above the gap , the heat conductivity will be even smaller than predicted here .
the same conclusion can be drawn at finite temperatures : particles with energy below the gap can not penetrate the superfluid and the conclusions of ref .
concerning the transmission coefficients will still be valid .
indeed , since it is known that the interface width increases with temperature as it varies between @xmath235 and @xmath236 near @xmath237 @xcite , @xmath231 is certainly of the order of @xmath238 at the experimentally relevant temperatures , that is below @xmath239 .
this can be seen in refs . and where the self - consistent calculations were performed for various temperatures and even for the strong coupling regime @xcite . in our model
, we do not take into account the temperature dependence of the gap .
however , this is well justified as the gap @xmath240 depends very weakly on temperature at low @xmath2 . in the bcs limit
it is well - known that @xmath240 goes as @xmath241 ( see for example in ref . ) .
moreover , this weak variation is also present in the strong coupling limit @xcite and can be understood within the single - channel model from the appearance of the factor @xmath242 in the finite - temperature gap equation .
we have studied the effects of a n - sf interface in experiments on trapped imbalanced fermion gases .
we clarify the nature of the possible processes that may occur when a particle is incident onto the interface .
such particles will be thermally excited at any finite temperature , and act as carriers of energy , thus eventually causing thermal equilibration between the n and sf phases .
we find that reflection ( both andreev and normal , or specular ) of low energy particles off the interface in experiments on trapped imbalanced fermion gases causes a suppression of energy transport from one phase to the other . at low enough temperatures ( comparable to the currently accessible temperatures in experiments ) , this suppression grows exponentially with decreasing temperature .
this , in turn , delays thermal equilibration of the system , which may result in a temperature difference between the n and the sf .
our estimates of the timescales for this equilibration are in the seconds , which indicates that this effect is experimentally relevant .
the incorporation of this temperature difference may allow current models to finally resolve the difficulties in explaining the experiments .
it is a pleasure to thank henk stoof and koos gubbels for discussions and useful suggestions .
we acknowledge partial support by project no . fwo g.0115.06 ; a.l .
is supported by project no .
goa/2004/02 and b.v.s . by the research fund k.u.leuven .
assume particles of species @xmath3 incident on the interface with momentum @xmath102 and energy @xmath117 .
define first @xmath243 and : if the coordinate @xmath245 with @xmath246 is positioned in regime vi of fig .
[ fig2a ] , the matching of the wave functions and their derivatives at @xmath70 leads to the coefficients : @xmath247 for regime v ( see fig . [ fig2a ] ) , we get : @xmath248 for regime iv ( see fig . [ fig2a ] ) , we get : @xmath249 for holes of species @xmath4 which are incident on the interface , and which necessarily must have energy and momentum in regime vi , the scattering amplitudes are : @xmath250 the transmission coefficients for the different regimes are : [ transmissions ] @xmath251/\mathcal{j}_{_0}^2.\nonumber\\ \text{for regime v : } s_{_\alpha}^{_p}=&\left[4k^{_p}k_{_p}|k^{_h}-k_{_h}|^2(\varsigma_{_h}^2-\widetilde{m})\right.\label{transmissions2}\\ & \left.+4k^{_h}k_{_p}|k_{_h } + k^{_p}|^2(\widetilde{m}-\varsigma_{_p}^2)\right]/|\mathcal{j}_{_1}|^2.\nonumber\\ \text{for regime iv : } s_{_\alpha}^{_p}=&\frac{4k^{_p}k_{_p}|k^{_h}+k_{_h}|^2(\varsigma_{_h}^2-\widetilde{m})}{|\mathcal{j}_{_2}|^2}.\label{transmissions3}\\ \text{for holes : } s_{_\alpha}^{_h}=&\left[4k^{_p}k_{_h}(k^{_h}-k_{_p})^2(\widetilde{m}-\varsigma_{_p}^2)\right.\label{transmissions4}\\ & \left.+4k^{_h}k_{_h}(k^{_p}+k_{_p})^2(\varsigma_{_h}^2-\widetilde{m})\right]/\mathcal{j}_{_0}^2.\nonumber\end{aligned}\ ] ] 70 g.b . partridge , w. li , r.i .
kamar , y.a .
liao and r.g .
hulet , science * 311 * , 503 ( 2006 ) .
partridge , w. li , y.a .
liao , r.g .
hulet , m. haque and h.t.c .
stoof , phys .
lett . * 97 * , 190407 ( 2006 ) .
zwierlein , a. schirotzek , c.h .
schunck and w. ketterle , science * 311 * , 492 ( 2006 ) .
zwierlein , c.h .
schunck , a. schirotzek and w. ketterle , nature * 442 * , 54 ( 2006 ) . y. shin , m.w .
zwierlein , c.h .
schunck , a. schirotzek and w. ketterle , phys .
lett . * 97 * , 030401 ( 2006 ) .
schunck , y. shin , a. schirotzek , m.w .
zwierlein and w. ketterle , science * 316 * , 867 ( 2007 ) .
y. shin , c.h .
schunck , a. schirotzek and w. ketterle , nature * 451 * , 689 ( 2008 ) .
m. houbiers , r. ferwerda , h.t.c .
stoof , w.i .
mcalexander , c.a . sackett and r.g .
hulet phys .
a * 56 * , 4864 ( 1997 ) .
d.e . sheehy and l. radzihovsky ,
( n.y . ) * 322 * , 1790 ( 2007 ) .
s. giorgini , l.p .
pitaevskii and s. stringari , rev .
phys . * 80 * , 1215 ( 2008 ) .
w. ketterle and m. w. zwierlein , rivista del nuovo cimento * 31 * , 247 ( 2008 ) .. d.e .
sheehy and l. radzihovsky , phys .
* 96 * , 060401 ( 2006 ) .
q. chen , j. stajic , s. tan and k. levin , phys .
rep . * 412 * , 1 ( 2005 ) . f.
chevy , phys .
lett . * 96 * , 130401 ( 2006 ) .
k.b . gubbels and h.t.c .
stoof , phys .
* 100 * , 140407 ( 2008 ) .
gubbels , m.w.j .
romans and h.t.c .
stoof , phys .
lett . * 97 * , 210402 ( 2006 ) .
bedaque , h. caldas and g. rupak , phys .
lett . * 91 * , 247002 ( 2003 ) .
parish , f.m .
marchetti , a. lamacraft and b.d .
simons , nature physics * 3 * , 124 ( 2007 ) .
m. haque and h.t.c .
stoof , phys .
a * 74 * , 011602 ( 2006 ) .
t.n . de silva and e.j .
mueller , phys .
lett . * 97 * , 070402 ( 2006 ) .
a. lazarides and b. van schaeybroeck , phys
a * 77 * , 041602(r ) ( 2008 ) . c. lobo , a. recati , s. giorgini and s. stringari , phys
* 97 * , 200403 ( 2006 ) .
j. tempere , s.n .
klimin and j.t .
devreese , arxiv:0807.4434 .
b. van schaeybroeck and a. lazarides , phys .
lett . * 98 * , 170402 ( 2007 ) .
de gennes , _ superconductivity of metals and alloys _ , ( addison - wesley , reading , ma , 1989 ) .
a.l . fetter and j.d .
walecka , _ quantum theory of many - particle systems _ , mcgraw hill , boston 1971 .
blonder , m. tinkham and t.m .
klapwijk , phys .
b * 25 * , 4515 ( 1982 ) .
j. tempre and j.t .
devreese , j.p hys .
* 39 * , s57 ( 2006 ) .
y. ohashi and a. griffin , phys .
a * 72 * , 013601 ( 2005 ) .
m. randeria , in _ bose einstein condensation _ , edited by griffin , stroke and stringari ( cambridge university press , 1995 ) .
a. bulgac and y. yu , phys .
lett . * 88 * , 042504 ( 2002 ) .
t. papenbrock and g.f .
bertsch , phys .
c * 59 * , 2052 ( 1999 ) .
su , phys .
a * 74 * , 063627 ( 2006 ) .
a. bulgac , phys .
a * 76 * , 040502 ( 2007 ) .
g. bertaina and s. giorgini , phys .
a * 79 * , 013616 ( 2009 ) .
m. haque and h.t.c .
stoof , phys .
* 98 * , 260406 ( 2007 ) .
r. haussmann and w. zwerger , phys .
a * 78 * , 063602 ( 2008 ) .
a. recati , c. lobo and s. stringari , phys .
a * 78 * , 023633 ( 2008 ) .
a.m. clogston , phys .
* 9 * , 266 ( 1962 ) .
m. urban , phys .
a * 75 * , 053607 ( 2007 ) .
i. zapata and f. sols , phys .
* 102 * , 180405 ( 2009 ) .
s.d . huber and g. blatter , phys .
b * 79 * , 174504 ( 2009 ) .
j. demers and a. griffin , can . j. phys . * 49 * , 285 ( 1971 ) .
landau and e.m .
lifshitz , _ quantum mechanics ( non - relativistic theory ) _ , ( pergamon , oxford , 1977 ) .
andreev , zh .
fiz . * 46 * , 1823 ( 1964 ) [ sov
jetp * 19 * , 1228 ( 1964 ) ] .
andreev , sov .
jetp * 49 * , 655 ( 1966 ) [ sov .
jetp * 22 * , 455 ( 1966 ) ] .
m. tinkham , _ introduction to superconductivity _ , 2nd ed .
( mcgraw - hill , new york , 1996 ) .
k. huang , _ statistical mechanics _
, first edition , ( wiley , ny , 1963 ) .
mcmillan , phys .
rev . * 175 * , 559 ( 1968 ) .
y. tanaka and m. tsukada , phys .
b * 42 * , 2066 ( 1990 ) .
van son , h. van kempen and p. wyder , phys . rev .
b * 37 * , 5015 ( 1988 ) .
j. stajic , j.n .
milstein , q. chen , m.l .
chiofalo , m.j .
holland and k. levin , phys .
a * 69 * , 063610 ( 2004 ) .
abrikosov , _ fundamentals of the theory of metals _
( elsevier , ny , 1988 ) . | recent experiments on imbalanced fermion gases have proved the existence of a sharp interface between a superfluid and a normal phase .
we show that , at the lowest experimental temperatures , a temperature difference between n and sf phase can appear as a consequence of the blocking of energy transfer across the interface . such blocking is a consequence of the existence of a sf gap , which causes low - energy normal particles to be reflected from the n - sf interface . our quantitative analysis is based on the hartree - fock - bogoliubov - de gennes formalism , which allows us to give analytical expressions for the thermodynamic properties and characterize the possible interface scattering regimes , including the case of unequal masses .
our central result is that the thermal conductivity is exponentially small at the lowest experimental temperatures . |
this presentation is based mainly on @xcite , work done in collaboration with c. j. horowitz , a. nogga , a. c. fonseca , c. hanhart , g. a. miller , j. a. niskanen , and u. van kolck .
the lagrangian of quantum chromodynamics ( qcd ) is almost symmetric under the exchange of the up and down quarks , which is called charge symmetry ( cs ) .
however , since the quark masses are different , this symmetry is broken , charge symmetry breaking ( csb ) . for hadrons , cs implies the invariance of the strong interaction under , e.g. , the exchange of the proton and neutron . experimental evidence for csb has been demonstrated ( before 2002 ) in @xmath1-@xmath2 mixing , the neutron - proton and other hadron mass splittings , the binding - energy difference of mirror nuclei such as @xmath3 and @xmath4 , the different scattering lengths of elastic @xmath5 and @xmath6 scattering , and in the difference between the proton and neutron analyzing powers of elastic @xmath7 scattering .
a review about csb and related issues can be found in ref .
@xcite .
the @xmath8-@xmath9 mass difference can be split into two parts , @xmath10 mev/@xmath11 , where @xmath12 is due to the strong interaction and stems from the up - down quark mass difference , while @xmath13 is due to electromagnetic effects .
the latter is known to be negative and various models gives typically @xmath14 mev/@xmath11 @xcite , but its precise value is not known .
how can we determine these constants ?
the lagrangian of chiral perturbation theory ( @xmath15pt ) contains two terms parametrized by @xmath12 and @xmath13 : @xmath16 the pion terms may give a different combination of @xmath12 and @xmath13 than for @xmath17 , which makes it possible to use reactions involving pions to extract @xmath12 and @xmath13 . a first suitable choice is the csb forward - backward asymmetry of @xmath18 , where originally a negative value @xmath19 was predicted from @xmath20-@xmath21 mixing @xcite .
when the @xmath12 and @xmath13 terms were included the sign changed and @xmath22@xmath23 @xcite .
this latter prediction is of the same sign and magnitude as the subsequent experimental result from triumf : @xmath24\times 10^{-4}$ ] @xcite .
another possiblity is the @xmath0 reaction , which is csb since the deuteron and @xmath25 particle are self - conjugate under cs , while the pion wave function changes sign .
thus this reaction could not occur if cs were conserved .
the cross section is proportional to the square of the cs amplitude , which is unique since all other csb observables involve interferences with cs amplitudes .
a long series of experimental work , culminating with the efforts at the saturne accelerator at saclay , lead only to decreasingly small upper limits @xcite . finally , the saturne group claimed to see @xmath0 at @xmath26 gev @xcite .
this was refuted by members of the same collaboration who argued that the reaction was responsible for the signal @xcite .
the importance of this background was later confirmed by calculation of the double - radiative capture @xcite , using a model based on a very successful treatment of @xmath27 at similar energies @xcite .
thus , the claimed signal is most likely a misinterpretation of a heavily - cut smooth @xmath28 background @xcite .
this realization influenced the design of the iucf experiment , which last year reported a very convincing @xmath0 signal near threshold ( @xmath29 pb at @xmath30 mev and @xmath31 pb at 231.8 mev ) , superimposed on a smooth @xmath28 background @xcite .
this background is roughly a factor two larger than calculations based on ref .
@xcite , but has the expected shape .
the data are consistent with the pion being produced in an @xmath32-wave , as expected from the proximity of the threshold ( @xmath33 mev ) .
it is hoped that these two experimental results , when analyzed in a consistent theoretical framework , can be used to extract the values of @xmath12 and @xmath13 @xcite .
our aim here is to provide the first study of csb in the near threshold @xmath0 reaction using chiral techniques .
previous models of @xmath0 have been at higher energies and have used @xmath20-@xmath21 mixing only @xcite .
we are guided in our quest by general symmetry considerations , relying on chiral power counting to give a list of possible amplitudes , and trusting that realistic wave functions can be constructed for the four - body states .
reaction ( simplified model ) , indicating the labeling of nucleons and defining basic kinematic variables .
the cross indicates the occurrence of csb and the dot a leading - order cs vertex.,width=226 ] the relevant scales are the typical momentum for pion production ( @xmath34 ) ordering the strong diagrams and the fine structure constant ( @xmath35 ) ordering electromagnetic effects . starting with the strong diagrams , at leading order ( lo )
there is only the @xmath36 rescattering diagram of fig .
[ fig : ddapi0 ] , which is driven by @xmath12 and @xmath13 .
at next - to - next - to - leading order ( nnlo ) there are several diagrams as displayed in fig .
[ fig : nnloeps ] . the first ( a ) ( one - body ) term is parametrized by @xmath20-@xmath21 mixing ( resonance saturation ) @xmath37 , unfortunately with large uncertainties in @xmath38 and @xmath39 .
the pion exchanges ( b)(d ) are smaller .
we have not yet included the pion loops ( e)(k ) .
the n@xmath40lo diagram of fig .
[ fig : nnloeps ] are in @xmath15pt given by low - energy constants ( lecs ) that should eventually be fitted to data . since the lecs are not known we chose to model diagram ( b ) by assuming resonance saturation , i.e. , heavy - meson ( @xmath41 , @xmath42 , @xmath2 ) exchanges .
also @xmath42-@xmath2 mixing and hard photon exchanges contribute .
this is very similar to the situation in @xmath43 where short - range mechanisms were also needed @xcite .
the csb parameter is again @xmath44 .
the @xmath45-excitations ( l ) has not been calculated .
the electromagnetic leading order causes csb in the wave functions .
higher orders contain csb pion production .
these amplitudes have not yet been included in our model .
as a first step toward a full model we chose to use simplified ( gaussian ) wave functions , without initial state distortions , to gain insight into the problem @xcite .
the diagram for the simplified model is given in fig .
[ fig : ddapi0 ] .
the deuterons need to be slowed down ( the c.m .
momentum is @xmath46 mev/@xmath47 ) in order to produce a pion and @xmath25 particle at rest .
thus we would like to have momentum sharing between the deuterons so that it does not push the nucleons far out in the high - momentum tail of the @xmath25-particle wave function . with spatially symmetric wave functions the spin - isospin parts are anti - symmetrized : @xmath48_0,\hspace{5mm}|d_{34}\rangle = ( 34)_1[34]_0 \\ |\alpha\rangle & = & \frac{1}{\sqrt{2}}\left\ { ( ( 12)_1(34)_1)_0[[12]_0[34]_0]_0-((12)_0(34)_0)_0[[12]_1[34]_1]_0 \right\}\end{aligned}\ ] ] where @xmath49 ( @xmath50_t$ ] ) are the spin ( isospin ) clebsch - gordan couplings , with magnetic quantum numbers suppressed , for nucleons , or nucleon pairs , @xmath51 and @xmath52 coupling to spin @xmath32 ( isospin @xmath53 ) .
amplitudes easily match @xmath54 with the first part of @xmath55 , while the second term is matched only if both spin and isospin of both nucleon pairs are flipped simultaneously .
thus symmetries decide about matching , but do they support momentum sharing ? the symmetries forbid lo pion exchange between the deuterons and only the nnlo recoil piece @xmath56 survives [ fig . [
fig : nnloeps](c ) ] . thus the lo is strongly suppressed .
the nnlo one - body can not allow for momentum sharing and only its recoil part can contribute to @xmath32-wave production , but it is supported by the symmetries and adds coherently over all nucleons .
it is quite small , of the order 1 pb .
the short range n@xmath57lo amplitudes survive the symmetries , allow for momentum sharing , and add coherently over all nucleons and with each other , dominating the cross section in our simplified model , as demonstrated in table [ tab : xs ] .
the diagrams calculated so far are summarized in table [ tab : xs ] , where a comparison is also made with the plane wave impulse approximation ( pwia ) using realistic wave functions , ( cd - bonn+urbana or av18+urbana ) .
the pion rescattering recovers somewhat , since new symmetries possible with realistic wave functions allow for momentum sharing .
the one - body term increases dramatically since it does not have any momentum sharing and the high-@xmath9 tail of the @xmath25 wave function is no longer a tiny gaussian .
.cross sections evaluated for @xmath0 at the lowest iucf energy , for simplified wave functions ( swf ) and realistic ( pwia ) , either cd - bonn or av18 . the experimental cross section is also given . [ cols="^,^,^,^ " , ] while the simplified model gets quite close to the experimental value , the realistic wave functions , without distortion , overshoot considerably .
there are several reasons why this may happen .
firstly , this can be because the initial state interactions are not included yet . secondly , the ignored diagrams ( e.g. , the pion loops at nnlo ) may cause destructive interferences .
thirdly , since the strength is mainly driven by the one - body term , the resonance saturation procedure may not be accurate ? to test this we turned off the @xmath20-@xmath36 mixing , which gave more reasonable cross sections .
another concern is the difference between the cd - bonn and av18 calculations .
however , our dependence on potential choices can be assessed only when the full model is in place .
we find that , at least in our simplified model , the lo diagrams are suppressed , while subleading amplitudes are important and dominate the cross section .
thus , these short - range terms may be important for @xmath18 and should be evaluated there as well .
we are working on the completion of our model , especially regarding the @xmath58 initial state . polarized @xmath59 data currently being analyzed at iucf will be used to test the scattering wave functions .
the remaining diagrams , the nnlo pion loops , photon two- and three - body terms , and possible @xmath45 ( @xmath9-wave pion ) contributions , will be included .
the dependence on the choice of wave function and potentials can then be tested . with realistic bound state wave functions
the cross section is overestimated , which may reflect our ignorance of the unknown lecs . to address this issue , new experiment are necessary , e.g. , @xmath60 and unpolarized or polarized @xmath0 , with angular distributions and possibly at higher energies . as a @xmath0 experiment
is being planned at cosy , hopefully new exciting data will be available within a few years . | the status of the theoretical effort to calculate charge - symmetry - breaking pion production reactions will be presented
. the main emphasis will be on the @xmath0 reaction , especially in relation to the recent iucf experiment .
suggestions for future theoretical and experimental work will be made .
address = ohio university , athens , ohio , u.s.a .
, [email protected] |
recent studies have started to tackle the formation and collapse of the first cosmic structures ( often referred to as popiii objects ) through numerical simulations ( abel 1998 ; bromm , coppi & larson 1999 ; abel , bryan & norman 2000 ; bromm 2001 ) based on hierarchical scenarios of structure formation .
these studies have shown that gravitational collapse induces fragmentation of pre - galactic units with an initial baryonic mass @xmath12 into smaller clumps with a typical mass of @xmath13 , which corresponds to the jeans mass set by molecular hydrogen cooling .
however , a considerable mass range ( @xmath14 ) for the clumps seems plausible .
tracking the subsequent gravitational collapse of these metal free clumps is a very challenging problem as it requires the simultaneous solution of hydrodynamics and of ( cooling lines ) radiative transfer ( omukai & nishi 1998 ; nakamura & umemura 1999 ; ripamonti 2001 ) .
preliminary attempts and several physical arguments indicate that these clumps do not fragment into smaller units as the evolution progresses to higher densities .
independent studies ( hernandez & ferrara 2001 ) comparing the observed number of metal poor stars with that predicted by cosmological models , also imply that the characteristic stellar mass sharply increases with redshift .
hence , there are grounds to believe that the first stars were very massive .
the evolution of massive , metal free stars is currently subject to a rapidly growing number of studies ( fryer 1999 ; fryer , woosley & heger 2001 ; heger & woosley 2001 ) which rejuvenate earlier activity ( fowler & hoyle 1964 ; carr , bond & arnett 1984 ; el eid 1983 ; fricke 1973 ; fuller 1986 ) .
as we discuss in detail later in the paper , stars more massive than about 260 @xmath15 collapse completely to black holes , therefore not contributing to the metal enrichment of the surrounding gas .
similar arguments apply to stars in a lower mass window ( 30 @xmath15 - 140 @xmath15 ) , which are also expected to end their evolution as black holes . hence ,
if supernovae from more standard progenitors ( stellar masses in the range 8 @xmath15 - 40 @xmath15 ) are either not formed efficiently or occur in negligible numbers , it appears that the initial cosmic metal enrichment had to rely on the heavy element yield from the so - called pair - unstable supernovae ( ) , whose explosion leaves no remnant .
this conclusion can potentially be tested by studying peculiar elemental abundances , for example of heavy _
r-_process elements ( oh 2001 ) .
metallicity is thought to noticeably affect the fragmentation properties of a gravitationally unstable gas .
for example , bromm ( 2001 ) have shown that the evolution of a collapsing proto - galaxy depends strongly on the level of gas pre - enrichment .
these authors argue that a critical metallicity @xmath16 exists such that , above that value vigorous fragmentation into relatively low mass ( @xmath17 ) clumps takes place , differently from what is discussed above under metal - free conditions .
the key question then concerns the interplay between the properties of the imf and the metal enrichment of the gas .
to better illustrate this , let us ideally suppose that the first stars are all formed as suggested by numerical simulations with masses above the mass threshold , thus leading to the formation of black holes . then , because metals are completely swallowed by the latter , the gas retains its primordial composition and star formation continues in the high - mass biased mode .
a solution to this `` star formation conundrum '' must evidently exist as at present stars form with a much lower characteristic mass ( @xmath18 ) . in this paper
, we describe in detail a possible solution to this conundrum utilizing various constraints : metal abundance patterns in clusters , the mass density of super - massive black holes , off - nuclear galactic x - ray sources and , to a more speculative extent , the nature of a particular class ( optically hidden ) of gamma - ray bursts - we attempt to infer the main properties of the early stages of cosmic star formation .
in this section , we focus on the first episode of star formation and discuss the relevant cooling criteria and timescales in detail .
we work within the paradigm of hierarchical cold dark matter ( cdm ) models for structure formation , wherein dark matter halos collapse and the baryons in them condense , cool , and eventually form stars . once a halo of mass @xmath19 collapses at redshift of @xmath20 , the baryons are shock - heated to the virial temperature given by @xmath21 where @xmath22 and @xmath23 is the molecular weight . if @xmath24 [ or equivalently @xmath25 , the baryonic gas cools due to the excitation of hydrogen ly@xmath26 line . in the absence of metals ,
as expected for the very first episode of star formation , objects with @xmath27 can cool only through the collisional excitation of molecular hydrogen .
hereafter , in this work , we refer to the former objects as ly@xmath26-cooling halos and to the latter as popiii objects .
the first stars form once the gas cools .
the typical initial mass function ( imf ) of this first generation of stars is still highly uncertain .
several authors have tackled this crucial issue through theoretical ( rees 1976 ; rees & ostriker 1977 ; silk 1977 , 1983 ; haiman , thoul & loeb 1996 ; uehara 1996 ) and numerical approaches ( omukai & nishi 1998 ; nakamura & umemura 1999 , 2001 ; abel , bryan & norman 2000 ; bromm , coppi & larson 2001 ; omukai 2000 , 2001 ; ripamonti 2001 ) .
recent multi - dimensional simulations of the collapse and fragmentation of primordial gas within popiiis , show preliminary indications that the imf could be either top - heavy with typical masses of order @xmath28 ( abel , bryan & norman 2000 ; bromm , coppi & larson 2001 ) or bimodal with peaks at @xmath29 and @xmath30 ( nakamura & umemura 2001 ) .
note , that these numerical treatments can not address the issue of the mass spectrum .
therefore , it is important to understand what physical processes set the scales of fragment masses and hence the stellar mass spectrum .
it is obvious that much hinges on the physics of cooling , primarily the number of channels available for the gas to cool and the efficiency of the process ( rees & ostriker 1977 ) . in general
, cooling is efficient when the cooling time @xmath31 is much shorter than the free - fall time @xmath32 , i.e. @xmath33 ; where @xmath34 ( @xmath35 ) is the gas number ( mass ) density and @xmath36 is the net radiative cooling rate [ erg @xmath37 .
this efficiency condition implies that the energy deposited by gravitational contraction can not balance the radiative losses ; as a consequence , temperature decreases with increasing density . under such circumstances ,
the cloud cools and then fragments . at
any given time , fragments form on a scale that is small enough to ensure pressure equilibrium at the corresponding temperature , the jeans length scale , r_f _
j c_s t_ff n^/2 -1 [ eq : rf ] where the sound speed @xmath38 , @xmath39 , where @xmath40 is the adiabatic index .
since @xmath41 varies on the cooling timescale , the corresponding @xmath42 becomes smaller as @xmath43 decreases .
similarly , the corresponding fragment mass is the jeans mass , m_f n r_f^ n^/2 + ( 1- ) , [ eq : jm ] with @xmath44 for filaments and @xmath45 for spherical fragments ( spitzer 1978 ) .
this hierarchical fragmentation process comes to an end when cooling becomes inefficient because _ ( i ) _ the critical density for lte is reached or _
( ii ) _ the gas becomes optically thick to cooling radiation ; in both cases , at that juncture @xmath46 . at this stage , the temperature can not decrease any further and it either remains constant ( if energy deposition by gravitational contraction is exactly balanced by radiative losses ) or increases .
the necessary condition to stop fragmentation and start gravitational contraction within each fragment is that the jeans mass does not decrease any further , thus favoring fragmentation into sub - clumps . from eq .
[ eq : jm ] , this implies the following condition : 2 , [ cond ] which translates into @xmath47 for a spherical fragment and @xmath48 for a filament .
thus , a filament is marginally stable and contracts quasi - statically when , @xmath49 , and the gas becomes isothermal .
finally , when @xmath50 or the fragments become optically thick to cooling radiation , the temperature increases as the contraction proceeds adiabatically . in this subsection , we follow the evolution of metal - free primordial clumps during the fragmentation process .
these results are based on the model of omukai ( 2001 ) .
the gas within the dark matter halo is given an initial temperature of @xmath51 and the subsequent thermal and chemical evolution of the gravitationally collapsing cloud is followed numerically until a central proto - stellar core forms .
the gas within the dark matter halo gets shock - heated to the virial temperature @xmath52 , which is typically @xmath53 .
however , after a short transient phase ( irrelevant for the present analysis ) , the evolutionary track in the @xmath54 plane shown in fig . 1 ( top curve , upper panel ) provides a good description of the thermal evolution of the gas .
the metal - free gas is able to cool down to temperatures of a few hundred kelvin regardless of the initial virial temperature .
this is the minimum temperature at which molecular line cooling becomes effective . in this analysis , external sources of heating
are not included ( external uv background or cmb radiation ) .
this is because at redshifts prior to reionization the uv background field is relatively weak and inhomogeneous ( ciardi , ferrara , & abel 2000 ) , hence we do not expect it to be important .
also , the cmb energy input is negligible for clouds with @xmath55 k ( or @xmath56 ) where the first stars presumably formed at redshifts @xmath57 .
it might affect the temperature evolution of more metal rich systems shown in fig . 1 , but does not modify the conclusions drawn here .
the gas within halos with @xmath58 k starts to cool through the hydrogen ly@xmath26 line .
it quickly reaches a temperature of @xmath59 and , at this stage , the fraction of molecular hydrogen formed is sufficient to activate h@xmath60 ro - vibrational line cooling .
thereafter , the gas within a ly@xmath26-cooling halo follows the same thermal evolution as the gas within popiiis .
independent of the virial temperature , the thermal evolution of the gas rapidly converges to the @xmath54 track corresponding to @xmath61 ( the zero metallicity track ) , shown in fig . 1 .
the temperature of the gas decreases with increasing density , thus favoring fragmentation into sub - clumps . as the number density increases
, it reaches the critical value @xmath62 ; and the corresponding jeans mass is @xmath63 ( empty dot in the upper panel of fig . 1 ) .
the cooling time at this critical point becomes comparable to the free fall time as a consequence of the levels being populated according to lte [ regime where the cooling rate @xmath64 and no longer according to nlte [ regime where @xmath65 .
the temperature then starts to rise slowly . at this stage
, the stability of the fragments toward further increase in the density needs to be investigated according to the condition eq .
[ cond ] above .
the lower panel of fig .
1 shows the density dependence of @xmath40 ( see eq .
[ eq : rf ] ) for each metallicity track . for a metal - free gas @xmath40 lies in the range @xmath66 , implying that further fragmentation is unlikely to occur unless the fragments are spherical .
although the gravitational evolution will probably favor a tendency towards spherical symmetry , this is not likely to occur until the central density has reached high ( @xmath67 ) values as seen from simulations ( abel 2000 ) .
the only two deviations from the above range occur ( see lower panel of fig .
1 ) around @xmath68 and @xmath69 . the former is a result of the thermal instability due to three - body h@xmath70 formation ( silk 1983 ; haiman et al . 1996 ) .
however , this instability is quite weak and does not lead to fragmentation ( abel et al . 2000 ) .
the latter ( at @xmath71 ) is caused by h@xmath70 collision - induced emission .
this instability is also weak and probably unimportant .
it is important to stress that , even if the fragments are nearly spherical , fragmentation will be modest and is likely to result only in a low multiplicity stellar system . as the density increases , quasi - static contraction takes place ( @xmath72 ) until the fragments become optically thick to lines , @xmath73 , and adiabatically collapse to increasingly higher central densities and temperatures . at this stage , @xmath74 and a central hydro - static core ( stellar core , filled circle in fig . 1 , upper panel ) is formed , with mass @xmath75 ( omukai & nishi 1998 ) .
we stress again that , as long as the gas is metal - free , these sequence of events and conclusions hold independent of the halo virial temperature , for popiiis as well as for ly@xmath26-cooling halos . in fig .
2 we show the evolution with temperature of the ionization fraction and of the fraction of molecular hydrogen for typical ly@xmath26-cooling halos of mass @xmath76 at three different redshifts , @xmath77 . we find that the evolution of these fractions is independent of the mass ( or , equivalently , @xmath52 ) and virialization redshift .
molecule formation in the post - shock flow which follows the virialization of the gas within a dark matter halo has been recently investigated by uehara & inutsuka ( 2000 ) .
if the gas is fully ionized , molecules form through non - equilibrium recombination , leading to overall fractions that are much higher than in the expanding homogeneous universe . as a consequence of enhanced and hd fractions
, the gas rapidly cools to well below @xmath78 and fragments .
molecular chemistry is important also for star formation in halos with @xmath79 k as assessed by susa ( 1998 ) and oh & haiman ( 2001 ) .
they find that initial atomic line cooling leaves a large residual free electron density which allows molecule formation up to a universal fraction of @xmath80 .
the newly formed molecules cool the gas further to @xmath81 k and the gas fragments on mass scales of a few @xmath82 .
once the critical density for ( @xmath83 or @xmath84 for hd , if this is assumed to be the main coolant ) has been reached , lte conditions for the level populations disfavor further cooling and fragmentation . at this stage , fragments have masses comparable to the jeans mass corresponding to the point @xmath85 and virialize , following the evolution described above .
therefore , independent of the initial virial temperature , fragments are formed with typical masses @xmath86 .
each fragment is characterized by a central core of @xmath87 surrounded by a large envelope of gravitationally unstable gas .
the core grows in mass due to gas accretion from the envelope .
the mass of the formed stars depends on the accretion rate as well as on the fragment mass ( larson 1999 ) . in the absence of metals , radiation pressure can not counter - balance mass accretion onto the core ( ripamonti 2001 , omukai & inutsuka 2001 ) and the mass of the resulting star is comparable to that of the original fragment , @xmath88 .
we will return to this crucial issue in section [ sec : accr ] .
of course , if the parent cloud is rotating , angular momentum , if not dissipated , might prevent the collapse of the entire cloud . however , for the arguments relevant to the present paper , it is sufficient that the overall efficiency is of a few % thus yielding a few hundreds solar mass stars out of the @xmath89 fragments predicted by fig . 1 .
the question that needs to be addressed now is what mechanism can finally lead to the transition to a conventional mode of star formation , to a standard imf , and what are the necessary conditions for this to occur .
we now consider the effects of the presence of heavy elements on the fragmentation process .
the discussion presented in this section is based on the analysis by omukai ( 2000 ) ( see the original paper for details ) .
1 , reproduced from omukai ( 2000 ) , shows the effects of metal enrichment on the @xmath54 tracks for the same initial conditions and different values of the mean metallicity
. in general , clouds with lower metallicity tend to be warmer because of their lower radiative cooling ability .
as long as the clouds are transparent , cooling and fragmentation occur .
clouds with a mean metallicity @xmath90 follow the same evolution as that of the gas with primordial composition in the @xmath54 plane .
however , at @xmath91 , formation on grain surfaces enhances cooling at low density .
dust grains are well known to condense out the ejecta of sne ( see todini & ferrara 2001 and references therein ) . when the lte - nlte transition occurs for , the cloud can still cool ( though less efficiently ) due to oi line cooling . at densities @xmath92 , heating due
to formation becomes larger than compressional work , i.e. @xmath93 and the temperature starts to increase until thermal emission from grains due to energy transfer between gas and dust dominates the cooling .
this occurs at a density @xmath94 , where the temperature drops to @xmath95 and a new fragmentation phase occurs .
this shows up in the lower panel of fig . 1 as the large dip in the @xmath40 evolution .
the minimum fragment mass is reached at the point indicated by the empty circle , when the density is @xmath96 and the corresponding jeans mass is of order @xmath97 . finally , as the density increases
, the gas becomes opaque to dust thermal emission , fragmentation stops and compressional heating causes the fragments to contract adiabatically ( @xmath98 ) .
therefore , a critical metallicity of @xmath99 can be identified , which marks the transition point between metal - free and metal - rich gas evolution .
when the metallicity is @xmath100 , at density @xmath101 cooling is driven by oi , ci and co line emission .
when the nlte - lte transition for the level populations of co occurs , fragmentation stops and the temperature increases due to formation . the larger concentration of dust grains ( assumed here to be proportional to the mean metallicity ) leads to a significant thermal emission which is responsible for cooling the gas and starting a new phase of fragmentation .
this stops when @xmath102 and thereafter the fragments contract quasi - statically until they become optically opaque to dust emission and adiabatic contraction occurs . due to the enhanced ability to cool , fragmentation stops at lower temperatures and densities for higher metallicity clouds ( see open circles in the lower panel of fig . 1 ) .
however , the jeans mass corresponding to the minimum fragmentation scale is always @xmath103 for the metallicity range @xmath104 , several orders of magnitude smaller than for a cloud with no metals . at the onset of adiabatic contraction , when @xmath40 becomes larger than 4/3 , an initial hydro - static core forms ( transient core ) with a mass @xmath105 regardless of the metallicity of the gas .
this transient core is fully molecular and is absent when the gas is metal - free .
the temperature of the transient core increases as its mass increases due to accretion of surrounding gas . eventually , the temperature reaches about @xmath106 , where h@xmath70 dissociation begins .
this softens the equation of state until the dissociation is almost complete .
then , @xmath40 falls below 4/3 in the density range @xmath107 .
note that the thermal evolution after h@xmath70 dissociation ( i.e. , @xmath108 ) is the same independent of the initial composition of the gas .
after @xmath40 once again exceeds 4/3 , a hydro - static core ( so - called stellar core ) forms .
its physical characteristics are independent of metallicity ( @xmath109 ; omukai 2000 ) .
proto - stars , whose mass is initially very low ( about @xmath87 ) , grow in mass by accretion of the envelope material .
the final mass of stars is , then , determined not only by the mass of the fragments , but also by when accretion stops .
the accretion rate onto the proto - star is related to the sound speed ( and therefore , temperature ) of the proto - stellar cloud by the relation , @xmath110 , ( stahler , shu , & taam 1980 ) . as seen in fig . 1
, the temperature of proto - stellar clouds decreases with metallicity .
therefore , the mass accretion rate is higher for proto - stars formed from lower metallicity gas .
if dust is present in the accretion flow , accretion onto the massive proto - star becomes increasingly difficult owing to the radiation pressure onto the dust .
in present - day interstellar gas , accretion onto stars more massive than @xmath111 is inhibited by this mechanism ( wolfire & cassinelli 1987 ) . in gas with lower metallicity , the mass bound is expected to be higher because of the higher accretion rate and lower radiation force . in particular , for metal - free gas , this mechanism does not work .
therefore , without dust , accretion is likely to continue until the ambient gas supply is exhausted ( ripamonti 2001 , omukai & inutsuka 2001 ) . in conclusion ,
the presence of metals not only enables fragmentation down to smaller mass scales @xmath97 , but also breaks the one - to - one correspondence between the mass of the formed star and that of the parent fragment by halting the accretion through radiation force onto the dust .
according to the scenario proposed above , the first stars that form within popiii and ly@xmath26-cooling halos out of gas of primordial composition tend to be very massive , with masses @xmath112 .
it is only when metals change the composition of the gas that further fragmentation occurs producing stars with significantly lower masses .
it is at this stage that we expect a transition to occur from a top - heavy imf towards a more conventional imf with a wide range of masses , such as the one observed locally ( scalo 1998 , kroupa 2001 ) . a growing body of observational evidence points to an early top - heavy imf ( see hernandez & ferrara 2000 and references therein ) .
compelling arguments for an early top - heavy imf can also be made from observations on various scales : _
( i ) _ the early enrichment of our galaxy required to solve the so - called g - dwarf problem , _
( ii ) _ the abundance patterns of metals in the intracluster medium ( icm ) , _ ( iii ) _ the energetics of the icm , _
( iv ) _ the non - detection of metal - free stars , and _ ( v ) _ the over - production of low mass stars at the present epoch and metals at @xmath113 for the submillimeter derived star formation histories using scuba detections for a standard imf can be resolved with an early top - heavy imf .
for instance , the icm metal abundances measured from chandra and xmm spectral data are higher than expected from the enrichment by standard imf sn yields in cluster galaxy members , which can be explained by a top - heavy early imf .
furthermore , the observed abundance anomalies ( oxygen ) in the icm can be explained by an early generation of popiii sne ( loewenstein 2001 ) .
there is observational evidence from the abundance ratio patterns of [ si / fe ] , [ mg / fe ] , [ ca / fe ] and [ ti / fe ] in the extremely metal - poor double - lined spectroscopic binary cs 22876 - 032 in the halo of our galaxy ( norris , beers & ryan 2000 ) for enrichment by a massive , zero - metallicity supernova on comparison with the theoretical models of woosley & weaver ( 1995 ) .
these issues , highly suggestive of top - heavy early star formation , have recently motivated a series of numerical investigations of the nucleosynthesis and final fate of metal - free massive stars ( heger & woosley 2001 ; fryer , woosley & heger 2001 ; umeda & nomoto 2002 ) . in their recent paper , heger & woosley delineate three mass ranges characterized by distinct evolutionary paths : 1 .
@xmath114 : the nuclear energy release from the collapse of stars in this mass range is insufficient to reverse the implosion . the final result is a very massive black hole ( vmbh ) locking up all heavy elements produced .
@xmath115 : the mass regime of the pair - unstable supernovae ( ) .
pre - collapse winds and pulsations result in little mass loss , the star implodes to a maximum temperature that depends on its mass and then explodes , leaving no remnant .
the explosion expels metals into the surrounding ambient ism .
@xmath116 : black hole formation is the most likely outcome , because either a successful outgoing shock fails to occur or the shock is so weak that the fall - back converts the neutron star remnant into a black hole ( fryer 1999 ) . stars that form in the mass ranges ( 1 ) and ( 3 ) above fail to eject most of their heavy elements .
if the first stars have masses in excess of @xmath117 ( in agreement with numerical findings ) , they invariably end their lives as vmbhs ( in the following we will refer to vmbhs as black holes of hundred solar mass or so ) and do not release any of their synthesized heavy elements .
however - as we have shown - as long as the gas remains metal - free , the subsequent generations of stars will continue to be top - heavy .
this _ star formation conundrum _ can be solved only if a fraction of the first generation of massive stars have masses @xmath118 . under such circumstances ,
these will explode as and enrich the gas with heavy elements up to a mean metallicity of @xmath119 , and as per arguments outlined in the previous section ( see fig . 1 )
, thereafter causing a shift over to an imf that is similar to the local one . in what follows , we further explore the implications of the above scenario and derive the conditions for the solution of the conundrum . as a consequence of the picture proposed above , vmbhs
are an inevitable outcome .
we now compute the expected mass density of metals and the mass density of remnant vmbhs produced in such a first episode of star formation . for a collapsed dark matter halo of total mass @xmath19
, the associated baryonic mass is assumed to be @xmath120 ( where @xmath121 is simply the baryon fraction ) . following the results of numerical simulations at different resolutions ( bromm 2001 , abel 2000 ) we assume that @xmath122 of the total available gas is utilized in star formation , the rest remaining in diffuse form .
the relative mass fraction of and vmbh progenitors are parametrized as follows : @xmath123 where @xmath124 ( @xmath125 ) is the total mass which ends up in ( vmbhs ) and @xmath126 is the mass processed into stars .
thus , only a fraction @xmath127 of the formed stars can contribute to gas metal enrichment .
the metal yields for the dominant elements have been computed using the results of heger & woosley ( 2001 ) , and are plotted in fig . 3 as a function of the mass of the progenitor star .
the bulk of the yield is contributed by o@xmath128 and the mass of metals ejected can be written as m_z m_/2 - 10 , where we have taken @xmath129 as a fiducial mass for progenitors .
next , the further assumption is made that metals are ejected from the parent galaxy into the igm and their cosmic volume filling factor is close to unity , therefore uniformly polluting the igm .
this comes from the results of madau , ferrara & rees ( mfr , 2001 ) and mori , ferrara & madau ( 2001 ) , where it is shown that for reasonable values of the star formation efficiency , metal bubbles produced by ( standard snii in ) proto - galaxies which result from 2-@xmath130 peaks of the density power spectrum at redshift 10 do overlap .
indeed , the kinetic energy released by is much higher than for ordinary snii ( see fig .
2 ) , hence making our assumption even more solid .
the temperature of the hot gas is expected to be somewhat higher than estimated by mfr : however , due to the higher redshifts and consequent stronger inverse compton cooling , ejected metals have enough time to cool by @xmath131 .
as long as the average metallicity is below the critical value , @xmath132 , we argue that the imf remains top - heavy and the redshift - dependent critical density of metals contributed by all at redshifts @xmath133 can be computed , _ z ( z ) = _ z^dz_m_min(z ) + dmn(m , z)m_z where @xmath134 is the number density of halos per unit mass predicted by the press - schechter formalism and the integration is performed from @xmath135 which is the minimum mass that can cool within a hubble time at the specified formation redshift @xmath136 , @xmath137 ( see ciardi 2000 ) .
we adopt a cosmological model with the following parameters to compute the abundance of halos : @xmath138 , @xmath139 , @xmath140 , @xmath141 ( latest predictions from bbn , see burles , nollett & turner 2001 ) and use the cobe - normalized power spectrum for fluctuations as described in efstathiou , bond & white ( 1992 ) . from the above expression
, we can estimate the transition redshift @xmath142 at which the mean metallicity is @xmath143 , = 10 ^ -4 z _ for various values of the fraction @xmath127 .
the results are plotted in fig . 4 alongwith the corresponding critical density @xmath144 contributed by the vmbhs formed , which is given by , _
vmbh(z_f ) = _
z_f^ dz_m_min(z ) dmn(m , z)m_*. several assumptions as to the fate of these vmbhs at late times will be examined in an attempt to obtain constraints on the value of @xmath127 from current observations in the next section .
the objective now is to set some limits on @xmath127 , the fraction of stars formed that result in providing the first metals , that are eventually responsible for the shift to a normal imf . in order to compare the predicted critical density of vmbh remnants to present observational data we need to make further assumptions about the fate of these vmbhs at late times .
the mass density contributed by remnant supermassive black holes @xmath145 can be estimated observationally denotes the present day value ] from the demography of local galaxies ( magorrian et al . 1998 ; merritt & ferrarese 2001 ) .
there are two extreme possibilities ( bounding cases for the values of @xmath127 ) concerning the relation between the inferred mass density of smbhs and vmbhs : [ a ] @xmath146 : all vmbhs have , during the course of galaxy mergers , in fact , been used to build up the smbhs detected today ; [ b ] @xmath147 there is no relation between the detected smbhs and the early vmbhs , implying that smbhs have grown primarily by gas accretion .
thus , in the first scenario [ a ] , we assume that all vmbhs are able to sink to the center of host systems to merge within a hubble time , giving birth to the smbhs in the nuclei of galaxies that we see today . note , that this is in fact unlikely ( as explained further below ) , nevertheless , this argument provides an extreme bound .
it is important to point out here , that in the current proposed theoretical models for the mass accretion history of black holes , the mass density seen today locally in smbhs can be built either predominantly by mergers or by accretion . in phenomenological models that attempt to tie in the current high redshift observations of the space density and luminosity function of quasars with that of the local space density of black holes
wherein it is assumed that quasars are _ ( i ) _ powered by bhs and _ ( ii ) _ are optically bright for a period of @xmath148 yr ( haiman & loeb 1998 ; haehnelt , natarajan & rees 1998 ; kauffmann & haehnelt 2000 ) , either picture ( mergers / accretion ) can adequately explain the mass assembly of the remnant smbhs .
this permits both scenarios [ a ] ( wherein smbhs grow primarily via mergers of vmbhs during galaxy assembly ) and [ b ] ( wherein smbhs grow primarily by accretion and @xmath149 is unrelated to @xmath144 ) to be plausible extreme cases .
consideration of cases [ a ] and [ b ] will provide bounds on the value for @xmath127 . for a given @xmath127 ,
the corresponding value for the transition redshift can be found on the @xmath142 curve shown in fig 4 .
we note here that vmbhs in the mass range expected from the first episode of metal - free star formation ( a few hundred solar masses ) are likely to be ubiquitous . using iso observations , gilmore & unavane ( 1998 )
have convincingly argued that low mass stars ( in fact hydrogen burning stars of any mass ) do not provide a substantial portion of the dark matter inferred in the halos of galaxies . in combination with the mass limits obtained on potential dark matter candidates from the macho ( alcock 2001 ) and eros ( lasserre 2000 ) experiments ( micro - lensing studies ) in our galaxy ,
low mass objects , with masses under @xmath150 are ruled out - thereby , making vmbhs ( @xmath3 bhs ) in halos ( case [ b ] ) plausible dark matter candidates . for case
[ a ] , we can simply equate @xmath151 to the measured mass density in smbhs found from the demography of nuclei of nearby galaxies ( magorrian 1998 ; gebhardt 2001 ) indicated by the horizontal line in fig .
4 ( top panel ) .
the most recent value reported by merritt & ferrarese ( 2001 ) is @xmath152 , which for @xmath153 and @xmath154 yields @xmath155 . from the top panel of fig . 4
, we see that this gives @xmath156 .
the corresponding value for the transition redshift can be found on the @xmath142 curve on the same plot , yielding a value @xmath157 .
this constraint implies that all but a small fraction of the mass involved in the first episodes of star formation - approximately 6% - went in objects outside the mass range @xmath158 , yielding remnant black holes with mass @xmath159 .
the scenario corresponding to [ a ] therefore implies that @xmath142 , the transition redshift from top - heavy to normal imf occurred at @xmath160 18.5 .
let us now look at the other extreme case [ b ] , wherein the assembled smbhs in galactic centers have formed primarily via accretion and are unrelated to vmbhs .
this leads us to two further possibilities : ( i ) vmbhs would still be in the process of spiralling into the centers of galaxies due to dynamical friction but are unlikely to have reached the centers within a hubble time due to the dynamical friction time - scale being long ( madau & rees 2001 ) , or ( ii ) vmbhs contribute the entire baryonic dark matter in galactic halos .
pursuing scenario ( i ) , therefore , at best some fraction of these vmbhs ( for instance , those in gas - rich regions of gas - rich systems ) might appear as off - center accreting sources that show up in the hard x - ray wave - band .
such sources have indeed been detected both by rosat ( roberts & warwick 2000 ) and more recently by chandra in nearby galaxies ( m84 : jones 2001 ; ngc 720 : buote 2001 ) . in a survey of archival rosat hri data to study the x - ray properties of the nuclei of 486
optically selected bright nearby galaxies ( ho , fillipenko & sargent 1995 ) , roberts & warwick ( 2000 ) found a large number of off - center x - ray sources .
the x - ray sources detected within the optical extent of these galaxies were classified either as nuclear or non - nuclear ( and therefore off - center ) depending on whether the source was positioned within 25 arcsec of the optical nucleus .
they detect a nuclear source in over 70% of the galaxy sample and a total of a 142 off - center sources .
roberts & warwick ( 2000 ) find that the non - nuclear sources follow a steep , near power - law x - ray luminosity distribution in the @xmath161 which leads to an @xmath162 ratio of = 1.1 10 ^ 39 ^ -1 ( 10 ^ 10 _ ) ^-1 .
[ eq : averagelx ] the median luminosity of the non - nuclear sources is found to be a factor of @xmath163 lower than that of the nuclear sources .
however , they estimate the incidence rate of off - center sources with @xmath164 ( which corresponds to the eddington luminosity for a 1.4 solar mass neutron star ) to be @xmath165 galaxy . the existence of accreting vmbhs might also help explain the following observation . the far - ir excess detected with dirbe at 60 and 100 @xmath23 m has been tentatively interpreted as an extra - galactic background with integrated intensity of @xmath166 nw @xmath167 in the range @xmath168 m ( finkbeiner , davis & schlegel 2000 ) . the energy required to produce such a fir background could derive from a highly obscured population of accreting vmbhs at moderate redshifts ( for an alternative to this see explanation of scenario ( ii ) below ) . in order to estimate the mass density contributed by these vmbhs
, we can compare the mass in vmbhs to that of the smbh in a typical galaxy of luminosity say @xmath169 for case [ b ] .
most of the vmbhs are likely to be en - route to galactic centers .
note however , that not all the inspiralling vmbhs will be accreting and be x - ray bright : many of these could , in fact , be low radiative efficiency adafs ( advection dominated accretion flows ) , and therefore be too faint to be detected by rosat or chandra or not accreting at all if they do not happen to be in gas - rich regions of the galaxy .
thus , using the following argument we can obtain a lower limit on the abundance of vmbhs .
the mass of the central smbh hosted by a galaxy with a luminosity @xmath170 ( assumed to be the fiducial galaxy for these purposes ) in the b - band is given by ( merritt & ferrarese 2001 ) : @xmath171 where the mass - to - light ratio in the b - band is taken to be @xmath172 and the galaxy luminosity is essentially dominated by the bulge luminosity . hence , the mass of the central smbh is @xmath173 .
now , we use the luminosity of the rosat off - center sources ( roberts & warwick 2000 ) and use the fact that they find @xmath160 0.7 off - center sources with luminosity @xmath174 per @xmath169 galaxy , to estimate the mass of vmbhs in the fiducial galaxy . if we assume that these vmbhs have a typical mass of @xmath175 , their eddington luminosity is , @xmath176 thus , vmbhs appear to be radiating at sub - eddington luminosities , the average rate ( given by eq .
[ eq : averagelx ] ) being @xmath177 of the eddington value .
this is consistent with the observed luminosities of the nuclear sources in the sample , which roberts & warwick ( 2000 ) find to be radiating at severely sub - eddington rates .
therefore , we expect that the typical accreting vmbh mass in a @xmath169 galaxy to be about 210@xmath178 .
little is known about the spatial distribution of such objects .
how can we take into account the fact that only a fraction of the sources are likely to be accreting ?
since the detected rosat off - center sources are within the optical radius of the galaxies , the number of non - accreting vmbhs in the halos of these galaxies can be large .
one can obtain a simple estimate of this number by pursuing the following argument .
assume that vmbhs , being collisionless particles , closely trace the dark matter ( which we take to obey a navarro , frenk & white [ nfw , 1997 ] profile ) and that the ratio of virial to optical radius is @xmath179 for a @xmath169 disk galaxy ( persic , salucci & stel , 1996 ) .
these hypotheses require that we scale up the total mass in vmbhs by a factor of @xmath180 , which gives @xmath181 .
now , the ratio @xmath182 is @xmath183 , implying that @xmath184 which in turn gives @xmath185 ( see dashed line in bottom panel of fig . 4 ) and @xmath186 .
now we explore scenario ( ii ) of case [ b ] , the instance when vmbhs constitute the entire baryonic dark matter content of galaxy halos ( but do not contribute to the disk dark matter ) .
it is important to point out here that cosmological nucleosynthesis arguments require both baryonic and non - baryonic dark matter ( pagel 1990 ) .
essentially , this is due to the fact that the mass density contributed by baryons @xmath187 , is well in excess of @xmath188 , the contribution to the mass density by visible baryons . using the luminosity dependent relation of visible to dark matter for spirals ( persic , salucci & stel , 1996 ) for a fiducial galaxy of @xmath189 we find , @xmath190 where @xmath191 is the visible mass and @xmath192 is the total dark matter mass , given as a sum of a baryonic and a non - baryonic component . the total baryonic mass is a fraction @xmath121 of the total mass of the system , @xmath193 from the two equations above and assuming a nfw density profile , we estimate that the total baryonic dark matter content as , @xmath194 moreover , 90% of this mass resides outside the optical radius and can be contributed by vmbhs : = = 2.1 10 ^ 3 , implying _
vmbh = 2.110 ^ 3 _ smbh = 6.4 10 ^ -3 . incorporating this constraint into fig .
4 ( bottom panel ) , we find @xmath195 and @xmath196 .
note that this scenario does not violate the constraint on the over - production of background light ( see carr 1998 ; bond , carr & hogan 1991 ; wright et al .
in fact , some fraction of the observed near - ir dirbe excess could be produced by vmbhs . in a recent estimate of the cosmic background at 1.25 @xmath23 m and 2.2 @xmath23 m ( corresponding to the j and k bands respectively ) using 2mass and the dirbe results cambresy et al .
( 2001 ) , also find an excess ( significantly higher than the integrated galaxy counts in the j and k bands ) suggesting the contribution of other sources .
pop iii stars and their vmbh remnants ( accreting at very high redshifts ) postulated here are likely candidates . according to scenario [ b ] ,
the average metal abundance at redshift @xmath197 should be @xmath132 , marking the transition from a top - heavy imf to a standard power - law imf .
this does not necessarily violate the observed metal abundances in damped ly@xmath26 systems , @xmath198 , as it is yet not clear to what extent the igm metallicity is spatially uniform at these intermediate redshifts .
therefore , scenario [ b ] is poorly constrained by the data , as the two options ( i ) and ( ii ) give large ranges for @xmath127 and @xmath142 : @xmath199 and @xmath200 .
consistently , scenario [ a ] gives a limit on @xmath127 and @xmath142 of @xmath201 and @xmath202 .
the actual data do not allow us at present to strongly constrain the above two quantities , but they provide interesting bounds on the proposed scenario .
these ranges also have implications for the expected detection rate of sne beyond @xmath142 with future instruments like ngst ( marri , ferrara & pozzetti 2000 ) .
we have proposed a scenario to solve a puzzling _ star formation conundrum _ : the first stars are now thought to be very massive and hence to lock their nucleosynthesis products into a remnant ( very massive ) black hole .
this high - mass biased star formation mode continues as long as the gas remains metal - free . during this phase
, metal enrichment can occur only if a fraction @xmath203 of the stars have mass in the window leading to pair - unstable sne ( @xmath204 ) which disperse their heavy elements into the surrounding gas .
such metals enrich the gas up to @xmath205 , when a transition to efficient cooling - driven fragmentation producing @xmath2 clumps occurs at redshift @xmath142 .
we argue that the remaining fraction of the first stars end up in @xmath206 vmbhs . by analyzing the evolutionary fate of such objects
, we argue that [ a ] they could end up in the smbhs in the centers of galactic nuclei , [ b ] ( i ) could be en - route to the center and hence identified with the x - ray bright , off - center rosat sources , or ( ii ) constitute the entire baryonic dark matter content of galaxy halos .
these possibilities are used to obtain constraints on the two quantities : @xmath207 and @xmath208 for [ a ] , and @xmath209 $ ] and @xmath210 $ ] for case [ b ] .
the value @xmath211 found here is admittedly somewhat uncertain .
for this reason we have investigated how the above results might be affected by a different choice , assuming @xmath212 . indeed ,
lower values of @xmath213 imply that less efficient metal enrichment is required in order to change to a more conventional star formation mode .
thus , comparable values for @xmath142 are found but the corresponding values for @xmath203 tend to be systematically smaller , being @xmath214 for scenario [ a ] and in the range @xmath215 for scenario [ b ] .
several uncertainties remain in the comparison of the inferred density of vmbhs to local observations .
for example , it is not obvious if smbhs could at all form out of vmbhs .
dynamical friction can effectively drag vmbhs towards the center of the host system , at least within a distance of @xmath81 pc ( madau & rees 2001 ) . unless most of the energy is radiated away in gravitational waves
, it could be difficult for a vmbh cluster to coalesce into a single unit .
furthermore , accretion onto isolated vmbhs could be too inefficient to explain most of the off - center sources observed by rosat and chandra .
higher accretion rates might be activated by the tidal capture / disruption of ordinary stars .
question remains if the frequency of such an event is in fact sufficiently high to explain the data . in spite of the many difficulties and uncertainties discussed above , our study represents a first attempt to link the first episode of cosmic star formation activity to present day observational evidence of their fossil remnants . a top - heavy imf for the early episodes of star formation in the universe might have other interesting observational consequences
, we speculate further on them below . the kinetic energy released during
the thermonuclear explosions powered by pair instability are @xmath216 larger than those of ordinary type ii sne .
this might cause the interaction with the circumstellar medium to be as strong as predicted for hypernovae ( woosley & weaver 1982 ) .
however , these explosions do not lead to the ejection of strongly relativistic matter ( fryer , woosley & heger 2001 ) and therefore can not power a gamma - ray burst ( grb ) . in popiii progenitors of vmbhs ,
the estimated angular momentum is sufficient to delay black hole formation and the system might develop triaxial deformations ( fryer , woosley & heger 2001 ) . if the instabilities have enough time to grow , the core might break into smaller fragments that would then collapse and merge to form the central vmbh . if not , the star might still develop a bar - like configuration .
both these scenarios lead to a significant emission of gravitational waves ( schneider 2000 ; fryer , holz & hughes 2001 ) .
furthermore , significant emission of gravitational radiation can occur as a result of the in - spiral and merger of vmbhs onto the smbhs in the center of host systems ( madau & rees , 2001 ) once the vmbhs have formed , accretion continues through a disk at a rate which can be as large as @xmath217 ( fryer , woosley & heger 2001 ) .
magnetic fields might drive an energetic jet which can produce a strong gamma - ray burst through the interaction with surrounding gas .
the properties of these popiii grbs would be considerably different from their more recent ( @xmath218 ) counterparts : depending on the uncertain interaction of the jet with the surrounding matter , the bursts would be probably longer [ @xmath219 s ] and the peak of emission , that in the rest frame is in @xmath40-rays , would be shifted into x - rays .
indeed , bepposax has revealed the existence of a new class of events , the so - called x - ray flashes or x - ray rich grbs , which emit the bulk of their energy in x - rays ( piro , private communication ) .
furthermore , since popiii grbs explode at very high redshifts , it is likely that their optical afterglow might be heavily absorbed by the intervening gas .
these systems might be the natural candidates for the significant number ( about 40 % of grbs for which fast follow - up observations were carried out ) of grbs which do not show an optical counterpart , the so - called ghost ( grb hiding optical source transient ) or dark grbs .
other explanations ascribe the failed optical detection to dust extinction within the host system but the ultimate nature of ghosts is still very debated ( lazzati , covino & ghisellini , 2001 ; djorgovski 2001 ) finally , the energetic jets generated by grbs engines produce , by photon - meson interaction , a burst of tev neutrinos while propagating in the stellar envelope ( mszros & waxman 2001 ) .
we investigate this aspect in a companion paper ( schneider , guetta & ferrara 2002 ) where we use the constraints set by the amanda - b10 experiment on the total integrated flux of tev neutrinos from popiii grbs .
we are grateful to p. coppi , k. nomoto & m. rees for comments and suggestions .
we wish to thank v. bromm and l. piro for discussions .
we thank the referee for his / her thoughtful comments . this work is partially supported ( rs ) by the italian cnaa ( project 16/a ) ; ko is supported by research fellowships of the japan society for the promotion of science for young scientists , grant 6819 . | recent studies suggest that the initial mass function ( imf ) of the first stars ( popiii ) was likely to be extremely top - heavy , unlike what is observed at present .
we propose a scenario to generate fragmentation to lower masses once the first massive stars have formed and derive constraints on the primordial imf .
we estimate the mass fraction of pair - unstable supernovae ( sn@xmath0 ) , shown to be the dominant sources of the first heavy elements . these metals enrich the surrounding gas up to @xmath1 , when a transition to efficient cooling - driven fragmentation producing @xmath2 clumps occurs .
we argue that the remaining fraction of the first stars ends up in @xmath3 vmbhs ( very massive black holes ) .
if we further assume that all these vmbhs are likely to end up in the centers of galactic nuclei constituting the observed smbhs , then @xmath4 of the first stars contributed to the initial metal enrichment and the imf remained top - heavy down to a redshift @xmath5 18 .
5 . at the other extreme
if none of these vmbhs have as yet ended up in smbhs , we expect them to be either ( i ) en - route towards galactic nuclei thereby accounting for the x - ray bright off - center sources detected locally by rosat or ( ii ) as the dark matter candidate composing the entire baryonic halos of galaxies . for case ( i )
we expect all but a negligible fraction of the primordial stars to produce metals causing the transition at the maximum possible redshift of @xmath6 , and for case ( ii ) @xmath7 - a very negligible fraction of the initial stars produce the metals and the transition redshift occurs at @xmath8 . in this paper
we present a framework ( albeit one that is not stringently constrained at present ) that relates the first episode of star formation to the fate of their remnants at late times .
clearly , further progress in understanding the formation and fragmentation of popiii stars within the cosmological context will provide tighter constraints in the future .
we conclude with a discussion of several hitherto unexplored implications of a high - mass dominated star formation mode in the early universe .
# 1@xmath9 # 1@xmath10 # 1equation ( [ # 1 ] ) # 1equation ( [ # 1 ] ) # 1 ( [ # 1 ] ) # 110^#1 # 1n _ # 1 # 1k_#1 2ph@xmath11 2ph_2^+ psfig |
pioneered by son @xcite , and balasubramanian and mcgreevy @xcite , the developments in the galilean holography have been seen lately .
they are aimed to be the non - relativistic generalizations of the ads / cft correspondence @xcite . the generators of the galilean algebra correspond to the temporal and spatial translations , rotational transformations , as well as to the galilean boost and mass operators .
one can extend the algebra by including the dilatation operator , and for the special case where the powers of the time coordinate scale twice as much as the spatial coordinates , the full non - relativistic conformal symmetry , that includes the special conformal transformation , can be obtained .
the non - relativistic conformal symmetry is also known as the schrdinger algebra @xcite .
the authors of references @xcite have discovered the @xmath0 dimensional spacetime geometry whose isometry group is the schrdinger symmetry of @xmath1 spatial dimensions , and based on the ads / cft correspondence , they proposed that the gravitational system of this schrdinger geometry is the holographic dual of the non - relativistic conformal field theories at strong couplings .
the expectation is that the duality would provide the important information about the strongly coupled real world non - relativistic systems ( see reference @xcite for the examples of such systems ) , as it has been the case for the ads / cft correspondence . among the subsequent developments ,
the finite temperature generalization of the galilean holography was discussed in references @xcite . in parallel to the finite temperature ads / cft correspondence @xcite , they have proposed a planar black solution , a black brane , in the schrdinger space as the holographic dual of the non - relativistic cft at finite temperature .
the authors embed the holographic set up in string theory .
references @xcite start with the near horizon geometry of d3-branes which is the ads@xmath2 black brane times @xmath3 . then they apply a solution generating procedure known as the null melvin twist @xcite to this system , and upon the kk - reduction of @xmath3 , they find that the resulting geometry is a black brane solution whose extremal and asymptotic limits reduce to the schrdinger geometry . of the resulting geometry
is @xmath4 .
the dynamical exponent @xmath5 determines the relative scaling of the time coordinate and others . for @xmath6 ,
the powers of the scaling for the time is twice that of translationally invariant coordinates and as noted , this is the special case where the galilean symmetry extends to the schrdinger . ]
notice that by construction , this procedure yields the planar black solutions .
the thermodynamic and transport properties were discussed in the references and references @xcite noticed the similarity to the usual schwarzschild - ads black branes . just as for the asymptotically ads case , the planar black solution does not exhibit the hawking - page phase transition @xcite .
the phase transition , in the ads case , is observed for the black holes , as opposed to the black branes , and the phase structure of the black holes tends to be richer than the planar solutions .
moreover , in the context of the ads / cft correspondence , the dual field theory is supposed to go through the confinement / deconfinement phase transition , corresponding to the gravitational hawking - page phase transition @xcite .
therefore , it is natural to search for the black hole solution in the schrdinger space and this is the topic of this paper .
our approach to this problem is not to embed the story in string theory but to directly deal with the @xmath7-dimensional action whose solution is supposed to be dual to the @xmath1 spatial dimensional non - relativistic conformal field theory .
( to be concrete , we will concentrate on the case with @xmath8 . )
such an action has been proposed by son @xcite and by the authors of references @xcite .
we seek for the black hole solution to this action and we indeed find one by mimicking the relationship between the black brane and black hole in ads . to examine the thermodynamics of a given black solution , we need to obtain the value of the action and the thermodynamic quantities . because of the asymptotic structure of the schrdinger space and the volume integral which must be carried out , the on - shell value of the action diverges and we must somehow regularize this . for this purpose ,
reference @xcite proposed boundary terms , similarly to the counterterm technique of asymptotically ads systems .
this proposal has a few problems .
one problem is that the resulting finite action seems to depend on arbitrary coefficients of the boundary terms introduced . in the reference ,
some minimal number of the boundary terms were considered , however , there are more terms that can contribute to the finite action [ _ e.g. _ , @xmath9 . this ambiguity may be eliminated by requiring the first law to be satisfied for the resulting thermodynamic quantities . however , given the action with the counterterms , it is natural to compute the thermodynamic quantities through the definition proposed by brown and york @xcite .
( reference @xcite computes thermodynamic quantities differently , as we will do in section [ subsec : bbthermo ] . )
this attempt fails unless one chooses unjustifiably awkward normalization for the time - like boundary killing vector field .
the most important problem for this paper is that the similar counterterm technique miserably fails for our black hole solution .
the counterterms in this case are not in the form to cancel the divergences coming from the bulk action and the gibbons - hawking surface term , mainly because of the nonzero scalar field ( dilaton ) at infinity .
another way to obtain the thermodynamic quantities was proposed in reference @xcite .
the idea was to transform the metric to the asymptotically ads form by using the symmetry of the system .
( the symmetry actually is that of the eight dimensional supergravity .
see the appendix of the reference for the details . )
then compute the relevant quantities by utilizing the established method @xcite for the ads case .
this procedure , however , does not directly apply for the black hole solution , mainly because the horizon of the black hole does not possess the spatial translation invariance .
( however , see the discussion in section [ sec : discussions ] where we argue that there should be a similar procedure that is applicable for the black hole . )
we therefore propose a new regularization method that is apt for all the black solutions discussed in this paper .
this actually is the oldest regularization technique proposed by gibbons , hawking and page @xcite , namely , the subtraction method , but with an unusual boundary matching .
this method proceeds as follows .
one needs to evaluate the action ( with gibbons - hawking surface term ) on the spaces with and without the black hole but due to the volume integral , each action is divergent . to regulate the divergences ,
one put the systems in a large box and the metrics on the wall of the box ( the first fundamental forms ) must be matched .
then one subtracts the regulated action from the other and the wall of the box is removed to infinity . in the schwarzschild - ads black hole case the difference is finite and the sign of the difference action determines the thermodynamically preferred solution ( that is , it is suitable in searching for the hawking - page phase transition ) . in our geometry , we find a problem for this procedure : the boundary matching can not be achieved because of the peculiar nature of one of the coordinates .
therefore , in this paper , we propose a kind of `` partial matching '' of the first fundamental forms where we scale the coordinates on the wall to match the metrics , except for the special coordinate . in section [ sec : bb ]
, the procedure is first applied to the known black brane case and checked to reproduce the thermodynamics of references @xcite .
it is noteworthy that reference @xcite and we find the finite action be the same as that of the schwarzschild - ads black brane .
next in section [ sec : bh ] , the black hole solution is presented and the same procedure is applied to this solution . remarkably , the resulting action turns out to be the same as the one for the schwarzschild - ads black hole .
the thermodynamic quantities resulting from the action are computed and the phase diagram of the system is obtained . among other structures in the phase diagram , we find the hawking - page phase transition . section [ sec : discussions ] is dedicated to the discussions .
the appendix presents the black hyperboloid solution and examines its thermodynamic properties .
in this section , we mainly review the results obtained in references @xcite .
we , however , propose a new way of computing the finite on - shell action in section [ subsec : bbdiffact ] .
the same method will be applied to the black hole solution in the next section and this section serves to show the validity of the method . in references
@xcite , the schrdinger geometry discovered in references @xcite was embedded in the frame work of string theory .
references @xcite achieve this by applying the null melvin twist procedure @xcite to the d3-brane geometry .
let us start with the non - extremal d3-brane geometry in the near horizon region ( schwarzschild - ads black brane ) : @xmath10 where @xmath11 is the ads scale , @xmath12 , @xmath13 , and @xmath14 is the line element of the unit five sphere .
the location @xmath15 is where the horizon is and notice that setting @xmath16 reduces the metric to the extremal case .
the fields @xmath17 and @xmath18 are the dilaton and nsns two - form , respectively .
we have the rr four - form potential as well but it does not play an important role in our discussion and hence omitted . for convenience , references @xcite adopt the @xmath19 hopf fibration over @xmath20 to write @xmath21 where @xmath22 is the one - form in the hopf fiber direction and
@xmath23 is the metric on @xmath20 ( see appendix a of reference @xcite for a little more details ) .
as always , @xmath24 is understood to be the symmetric tensor product .
( when we intend to mean alternating projections , we write wedges explicitly . )
the coordinate @xmath25 is arbitrarily singled out from the three translationally invariant spatial directions ( the other directions are denoted by @xmath26 ) and it is the direction that plays a special role in the null melvin twist procedure .
the procedure yields the system , @xmath27 \nonumber\\ & + \bigg(\frac{r}{r}\bigg)^2 h^{-1}dr^2 + k^{-1}r^2\eta^2 + r^2ds_{\mathbb{p}^2}^2 \;,\nonumber\\ \phi = & -\frac{1}{2}\ln k \;,\nonumber\\ b = & k^{-1}\bigg(\frac{r}{r}\bigg)^2 b\ , ( \,h\ , dt + dy)\wedge\eta \;,\end{aligned}\ ] ] where @xmath28 and the parameter @xmath29 has the dimension of inverse length and is related to the twist factor given in @xmath22-direction ( see references @xcite ) .
the kk reduction of the five - sphere is a consistent truncation @xcite and this provides a five dimensional system . in doing so , following herzog _
@xcite , we introduce the light - cone coordinates @xmath30 the particular choice of the normalization will be explained shortly .
we then have the five dimensional system @xmath31 + k^{1/3 } \bigg(\frac{r}{r}\bigg)^2 h^{-1}dr^2 \;,\nonumber\\ \phi = & -\frac{1}{2}\ln k \;,\nonumber\\ a = & k^{-1}\bigg(\frac{r}{r}\bigg)^2 b \ , \bigg\ { \frac{h+1}{2br } dx^+ + br(h-1 ) dx^- \bigg\ } \;,\end{aligned}\ ] ] where we changed to the einstein frame and renamed the kk - reduced two - form field @xmath18 to @xmath32 which is a one - form field . [ the reader is warned for the slightly confusing notations here : @xmath33 and @xmath34 . ]
as suggested by son @xcite , @xmath35 is reinterpreted as the time coordinate and @xmath36 is assumed to be a compactified direction to yield discrete mass spectrum of the system .
the extremal case is given by the value @xmath37 and the non - extremal case approaches this at asymptotically large @xmath38 .
notice that the normalization of the light - cone coordinates in equations ( [ eq : lc ] ) is designed to yield the extremal system which is free of the parameter @xmath29 .
this means that the parameter is not physical in the extremal case and this fact makes the normalization of the light - cone coordinates ( [ eq : lc ] ) unique , even for the non - extremal case because that should be fixed by the boundary metric .
+ the five dimensional system ( [ eq : k0nonext ] ) can be obtained from the equations of motion that follow from a five dimensional action .
such an action was originally proposed by son @xcite for the extremal system and references @xcite find the action that also supports the non - extremal system : @xmath39 \;,\ ] ] where @xmath40 is the five dimensional newton constant , @xmath41 is the determinant of the five dimensional metric , @xmath42 is the scalar curvature , @xmath43 and the potential @xmath44 is defined to be @xmath45 son s original action is recovered by setting @xmath17-field to zero .
+ we wrap up this subsection by briefly discussing the geometric properties of the metric , especially the causal development of it . because of the nonzero @xmath46 component
, the geometry is stationary but not static .
we interpret this as the rotating black brane in the compactified @xmath36 direction .
let us rewrite the metric in the adm form ; @xmath47 + k^{1/3 } \bigg(\frac{r}{r}\bigg)^2 h^{-1 } dr^2 \;.\end{aligned}\ ] ] from this form , we can pick up some information about this geometry .
first , recall that @xmath48 , so we clearly see the existence of the ergo - region , as typical for a rotating black hole .
second , the angular velocity of the horizon @xmath49 measured in the units of @xmath36-circumference is given by @xmath50 and the coordinate angular velocity diverges at the boundary where @xmath51 . in the usual notion of geometrodynamics
, @xmath35 is supposed to describe the cauchy development of the space - like four - surface with the metric @xmath52 this poses a problem for the extremal case with @xmath37 , because the @xmath53 component is degenerate .
the surface of constant @xmath35 is not space - like and the four - surface has zero measure .
this fact stems from the light - like nature of the coordinate @xmath36 before we reinterpret @xmath35 as the time coordinate .
therefore , we see that @xmath35 is not an appropriate time function in the five dimensional spacetime point of view .
one can , actually , see that this is not just due to the particular choice of the time coordinate .
for example , if we use time @xmath54 as in ( [ eq : k0tybasis ] ) , we see that the lapse function squared always becomes negative at sufficiently large @xmath38 .
the fact is that this spacetime is not globally hyperbolic , just like ads and pp - wave spacetimes and to the latter , the schrdinger geometry is conformal .
the causal pathology of the geometry ( [ eq : k0nonext ] ) was pointed out in reference @xcite and argued to be the evidence of the holographic dual of non - relativistic cft which should allow the action - at - a - distance . though we feel that this interesting point should be investigated further
, we will not dwell on this issue in this paper and leave it for the future consideration .
as noted before , the black solution ( [ eq : k0nonext ] ) found in references @xcite approaches the extremal solution near the boundary and therefore naturally interpreted as the finite temperature generalization of the galilean holography .
this is similar to the finite temperature ads / cft correspondence @xcite .
we thus would like to examine the thermodynamic properties of this finite temperature system .
we have mentioned in the introduction that the methods of computing the regularized action and thermodynamic quantities adopted in references @xcite are not directly applicable to non - planar black solutions .
we , therefore , use the subtraction method which we describe the details here . in what follows ,
`` the action '' is meant to be the sum of the bulk action and the gibbons - hawking surface term .
we first analytically continue @xmath35 to @xmath55 and put the system into a box by the cutoff @xmath56 .
the cutoff @xmath57 is assumed to be much larger than the scale @xmath11 but finite .
we subtract the action of the extremal solution from the non - extremal one . in doing so
, it is instructed in reference @xcite to match the metrics of those geometries at the wall @xmath56 . in our case
, we have a problem with the boundary matching : since the @xmath58 component of the extremal metric ( with @xmath37 ) in equation ( [ eq : k0nonext ] ) is degenerate , there is no way we can match the metrics at the wall .
as noted before , the @xmath35-constant box boundary slice has zero measure for the extremal case .
therefore , we propose to match the boundary metric of the extremal geometry to the non - extremal one only for the @xmath36-constant three dimensional slices parametrized by @xmath59 .
we can achieve this by appropriately scaling those three coordinates .
as for the @xmath36 direction , we scale this coordinate by a constant and adjust the constant so that the resulting difference action is finite in the limit @xmath60 .
it turns out that this constant is just unity for all the black solutions discussed in this paper , so we do not include the constant in the computations below . concretely ,
we have the analytically continued and scaled extremal metric and the fields @xmath61+\bigg(\frac{r}{r}\bigg)^2 dr^2 \;,\nonumber\\ \phi = & 0 \;,\nonumber\\ a = & i\bigg(\frac{r}{r}\bigg)^2 \frac{h_b}{r } dx^+ \;,\end{aligned}\ ] ] where we have defined the scaling factors @xmath62 we then compute the action @xmath63 where @xmath64 is the action ( [ eq : theaction ] ) evaluated on the solution ( [ eq : k0scaled ] ) . the second term
@xmath65 is the gibbons - hawking term @xmath66 where @xmath67 is the determinant of the boundary first fundamental form and @xmath68 is the trace of the second fundamental form with respect to the outward pointing unit normal vector .
since the invariants in the action are not affected by the scaling , the factors of @xmath69 and @xmath70 come in only from the metric determinants .
similarly , we compute the action @xmath71 evaluated on the non - extremal solution ( [ eq : k0nonext ] ) .
the difference @xmath72 is now finite in the limit @xmath60 and we have @xmath73 this result is in agreement with reference @xcite whose authors have noticed that this is the same result as the schwarzschild - ads@xmath2 black brane . in the discussion of the black hole solution , we will find that the action is the same as the schwarzschild - ads@xmath2 black hole .
we now reproduce the thermodynamics of references @xcite that follows from the regulated action . for this purpose
, we need to identify the temperature , chemical potential , energy , charge and entropy of the system .
let us start with the temperature .
we identify the temperature in the traditional way , that is , we set @xmath74 where @xmath75 is the horizon surface gravity . in the computation of @xmath75 , the killing vector field is taken as @xmath76 where the horizon angular velocity @xmath49 is shown in equation ( [ eq : omegah ] ) and it is measured in the units of the compact @xmath36 circumference in the coordinate length .
we then have @xmath77 and we set the ( coordinate ) circumference of @xmath35 to this value .
one can also obtain the same result by requiring the analytic continuation of the metric ( [ eq : k0nonext ] ) to have the smooth geometry in the @xmath35-@xmath38 slice .
[ in doing so , one must bring the metric to the adm form ( [ eq : admform ] ) , just as in the kerr black hole case . ]
let us now discuss the chemical potential of the system . as discussed before , our metric ( [ eq : k0nonext ] )
is stationary but not static and it describes the black brane rotating in the compactified @xmath36 direction .
we then naturally interpret the angular momentum and velocity as the charge and the conjugate chemical potential of the system , respectively .
this offers @xmath78 where the hat on @xmath79 reminds us that the angular velocity is measured in the units of the @xmath36 circumference .
this identification , however , is subtle . in the usual rotating
black hole systems , a chemical potential is taken to be the _ difference _ between the angular velocities at the horizon and the boundary . however , as we saw previously , our horizon angular velocity is @xmath49 and the boundary coordinate angular velocity diverges . hence the usual identification does not work . despite this subtlety , we will see shortly that the definition ( [ eq : chemp ] ) yields the consistent entropy , suggesting that the identification is correct . to obtain the energy , the charge ( conjugate to the chemical potential
just defined ) and the entropy , we define the zero - loop saddle point free energy @xmath80 where @xmath81 represents the integration over @xmath36 and @xmath26 .
notice that the product @xmath82 is dimensionless , hence so is @xmath83 . from this
, we have @xmath84 these results have been derived in references @xcite .
these quantities , by construction , satisfy the first law .
one non - trivial result , though , is that the horizon area [ the three - volume of @xmath35-constant and @xmath15 slice of the metric ( [ eq : k0nonext ] ) ] is exactly four times the entropy derived above , apart from the overall factor in the definition of the free energy ( [ eq : k0free ] ) .
this makes us confident of the chemical potential identified above . to examine the local thermodynamic stability of the system
, one can compute the hessian of @xmath85 with respect to the thermodynamic variables @xmath86 and evaluate it at the on - shell values of @xmath87 .
this gives @xmath88 , which is always positive .
therefore , the system is thermodynamically stable .
finally , we note that the free energy ( [ eq : k0free ] ) is negative , implying that the non - extremal black brane solution is always thermodynamically preferred to the thermal schrdinger space without the black brane .
this means that the system does not possess the hawking - page phase transition .
it is common that a single gravitational action supports the black solutions whose horizon geometries are @xmath89 , @xmath90 and @xmath91 .
we find that it is also the case for the five dimensional action ( [ eq : theaction ] ) . in this section ,
we present the spherical solution and the hyperbolic case is worked out in the appendix .
we will see that the spherical case has a richer phase structure than the other solutions such as the hawking - page phase transition .
consider the schwarzschild - ads black hole solution of type iib supergravity compactified on @xmath3 @xmath92 where @xmath93 and @xmath94 is the non - extremality parameter in that the extremal solution is given by @xmath95 .
this parameter can be re - expressed in terms of @xmath96 by solving @xmath48 .
the metric of the three sphere @xmath97 can be written as @xmath98 where we have defined @xmath99 then the metric above can be trivially rewritten as @xmath100 one surely notices the similarity of this metric to the metric ( [ eq:1stmetric ] ) . guided by this analogy , consider the following set of a metric and fields which is very similar to the truncated version of equations ( [ eq : k0tybasis ] ) , @xmath101 \nonumber\\ & + k^{1/3}\bigg(\frac{r}{r}\bigg)^2 h^{-1}dr^2 \;,\nonumber\\ \phi = & -\frac{1}{2}\ln k \;,\nonumber\\ a = & k^{-1}\bigg(\frac{r}{r}\bigg)^2 b\ , ( \,h\ , dt + \eta ) \;,\end{aligned}\ ] ] where @xmath28 .
one can check that this set indeed is a solution to the equations of motion that follow from the five dimensional action ( [ eq : theaction ] ) .
notice that this solution reduces to the schwarzschild - ads black hole solution by setting @xmath102 , like the kk - reduced solution ( [ eq : k0tybasis ] ) becomes the schwarzschild - ads black brane .
also , just as in the ads case , this solution s isometry group is the maximal compact subgroup of the schrdinger group .
thus , we see that this is the ( asymptotically ) schrdinger space which has @xmath103-foliation for the slice of constant @xmath54 and @xmath22 . to identify the hamiltonian generator and the compactified direction , we need to define the light - cone - like basis and we introduce the following sets of the coframe @xmath104 and the frame @xmath105 note that we have @xmath106 , as required for a frame and the corresponding coframe and also observe that @xmath107 are defined similar to the light - cone coordinates of equations ( [ eq : lc ] ) .
this clearly is a non - coordinate basis and we have the non - vanishing lie product @xmath108 = 2b\,e_+ - \frac{1}{br^2}e_- -\frac{2}{r}\cot\theta\ , e_\phi \;.\ ] ] because of this non - vanishing lie bracket , the usual definitions of geometric quantities based on the coordinate basis must be modified ( see mtw @xcite ) . among other things ,
the christoffel symbols @xmath109 defined with respect to a coordinate basis are modified and we no longer necessarily have the symmetry @xmath110 in the non - coordinate basis .
given the new coframe , we can now write the solution as @xmath111 + k^{1/3 } \bigg(\frac{r}{r}\bigg)^2 h^{-1}\omega^{r2 } \;,\nonumber\\ \phi = & -\frac{1}{2}\ln k \;,\nonumber\\ a = & k^{-1}\bigg(\frac{r}{r}\bigg)^2 b \ , \bigg\ { \frac{h+1}{2br } \omega^+ + br(h-1 ) \omega^- \bigg\ } \;,\end{aligned}\ ] ] where we have defined @xmath112 . here
, one sees that the coframe is chosen so that the solution appears congruous to the black brane solution ( [ eq : k0nonext ] ) . in parallel to the black brane case , we interpret @xmath113 as the direction of time and @xmath114 as the compactified direction with some circumference .
though obvious , it is an interesting exercise to check that the equations of motion are satisfied in the new frame and coframe with the modified geometric quantities .
and we may not replace the semi - colons to just commas , because the symmetry of the christoffel symbols is lost .
another subtle point is that the definition of riemann tensor involves the non - trivial derivatives with respect to @xmath115 which is different from @xmath116 by the factor of @xmath117 and this must be done with care . ]
we remark that for this solution to make sense , we must impose the restriction @xmath118 this is because we have @xmath119 , and without the restriction , @xmath120 can become zero or negative .
this is a special condition for the black hole solution and we do not have it for the black brane case . as in the previous section
, the geometry ( [ eq : k1nonext ] ) has the ill - defined causal structure .
we can bring the metric to the adm form identical to equation ( [ eq : admform ] ) with the obvious replacements .
a quick inspection reveals that with the function @xmath121 of the black hole solution , the adm form is highly pathological .
for example , the extremal case ( @xmath95 ) with the condition ( [ eq : restriction ] ) has negative definite lapse function squared .
therefore , as in the black brane case , we must give up on the causal development of the hypersurface defined by constant @xmath113 .
currently the consequences of this observation are unclear and it must be investigated in future whether this has something to do with the holographic dual of _ non - relativistic _
cft on @xmath103 .
let us now proceed to compute the on - shell difference action following the procedure described in the previous section .
unlike the previous case , the @xmath114 direction is in general not degenerate for the extremal ( @xmath95 ) nor non - extremal ( @xmath122 ) black hole solutions .
however , simple scaling of the cobasis can not achieve the full matching of the extremal @xmath56 boundary metric to the non - extremal one .
therefore , we match the metrics only for the @xmath114-constant three - slices parametrized by @xmath123 , just as was done in the previous section .
we analytically continue @xmath113 to @xmath124 and set the cutoff at @xmath56 .
we then have the scaled extremal metric and fields @xmath125 + k_0^{1/3 } \bigg(\frac{r}{r}\bigg)^2 h_0^{-1}\omega^{r2 } \;,\nonumber\\ \phi = & -\frac{1}{2}\ln k_0 \;,\nonumber\\ a = & k_0^{-1}\bigg(\frac{r}{r}\bigg)^2 b \ ,
\bigg\ { \frac{h_0 + 1}{2br } h_b\ , \omega^+ + br(h_0 - 1 ) \omega^- \bigg\ } \;,\end{aligned}\ ] ] where we have defined @xmath126 next we compute the action @xmath63 where @xmath64 is the action ( [ eq : theaction ] ) evaluated on the solution ( [ eq : k1scaled ] ) and the second term @xmath65 is the gibbons - hawking term . notice that due to the scaling in ( [ eq : k1scaled ] ) , the coframe ( [ eq : coframe ] ) and frame ( [ eq : frame ] ) must be scaled accordingly , including the factor of `` @xmath127 '' that comes from the analytic continuation .
this results in the modified lie bracket ( [ eq : lie ] ) which in turn affects the non - coordinate based geometric quantities . using those quantities
, one can check that the invariants in the action @xmath128 are independent of the scaling and the scaling factors come in only from the metric determinants , as expected .
similarly , we compute the action @xmath71 evaluated on the non - extremal solution ( [ eq : k1nonext ] ) . the difference @xmath72 is finite in the limit @xmath60 and we have @xmath129 rather surprisingly , this action is the same as the schwarzschild - ads@xmath2 black hole . in this subsection , we examine the thermodynamics following from the action calculated in equation ( [ eq : k1action ] ) .
we proceed in parallel with the previous section .
first , we identify the temperature .
we can compute this just as in the previous section , either through the surface gravity [ using the same killing vector field as equation ( [ eq : killingvf ] ) with the replacements @xmath130 or by requiring the smooth geometry .
the result is @xmath131 now , it is a little naive to identify this to the `` circumference '' of @xmath113 , because it is not a coordinate basis .
however observe that at each fixed point on @xmath103 , parametrized by ( @xmath132 ) , @xmath113 _ is _ a coordinate basis .
we thus write @xmath133 of equation ( [ eq : k1action ] ) as @xmath134 where @xmath81 includes the `` circumference '' of the @xmath114 direction . second
, we identify the chemical potential to the angular velocity of the horizon as before and it turns out that it is the same @xmath135 to obtain the thermodynamic quantities , we define the zero - loop saddle point free energy @xmath136
\big [ f(\beta,\hat\mu)-6\beta^2\hat\mu \big ] \;,\end{aligned}\ ] ] where @xmath137 and this is defined purely for the presentation purpose and we do not mean anything deeper in this function @xmath138 . then we compute the thermodynamic quantities as in equations ( [ eq : k0eqs ] ) and obtain @xmath139 again , by construction , these quantities satisfy the first law .
the entropy , thus calculated , is consistent with the quarter - area law which is a non - trivial result and indicates the right identification of the chemical potential .
the local thermodynamic stability can be examined by computing the hessian of @xmath85 with respect to the thermodynamic variables @xmath86 and evaluate it at the on - shell values of @xmath87 .
the result is @xmath140 this implies the stability threshold @xmath141 which translates to the following curve in the @xmath142-@xmath143 parameter space : @xmath144 the mechanism of this instability is very similar to the usual schwarzschild - ads black hole , namely , it is the merger point of the large and small black holes in the parameter space . though we have not explicitly discussed the small black hole , it exists .
a quick way to see this is to note that there are two solutions to equation ( [ eq : k1beta ] ) with respect to the variable @xmath96 and these solutions degenerate at the threshold .
let us now discuss the hawking - page phase transition @xcite .
it is clear from equation ( [ eq : k1free ] ) that the free energy changes sign at @xmath145 .
this implies that the thermodynamically preferred solution changes from the thermal space without black hole to the space with black hole , and _
vice versa _ , at the threshold .
this is the hawking - page phase transition .
the threshold @xmath145 can be expressed in the parameter space of @xmath142 and @xmath143 as @xmath146 and the higher temperature side of this curve is the black hole phase . comparing equations ( [ eq : instability ] ) and ( [ eq : hptrans ] ) , we see that the black hole instability line is inside the thermal schrdinger geometry phase . finally , we note that the restriction ( [ eq : restriction ] ) requires @xmath147 where the critical temperature coincides with the crossing point of the @xmath148 line and the local thermodynamic instability curve ( [ eq : instability ] ) .
recall that the chemical potential @xmath142 is measured in the units of the circumference of the compactified direction .
therefore , this restriction may be alternatively interpreted as the minimal size of the compactification .
+ the discussions above are summarized in the phase diagram figure [ fig : schrphase ] .
in this paper , we have discussed the thermodynamics of the black solutions to the action that describes the space with schrdinger symmetry .
the black hole and black hyperboloid solutions are newly found solutions and also we have proposed a way to compute the finite actions . for the black hole case , we have found the hawking - page phase transition , as well as the instability lines in the phase diagram .
we have shown that the on - shell actions of all solutions discussed in this paper are identical to those of the black solutions in ads .
this seems to imply that there is a symmetry that transforms the asymptotically schrdinger solutions to the corresponding asymptotically ads solutions . for the planar case , reference @xcite explicitly show such a symmetry .
therefore , it is likely that the symmetry transformation of the reference can be generalized to the non - planar solutions discussed in this paper .
if this expectation is borne out , the method similar to the one proposed in appendix c.2 of reference @xcite can be applied to the solutions other than the black branes and the resulting thermodynamic quantities computed this way would agree with our results in this paper .
it is undeniable that the matching procedure proposed in this paper is highly _
ad hoc_. though this procedure produces the reasonable results , further understanding of this is required .
this issue is closely related to the ( so far not so clear ) physical meaning of the @xmath36 coordinate , because the procedure gives up on the matching of the boundary first fundamental forms for this component .
perhaps , this is related to the fact that the boundary cft is defined on the coordinate that is not including @xmath36 .
we have mentioned in section [ subsec : bbthermo ] that there is a certain subtlety in identifying the chemical potential , because the difference of the angular velocities at the horizon and the boundary diverges .
however , we have shown that our identification of the chemical potential yields the entropy that is consistent with the quarter - area law .
on the contrary , it is advocated in references @xcite that the chemical potential should be identified with the coefficient of @xmath149 term in the @xmath150 component .
this identification yields the entropy that is _ not _ consistent with the quarter - area law .
however , it is not clear to us what is inappropriate with their identification .
clarification on this issue is desired .
as pointed out in the ends of sections [ subsec : bbsoln ] and [ subsec : bhsoln ] , the spacetimes discussed in this paper are causally pathological .
the problem is clear when we bring the metric into the adm form with respect to the @xmath35 direction which the galilean holography seems to uniquely selects as the time .
it is likely that this behavior of the spacetime is closely related to the fact that this is the ( conjectured ) dual of non - relativistic conformal field theories .
we expect that investigation into this issue would deepen our understanding of the holography .
it is known that the r - charged black holes possess a rich phase structure @xcite .
it is possible to include the r - charge into the galilean holography by applying the null melvin twist to the spinning d3-brane geometry @xcite and taking the near horizon limit of the resulting geometry .
( the rr potential , in this case , becomes non - trivial and this itself is interesting to investigate . ) from this solution , one could obtain the five dimensional action and its spherical ( and hyperbolic ) solution .
we then able to observe the interplay between the chemical potential discussed in this paper and the one for the r - symmetry .
it would be very interesting to see how the phase structure of the r - charged black holes would be extended .
in this paper , we have exclusively worked out the gravity side of the proposed galilean holography .
it is of natural interest to investigate into the cft side of the corresponding phenomena found in this work .
most importantly , we would like to see in detail how the non - relativistic conformal field theory behaves in accordance with the hawking - page phase transition found in this paper .
learning from the ads / cft correspondence , it is likely that the phase transition corresponds to the confinement / deconfinement phase transition in the cft side
. it would be intriguing to see how this actually works in the non - relativistic set up .
meanwhile , we should always keep in mind that this holography is initiated in hope to understand the real world strongly coupled non - relativistic field theories .
the gravity duals being discussed in the literature ( including this work ) most likely do not have the real world field theory duals .
but as in the ads / cft correspondence , we expect the galilean holography to provide important universal properties of non - relativistic field theories .
the fact is that the experimental data exists on the field theory side of the proposed duality .
it is crucial to see if one could come up with the gravity dual that is consistent with the experimental data .
i would like to thank andrei kryjevski for drawing my attention to son s original work @xcite .
i benefited from the discussion session held at weizmann institute , especially from ofer aharony and zohar komargodski .
i would also like to thank umpei miyamoto for valuable discussions .
i thank a. kryjevski and m. hanada for the partial involvements in this project .
this work was supported by the lady davis fellowship .
in this appendix , we present the black hyperboloid solution and examine its thermodynamics . the discussion follows the structure of the main text .
we would like to find the solution to the action ( [ eq : theaction ] ) with @xmath151 horizon .
for this purpose , we find the appropriate form of the metric for @xmath152 as @xmath153 with @xmath154 we then find the desired solution @xmath155 + k^{1/3 } \bigg(\frac{r}{r}\bigg)^2 h^{-1}\omega^{r2 } \;,\nonumber\\ \phi = & -\frac{1}{2}\ln k \;,\nonumber\\ a = & k^{-1}\bigg(\frac{r}{r}\bigg)^2 b \ , \bigg\ { \frac{h+1}{2br }
\omega^+ + br(h-1 ) \omega^- \bigg\ } \;,\nonumber\\ h : = & 1 - \bigg(\frac{r}{r}\bigg)^2 - \bigg(\frac{r_0r}{r^2}\bigg)^2 \;,\nonumber\\ k : = & 1-(h-1)b^2r^2 = 1+(br)^2 + \bigg(\frac{r_0br}{r}\bigg)^2 \;,\end{aligned}\ ] ] where we have defined the coframe @xmath156 and the frame @xmath157 with this non - coordinate basis , the non - vanishing lie bracket is @xmath158 = -2b\,e_+ + \frac{1}{br^2}e_- -\frac{2}{r}\coth\chi\ , e_\phi \;.\ ] ] notice that the non - extremality parameter @xmath94 is expressed as @xmath159 so we have the extremality at @xmath160 and the horizon radius must be larger than this value .
one can proceed to compute the difference action just as in the main text and find that @xmath161 this is the same as the schwarzschild - ads@xmath2 black hyperboloid .
the temperature can be computed in the usual way and one obtains @xmath162 since we have the restriction @xmath163 , we see that this solution does not have zero temperature limit .
the chemical potential is the same as other cases : @xmath164 to obtain the thermodynamic quantities , we define the zero - loop saddle point free energy @xmath165
\;,\end{aligned}\ ] ] where @xmath166
. then we compute the thermodynamic quantities as in equations ( [ eq : k0eqs ] ) and obtain @xmath167 by construction , these quantities satisfy the first law and the entropy is consistent with the quarter - area law .
the local thermodynamic stability can be examined by computing the hessian of @xmath85 with respect to the thermodynamic variables @xmath86 and evaluate it at the on - shell values of @xmath87 .
the result is @xmath168 because of the restriction @xmath163 , we see that the system is always locally stable . notice that for the range of the horizon radius @xmath169 , we have @xmath170 , but the system is stable .
aaaa d. t. son , _ toward an ads / cold atoms correspondence : a geometric realization of the schroedinger symmetry _ ,
d * 78 * , 046003 ( 2008 ) [ arxiv:0804.3972 [ hep - th ] ] .
k. balasubramanian and j. mcgreevy , _ gravity duals for non - relativistic cfts _ , phys .
* 101 * , 061601 ( 2008 ) [ arxiv:0804.4053 [ hep - th ] ] .
j. m. maldacena , _ the large n limit of superconformal field theories and supergravity , _ adv .
* 2 * , 231 ( 1998 ) [ int .
j. theor .
phys . * 38 * , 1113 ( 1999 ) ] [ arxiv : hep - th/9711200 ] .
s. s. gubser , i. r. klebanov and a. m. polyakov , _ gauge theory correlators from non - critical string theory , _ phys .
b * 428 * , 105 ( 1998 ) [ arxiv : hep - th/9802109 ] .
e. witten , _ anti - de sitter space and holography , _ adv .
theor . math .
* 2 * , 253 ( 1998 ) [ arxiv : hep - th/9802150 ] . c. r. hagen , _ scale and conformal transformations in galilean - covariant field theory _ , phys .
d * 5 * , 377 ( 1972 ) .
u. niederer , _ the maximal kinematical invariance group of the free schrodinger equation _ , helv .
acta * 45 * , 802 ( 1972 ) . y. nishida and d. t. son , _ nonrelativistic conformal field theories _ , phys . rev .
d * 76 * , 086004 ( 2007 ) [ arxiv:0706.3746 [ hep - th ] ] .
c. p. herzog , m. rangamani and s. f. ross , _ heating up galilean holography _ , arxiv:0807.1099 [ hep - th ] .
j. maldacena , d. martelli and y. tachikawa , _
comments on string theory backgrounds with non - relativistic conformal symmetry _ , arxiv:0807.1100 [ hep - th ] .
a. adams , k. balasubramanian and j. mcgreevy , _ hot spacetimes for cold atoms _ , arxiv:0807.1111 [ hep - th ] .
e. witten , _ anti - de sitter space , thermal phase transition , and confinement in gauge theories _ , adv .
* 2 * , 505 ( 1998 ) [ arxiv : hep - th/9803131 ] .
m. alishahiha and o. j. ganor , _ twisted backgrounds , pp - waves and nonlocal field theories _ , jhep * 0303 * , 006 ( 2003 ) [ arxiv : hep - th/0301080 ] .
e. g. gimon , a. hashimoto , v. e. hubeny , o. lunin and m. rangamani , _ black strings in asymptotically plane wave geometries _ , jhep * 0308 * , 035 ( 2003 ) [ arxiv : hep - th/0306131 ] .
s. w. hawking and d. n. page , _ thermodynamics of black holes in anti - de sitter space _ , commun .
phys . * 87 * , 577 ( 1983 ) .
j. d. brown and j. w. york , _ quasilocal energy and conserved charges derived from the gravitational action _ , phys .
d * 47 * , 1407 ( 1993 ) [ arxiv : gr - qc/9209012 ] . v. balasubramanian and p. kraus , _ a stress tensor for anti - de sitter gravity _
, commun .
phys . *
208 * , 413 ( 1999 ) [ arxiv : hep - th/9902121 ] .
p. kraus , f. larsen and r. siebelink , _ the gravitational action in asymptotically ads and flat spacetimes _ , nucl .
b * 563 * , 259 ( 1999 ) [ arxiv : hep - th/9906127 ] .
s. de haro , s. n. solodukhin and k. skenderis , _ holographic reconstruction of spacetime and renormalization in the ads / cft correspondence _ , commun .
phys . * 217 * , 595 ( 2001 ) [ arxiv : hep - th/0002230 ] .
m. bianchi , d. z. freedman and k. skenderis , _ holographic renormalization _
b * 631 * , 159 ( 2002 ) [ arxiv : hep - th/0112119 ] .
g. w. gibbons and s. w. hawking , _
action integrals and partition functions in quantum gravity _ , phys .
d * 15 * , 2752 ( 1977 ) .
m. cvetic and s. s. gubser , _ phases of r - charged black holes , spinning branes and strongly coupled gauge theories _ , jhep * 9904 * , 024 ( 1999 ) [ arxiv : hep - th/9902195 ] .
d. yamada , _ metastability of r - charged black holes _
, class . quant .
* 24 * , 3347 ( 2007 ) [ arxiv : hep - th/0701254 ] .
p. kraus , f. larsen and s. p. trivedi , _ the coulomb branch of gauge theory from rotating branes _ , jhep * 9903 * , 003 ( 1999 ) [ arxiv : hep - th/9811120 ] .
j. g. russo and k. sfetsos , _
rotating d3 branes and qcd in three dimensions _ , adv .
phys . * 3 * , 131 ( 1999 ) [ arxiv : hep - th/9901056 ] .
m. cvetic _ et al .
_ , _ embedding ads black holes in ten and eleven dimensions _ , nucl .
b * 558 * , 96 ( 1999 ) [ arxiv : hep - th/9903214 ] . | a black hole and a black hyperboloid solutions in the space with the schrdinger isometries are presented and their thermodynamics is examined .
the on - shell action is obtained by the difference between the extremal and non - extremal ones with the unusual matching of the boundary metrics .
this regularization method is first applied to the black brane solution in the space of the schrdinger symmetry and shown to correctly reproduce the known thermodynamics .
the actions of the black solutions all turn out to be the same as the ads counterparts .
the phase diagram of the black hole system is obtained in the parameter space of the temperature and chemical potential and the diagram contains the hawking - page phase transition and instability lines . |
when a group of particles flow through a narrow bottleneck an opening not much larger than the particle size the dissipative interactions among the particles may lead to the spontaneous development of clogs . these kind of jams can be observed in grains discharging from a silo @xcite , people escaping from a room @xcite or vehicle traffic in a highway .
although the nature of the particles is wildly different in each one of these instances , they share interesting resemblances which suggest that a general theory could describe at least the most important features of the phenomenon @xcite .
one of the most striking similarities among people escaping from a room and beads passing through a narrowing is that the distribution of avalanche sizes ( or bursts ) displays an exponential behavior .
generally speaking , the avalanche size is defined as the number of particles passing through the bottleneck between two consecutive clogs .
for the case of inert beads , the avalanche size is easily determined , as the flow is halted permanently and for good by arches formed just above the exit orifice @xcite .
the flow can only be restored by breaking such arches with an external energy input . at that moment ,
a new avalanche begins .
therefore , the avalanche size is defined as the number of particles that get out of the silo from the instant when such an energy input is applied , till the moment when an arch blocks the orifice and ends the outpouring . for the case of people , or other live beings ,
the definition of avalanche is more problematic due to the fact that the clogs , left to their own , last just for a short time .
hence , the only way to define a clog is to choose a certain time lapse during which no individual has come out from the enclosure @xcite .
once the clogs are thus specified , the avalanche sizes are just measured as the number of individuals that get out between two consecutive clogs .
obviously , in this case the avalanche sizes depend on the time lapse chosen to determine whether clogging has occurred , whereas in the case of inert particles this arbitrariness is absent .
this difference in the nature of the clogs may be at the hearth of a remarkable fact : while the exponential behavior is general in the case of granular flows @xcite , it is only observed for very small door sizes in crowd dynamics @xcite .
the exponential distribution of avalanche sizes for inert beads was explained in @xcite in terms of the probability @xmath0 that a particle passes through the outlet without getting stuck . assuming that this probability remains constant during the whole avalanche , the distribution of avalanche sizes was written as @xmath1 where @xmath2 is , of course , the probability that a particle blocks the orifice .
the mean avalanche size @xmath3 , i.e. the first moment of the distribution described by eq .
[ eq1 ] , can be written as : @xmath4 as expected , if the probability of clogging @xmath2 increases , then the mean avalanche size decreases . in both cases
people escaping from a room and particles outpouring from a silo the enlargement of the outlet leads to an increase in the size of bursts ( or avalanches ) .
a lingering question is whether there exists an outlet size above which clogging is impossible . for the case of a silo filled with grains
, debate still goes @xcite .
nevertheless , outlet sizes larger than @xmath5 or @xmath6 times the particle size generate immense avalanches , so in practical terms two regimes can be distinguished : one of _ clogging _ for small outlet sizes , and another one of _ no clogging _ for big outlet sizes .
another feature of clogging that is shared by people in a crowd and particles in a silo is the role of pressure . in humans ,
an increase of pressure caused by panic seems to be a key ingredient for the appearance of clogging @xcite .
parisi and corso @xcite have shown that the evacuation time for a room as a function of the pedestrians desired velocity ( which is a variable considered in the model ) presents a minimum for intermediate velocities .
a very high desired velocity leads to an increase in the evacuation time due to clogs . in granular matter , a low pressure ( which can be attained in shallow layers of grains ) is found to effectively prevent clogging @xcite .
low pressure is also behind an ingenious idea sometimes used when trying to improve the flow of pedestrians through a bottleneck : the placement of an obstacle _ before _ the exit @xcite .
the size of the column the obstacle is typically of the order of a pedestrian and the position varies from one study to another , although it is generally close to the exit ( at most at a distance of 2 or 3 pedestrian sizes ) . in the case of silos ,
the placement of an obstacle above the outlet has also been used , but its relationship with clogging remains scantily explored . instead
, the placement of obstacles is usually aimed to improve the flow and to the reduction of undesirable wall stresses .
indeed , most of the studies about silos and obstacles are performed with such a large outlet size that clogging is practically impossible @xcite . in a previous work we reported experimental evidence of the silo clogging reduction induced by the presence of an obstacle @xcite .
it was shown that the clogging probability could be reduced by almost two orders of magnitude if the obstacle position is properly selected .
this important effect was attributed to a pressure reduction in the outlet neighborhood .
notably , the presumed pressure variation induced by the obstacle affects very little the flow rate .
this result is in agreement with previous works , where it was proved that the pressure in a silo does not have a significant influence on the flow rate @xcite . in this manuscript
, we extend the range of previous results by analyzing the effect of the obstacle position for different sizes of both the outlet and the obstacle .
furthermore , we investigate the effect of the obstacle in the features of the flow rate and its fluctuations . finally , the comparison between the arches formed in silos with and without an obstacle reveals meaningful geometrical differences which seem to be related to the pressure variation .
the manuscript is structured as follows .
first , we will explain the experimental setup and procedure .
then , we will show the generality of the effect caused by the obstacle regardless of outlet size . in the next two sections we will analyze the flow rate properties , in connection with clogging .
finally , we will relate the pressure reduction induced by the obstacle with the arch shape and we will draw some conclusions .
the experimental setup consists of a two dimensional rectangular silo 800 mm high and 200 mm wide .
it is made of two glass sheets separated by two flat metal strips @xmath7 thick and 800 mm high .
these metal bands are the side walls , so the distance between them ( 200 mm ) fixes the width of the silo .
the silo bottom is flat and formed by two facing metal flanges , so that their edges define the outlet size @xmath8 , which can be varied at will ( fig .
[ fig : exp ] ) . above the bottom
, a disk of diameter @xmath9 is placed vertically above the outlet center . in most of the experiments @xmath9
is set at @xmath10 , although in chapter [ sect : obstaclesize ] this value has been modified in order to study its effect in the discharge phenomenon . the distance @xmath11 from the bottom of the obstacle to the outlet ( see fig . [
fig : exp ] ) can be varied and is carefully measured with an accuracy of @xmath12 . as in a previous work @xcite , the case of a silo without obstacle will be referred to as @xmath13 . the silo is filled with a sample of monodisperse stainless steel beads with a diameter of @xmath14 . hence , the grains are disposed between the two glass sheets conforming a monolayer .
the silo filling is performed by pouring the grains along its whole width through a hopper at the top . after the silo filling
, grains start to flow through the outlet until an arch blocks it .
the particles are collected in a cardboard box placed on top of a balance . as the weight of one particle
is known , the size of the avalanche @xmath15 the number of particles fallen between two consecutive clogs is easily calculated .
then , a picture of the region above the orifice is taken with a standard video camera and further analyzed in order to detect the position of every particle in the image . from these , the particles forming the blocking arch are obtained as explained in @xcite .
the experiment is resumed by blowing a jet of compressed air aimed at the orifice that starts a new avalanche .
the experimental setup is automated and controlled by a computer .
this allows us to register a large number of avalanches ( between @xmath16 and @xmath17 ) and the corresponding arches at each run .
let us note that the silo is refilled whenever the level of grains falls below a preset lower limit of around @xmath18 ( @xmath19 times the width of the silo ) .
the reason for this is to avoid pressure variations at the bottom due to the amount of grains in the silo ; recall that the pressure at the base of a silo saturates and is therefore independent on the filling level as long as the height of the granular layer exceeds a certain level @xcite .
photograph of an arch formed above the outlet .
the dashed horizontal segment marks the line where the crossing of each bead is scored to compute passage times .
@xmath8 is the length ( size ) of the outlet and @xmath11 the distance from the bottom of the obstacle to the outlet .
particles forming the arch are indicated by crosses and particles forming the base of the arch are indicated by circles.,title="fig : " ] additionally , for each experimental condition , a number of movies were recorded of the region above the outlet with a high speed camera at @xmath20 frames per second during a total time lapse of @xmath21 seconds .
these recordings allowed to accurately measure the moment at which each particle crosses the outlet ( with a precision better than @xmath22 ms ) . from this , the flow rate @xmath23 ( in number of particles per second ) is calculated within time intervals of @xmath24 ms .
it should be noted that the flow rate measurements were always performed well inside the avalanches . in particular
, we always wait @xmath25 seconds after the beginning of the avalanche and we stop the measurement at least @xmath22 second before the end of the avalanche . in this way
we intend to avoid any possible influence of a transient regime at the beginning or the end of the avalanche .
we remark that , as reported before @xcite , it is possible to find values of @xmath26 within the avalanche .
such events were proposed to correspond to unstable clogs : arches that interrupt the flow of the particles for a short time but were not strong enough to durably clog the silo . from all the values of @xmath23 obtained ( typically @xmath27 measurements for each experimental conditions ) we calculated the average flow rate @xmath28 as well as its coefficient of variation .
( color online ) a ) mean avalanche size versus @xmath11 for @xmath29 mm ( @xmath30 ) , @xmath31 mm ( @xmath32 ) and @xmath33 mm ( @xmath34 ) .
horizontal lines indicate the values of @xmath3 when @xmath13 .
b ) same results than in ( a ) , but dividing the mean avalanche size by the mean avalanche size at @xmath13 corresponding to each value of @xmath8.,title="fig : " ] in a previous work , it was shown that the avalanche size distribution displays an exponential decay for all the obstacle positions @xcite .
hence , the mean avalanche size @xmath3 can be defined and used as a characteristic parameter of the distribution .
let us now focus on the behavior of @xmath3 as a function of the obstacle height above the outlet ( @xmath11 ) for different outlet sizes . in fig .
[ fig : meanavas ] ( a ) we plot the experimental values of @xmath35 versus @xmath11 for three different outlet sizes ( @xmath29 , @xmath36 , and @xmath37 mm ) . for these values of @xmath8 , the mean avalanche size
if the obstacle is absent ( @xmath38 ) extends over a wide range : from @xmath39 to @xmath40 particles .
these values are represented by dashed horizontal lines in fig .
[ fig : meanavas ] ( a ) .
this plot manifests that the mean avalanche size as a function of @xmath11 displays the same trend independently of the outlet size .
in all the cases , for positions of the obstacle above @xmath41 mm or so , the value of @xmath3 is very similar to the value obtained without obstacle .
this indicates that for such values of @xmath11 the obstacle effect nearly vanishes .
when the obstacle approaches the orifice from @xmath13 , @xmath3 increases , displaying a maximum for @xmath42 mm . for smaller values of @xmath11 the mean avalanche size decreases , because arches begin to build up between the obstacle and the bottom ( before that , arches just span over the exit orifice without touching the obstacle ) @xcite . despite the similar behavior displayed in fig .
[ fig : meanavas ] ( a ) for different outlet sizes , it is evident that the consequence of the peak is enhanced as @xmath8 increases .
this result is more obvious if we plot @xmath3 divided by the values of @xmath43 corresponding to each @xmath8 ( fig .
[ fig : meanavas]b ) .
this behavior makes sense , because as @xmath8 increases and approaches values of ` no clogging ' , arches are composed of more particles and are weaker @xcite .
hence , a similar decrease of pressure can result in a stronger reduction of the clogging probability ( or enhancement of the mean avalanche size ) as @xmath8 increases .
another interesting fact observed in fig .
[ fig : meanavas]b is that the position of the peak does not depend significatively on the outlet size .
this result can be understood if we assume that the decay of @xmath3 for small values of @xmath11 is due to clogs developed between the orifice and the obstacle .
thus , in this region , the parameter that governs clogging is not @xmath8 but the distance between the bottom of the obstacle and the bottom of the silo . in any case , the mechanisms causing clogging for such small values of @xmath11 which should be closely related to clogging in inclined orifices @xcite are not the focus of this paper . in brief , if we just consider the domain where arches are not formed between the outlet and the obstacle ( @xmath44 ) , it can be stated that the bigger the outlet , the stronger the effect of the obstacle in the clogging reduction .
( color online ) a ) average flow rate versus @xmath11 for @xmath29 mm ( @xmath30 ) , @xmath31 mm ( @xmath32 ) and @xmath33 mm ( @xmath34 ) .
error bars are confidence intervals at @xmath45 .
horizontal lines indicate the values of @xmath46 when @xmath13 .
b ) same results than in a ) but dividing the average flow rate by the average flow rate at @xmath13 corresponding to each value of @xmath8.,title="fig : " ] in this section we will analyze the effect of the obstacle on the flow rate ( calculated as explained in section [ sect : exp ] ) .
we first show in fig .
[ fig : flowrate]a the results of the average flow rate @xmath28 obtained for different obstacle positions @xmath11 .
clearly , the behavior is similar for the three outlet sizes ( @xmath8 ) studied in this work .
when @xmath11 is too small , the flow rate is smaller than the one obtained without obstacle ( which is marked with dashed horizontal lines in fig .
[ fig : flowrate]a ) .
this is due to the short distance between the outlet and the obstacle , that strongly affects the flow rate .
if the obstacle position is moved upwards , far from the orifice , the flow rate can be increased , reaching values up to @xmath47 higher than in the silo without obstacle .
this result is in agreement with recent results obtained for large outlet sizes where clogging did not appear @xcite .
the measurements displayed in fig .
[ fig : flowrate]a also indicate that the flow rate enhancement is more conspicuous for the smallest outlet size . in addition
, it seems that the transition point from flow rate reduction ( at small values of @xmath11 ) to flow rate enhancement ( at high values of @xmath11 ) moves towards higher values of @xmath11 as @xmath8 is increased .
this is more clearly seen if we represent the results of the flow rate rescaled by the flow rate obtained without obstacle corresponding to each outlet size , _
i. e. _ @xmath48 ( see fig . [
fig : flowrate ] b ) .
we remark that no obvious relationship can be perceived between the average flow rate and the mean avalanche size measurements .
indeed , the effect of the obstacle on the avalanche size is more prominent as @xmath8 is increased , whereas the effect on the flow rate is stronger for small @xmath8 .
at the same time , the obstacle positions at which the maximum flow rate is obtained do not coincide with the positions at which the avalanche size is maximized .
all these facts suggest the different nature of two processes : the flow of particles through the outlet , and the clogging due to arch formation @xcite . on the other hand
, it has been recently proposed that some connection does exist between the mean avalanche size and the fluctuations of the flow rate @xcite .
indeed , it was shown that in a 2d silo the values of @xmath23 display a gaussian distribution if the outlet size is large .
on the contrary , as @xmath8 was reduced and the region of clogging approached , there was an increase on the number of events with @xmath49 , so that the distribution was no longer gaussian .
those events were attributed to the existence of partial clogs that were not strong enough to permanently halt the flow . in fig .
[ fig : flowratefluctus]a we present the time series of @xmath23 obtained for a silo with orifice size @xmath50 without obstacle . in agreement with @xcite ,
the trace displays downward spikes in which the flow goes to zero .
these events are consequence of temporal ( not definitive ) interruption of the flow . if the same results are presented for the case of a silo with an obstacle of 10 mm diameter placed at @xmath51 ( fig .
[ fig : flowratefluctus ] b ) it becomes clear that the downward spikes disappear .
recall that @xmath51 is the obstacle position for which the avalanche size is maximum and hence , the clogging probability is minimum .
in addition , from fig . [
fig : flowratefluctus ] b , it seems that placing an obstacle also minimizes the upward fluctuations making the flow rate more homogeneous while its average value is practically unaltered . in figs . [
fig : flowratefluctus]c - f we present the normalized distributions of @xmath23 for four different obstacle positions . fig .
[ fig : flowratefluctus]a shows the distribution for @xmath23 without obstacle , where the existence of a large number of events with @xmath49 is rather obvious . as the position of the obstacle
is moved toward the outlet , the number of events corresponding to @xmath49 is reduced , to the point of being almost absent when @xmath51 ( fig .
[ fig : flowratefluctus]d ) .
in addition , decreasing @xmath11 , the histograms become narrower implying that the placement of the obstacle helps to avoid fluctuations of the flow rate .
the results obtained for the distributions of @xmath23 for the two other outlet sizes ( not shown ) display the same behavior . time series of the flow rate q ( beads per unit time as measured at nonoverlapping time windows of 30 ms ) for an orifice size @xmath50 and an obstacle of @xmath41 mm diameter placed at a ) @xmath13 and b ) @xmath52 mm . in figures
c - e the normalized histograms of the flow rate values are presented for different obstacle positions , namely : c ) @xmath13 , d ) @xmath53 mm , e ) @xmath54 mm and f ) @xmath52 mm.,title="fig : " ] ( color online ) coefficient of variation , _ i.e. _ standard deviation of the values obtained for the flow rate @xmath23 rescaled by @xmath28 , versus the obstacle position for @xmath29 mm ( @xmath30 ) , @xmath31 mm ( @xmath32 ) and @xmath33 mm ( @xmath34 ) .
the horizontal lines indicate the values of @xmath55 for @xmath13 for each value of @xmath8.,title="fig : " ] the effect of the obstacle position in the flow rate fluctuations is quantified by means of the standard deviation of the distribution divided by the average , _
i. e. _ the coefficient of variation @xmath56 .
the results obtained for different obstacle positions and the different values of @xmath8 are presented in fig .
[ fig : sigmaflowrate ] .
interestingly , the global behavior obtained can be easily related to the one reported in fig .
[ fig : meanavas ] a for the mean avalanche size .
for all the outlet sizes studied , high values of @xmath11 display values of @xmath57 close to the ones seen in the absence of an obstacle . as @xmath11 is reduced , the fluctuations decrease just as the probability of clogging does ( which implies an increase in the mean avalanche size ) . in all the cases ,
the minimum of @xmath57 is reached at values of @xmath11 similar to those for which the maximum avalanche size is obtained .
these results exhibit the close relationship between clogging reduction and the fluctuations of the flow rate : reducing the clogging probability is always associated with a reduction in the fluctuations of the flow rate .
let us finally stress that by placing an obstacle above the outlet , the flow rate fluctuations can be reduced without an associated increase of the outlet size and average flow rate .
this implies a clear advantage with respect to the work of janda et al .
@xcite , where the reduction of fluctuations was always associated with an increase of the flow rate .
in order to further explore the possible relationship between clogging and flow rate fluctuations , we have performed some experiments where the diameter of the obstacle ( @xmath9 ) has been varied . we choose a fixed outlet size ( @xmath58 mm ) and obstacle position ( @xmath59 mm ) .
the election of @xmath8 is based in practical terms as the mean avalanche size is small , with the consequent time savings that this implies .
the reason for choosing @xmath59 mm is that , for such values of @xmath11 , both the avalanche size and the average flow rate depend on the obstacle position ( fig .
[ fig : meanavas ] and fig .
[ fig : flowrate ] respectively ) .
therefore , we expect that the change of obstacle size has also an effect on these variables .
we suspect that choosing a higher value of @xmath11 could lead to values of the flow rate and avalanche size less dependent on the obstacle size . in any case
, this hypothesis remains to be confirmed in future works . in fig .
[ fig : obstaclesize]a the results of the mean avalanche size are displayed for five different diameters of the obstacle and compared with the case without obstacle ( which is presented as @xmath60 ) .
the first result that becomes evident is that the obstacle , whatever its size , causes an important increase of the mean avalanche size .
the mere presence of the obstacle , even when it is small , prevents the clogging of particles passing through a bottleneck at least for the outlet size and the obstacle position used here .
in addition , it seems that there is a obstacle size ( @xmath61 mm ) for which the avalanche size is maximum .
this result can be understood as follows : for small obstacle sizes , the bigger the obstacle , the stronger its effect preventing clogging .
hence , enlarging the obstacle leads to an increase of the avalanche size .
this tendency is reversed for very big obstacle sizes , a fact which is attributed to an enhancement of the clogging between the obstacle and the bottom of the silo .
intuitively , an exceedingly large obstacle would lead to a situation where the flow would be impossible , as the angle of repose of the material imposes a limit at which the particles are sustained by themselves @xcite .
we also analyzed if the flow rate depends on the obstacle size ; measurements are presented in fig .
[ fig : obstaclesize]b .
it can be seen that the larger the diameter of the obstacle , the smaller the average flow rate .
this behavior can be understood if we think that increasing the obstacle size leads to a reduction of the distance between the obstacle and the static region of grains at the bottom corners of the silo imposed by the angle of repose .
this process is analogous to moving the obstacle towards the orifice ( decreasing @xmath11 ) .
let us recall that the obstacles of different diameter were placed at a position ( @xmath59 mm ) for which the average flow rate decreases as the obstacle approaches the outlet ( as shown in fig .
[ fig : flowrate ] ) . finally , in fig .
[ fig : obstaclesize]c we show the variation coefficient as a function of @xmath9 .
interestingly , the highest value of @xmath55 is obtained for the silo without obstacle , where the minimum value of the mean avalanche size is also found .
in addition , the coefficient of variation also displays the non - monotonic dependence on @xmath9 that was observed for the mean avalanche size , supporting the suggestion of sect .
[ sect : flowratefluctus ] about the relationship of reduced clogging probability and lessened flow rate fluctuations . on the contrary ,
the average flow rate and the mean avalanche size ( or the clogging probability ) do not seem to be related in any recognizable way .
from the pictures recorded in the experiment ( fig . [ fig : exp ] ) the position of the beads in the arches that block permanently the silo can be accurately obtained . as reported in a previous paper @xcite ,
the particles forming the base of the arch are not considered to belong to it , in compliance with the definition of arch given in numerical works @xcite . in practical terms , the particles in the base are those whose centers are at the extreme positions in the horizontal direction . in fig .
[ fig : exp ] the centers of the particles belonging to the arch are marked with crosses and the particles of the base are marked with circles .
once the particles forming the arch have been identified , a number of arch properties can be analyzed ( the size , measured in number of beads ; span ; height ; and aspect ratio ) .
the span is defined as the distance , projected on the horizontal direction , between the two outermost particles of the arch .
accordingly , the height is defined as the distance between the vertical coordinates of the centers corresponding to the highest and the lowest beads of the arch .
finally , the aspect ratio ( @xmath62 ) is calculated as the quotient between half the span and the height of the arch .
all these parameters will allow to explore if the presence of an obstacle induces any kind of difference in the geometrical properties of the arches . in this work
, we will compare the arch features in a silo with an orifice of @xmath50 mm in two different situations : without obstacle and with an obstacle at @xmath51 mm .
we note that for this value of @xmath8 , the obstacle placed at @xmath51 mm yields the maximum avalanche size observed ( fig .
[ fig : meanavas ] a ) .
( color online ) ( a ) normalized histogram of the size of the arches measured in number of beads ( @xmath63 ) .
( b ) normalized histogram of the arch height ( note the semilogarithmic scale ) .
( c ) normalized histogram of the arch span ( in semilogarithmic scale ) .
( d ) normalized histogram ( in semilogarithmic scale ) of the aspect ratios ( a ) of the arches , defined as half the span divided by the height . in all the graphs measurements
have been carried out in a silo with an outlet size @xmath50 mm , without an obstacle ( @xmath34 ) and with a @xmath41 mm diameter obstacle placed at @xmath51 mm ( @xmath32).,title="fig : " ] for these two cases , namely , no obstacle and an obstacle placed at @xmath51 mm , let us compare the arch size , measured in number of beads ( @xmath63 ) . in fig .
[ fig : shapearches]a we present the two arch size distributions . clearly , smaller arches ( with fewer particles ) are formed when there is an obstacle . the arch height distribution ( fig .
[ fig : shapearches]b ) and the arch span distribution ( fig .
[ fig : shapearches]c ) confirm this result : in the presence of an obstacle the arches are shorter and narrower .
the aspect ratio @xmath62 , plotted in fig .
[ fig : shapearches]d , shows how arches are flatter ( higher @xmath62 ) when there is an obstacle .
in addition , it seems that both distributions present an exponential decay in the aspect ratio probability .
the aspect ratio of an arch is a especially important variable , as it is used as a key ingredient in one of the few models linking the silo clogging probability to the arch properties @xcite . in this model
the arch is proposed to be semicircular , i.e. an aspect ratio equal to one . in an experimental work performed afterwards , the validity of this assumption was demonstrated for large arches ( broader than the outlet by more than one bead diameter ) @xcite .
in addition , the aspect ratio of the arches is appealing because it may reflect interesting features related with the loads sustained by them .
an aspect ratio smaller than one ( pointed arch ) would indicate that the arch is optimized to sustain a vertical pressure . on the contrary , an arch optimized to sustain a horizontal load would be flatter , exhibiting an aspect ratio larger than one . a semicircular arch ( aspect ratio one ) is the preferred shape to optimize an isotropic pressure @xcite . based on these assumptions , the results presented in fig .
[ fig : shapearches]d are consistent with the fact that an obstacle above the outlet screens the pressure of the particles in the silo .
if so , the flatter arches ( higher aspect ratio ) obtained when an obstacle is present could be the consequence of the load reduction in the vertical direction caused by the obstacle .
( color online ) average aspect ratio for arches as a function of their number of particles ( @xmath63 ) . as in fig .
[ fig : shapearches ] , the measurements have been carried out in a silo with an outlet size @xmath50 mm , without an obstacle ( @xmath34 ) and with a @xmath41 mm diameter obstacle placed at @xmath51 mm ( @xmath32 ) .
the error bars are the @xmath45 confidence intervals.,title="fig : " ] once we have seen that the arches developed in the presence of an obstacle are smaller ( in number of particles , height and span ) and flatter , we can consider if those are independent effects or else if one effect is caused by the other . indeed , in a previous work @xcite it was suggested that all these variables are strongly related . in fig .
[ fig : avsnumberparticles ] we present the values of the aspect ratios averaged over all the arches formed by a given number of particles .
the results obtained for silos without an obstacle and with an obstacle at @xmath51 mm are almost identical and confirm that the aspect ratio of the arches is one when they are sufficiently large .
the only difference between both situations is that arches in the silo without obstacle are formed by a higher number of particles reaching maximum values of 10 instead of 9 something that was expected from fig.[fig : shapearches ] .
the fact that the obtained results in both situations are so far indistinguishable , seems to indicate that the arches of a given number of particles have the same geometrical properties regardless of the presence of an obstacle .
the only effect of placing an obstacle is to prevent the formation of pointed arches ( aspect ratio smaller than one ) which , in general , consist of more particles .
it can not be conclusively asserted whether the aspect ratio is responsible for this behavior the cause could instead be ascribed to other variable , but the above mentioned relationship between the aspect ratio and the load distribution makes it a likely candidate .
we have presented a detailed analysis of the effect that placing an obstacle above the outlet of a silo has on the flow and clogging processes .
the measurement of the dependence of the avalanche size on the obstacle position , and this for different outlet sizes , reveals that the clogging reduction caused by the obstacle is enhanced as the outlet size is increased , thereby approaching the region of ` no clogging ' .
this phenomenon can be understood in terms of the robustness of the arches .
if we assume that arches become weaker as their size increases , it seems plausible that a similar pressure reduction has a stronger effect in greater arches than in smaller ones . although it is still not clear if there is a critical outlet size above which clogging is forbidden @xcite , if we assume that this boundary does exist in practical terms , then it makes sense that small perturbations have more consequence as the critical point is approached .
we have also presented results of the flow rate properties for different outlet sizes and obstacle positions . concerning the average flow rate
, it is shown that the obstacle can cause an increase up to around @xmath47 .
this increase of the flow rate is shown to be relatively more pronounced as the outlet size is reduced .
in other words , the placement of the obstacle has a stronger effect in the flow rate for small outlets .
this trend is opposite to the one observed for the avalanche size ( or the clogging probability ) .
indeed , it seems that there is not any relationship between the average flow rate and the probability of clogging . on the contrary
, we report a clear and strong relationship between the clogging reduction ( increasing of avalanche size ) and a decrease of the coefficient of variation in the flow rate fluctuations .
this could be of practical interest as it suggests that the measurement of this coefficient of variation would be enough to estimate the mean avalanche size .
the relationship among these two parameters may be signaling that the number of stable and unstable clogs is correlated . increasing the number of stable clogs leads to a decrease of the avalanche size , while increasing the number of unstable clogs provokes an enhancement of the fluctuations .
this behavior was already observed in previous works @xcite by varying the outlet size in a silo without an obstacle : by increasing the outlet size , the probability of clogging was reduced , as well as the fluctuations , but the price to pay was a corresponding increase of the average flow rate .
the interesting fact observed when placing an obstacle is that fluctuations and average flow rate are not necessarily linked .
the use of different obstacle sizes has allowed us to check that this effect is robust .
finally , we have performed an exhaustive comparison of the shape of the arches developed in a silo in the two situations where the pressure difference is apparently more significant : the silo without an obstacle and the silo with an obstacle placed at @xmath51 mm .
the results reveal that smaller and flatter arches are obtained in the silo with an obstacle the case in which the pressure at the outlet is reduced .
this can be qualitatively understood if we recall that pointed arch shapes optimize vertical loads ( which seem to be more important in the absence of an obstacle ) and flatter arches are the optimal response to horizontal loads ( which seem to be more important in the silo with obstacle ) . in summary ,
the experimental results presented in this work show that the placement of an obstacle has a robust effect in the silo clogging reduction .
in addition , we report convincing evidences that the mechanism by which this effect is attained is a pressure reduction near the orifice . however , other possibilities such as a modification of the flow streamlines similar to the one reported in @xcite should not be discarded .
another interesting analogy of this work can be drawn with experiments where a big obstacle moves slowly within a granular media @xcite . indeed in @xcite it is suggested that the stress fluctuations induced by an obstacle in a dense granular flow close to the jamming transition , may help to overcome the flow threshold .
further experiments and simulations would be interesting in order to characterize the pressure exerted by the granular media on the obstacle , as well as its fluctuations .
this work has been financially supported by projects fis2008 - 06034-c02 - 01 and fis2011 - 26675 ( spanish government ) and by piuna ( universidad de navarra ) .
thanks asociacin de amigos de la universidad de navarra for a scholarship .
i. zuriguel , a. garcimartn , d. maza , l. a. pugnaloni and j. m. pastor , phys .
e * 71 * , 051303 ( 2005 ) ; i. zuriguel , l. a. pugnaloni , a. garcimartn , and d. maza , phys . rev .
e * 68 * , 030301 ( r ) ( 2003 ) .
m. a. aguirre , j. g. grande , a. calvo , l. a. pugnaloni , and j .- c .
gminard , phys .
lett * 104 * , 238002 ( 2010 ) ; m. a. aguirre , j. g. grande , a. calvo , l. a. pugnaloni , and j .- c .
gminard , phys .
e * 83 * , 061305 ( 2011 ) . | in a recent paper [ zuriguel et al . , phys . rev .
lett . * 107 * , 278001 ( 2011 ) ] it has been shown that the presence of an obstacle above the outlet can significatively reduce the clogging probability of granular matter pouring from a silo .
the amount of this reduction strongly depends on the obstacle position . in this work ,
we present new measurements to analyze different outlet sizes , extending foregoing results and revealing that the effect of the obstacle is enhanced as the outlet size is increased . in addition , the effect of the obstacle position on the flow rate properties and in the geometrical features of arches is studied .
these results reinforce previous evidence of the pressure reduction induced by the obstacle .
in addition , it is shown how the mean avalanche size and the average flow rate are not necessarily linked .
on the other hand , a close relationship is suggested between the mean avalanche size and the flow rate fluctuations . |
supernovae ( sne ) play a central role in astrophysics , not only as distance indicators for cosmology ( e.g. , riess et al .
1998 ; perlmutter et al . 1999 ) , but as prime synthesizers of heavy elements ( e.g. woosley 2007 ) , sources of kinetic energy , and accelerators of cosmic rays ( e.g. helder et al . 2009 ) .
however , many of the most basic facts about these events are still poorly understood .
core - collapse ( cc ) sne are descended from massive stars , roughly in the range @xmath13 ( e.g. gal - yam & leonard 2009 ) , but the exact limits are not known , neither from theory nor from observation .
it is not yet clear what are the physical parameters ( mass , binarity , rotation , magnetic field , and more ) that determine the variety of observed cc - sn subtypes ( iip , ib , ic , iin , iil ; see filippenko 1997 , for a review ) .
type - ia sne ( sne ia ) are linked by indirect evidence to the thermonuclear detonations of carbon - oxygen white dwarfs ( wds ) whose mass has grown to near the chandrasekhar limit ( hoyle & fowler 1960 ) . however ,
competing scenarios exist for the initial conditions and evolutionary paths that lead to this mass growth . in the single degenerate ( sd ) model
( whelan & iben 1974 ) a wd grows in mass through accretion from a non - degenerate stellar companion . in the double degenerate ( dd ) model ( iben & tutukov 1984 ; webbink 1984 ) , two wds merge after losing energy and angular momentum to gravitational waves . in both scenarios ,
many question remain regarding the ignition and development of the explosion itself .
it is only recently that the relative and absolute rates of cc sne and sne ia as a function of environment and cosmic time are starting to be measured accurately , and therefore the respective quantity , types , and times of their contributions to metal enrichment history are still poorly constrained .
a fundamental function that could shed light on all of these issues is the sn delay time distribution ( dtd ) .
the dtd is the hypothetical sn rate versus time that would follow a brief burst of star formation .
the dtd is directly linked to the lifetimes ( i.e. , the initial masses ) of the progenitors and to the binary evolution timescales up to the explosion , and therefore different progenitor scenarios predict different dtds .
furthermore , the dtds of different sn types dictate directly the mix of contributions of different sn types to metal enrichment throughout cosmic history .
until recently , only few , and often - contradictory , observational constraints on the dtd existed .
one observational approach has been to compare the sn rate in field galaxies , as a function of redshift , to the cosmic star formation history ( sfh ) . given that the dtd is the sn `` response '' to a short burst of star formation , the sn rate versus cosmic time will be the convolution of the full sfh with the dtd .
gal - yam & maoz ( 2004 ) carried out the first such comparison , using a small sample of sne ia out to @xmath14 , and concluded that the results were strongly dependent on the poorly known cosmic sfh , a conclusion echoed by forster et al .
( 2006 ) . with the availability of sn rate measurements to higher redshifts , barris & tonry ( 2006 ) found a sn ia rate that closely tracks the sfh out to @xmath15 , and concluded that the dtd must be concentrated at short delays , @xmath16 gyr .
similar conclusions have been reached , at least out to @xmath17 , by sullivan et al .
in contrast , dahlen et al . ( 2004 , 2008 ) and strolger ( 2004 ) have argued for a dtd that is peaked at a delay of @xmath18 gyr , with little power at short delays , based on a decrease in the sn ia rate at @xmath19 .
however , kuznetzova et al .
( 2007 ) have re - analyzed some of these datasets and concluded that the small numbers of sne and their potential classification errors preclude reaching a conclusion .
similarly , poznanski et al . (
2007 ) performed new measurements of the @xmath19 sn ia rate , and found that , within uncertainties , the sn rate could be tracking the sfh .
this , again , would imply a short delay time .
greggio et al .
( 2008 ) pointed out that underestimated extinction of the highest-@xmath8 sne , observed in their rest - frame ultraviolet emission , could be an additional factor affecting these results . a second approach to recovering the dtd has been to compare the sn rates in galaxy populations of different characteristic ages . using this approach , mannucci et al .
( 2005 , 2006 ) , scannapieco & bildsten ( 2005 ) , and sullivan ( 2006 ) all found evidence for the co - existence of two sn ia populations , a `` prompt '' population that explodes within @xmath20 myr , and a delayed channel that produces sne ia on timescales of order 5 gyr .
naturally , these two `` channels '' may in reality be just integrals over a continuous dtd on two sides of some time border ( greggio et al .
totani et al .
( 2008 ) have used a similar approach to recover the dtd , by comparing sn ia rates in early - type galaxies of different characteristic ages , seen at @xmath21 as part of the subaru / xmm - newton deep survey ( sxds ) project .
they find a dtd consistent with a @xmath22 form , which is roughly generic for dd models ( e.g. greggio 2005 ; see [ sspowerlawdtd ] , below ) .
additional recent attempts to address the issue with the `` rate vs. age '' approach have been made by aubourg et al .
( 2008 ) , raskin et al .
( 2009 ) , yasuda & fukugita ( 2009 ) , and cooper et al .
( 2009 ) .
both of the above approaches involve averaging , and hence some loss of information . in the first approach , one averages over large galaxy populations , by associating all of the sne detected at a given redshift with all the galaxies of a particular type at that redshift . in the second approach , a characteristic age for a sample of galaxies
replaces the detailed sfh of the individual galaxies in a sn survey .
maoz et al .
( 2010 ) recently presented a method for recovering the dtd , which avoids this averaging . in the method ,
the sfh of every individual galaxy , or even galaxy subunit , is compared to the number of sne it hosted in the survey ( generally none , sometimes one , rarely more ) .
dtd recovery is treated as a linearized inverse problem , which is solved statistically .
maoz et al .
( 2010 ) applied the method to a subsample of the galaxies in the lick observatory sn search ( filippenko et al .
2010 ; leaman et al . 2010
; li et al .
2010a , b ) , having sfh reconstructions by tojeiro et al . ( 2009 ) based on data from the sloan digital sky survey ( sdss ; york et al , 2000 ) .
maoz et al .
( 2010 ) find a significant detection of both a prompt sn ia component , that explodes within 420 myr of star formation , and a delayed sn ia with population that explodes after @xmath23 gyr .
maoz & badenes ( 2010 ) applied this method also to a sample of sn remnants in the magellanic clouds , compiled by badenes , maoz , & draine ( 2010 ) . treating the remnants as an effective sn survey conducted over @xmath24 kyr
, they also find a significant detection of a prompt ( this time @xmath25 myr ) sn ia component .
a related dtd reconstruction method has been applied by brandt et al .
( 2010 ) to the sne ia from the sdss ii survey . like maoz et al .
( 2010 ) , they detect both a prompt and a delayed sn ia population .
yet another approach for recovering the dtd , which is at the focus of this paper , is to measure the sn rate vs. redshift in massive galaxy clusters .
the deep potential wells of clusters , combined with their relatively simple sfhs , make them ideal locations for studying both the dtd and the metal production of sne .
optical spectroscopy and multiwavelength photometry of cluster galaxies has shown consistently that the bulk of their stars were formed within short episodes ( @xmath26 myr ) at @xmath27 ( e.g. , daddi et al . 2000 ; stanford et al . 2005 ; eisenhardt et al .
thus , the observed sn rate vs. the elapsed cosmic time since the stellar formation epoch provides an almost ( see [ sscomparisontopred ] ) direct measurement of the form of the dtd .
furthermore , the record of metals stored in the intracluster medium ( icm ) constrains the number of sne that have exploded , and hence the normalization of the dtd .
renzini et al .
( 1993 ) and renzini ( 1997 ) first pointed out that the large mass of iron in the icm could not have been produced solely by cc sne from a stellar population with a standard initial mass function ( imf ) .
either a `` top - heavy '' imf , or a dominant contribution by sne ia is required .
maoz & gal - yam ( 2004 ) revisited the problem , and tested the hypothesis that the main source of iron is sne ia .
accounting for the cc sn contribution to the observed iron mass , they calculated the number of sne ia , per present - day stellar luminosity , needed to have exploded over the entire past history of a cluster , in order to produce the observed iron mass .
finally , they compared some simple theoretical dtds , normalized to produce this number of sne , to the then - available cluster sn ia rates ( gal - yam , maoz , & sharon 2002 ; reiss 2000 ) .
the low observed sn ia rates out to @xmath15 implied that the large number of events , needed to produce the bulk of the iron , occurred at even higher redshifts , beyond the range of the then - existing observations .
maoz & gal - yam ( 2004 ) therefore concluded that sne ia can be the main source of iron only if most of them explode during the relatively brief time interval between star formation in massive clusters ( at @xmath28 ) and the highest - redshift cluster sn rate measurements ( at @xmath15 ) .
in other words , the majority of sne ia ( at least those that occur in a galaxy cluster environment ) must have a short time delay , @xmath29gyr .
a shortcoming of the work by maoz & gal - yam ( 2004 ) was the paucity and quality of the cluster sn rate data it was based on .
the low-@xmath8 cluster sn rate by reiss ( 2000 ) was never published in the refereed literature .
the cluster sn rates at @xmath30 by gal - yam et al .
( 2002 ) , were based on very few events two or three discovered in archival _
hubble space telescope _ ( hst )
wfpc2 images of clusters .
the large poisson and systematic uncertainties in the resulting rates precluded any detailed discrimination among model dtds . over the past few years
, the observational situation has improved dramatically .
accurate cluster sn ia rates have been measured in the redshift range of 0 to 1.5 , with smaller errors than the previous sole published study at @xmath31 by gal - yam et al .
( 2002 ) . in this paper
, we utilize this wealth of new measurements for a renewed analysis of the cluster sn rate and its implications for the sn ia dtd and for cluster metal - enrichment history .
we show that the new data permit a direct comparison of the observations with the functional behavior of various dtds that have been proposed in the literature , whether parametrized - mathematical or phenomenological dtds , or dtds resulting from physical considerations at different levels of sophistication .
we find that few of these models , in their simplest forms , are compatible with the emerging observational picture , and therefore we have reached the point where the observations discriminate among models . in 2 , below , we compile the existing cluster sn ia measurements . in 3 , we review the existing measurements of the properties of clusters , particularly the gas - to - stellar mass ratio , that are relevant for estimating the present - day cluster iron - to - stellar mass ratio .
this ratio , in turn , sets the number of sn ia that have exploded in clusters , per unit stellar mass formed , which then fixes the normalization of the dtd . in
4 we use the rates and the normalization to recover the observational dtd . we take a forward - modeling approach starting in 5 , where we compare the observed rates to the dtd predictions of various `` single - component '' models that invoke a single mathematical form or physical progenitor formation channel . in [ ss.noninst ] , we relax the assumption of a single , instantaneous , star - formation episode in clusters , and examine whether an extended star - formation history , combined with some single dtd , can reproduce the measurements . in 7 , we examine the ability of composite models , which mix diverse dtds , or multiple episodes of star formation , to match the observations .
we summarize our results in 8 . throughout the paper
we assume a cosmology with parameters @xmath32 , @xmath33 , and @xmath34 km s@xmath35 mpc@xmath35 .
to perform our analysis , we first compile the cluster sn ia rate measurements currently in existence .
the recent measurements are at mean redshifts of @xmath36 ( mannucci et al 2008 ) ; @xmath37 ( sharon et al . 2007 ;
gal - yam et al .
2008 ) ; @xmath38 ( dilday et al . 2010 ) ; @xmath39 ( graham et al . 2008 ) ; @xmath40 ( sharon et al . 2010 ) ; and @xmath41 ( barbary et al .
the mean redshifts are visibility - time - weighted means for the clusters monitored by each survey , i.e. , in the calculation of the mean , the redshift of each cluster in the survey sample is weighted by the effective time during which a sn ia would have been visible in the survey ( see , e.g. , sharon et al . 2007 ) . before these recent measurements ,
the sole modern published cluster rates were at @xmath31 ( gal - yam et al .
table 1 summarizes , for each of these measurements , the mean redshift of the sample , the redshift range , the cosmic times corresponding to these redshifts in our assumed cosmology , and the sn ia rate , normalized by stellar mass .
the stellar masses , at the cosmic time corresponding to the redshifts of the clusters , have generally been estimated consistently in the respective papers , based on the observed stellar luminosities and colors of the monitored cluster galaxies , and assuming the same imf the `` diet salpeter '' imf of bell et al .
exceptions to this are : dilday et al .
( 2010 ) , who converted their luminosity - normalized rates to mass - normalized rates assuming that the mean @xmath42 ratio found by sharon et al .
( 2007 ) is valid for their cluster sample , which spans a similar redshift range ; gal - yam et al .
( 2002 ) , whose rates were mass normalized by sharon et al .
( 2010 ) , using @xmath42 from sharon et al .
( 2007 ) at low @xmath8 and @xmath42 from sharon et al .
( 2010 ) at high @xmath8 ; and graham et al .
( 2008 ) , who estimate masses for the galaxies in their sample using the spectral population synthesis of buzzoni ( 2005 ) who , in turn , assumed a salpeter ( 1955 ) imf .
we therefore scale up the graham et al .
( 2008 ) rate by a factor of 1.77 , which is the ratio between the remaining stellar mass in an old and inactive quiescent population in a salpeter ( 1955 ) imf , and that in a diet - salpeter imf ( see [ sscccont ] , below , for details ) .
we follow barbary et al .
( 2010 ) in scaling up the rate of sharon et al .
( 2010 ) by 35% , to account for the offset in the @xmath43 vs. @xmath44 relation of bell et al .
( 2003 ) , expected in the younger galaxies that exist at @xmath45 .
crcrrl 0.020@xmath46&13.2@xmath47&0.066@xmath48 & 11.1@xmath49 & 3.6@xmath50&m08 + 0.084@xmath51&12.4@xmath52&0.060@xmath53 & 10.3@xmath54 & 3.3@xmath55&d10 + 0.150@xmath56&11.6@xmath57&0.098@xmath58 & 9.5@xmath59 & 5.5@xmath60&s07 + 0.225@xmath61&10.8@xmath62&0.088@xmath63 & 8.7@xmath64 & 4.9@xmath65&d10 + 0.250@xmath66&10.5@xmath67&0.110@xmath68 & 8.4@xmath69 & 6.2@xmath70&gy02 + 0.460@xmath71 & 8.7@xmath72&0.177@xmath73 & 6.6@xmath74 & 10.1@xmath75&g08 + 0.600@xmath76 & 7.8@xmath77&0.151@xmath78 & 5.7@xmath79 & 8.7@xmath80&s10 + 0.900@xmath81 & 6.2@xmath82&0.220@xmath83 & 4.1@xmath84 & 12.9@xmath85&gy02 + 1.120@xmath86 & 5.3@xmath87&0.364@xmath88 & 3.2@xmath89 & 21.6@xmath90&b10 + & & & 1.1@xmath91 & 230 @xmath92&txt + .
figure [ figobsrates ] shows the rates we have compiled , and which we analyze in this paper .
it suggests only a mild rise in the rates , by a factor of @xmath93 , over the redshift interval probed .
remarkably , the cluster sn rate appears not to change much from @xmath0 to @xmath1 , corresponding to a lookback time of 9.8 gyr .
this is only about 2 gyr after the formation of the cluster stars , assuming this occurred at a redshift @xmath94 .
as pointed out by maoz & gal - yam ( 2004 ) and sharon et al .
( 2010 ) , and now further suggested by the results above , the integral over the sn ia rate from @xmath0 to @xmath1 , times the mean iron yield of sne ia ( @xmath95 ; see below ) , gives just a small fraction , roughly 10% , of the present - day ratio of iron mass to stellar mass observed in clusters .
it is then unavoidable that the large majority of cluster metals were produced within the first 1 - 2 gyrs after the formation of the cluster stars .
the metals were produced by cc sne and perhaps by `` prompt '' sne ia , but they were not produced by any sne with a larger delay , as the rate measurements show that only few such explosions took place at those later times . in order to juxtapose the new cluster rates , compiled above , with model expectations , we revisit the question of the iron - to - stellar mass ratio in clusters , which dictates the normalization of the dtd .
in addition to considering the latest data on cluster properties , our treatment will differ from the one in maoz & gal - yam ( 2004 ) in several respects , most notably in that we will use mass - normalized , rather than luminosity - normalized sn rates , and we will account for mass evolution due to stellar mass loss .
stellar mass is , of course , more closely related to the number of stars than is blue luminosity , and is a more stable quantity .
the mass of iron in clusters that is attributable to sne ia is the sum of the iron masses in the icm and in stars , minus the mass of iron produced by cc sne .
thus , the ratio between sn ia - produced iron mass and the present - day stellar mass is @xmath96 @xmath97.\ ] ] we discuss below each of the quantities that enter this equation .
the value of the photospheric solar abundance of iron , @xmath98 , is from anders and grevesse ( 1989 ) .
although the correct photospheric iron abundance has been revised down to 0.00176 ( grevesse & sauval 1999 ) , the icm and stellar abundances we quote below , and which are still generally used in the literature , relate to the anders and grevesse ( 1989 ) value lower than we find here . ] .
the present - day mass ratio of baryons in the icm gas and in stars , @xmath99 , has been recently re - evaluated by gonzalez , zaritsky , & zabludoff ( 2007 ) , taking into account the contribution to @xmath100 by the intergalactic stellar population of a cluster .
contrary to lin et al .
( 2003 ) , who argued for @xmath99 that is independent of total cluster mass , gonzalez et al .
( 2007 ) find that @xmath101 rises monotonically with cluster mass , from a value of about 5 for low - mass clusters to about 15 for the most - massive systems . to find the suitable values for the clusters
from which were derived the sn rates that we analyze , we now examine the typical masses of those clusters .
the most massive cluster sample in our sn rate compilation is that of sharon et al .
( 2010 ) , which is largely coincident with the @xmath102 macs cluster sample of ebeling et al .
the clusters in this sample have total masses @xmath103 , where @xmath104 is the mass within a spherical volume in which the mean density is 500 times the critical closure density . in the sharon et al .
( 2010 ) sample , the part of the rest - frame @xmath105-band luminosity of each cluster that is included in the hst field of view ( and which is therefore monitored for sne ) is @xmath106 , and the corresponding stellar mass range is @xmath107 , with a mean of @xmath108 .
such masses also apply to the massive , x - ray - selected clusters at @xmath31 studied by gal - yam et al .
for the @xmath37 clusters monitored by gal - yam et al .
( 2008 ) , the typical stellar mass is @xmath109 ( sharon et al .
2007 ) . from gonzalez et al .
( 2007 ) , this corresponds to total masses of @xmath110 .
barbary et al . (
2010 ) find , for their @xmath41 clusters , a mean @xmath105-band luminosity , of @xmath111 , and typical stellar masses of @xmath112 . as a check , the bulk properties of the barbary et al .
( 2010 ) clusters , such as x - ray luminosities , x - ray temperatures , and velocity dispersions , also suggest masses about one - half as large as those of the cluster sample of sharon et al .
( 2010 ) at @xmath40 .
well - measured x - ray temperatures exist in the literature for five of the barbary et al .
sample clusters , and are all around 6 - 7 kev ( rosati et al .
2004 ; stanford et al .
2002 , 2006 ; boehringer et al . 2008
; gilbank et al . 2008 ) .
this compares to the mean temperature , @xmath113 kev , of the @xmath102 macs cluster sample of ebeling et al .
x - ray luminosities , reported for these five barbary et al .
clusters , and for an additional four of their clusters ( rosati et al . 1999 ; postman et al . 2001 ; bremer et al .
2006 ; andreon et al .
2008 ) are in the range @xmath114 , with a mean of @xmath115 , compared to @xmath116 for the ebeling et al .
( 2007 ) sample .
galaxy velocity dispersion best - fit measurements for seven of the barbary et al .
clusters are in the range of @xmath117 ( references above , plus eisenhardt et al .
2008 ; cain et al .
2008 ) , with a mean of about @xmath118 , compared to @xmath119 for the @xmath102 macs clusters .
since total mass depends ( e.g. , hicks et al .
2008 ) on x - ray temperature roughly as @xmath120 , on x - ray luminosity roughly as @xmath121 , and on velocity dispersion as @xmath122 , all of these observables suggest that the typical cluster in the barbary et al .
sample has about one - half the mass of the typical cluster in the sharon et al .
( 2010 ) sample . assuming a constant fraction of the cluster mass in stars , the stellar mass in the barbary et al
. sample would also be of order one - half that in the sharon et al .
( 2010 ) sample .
however , the stellar mass fraction in clusters depends on total cluster mass .
for example , andreon ( 2010 ) has recently found a dependence of roughly @xmath123 .
if this relation held also for clusters at redshifts @xmath19 , the stellar mass in the barbary et al .
clusters would be only about 1.4 times lower than in the sharon et al .
( 2010 ) sample , and thus more similar to the @xmath37 clusters monitored by gal - yam et al .
( 2008 ) , the optically selected clusters monitored for sne by graham et al .
( 2008 ) and dilday et al .
( 2010 ) have masses comparable to those of the samples discussed above , though somewhat lower .
graham et al .
( 2008 ) derived a total stellar mass of @xmath124 for the 30 clusters they used for their sn rate measurement , or @xmath125 for a typical cluster .
dilday et al .
( 2010 ) report a total @xmath126-band luminosity of @xmath127 for the 71 clusters in their low-@xmath8 , `` c4 '' , cluster subsample . taking @xmath128 ( sharon et al .
2007 ) , this gives a mean stellar mass per cluster of @xmath129 .
for the 492 clusters in the @xmath130 `` maxbcg '' sample of dilday et al .
( 2010 ) , the integrated luminosity is @xmath131 , and hence the mean stellar mass per cluster is @xmath132 .
thus , this one subsample of dilday et al .
( 2010 ) has clusters that are undermassive , compared to the other cluster samples we analyze .
the other samples have only about a factor-3 difference in mean cluster mass between the higher- and lower - mass extremes . from the fits by gonzalez et al .
( 2007 ) , the lower and higher typical masses considered above ( excluding the lower - mass maxbcg sample of dilday et al .
2010 ) correspond to @xmath99 values of 6.5 and 14.5 , respectively .
however , gonzalez et al . ( 2007 ) used the relation obtained by cappellari et al .
( 2006 ) between the kinematically measured total - mass - to - luminosity ratio , @xmath133 of nearby ellipticals and their @xmath134-band luminosity .
cappellari et al . , when comparing their results to @xmath133 estimates for the same galaxies based on spectral population synthesis , concluded that 30% of the contribution to their kinematic @xmath133 could be due to dark matter within the central regions .
if so , the stellar masses in clusters found by gonzalez et al .
( 2007 ) would be scaled down by 0.7 , and hence @xmath99 would increase to 9.3 for , e.g. , the @xmath37 cluster sample , and to 21 for the @xmath40 cluster sample .
lagan et al . (
2008 ) re - analyzed x - ray and optical data for five clusters to deduce the stellar and gas mass fractions . to obtain stellar masses
, they used the kauffmann et al .
( 2003 ) @xmath133 ratios , which are consistent with the bell et al .
( 2003 ) @xmath133 ratios , used in deriving the sn rates analyzed here . for clusters with total masses , @xmath104 , similar to the two ends of the typical mass range considered here , lagan et al .
( 2008 ) find @xmath99 of about 8 and 17 , respectively .
giodini et al .
( 2009 ) also recently examined the dependence of stellar and gas mass fractions on @xmath104 .
gas mass fractions were estimated for 41 clusters compiled by pratt et al .
( 2009 ) from the literature .
stellar mass fractions were obtained for a sample of 91 groups and poor clusters from the cosmos survey , based on spectral population synthesis fitting of their spectral energy distributions .
these were combined with the re - evaluated stellar masses of 27 clusters from lin et al .
we scale down their stellar masses by 0.7 , to transform from the salpeter ( 1955 ) imf they assumed to our assumed `` diet salpeter '' imf ( see below ) .
for the two ends of the cluster mass range considered above , the best - fit relations of giodini et al .
( 2009 ) then give @xmath99 of about 7 and 11 , respectively . finally , a recent analysis by andreon ( 2010 ) of 52 clusters , chosen to have accurate measurements , results in similar relations , with values of @xmath99 of about 7 and 13.5 for the two typical cluster masses , respectively .
considering all of the above studies , we see that the estimates of @xmath99 are in the range of @xmath135 for the lower - mass clusters of the type monitored by most of the low-@xmath8 surveys whose sn rates we analyze , @xmath136 for the massive , intermediate-@xmath8 clusters monitored by gal - yam et al .
( 2002 ) and by sharon et al .
( 2010 ) , and somewhere inside this range for the high-@xmath8 cluster sample of barbary et al .
( we also note that these ratios appear to be independent of redshift ; e.g. giodini et al .
when considering the iron mass in clusters and the number of sne ia needed to produce it , we therefore choose @xmath137 as our `` optimal '' fiducial value , but we will also consider the consequences of a `` minimal iron '' value of @xmath138 .
a `` canonical '' value of @xmath139 is by now fairly well - established for the icm gas iron abundance , relative to solar .
balestra et al .
( 2006 ) , maughan et al .
( 2008 ) , and anderson et al .
( 2009 ) argue for values that have evolved even higher , to @xmath140 , at low redshifts ( but see ehlert & ulmer 2009 ) .
to err conservatively , we adopt @xmath141 for out fiducial value .
these iron abundances typically relate to the central @xmath142 mpc radius of massive clusters , the same region in which the stellar luminosity is measured and the sne are detected by the cluster sn surveys we consider .
nevertheless , because the x - ray emission is strongly centrally peaked , there is some concern that the above iron abundance could be strongly biased by a high abundance in the very core region , which might be unrepresentatative of a much lower abundance existing over most of the volume within 1 mpc .
however , maughan et al . ( 2008 ) have shown that measuring iron abundances within an annulus that excludes the central 15% in radius reduces the results only mildly , to values still consistent with the canonical @xmath141 .
for the stellar iron abundance relative to solar , we follow maoz & gal - yam ( 2004 ) and lin et al .
( 2003 ) in adopting the value @xmath143 from the study by jorgensen et al .
( 1996 ) .
@xmath144 is the ratio between the initial mass in stars of mass above @xmath145 , which will explode as cc sne , and the present - day mass in stars in clusters .
maoz & gal - yam ( 2004 ) examined the dependence of this ratio on the choice of imf for a large range of standard and non - standard imfs . in the present work
, we will consistently assume the `` diet salpeter '' imf of bell et al .
( 2003 ) , as this is the imf that has been assumed when determining rates per unit mass in the sn surveys that we analyze .
this imf is like the salpeter ( 1955 ) single - power - law imf with slope @xmath146 between 0.1 and 100@xmath147 .
however , when calculating observables involving mass , such as @xmath133 , the total initial mass in stars is scaled down by a factor of @xmath148 , to simulate the deficit , relative to the salpeter imf , of low - mass stars in realistic imfs .
the ratio of mass in stars of mass above @xmath145 to the total initial mass , in such an imf , is 0.20 . from bruzual & charlot ( 2003 ) , for a salpeter ( 1955 ) imf , during the stellar evolution of a 10 gyr - old population 31% of the stellar mass is returned to the interstellar medium ( ism ) via stellar winds and sn explosions . for a diet salpeter imf , this mass - loss fraction is @xmath149 , and hence the present - day ratio @xmath150 .
this is , in the present context , a conservative estimate , in that it maximizes the contribution of cc sne , relative to sne ia , to cluster iron production . if , as generally believed , stars in some mass ranges collapse directly into black holes , without a sn explosion and its contribution to iron production , then the appropriate mass ratio to be used here would be lower .
for example , if only stars up to @xmath151 or @xmath152 explode as cc sne , @xmath144 will be reduced to @xmath153 or @xmath154 , respectively .
the diet salpeter imf gives a very similar @xmath144 ratio to that of the kroupa ( 2001 ) and gould et al .
( 1997 ) imfs ( see maoz & gal - yam 2004 and bell et al .
2003 ) . in any case ,
since both the sn rate and the iron mass in clusters are normalized relative to stellar mass , as derived from observed stellar luminosity , the conclusions are insensitive to the assumed form of the imf in the low - mass range , as long as it is assumed consistently for both .
the ratio @xmath144 also depends weakly on the assumed age of the population , as long as it is of order several gyr or more , since most of the return of mass to the ism occurs early on .
finally , we follow maoz & gal - yam ( 2004 ) in assuming that the iron yield of a cc sn , @xmath155 , is 1% of the initial mass of the progenitor star . as discussed there , this is likely to be a conservative overestimate as well . for @xmath156 and
the fiducial values of the metallicities , this means about 1/3 of the cluster iron mass was produced by cc sne . for @xmath157 ,
less than 1/5 of the iron mass is from cc sne .
alternatively , for the `` minimal iron '' value of @xmath138 , the core - collapse contribution is between 1/2 and 1/4 . dividing eq .
[ irontostar ] by the mean iron yield of sne ia gives the time - integrated number of sne ia that have exploded in clusters , per present - day unit stellar mass , @xmath158 .
the mean iron yield of a sn ia is @xmath159 ( e.g. , mazzali et al .
2007 ) . in line with our conservative approach
, we will assume the higher value of @xmath160 , thus lowering the integrated number of sne ia required to explain the observed abundances .
the time - integrated number of sne ia per unit _ formed _ stellar mass , before the 44% mass loss that occurs over the course of @xmath161 gyr ( see [ sscccont ] , above ) , will be 0.56 times lower .
multiplying eq .
[ irontostar ] by @xmath162 , we thus obtain the time - integrated number of cluster sne ia per unit formed stellar mass , @xmath163 assuming the `` optimal '' value of @xmath137 , and the other fiducial values in eq .
[ irontostar ] .
alternatively , for the `` minimal '' value of @xmath138 , we obtain @xmath164
in this section , we combine the observed cluster sn rates , @xmath165 ( from [ ssobsrates ] ) and the integrated sn ia number per formed stellar mass , @xmath166 ( from [ ssnoverm ] ) , to recover the sn ia dtd . in the next section ,
we take an alternative , forward - modeling , approach , to examine the predictions of a variety of specific model dtds . throughout this section ,
we assume that all stars in clusters formed in a single instantaneous burst , neglecting any subsequent star formation . as already noted above , this is a fair approximation to the results of spectral synthesis of the mass - dominant stellar populations in cluster ellipticals , indicating a burst lasting @xmath167 gyr at @xmath94 ( corresponding to cosmic time @xmath168 gyr ) , i.e. , some 11.5 gyr ago .
the consequences of relaxing the instantaneous - burst assumption are studied in [ ss.noninst]-[ssdoublecomp ] , below .
if , in fact , the burst assumption is valid , recovery of the dtd is straightforward .
the sn rate since the epoch of star formation differs from the dtd only in the stellar - mass normalization
the dtd is the sn rate normalized by unit stellar mass at the formation epoch , @xmath169 , while the sn rate is normalized by the remaining stellar mass , @xmath170 , at the cosmic epoch of the rate measurement . here , @xmath171 is the remaining fraction of the initially formed stellar mass at cosmic time @xmath172 ( since the big bang ) .
the observed sn rates per unit mass at cosmic times @xmath172 are the values of the dtd at delays @xmath173 , up to the correction , easily applied , to convert from the existing mass at time @xmath172 to the formed mass at @xmath174 , accounting for mass loss during stellar evolution .
the sn rate and the dtd are thus related by @xmath175 bruzual & charlot ( 2003 ) tabulate the relative accumulated stellar mass loss versus time , @xmath176 , following a burst of star formation , assuming a salpeter ( 1955 ) imf . for the `` diet salpeter '' imf ,
the remaining mass fraction is @xmath177 . starting 2.5 myr after star formation
, the remaining mass @xmath171 is well approximated as @xmath178 .
the predicted sn rate will therefore decline slightly less steeply than the dtd .
we convert the observed sn rates , @xmath179 , to the dtd , @xmath180 , using eq .
[ ddtmassloss ] with the full bruzual & charlot ( 2003 ) mass loss ( rather than the power - law approximation ) . to estimate the dtd at delay times earlier than those corresponding to the redshift , @xmath181 , of the most distant clusters monitored
, we can use the @xmath182 constraints from the iron observations .
the total number of sne per unit formed stellar mass between @xmath183 and @xmath181 is @xmath184 , where @xmath185 is the total number between @xmath0 and @xmath181 , obtained by integrating the dtd over that range . the mean dtd value in the time interval @xmath186 $ ]
, corresponding to the unobserved redshift range @xmath183 to @xmath181 , is just @xmath187 . to estimate @xmath185 , we integrate , using small time steps , over the mass - loss - corrected sn rates . at times that are covered by several measurements ,
we take a mean rate , weighted by the relative measurement errors . out to @xmath1
, we find an integrated sn to stellar mass ratio of @xmath188 .
subtracting this from the minimal - iron value of @xmath189 ( eq . [ eqminiron ] ) leaves @xmath190 in the time interval of 2.2 gyr between @xmath191 to @xmath192 , or a mean dtd value of 0.012 sne yr@xmath193 in this delay time interval of @xmath194 gyr . for the optimal - iron value ( eq .
[ eqoptiron ] ) of @xmath195 , the mean dtd value in this time bin is 0.023 sne yr@xmath193 .
we take this range of dtd values as the uncertainty , which is driven in this case mainly by the systematic uncertainty in the cluster gas - to - stellar mass ratio . for our best estimate
we take the optimal - iron value .
we note that this exercise implies that , for the minimal and optimal iron assumptions , respectively , 79% to 88% of the sne ia in clusters exploded before @xmath1 . .
figure [ figdtd ] shows the recovered dtd , whose values are also listed in table 1 .
the horizontal error bars mark the limits of each dtd time bin , while the vertical error bars show the summed ( not in quadrature ) and propagated statistical and sytematic errors . also shown in fig .
[ figdtd ] are three other recently recovered observational sn ia dtds .
one ( empty circles ) is from maoz et al .
( 2010 ) , obtained for a subsample of the local galaxies in the lick observatory sn search having individual sfh reconstructions based on sdss spectra .
we have plotted here the maoz et al .
( 2010 ) dtd from a subsample that excluded galaxies of hubble types sa to sbc , in order to reduce cross talk between time bins . due to the limited aperture of the sdss spectrograph fibers , there remains a `` leak '' of signal from the first to the second time bins , and therefore the plotted dtd level in the @xmath196 myr time bin is likely an underestimate of the true level ( see maoz et al .
another dtd shown ( empty triangles ) was found by maoz & badenes ( 2010 ) , using the sn remnants in the magellanic clouds as an effective sn survey . in this dtd
, there is a detection in one time bin , of @xmath197 myr , but only an upper limit on the dtd value between 330 myr to a hubble time , which is indicated .
finally , we show ( empty squares ) the dtd recovered by totani et al .
( 2008 ) , using sn ia candidates in sxds field elliptical galaxies at @xmath21 .
we have divided the dtd values of totani et al . by a factor @xmath198 , given by them , to convert from their normalization unit , of present - day @xmath199-band luminosity , to ours , of formed stellar mass .
we have then divided by a further factor 0.7 to convert from their assumed salpeter ( 1955 ) imf to our diet - salpeter imf .
we also show , for reference , two power laws , @xmath200 and @xmath201 , which are discussed further in [ sspowerlawdtd ] , below .
several facts are apparent in fig .
[ figdtd ] .
first , considering the different time bins of the different measurements , and the known systematics , the various measurements are generally consistent with each other in regions of overlap .
second , the dtd is a monotonically decreasing function , peaking at the earliest delays probed by the observations . finally all the plotted points appear to be generally consistent with the illustrated power - law dependences , which , remarkably , pass through each and every error box .
this consistency is tested quantitatively in [ sspowerlawdtd ] . with this new dtd derivation , combined with those by maoz et al .
( 2010 ) and maoz & badenes ( 2010 ) , we thus confirm and extend the results of totani et al .
( 2008 ) , who found in their field elliptical sn sample a best - fit dtd dependence of a power law , @xmath202 , with @xmath203 .
this is also further discussed below , in [ sspowerlawdtd ] .
in this section , we take a forward - modeling approach , to compare the predictions of specific model dtds to the observations . foward modeling is advantageous in that the actual data are not manipulated and errors need not be propagated .
furthermore , this approach will permit us later ( [ ss.noninst]-[ssdoublecomp ] ) to relax the assumption of a single short burst of star formation in clusters .
a model dtd that we test , @xmath180 , in order to be consistent with observed cluster iron abundances , needs to have a normalization such that its integral over a cluster stellar age , @xmath204 , agrees with the above , time - integrated , sn numbers : @xmath205 this is one constraint on the dtd , imposed by the cluster iron abundances .
a second set of constraints is imposed by the observed time dependence of the sn rate , which can be compared to the model dtd predictions .
the predictions will be tested with the @xmath206 figure of merit .
the sn rate at each visibility - time - weighted mean redshift is compared with the model prediction , with the latter averaged over the redshift interval spanned by the cluster sample used for deriving each sn rate .
errors on the observed rates are generally asymmetric . for the @xmath206 calculation
, we use the error that is on the model side of each data point . throughout this section
, we assume that all stars in clusters formed in a single instantaneous burst , neglecting any subsequent star formation .
as already noted above , this is a fair approximation to the results of spectral synthesis of the mass - dominant stellar populations in cluster ellipticals , indicating a short burst at @xmath94 . as noted in
[ ssdtd ] , the sn rate since the epoch of star formation differs from the dtd only in the stellar - mass normalization , whether formed mass , @xmath169 , or remaining stellar mass , @xmath170 , at the epoch of the rate measurement , and we again use eq . [ ddtmassloss ] , this time to convert the dtd model , @xmath180 , to a rate prediction , @xmath165 .
various forms have been proposed for the dtd , some derived from detailed binary population synthesis calculations ( e.g. , yungelson & livio 2000 ; han & podsiadlowski 2004 ; ruiter et al .
2009 ; bogomazov & tutukov et al . 2009 ; mennekens et al . 2010 ) ; some physically motivated mathematical parameterizations , with varying degrees of sophistication ( e.g. , madau et al . 1998 ; greggio 2005 ; totani et al . 2008 ) ; and some ad hoc formulations intended to reproduce the observed field sn rate evolution ( e.g. , strolger et al . 2004 ) .
we now attempt to fit the observed cluster sn ia rates with some of these proposed dtds , using eq .
[ ddtmassloss ] , while simultaneously satisfying the dtd normalization constraint , eq .
[ normconstraint ] , set by the iron abundances in clusters .
a first and simple mathematical parametrization of the dtd that we compare to the observed cluster sn rates is a power law in time , @xmath207 .
power laws have been long considered as possible forms of the dtd ( e.g. , ciotti et al .
1991 ; sadat et al .
1998 ) . as noted by previous authors ( e.g. ,
greggio 2005 ; totani et al . 2008 ) a power - law dependence is generic to models ( such as the dd model ) in which the event rate ultimately depends on the loss of energy and angular momentum to gravitational radiation by the progenitor binary system .
if the dynamics are controlled solely by gravitational wave losses , the time @xmath172 until a merger depends on the binary separation @xmath208 as @xmath209 if the separations are distributed as a power law @xmath210 then the event rate will be @xmath211 for a fairly large range around @xmath212 , which describes well the observed distribution of initial separations of non - interacting binaries ( see maoz 2008 for a review of the issue in the present context ) , the dtd will have a power - law dependence with index not far from @xmath213 .
however , in reality , the binary separation distribution of wds that have emerged from their common envelope phase could be radically different , given the complexity of the physics of that phase .
thus , the @xmath214 dtd dependence of the dd channel can not be considered unavoidable .
be that as it may , the dtd reconstructions by totani et al .
( 2008 ) , maoz et al .
( 2010 ) , maoz & badenes ( 2010 ) , and in this work ( [ ssdtd ] ) , all point to a @xmath9 power - law .
a different power - law dtd dependence , with different physical motivation , has been proposed by pritchet et al .
if the time between formation of a wd and its explosion as a sn ia is always brief compared to the formation time of the wd , the dtd will simply be proportional to the formation rate of wds .
assuming that the main - sequence lifetime of a star depends on its initial mass , @xmath215 , as a power law , @xmath216 and assuming the imf is also a power law , @xmath217 then the wd formation rate , and hence the dtd , will be @xmath218 for the commonly used value of @xmath219 and the salpeter ( 1955 ) slope of @xmath220 , the resulting power - law index is @xmath221 , or roughly @xmath222 .
pritchet et al .
( 2008 ) have argued that such a @xmath7 form for the dtd can explain the trend of sn ia rate versus specific star - formation rate of the host population in the supernova legacy survey .
we test model power - law dtds with these two particular power - law indices , @xmath213 , and @xmath222 , as well as a continuous range of indices , @xmath223 , against the cluster data .
we assume @xmath224 for @xmath225 myr , corresponding to the lifetime of @xmath145 stars ( e.g , girardi et al .
2000 ) , at the border between cc and wd formation .
we start by setting the star - formation redshift at @xmath191 , i.e. , at a cosmic time @xmath226 gyr .
fitting the observed cluster sn rates with a power law dtd having a free normalization and index @xmath223 , the best fit is obtained for @xmath227 .
the time - integrated sn number is @xmath228 for a power law of this slope , which is intermediate to the `` minimal iron '' value we have derived above ( eq . [ eqminiron ] ) and the `` optimal '' value ( eq . [ eqoptiron ] ) .
however , with the normalization thus unconstrained , @xmath206 depends weakly on @xmath223 , and there is a wide range of indices , @xmath229 , that give acceptable fits to the observed rates.(we will deem as `` acceptable '' those model with @xmath230 , which corresponds to a probability of @xmath231 for 7 d.o.f .
9 data points minus two free parameters , @xmath223 and the normalization . ) the steeper ones among these power laws also have sufficiently large integrals .
specifically , power laws with @xmath232 satisfy the minimal iron constraint , while shallower power laws , e.g. , @xmath233 , do not .
conversely , an @xmath233 power - law normalized to have the correct integral , when compared to the data , grossly overpredicts the observed rates , giving an unacceptably high reduced chi - square , @xmath234 , of 17 and 69 per degree of freedom ( d.o.f . ) , for the minimal and optimal normalizations , respectively .
( in the latter two fits , there are no free parameters that are adjusted to fit the data , and hence we have 9 degrees of freedom , as the number of independent data points . ) figure [ figplm0.5 ] shows these three fits .
thus , a @xmath7 power law , while it can describe well the time dependence of the cluster rates at @xmath235 , can not simultaneously produce the required time - integrated sn ia numbers .
if we force the minimal - iron constraint , the best - fit power law is @xmath236 , with an acceptable @xmath237 , as shown in fig .
[ figplm1 ] .
the low @xmath206 value indicates that the errors on at least some of the rates have been conservatively overestimated .
the combined constraints of the sn rate data and the minimal iron abundances limit the acceptable values of the power law index to @xmath238 . for the optimal iron constraint , the acceptable range of indices is @xmath239 , with @xmath240 for the best fit .
smaller , more realistic , rate uncertainties would reduce these allowed ranges .
we thus see that a simple parametrized model dtd of the form @xmath22 , or slightly steeper , can provide a good fit to the cluster sn ia rates while simultaneously providing a sufficient time - integrated number of sne ia to satisfy the iron - based constraints .
we note that these conclusions depend weakly on the chosen epoch of cluster star formation , @xmath174 . for a given power - law slope ,
acceptable fits are obtained for @xmath174 in the range 1.2 to 2.3 gyr , corresponding to @xmath241 to 2.8 , with a best fit that has a constant minimum @xmath206 in the range @xmath242 .
varying @xmath174 can not salvage power - law dtds such as @xmath7 , that are strongly ruled out because of the shallowness of their slope .
however , moving @xmath174 back can slightly increase the allowed range of slopes .
for example , for the optimal iron value , the maximal acceptable power - law slope can be raised from @xmath243 to @xmath244 by shifting @xmath183 back to 3.6 .
it is arguable that , instead of a single , @xmath9 power law , motivated by binary mergers , with this power law extending back to delays as short as 40 myr , there could be a `` bottleneck '' in the supply of progenitor systems below some delay .
such a bottleneck could be due to the birth rate of wds , which behaves as @xmath245 .
one possible result would then be a broken - power - law dtd , with @xmath246 up to some characteristic time , @xmath247 , and @xmath248 thereafter .
a possible value could be @xmath249 myr , corresponding to the lifetimes of @xmath250 stars .
if that were the lowest initial mass of stars that can produce the wd primary in a dd sn ia progenitor , then beyond @xmath247 the supply of new systems would go to zero , and the sn ia rate would be dictated by the merger rate . we have therefore attempted to fit the data with such a @xmath7 , @xmath22 broken power law . with
the dtd normalization fixed to produce the required mimimum iron integrated sn numbers , this model gives an acceptable @xmath251 , though this is considerably worse than the single power - law fit , and could become unacceptable with more realistic errors .
we test the optimal iron value with a @xmath7 , @xmath252 broken power law , that is steeper at late times ( since even @xmath22 alone is already rejected , see above ) . here , too , an acceptable @xmath253 is found . for the minimal - iron normalization , a @xmath7 , @xmath236 dependence is acceptable , as long as @xmath254 gyr .
such a late break time is interesting in the context of sub - chandra merger models , in which the mergers of white dwarfs of initial masses smaller than @xmath255 produce sne ia ( sim et al .
2010 ; van kerkwijk et al .
2010 ) . to summarize
, the cluster sn rates plus iron abundances can be fit with a power - law dtd , under some conditions . assuming the minimal iron mass value indicated by cluster observations , a range of slopes with values of @xmath256 , or somewhat steeper , is allowed .
a break to a shallower @xmath7 dependence at @xmath257 is also permitted , provided the power - law is steep enough at longer delays , and the break does not occur too late . for the acceptable range of power - law dtds ,
between 50% and 85% of sne ia explode during the first gyr after star formation . over the past decade , a number of groups ( e.g. , yungelson & livio 2000 ; han & podsiadlowski 2004 ; belczynski et al . 2008 ; ruiter et al . 2009 ; bogomazov & tutukov 2009 ; wang , li , & han 2010 ; mennekens et al .
2010 ) have calculated dtds using binary population synthesis ( bps ) , in which a large number of binaries with a chosen distribution of initial conditions are followed through the stages of stellar and binary evolution , to the point that some of them reach the conditions for explosion as sne ia . in our comparison to cluster sn rates , we focus here , as examples , on the models presented by yungelson & livio ( 2000 ) , and on the more recent ones by mennekens et al .
( 2010 ) . as opposed to the simple parametrized models , considered above
, the bps models can make absolute predictions of sn rates vs. time , i.e. , their normalizations are set .
therefore , in our comparisons of these models to the observed rates and to the integrated iron mass in clusters , we first consider the `` raw '' predictions of the models , without any scalings , and then proceed to test scaled versions .
yungelson & livio ( 2000 ) studied four different evolutionary paths to a sn ia : a dd model ; an sd model with accretion of he from a giant companion and detonation at sub - chandrasekhar mass , through an edge - lit detonation caused by ignition of the he layer ( he - eld ) ; and sd models with accretion from a sub - giant companion and detonation at the chandrasekhar mass ( sg - ch ) , or through an edge - lit detonation ( sg - eld ) . the dtds for these different paths can be seen in their fig .
we have scaled up these dtds by a small factor of 1.05 , to convert from the imf assumed by yungelson & livio ( 2000 ) to our adopted diet salpeter imf . their assumed imf is a broken power law , of index @xmath258 from @xmath259 to @xmath260 , and with @xmath261 from @xmath262 to @xmath259 ( l. yungelson , private communication ) .
( for a fully self - consistent comparison of the shapes of the dtds , their models would need to be re - calculated with our adopted imf , but it is unlikely that the different imf slopes , @xmath263 and @xmath146 , over the limited range of masses that contribute to sn ia progenitors , would lead to major changes in the dtd ) .
two of the models , he - eld and sg - ch , predict no sne beyond @xmath264 gyr after star formation , where all the measured cluster sn rates are . these models are obviously inconsistent with the observed rates , although they could play a role in a multi - component dtd scenario ( see [ ssdoublecomp ] , below ) .
the dd and sg - eld models , on the other hand , do predict sn events on long time scales , and can be tested .
figure [ figyungdd ] compares the predictions of the dd model of yungelson & livio ( 2000 ) to the observations . the `` raw '' version of this model , without any re - scaling of the dtd ( beyond the adjustment above for conversion between the imfs ) , underpredicts the observed cluster sn rates , with a poor @xmath265 .
its integrated number of sne per formed stellar mass is @xmath266 , just 1/5 of the minimal - iron value , and only 12% of the optimal value .
this confirms previous assertions by maoz ( 2008 ) , ruiter et al .
( 2008 ) , and mennekens et al .
( 2010 ) , that bps models underpredict observed sn rates by at a factor of at least a few , and likely by more . as seen in fig .
[ figyungdd ] , if we scale up the model by a multiplicative factor of 5 , so as to integrate to the minimal - iron value , we obtain an acceptable @xmath267 for the rates .
although bps models have many free parameters , it is not clear that such a level of scaling - up of the model could be achieved easily .
if we force the dtd integral to the optimal - iron value , the predicted rates are too high , giving an unacceptable @xmath268 . proceeding to the sg - eld model of yungelson & livio ( 2000 ) , this model ( fig .
[ figyungsgeld ] ) always gives a poor fit to the sn rates , whether in its raw form ( which again produces only 23% of the minimum iron value ) , or scaled to satisfy the iron constraints , or even if scaled arbitrarily .
the sg - eld model begins making sne ia only about 800 myr after star formation , and thus misses the opportunity of producing the bulk of the iron mass during that time .
furthermore , this model then predicts sn rates that fall too steeply with time .
these problems can not be alleviated by a change in the star - formation epoch , @xmath174 , which has little effect on the fits , as was the case for parametrized power - law dtds .
we turn now to the bps models of mennekens et al .
( 2010 ) , who have examined both dd and sd models . they have assumed a kroupa et al .
( 1993 ) imf , consisting of a three - part broken power law , with index @xmath269 from @xmath262 to @xmath270 , @xmath271 from @xmath272 to @xmath273 , and @xmath274 from @xmath275 to @xmath260 .
the mass ratio between this imf and the diet - salpeter imf is 0.92 , and we therefore scale down the mennekens et al .
( 2010 ) dtds by this factor . in their dd models that we examine , mennekens et al .
( 2010 ) introduce the possibility of non - conservative roche - lobe overflow , which they parametrize with a @xmath276 parameter , the fraction of material lost by the donor star that is accepted by the accreting star .
values of @xmath277 , and 1.0 are considered . the @xmath278 parameter ( webbink 1984 ) , which is the fraction of the orbital energy lost during the common - envelope phase that is transferred to kinetic energy of the envelope , is set to 1 .
an additional dd model is calculated with @xmath279 , but treating the common - envelope phase , instead of with the @xmath278 model , by using the @xmath280 parametrization of nelemans & tout ( 2005 ) , with @xmath281 .
the @xmath280 approach quantifies the change in angular momentum during the common envelope phase .
two sd models with @xmath279 have also been calculated by mennekens et al .
( 2010 ) , one with @xmath282 and one with @xmath281 . fitting these models to the cluster rates , with or without the cluster iron constraints , we find the following . none of the raw dtds of mennekens et al .
( 2010 ) produce enough time - integrated sn numbers to reproduce the observed iron abundances . at best , the @xmath283 and @xmath284 dd models produce 12% of the minimal - iron value , and the sd @xmath285 model makes 16% .
the other models make only a few percent or less of the minimal number of sne required by cluster abundances . if we scale up the models , forcing the minimal - iron normalization , the mennekens et al .
( 2010 ) sd models , qualitatively like the yungelson & livio ( 2000 ) sd models discussed above , predict no sne @xmath286 gyr after star formation , and hence can not match the observed cluster rates .
this is illustrated in fig .
[ figmensd ] .
two of the scaled mennekens et al .
( 2010 ) dd models give acceptable fits to the observed cluster sn rates : dd with @xmath285 and @xmath284 ; and dd with @xmath281 and @xmath279 .
these fits are shown in fig .
[ figmendd ] . to summarize our analysis of bps models from these two teams ,
the emerging picture is that sd models always fit the observations poorly in terms of both the absolute `` raw '' numbers of sne they predict , and the time dependence of the cluster sn rates they predict .
the raw dd models also underpredict the absolute sn numbers , but by lower factors , of @xmath287 .
if we treat the bps models as scalable , then for some of the dd models it is possible to simultaneously satisfy the minimal - iron constraints and the observed cluster sn rate dependence on redshift .
the higher , optimal - iron , constraint can also be satisfied by two scaled dd models among those of mennekens et al .
( 2010 ) .
another approach to making dtd predictions , followed by greggio ( 2005 ) , is to calculate analytical dtd models , based on stellar evolution arguments and on various parametrizations of the possible results of the complex common - envelope phases through which sn ia progenitor systems must pass . for each of several sn ia channels , she calculated the dtds that emerge when varying the values for the parameters describing the initial conditions , and the mass and separation distributions and limits of the systems that eventually explode .
mennekens et al .
( 2010 ) have criticized this approach in that it overlooks the changes in stellar - evolution timescales that occur as a result of the `` rejuvenation '' due to mass transfer between stars . on the other hand ,
an advantage of the analytic approach , compared to the bps approach , is that predictions for a large range of parameters can be made quickly .
this successful range can then be investigated in more detail with actual bps simulations .
we focus on a selection of representative models shown in fig . 1 of greggio et al .
these include one sd model and four dd models , computed under different assumptions for these parameters .
the `` wide '' and `` close '' labels of the dd models refer to two possible parametric schemes used by greggio ( 2005 ) to describe the wd separation distribution after the common envelope phase .
the greggio models do not predict the absolute levels of the sn rates ( i.e. , the normalization of the model dtds ) , as this is another free parameter in the models .
we therefore examine only various scaled versions of the greggio models .
the dd - close model with a minimum secondary initial mass of @xmath288 fits the observed sn rate redshift dependence , while satisfying either the minimal iron constraint ( with @xmath289 ) or the optimal iron constraint ( with @xmath290 ) .
maoz et al .
( 2010 ) have already noted the good agreement between this greggio ( 2005 ) model and a dtd reconstructed from the sne in the loss - sdss subsample , and the fact that this greggio model is similar to a @xmath7,@xmath252 , broken power - law with break at @xmath291 myr . such a broken power law was shown to fit well the data and the minimal - iron constraint in [ sspowerlawdtd ] , above .
the resemblance is seen in fig .
[ figgreggio ] . the dd - close model with a minimum secondary mass of @xmath292 , also shown in fig .
[ figgreggio ] , can fit the rates as well , but only with the minimal - iron constraint , due to its shallower slopes at both early and late times .
the two dd - wide models of greggio ( 2005 ) we test , with minimal masses of @xmath292 and @xmath293 , are both too shallow to fit the rates and either of the normalizations , and give @xmath294 . finally , the sd model of greggio ( 2005 ) ,
while having a dtd that is more extended in time than the sd models from bps that we have examined above , suffers from similar problems , particularly a steep drop in rates beyond @xmath295 gyr . the minimal - iron normalized sd model , shown in fig .
[ figgreggio ] , is formally acceptable , with @xmath296 , but its low predicted rates at low redshifts are in conflict with the measurements .
( poisson statistics would be more appropriate to compare the low prediction to the accurately measured , non - zero , rates at low redshift . )
thus , only the two dd - close greggio ( 2005 ) models are consistent with the rates and the iron data .
this is simply the result of the fact that dd models generically produce power - law dtds , and the power - laws required to fit simultaneously the observed rates and the normalizations need to have indices of @xmath297 or steeper .
such indices can be obtained by emerging from the common - envelope phase with a steep power - law separation distribution , i.e. , with relatively more close pairs , as discussed in [ sspowerlawdtd ] and seen in eq .
[ dddependence ] .
naturally , we have examined a limited range of the greggio ( 2005 ) models , and it would be interesting to see if there are others ( e.g. greggio 2010 ) that do fit the data , and what are their parameters .
so far , we have assumed the cluster sfh to be a single , instantaneous burst at @xmath183 .
we now test whether relaxing this assumption can improve the fit of any of the models to the data , in terms of reproducing both the time dependence of the sn ia rate and the normalization , as required by the iron abundance .
furthermore , a non - instantaneous burst may be a more realistic description of cluster sfh .
the predicted sn rate will now be a convolution of the sfh , @xmath298 , with the dtd , after correction for stellar mass loss , @xmath299 we consider a single , but non - instantaneous , burst of star formation in galaxy clusters in the form of an exponentially decaying sfh , @xmath300 $ ] , starting at time @xmath174 .
more complex sfhs are considered in [ ss.compositesfh ] , below .
we have re - fit the cluster rates with the iron - mass normalizations using all the dtd models discussed in
[ sscomparisontopred ] , above , but convolved with this exponential sfh , with values of @xmath301 gyr . the effect on the sn rate of the convolution between the dtd and any temporally extended star formation is always to transfer some fraction of the sn events to later times . since all the dtds that we consider peak at short delays , this means that the sn rate rises more slowly at early times than in the instantaneous burst approximation
this , in turn , means that a smaller fraction of the time - integrated sn number , dictated by the iron abundances , can be produced at early times . raising the model normalization , such that the iron constraints are met
, then results in a poor fit to the sn rate data , compared to the instantaneous case . to try to mitigate this effect
, we shift back by 1 gyr the time of initial star formation to @xmath302 gyr , corresponding to @xmath303 . nonetheless , in every case , the fit with an extended sfh is worse than in the case of an instantaneous burst .
these results are illustrated with two examples .
figure [ figplsfhexp ] shows the predictions of a minimal - iron - normalized @xmath236 power - law dtd , with the exponential sfh starting at @xmath303 , and several characteristic exponential times @xmath304 .
also shown , for comparison , is the @xmath191 instantaneous burst , shown before in fig [ figplm1 ] . at late times , all the predictions have similar slopes , while at progressively shorter times , the rate dependence is shallower for the more extended sfhs .
this leads to progressively greater overprediction of the observed rates .
formally , @xmath305 gyr is ruled out , based on @xmath306 .
conversely , if we ignore the iron constraints and find the best - fit normalization , then the integrated number of sne ia is significantly lower than required by the minimal constraint ( e.g. , by a factor 2 , for @xmath307 gyr ) . in fig .
[ figyungsfhexp ] , we show the same exercise for the sg - eld model of yungelson & livio ( 2000 ) . it might have been hoped that the convolution with an extended sfh could moderate the steep fall at late delays of this dtd , and thus provide a better fit to the observed weak time dependence of the rates . although the steep fall , predicted in the instantaneous burst case , is indeed moderated , the smoothing at short delays lowers the contribution from early times to the integrated sn numbers , forcing a higher overall normalization and a poor fit . to summarize this section , the moderate slope of the observed sn rate redshift dependence at @xmath308 . , combined with the large time - integrated number of sne ia indicated by the iron abundance , together call for a dtd that is sharply peaked at short delays , but with a low tail out to long delays .
convolution of any of the few single - dtd models that satisfy these constraints with any simple sfh that is extended on timescales @xmath309 gyr only degrades the fits .
we now examine to what degree the challenges of reproducing the observations can be overcome by the addition of free parameters that is implicit in the combination of multiple components either two dtds ( as could be expected from the co - existence of two distinct physical sn ia channels , e.g. , dd and sd ) , or two components of cluster sfh , as opposed to the single bursts assumed so far .
the idea of two co - existing sn ia channels , prompt and delayed , emerged several years ago from the observation of , on the one hand , a proportionality between star - formation rate ( sfr ) and sn ia rate per unit stellar mass in star forming galaxies , and on the other hand , a non - zero sn ia rate in early - type galaxies with no current star formation ( mannucci et al .
2005 , 2006 ; scannapieco & bildsten 2005 ; sullivan et al .
2006 ) . as noted in [ ss.intro ] , this observation does not necessarily imply the existence of two separate physical channels .
instead , it could be the manifestation of a dtd from a single channel but with delays spread over a wide range of timescales .
indeed , our analysis , above , shows that single parametrized dtds of power - law form , with indices somewhat steeper than @xmath213 can match the observational constraints ( as can some dd models that produce dtds of this type , if they are suitably scaled up ) .
however , some of the other theoretical dtds , that individually are incompatible with the data because they predict no delayed sne , can be `` saved '' by incorporating them into a two - channel picture .
for example , two of the models of yungelson & livio ( 2000 ) , he - eld and the sg - ch , predict no sne at long time delays .
figure [ figheeldpdd ] shows the predicted cluster rates from the combination of the he - eld model and the dd model , discussed in [ ss.yungel ] .
the raw sum of these two models , as before , produces a too - small fraction ( less than half ) of the total number of sne ia indicated by the minimal iron constraint , and underpredicts all the sn rates below @xmath310 ( by a factor of 5 ) .
however , as seen in fig .
[ figheeldpdd ] , scaling up this composite dtd by factors of 2.1 or 3.6 can solve these problems for the minimal and optical iron cases , respectively . in this example , 60% of the sne ( and the iron ) are from the prompt he - eld sd component , and the rest from the dd component .
similar combinations of two components can work using the other dtds we have considered , as long as one of the components is a dd model ( or a similar power law ) that can provide the sne with long delays .
in addition to the choice of components , one can choose the relative scaling between them , providing a further adjustable parameter .
the current data of course can not discriminate between the various possible prompt components as they make their contributions beyond the redshifts at which rate measurements exist .
our experiments at combining different dtd are obviously not exhaustive , but the emerging picture is nonetheless clear .
the observed sn rates and the iron mass in clusters can be explained simultaneously by combining `` prompt '' and `` delayed '' dtds .
the sne ia from the prompt component produce the majority of the iron mass in clusters .
the sne from the delayed component produce only a fraction of the metals , and they are the events detected by current sn rate measurements , with their weak time dependence .
a final scenario we examine is that of a single dtd , but with a composite cluster sfh , consisting of a short starburst , beginning at @xmath183 , combined with a more extended sfr . with the additional free parameters introduced in this scheme
, it is easy to find combinations of dtds with such composite sfhs that reproduce the sn ia rate versus time , while simultaneously providing a sufficient time - integrated number of sne to produce the observed ratio of iron mass to stellar mass . figure [ figmensfexpplusconst ] shows an example . here
, we have used the @xmath285 , @xmath279 sd model , discussed previously in
[ ss.yungel ] , from the bps calculations of mennekens et al .
we recall that this dtd produces only sne ia with short delays . in the context of the present exercise , this is desirable , as the observed sn rate dependence in generated by the sfh rather than by the dtd . in this example , for the two sfh components we take the sum , @xmath311 , of an exponential sfr , @xmath312 $ ] as in
[ ss.noninst ] , and a constant sfr , @xmath313 .
the relative levels of the two components are adjusted to fit the iron - based normalization constraints . for this choice of functions ,
@xmath314 gyr is required .
a more prolonged exponential component either overpredicts the high-@xmath8 rates , or forces the constant component to overpredict the low-@xmath8 rates .
this scenario can , in principle , explain the presently discussed data in terms of `` prompt '' dtds such as those from sd models , combined with residual star formation . however , it is at odds with many other observations of clusters .
star - formation activity in clusters avoids the cluster cores ( @xmath16 mpc ) , and increases progressively with radius ( e.g. , hansen et al 2008 ; porter et al .
2008 ; bai et al .
2009 ; saintonge et al . 2008 ; loh et al .
in contrast , all of the cluster - sn candidate hosts in sharon et al .
( 2010 ) are found at projected distances @xmath315 mpc from the brightest cluster galaxy . in the sample of barbary et al .
( 2010 ) all are within @xmath316 mpc , or even @xmath317 mpc if excluding the outermost one .
furthermore , the sne from the various surveys we analyze are almost always found in early - type , red - sequence ( and hence apparently quiescent ) , galaxies .
for example , in the sample of gal - yam et al .
( 2008 ) , for 4 out of the 5 cluster sne ia that have host galaxies , the galaxies are early - types , based on colors and spectra . in the sample of sharon et al .
( 2010 ) , 6 out of the 7 most likely cluster sn ia hosts are red - sequence galaxies . in barbary et al .
( 2010 ) , this figure is 8 out of 9 .
thus , the cluster sne ia , on which are based the rates that we analyze , are neither at the locations nor in the types of galaxies where star - formation in clusters is actually observed to occur .
quantitatively , in the above version of the burst+constant sfh scenario , as shown in fig . [ figmensfexpplusconst ] , for , e.g. , @xmath318 , @xmath319 of the present - day cluster stellar mass
is formed before @xmath320 , and the remaining 7% is from the constant sfr between @xmath320 and 0 .
a typical monitored stellar mass in the clusters we consider is @xmath321 m@xmath322 ( see [ ssgastostar ] ) or , correcting to formed mass before stellar evolution mass losses , @xmath323 m@xmath322 . with a lookback time to @xmath320 of 8 gyr ,
the 7% fraction implies a constant sfr between @xmath0 and 1 of @xmath324175 m@xmath322 yr@xmath35 , just in the central regions of clusters . by comparison , several recent studies ( e.g. , krick et al .
2009 , bai et al . 2009 ; koyama et al .
2010 ) find sfrs , integrated over the galaxies within a cluster , between @xmath4 and a few hundred m@xmath322 yr@xmath35 .
some clusters thus do have the high integrated sfrs required in the scenario of a composite sfh + prompt dtd ( ignoring for the moment that this sfr is not in the central , quiescent galaxies that are seen to host the sne , but rather in the outer regions , e.g. , haines et al .
however , the observed cluster sfr , normalized by cluster mass , rises steeply with redshift , as @xmath325 , with @xmath326 ( bai et al .
2009 ) , @xmath327 ( haines et al .
2009 ) , or @xmath328 ( koyama et al . 2010 ) .
this corresponds to one or two orders of magnitude increase in sfr over the @xmath12 range , and is in contrast to the roughly constant mass - normalized sfr that is required in this range , in the composite sfh scenario , to reproduce the cluster sn rates .
we also note that , if the dtd were universal in environment and in time , and the sfh + dtd scenario were true , then it would apply to early - type galaxies in general .
sne ia in all early types , whether in the field or in clusters , and at low or high redshifts , would be the prompt outcome of low levels of recent but unseen star - formation in these galaxies .
however , a study by foerster & schawinski ( 2008 ) of the early - type hosts of sne ia argues against this possibility .
furthermore , the dtd reconstruction in nearby galaxies by maoz et al .
( 2010 ) demonstrates a @xmath329 detection of a delayed sn ia component , with delays of @xmath330 gyr .
we therefore conclude that the scenario of a burst+constant sfh , combined with a prompt single - component dtd , is probably not a viable model .
while it is capable of explaining the cluster sn rate data and the observed iron to stellar mass ratio , it is incompatible with other cluster observations .
those observations indicate that , although about 10% of the stellar mass of clusters formed at @xmath331 , this activity did not occur in the central , quiescent , galaxies hosting the sne found by current surveys , and that the sfr falls steeply with cosmic time , in contrast to the flat sfh required to reproduce , under this scenario , the flat observed sn rate .
measurements of galaxy - cluster sn ia rates as a function of cosmic time provide a particularly direct avenue to obtain the dtd of sne ia , with its implications for progenitor models and cosmic history .
such measurements can constrain both the functional form and the normalization of the dtd .
we have combined recently completed measurements of cluster rates between @xmath0 and @xmath1 with the latest results on the iron mass content of clusters , to recover the sn ia dtd , and to test a variety of model dtds , from purely mathematical parametrized forms , to dtds obtained from more detailed physical considerations .
our main results are as follows .
+ 1 . assuming that the bulk of the stars in clusters formed with a normal imf in a brief burst at @xmath94 , as indicated by optical spectroscopy of cluster galaxies , the sn ia dtd can be directly recovered from the observations .
the resulting dtd peaks at the shortest delays probed , @xmath3 gyr , decreases steeply at longer delays , and extends out to 11 gyr .
the recovered dtd agrees remarkably well , both in shape and in absolute normalization , with dtds recently recovered using different techniques , in different environments : in field ellipticals ( totani et al .
2008 ) ; in nearby galaxies ( maoz et al .
2010 ) ; and in the magellanic clouds ( maoz & badenes 2010 ) .
the current derivation is complementary to previous ones in that it recovers the dtd over its full time range , with good time resolution at delays @xmath332 gyr .
the emerging picture is of a universal dtd that is not strongly dependent on environment or cosmic time , and can be well represented by a single power law , of index @xmath333 .
+ 2 . comparing the data to theoretical or phenomenological dtds in a forward - modeling process , the best parametric description of the dtd is a power law , @xmath334 with index @xmath335 or @xmath336 , depending on whether we adopt a minimal - iron constraint or an optimal - iron constraint , respectively , based on cluster data .
single - component dtds consisting of shallower power laws , such @xmath246 , can not simultaneously match the observed sn rates and the integrated sn ia numbers dictated by the iron - to - stellar mass ratio .
physical dd models from the bps simulations we have examined can match the observations , provided they are scaled up in numbers by factors of @xmath287 . on the other hand , sd models , on their own , fail because they do not produce sne at late delays , as implied by the data .
our results thus provide strong support for the double - degenerate sn ia progenitor scenario .
the above conclusions are insensitive to the exact epoch of star formation in clusters , in the range @xmath337 .
the conclusions also do not change if one replaces the instantaneous starburst with an exponentially decaying sfh , except that the model fits deteriorate with increasing star - formation timescale , and so the range of acceptable models shrinks .
multi - component models , that either combine freely scaled versions of two dtds ( e.g. sd and dd ) , or two sfh models ( such as a burst plus a constant ) with a prompt dtd , have enough freedom to permit many combinations that can match the existing data .
however , apart from the loss of simplicity involved , such combinations face other problems . for the composite dtd models , it remains to be seen if the required combinations and scalings are physically plausible . for the composite sfhs , a high sfr in the central regions of clusters , and which remains constant between @xmath12 ,
is required , in conflict with observations of clusters at these redshifts .
all of the successful dtd models produce just a fraction of the time - integrated numbers of sne ia at redshifts below 1.4 .
a clear prediction of these models is , therefore , that sn surveys of clusters or proto - clusters at even higher redshifts , approaching the stellar formation redshifts of the clusters , should reveal a sharply rising rate of sne ia .
alternatively , observation of a non - rising sn ia rate would raise the possibility that the bulk of metals in clusters was produced by cc sne from a top - heavy imf that exploded even earlier in cluster environments .
we thank k. barbary , m. graham , f. mannucci , n. mennekens , e. ofek , d. poznanski , c. pritchet , l. yungelson , d. zaritsky , and the anonymous referee for providing useful input . d.m .
acknowledges support by a grant from the israel science foundation .
acknowledges support by the grant 07ast - f9 from the ministry of science , culture and sport , israel , and the ministry of research , france .
is also supported by the israel science foundation , the eu fp7 marie curie program via an irg fellowship , the benoziyo center for astrophysics , weizmann - uk , and the peter and patricia gruber awards .
thanks the benoziyo center for astrophysics for its hospitality in the course of this work .
this work was supported by grants go-10493 and go-10793 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 .
balestra , i. , tozzi , p. , ettori , s. , rosati , p. , borgani , s. , mainieri , v. , norman , c. , & viola , m. 2007 , , 462 , 429 barbary , k. , et al .
2010 , apj , submitted barris , b. j. , & tonry , j. l. 2006 , , 637 , 427 boehringer , h. , mullis , c. , rosati , p. , lamer , g. , fassbender , r. , schwope , a. , & schuecker , p. 2005
, the messenger , 120 , 33 bogomazov , a.i & tutukov , a.v .
, 2009 , astronomy reports , 53 , 214 brandt , t. d. , tojeiro , r. , aubourg , e. , heavens , a. , jimenez , r. , & strauss , m. a. 2010 , arxiv:1002.0848 bremer , m. n. , et al . 2006 , , 371 , 1427 bruzual , g. , & charlot , s. 2003 , , 344 , 1000 buzzoni , a. 2005 , , 361 , 725 cain , b. , et al .
2008 , , 679 , 293 cappellari , m. , et al .
2006 , , 366 , 1126 ciotti , l. , dercole , a. , pellegrini , s. , & renzini , a. 1991 , , 376 , 380 cooper , m. c. , newman , j. a. , & yan , r. 2009 , , 704 , 687 daddi , e. , cimatti , a. , & renzini , a. 2000 , , 362 , l45 dahlen , t. , et al .
2004 , , 613 , 189 dahlen , t. , strolger , l .- g .
, & riess , a. g. 2008 , , 681 , 462 dawson , k. s. , et al .
2009 , , 138 , 1271 de grandi , s. , & molendi , s. 2001 , , 551 , 153 dilday , b. , et al .
2010 , , 715 , 1021 filippenko , a. v. 1997 , , 35 , 309 filippenko , a.v . ,
2010 , in preparation frster , f. , wolf , c. , podsiadlowski , p. , & han , z. 2006 , , 368 , 1893 frster , f. , & schawinski , k. 2008 , , 388 , l74 gal - yam , a. , maoz , d. , & sharon , k. 2002 , , 332 , 37 gal - yam , a. , & maoz , d. 2004 , , 347 , 942 gal - yam , a. , maoz , d. , guhathakurta , p. , & filippenko , a. v. 2008 , , 680 , 550 gal - yam , a. , & leonard d. c. 2009 , , 458 , 865 gilbank , d. g. , yee , h. k. c. , ellingson , e. , hicks , a. k. , gladders , m. d. , barrientos , l. f. , & keeney , b. 2008 , , 677 , l89 giodini , s. , et al .
2009 , , 703 , 982 girardi , l. , bressan , a. , bertelli , g. , & chiosi , c. 2000 , , 141 , 371 gonzalez , a. h. , zaritsky , d. , & zabludoff , a. i. 2007 , , 666 , 147 gould , a. , bahcall , j. n. , & flynn , c. 1997 , , 482 , 913 graham , m. l. , et al . 2008 , , 135 , 1343 greggio , l. 2005 , , 441 , 1055 greggio , l. , renzini , a. , & daddi , e. 2008 , , 388 , 829 greggio , l. 2010 , , 406 , 22 grevesse , n. , & sauval , a. j. 1999 , , 347 , 348 haines , c. p. , et al.2009 , , 704 , 126 han , z. , & podsiadlowski , p. 2004
, , 350 , 1301 hansen , s. m. , sheldon , e. s. , wechsler , r. h. , & koester , b. p. 2009
, , 699 , 1333 helder , e. a , et al .
2009 , sceince , 325 , 719 hicks , a. k. , et al .
2008 , , 680 , 1022 hoyle , f. , & fowler , w. a. 1960 , , 132 , 565 iben , i. , jr . , & tutukov , a. v. 1984 , , 54 , 335 jorgensen , h. e. , lipunov , v. m. , panchenko , i. e. , postnov , k. a. , & prokhorov , m. e. 1997 , , 486 , 110 kauffmann , g. , et al .
2003 , , 341 , 33 koyama , y. , kodama , t. , shimasaku , k. , hayashi , m. , okamura , s. , tanaka , i. , & tokoku , c. 2010 , , 403 , 1611 krick , j. e. , surace , j. a. , thompson , d. , ashby , m. l. n. , hora , j. l. , gorjian , v. , & yan , l. 2009 , , 700 , 123 kroupa , p. , tout , c. a. , & gilmore , g. 1993 , , 262 , 545 kroupa , p. 2001
, , 322 , 231 kuznetsova , n. , et al . 2008 , , 673 , 981 lagan , t. f. , lima neto , g. b. , andrade - santos , f. , & cypriano , e. s. 2008 , , 485 , 633 leaman , j. , li , w. , chornock , r. , filippenko , a. 2010 , apj , submitted , arxiv:1006.4611 li , w. , et al . 2010a , apj , submitted , arxiv:1006.4612 li , w. , chornock , r. , leaman , j. , filippenko , a. v. , poznanski , d. , xiaofeng , w. , ganeshalingam , m. , mannucci , f. 2010b , apj , submitted , arxiv:1006.4613 lin , y .- t . , mohr , j. j. , & stanford , s. a. 2003 , , 591 , 749 loh , y .- s . , ellingson , e. , yee , h. k. c. , gilbank , d. g. , gladders , m. d. , & barrientos , l. f. 2008 , , 680 , 214 madau , p. , della valle , m. , & panagia , n. 1998 , , 297 , l17 mannucci , f. , della valle ,
m. , panagia , n. , cappellaro , e. , cresci , g. , maiolino , r. , petrosian , a. , & turatto , m. 2005 , , 433 , 807 mannucci , f. , della valle , m. , & panagia , n. 2006 , , 370 , 773 mannucci , f. , maoz , d. , sharon , k. , botticella , m. t. , della valle , m. , gal - yam , a. , & panagia , n. 2008 , , 383 , 1121 maoz , d. 2008 , , 384 , 267 maoz , d. , mannucci , f. , li , w. , filippenko , a. v. , della valle , m. , & panagia , n. 2010 , arxiv:1002.3056 , mnras , in press maoz , d. , & badenes , c. 2010 , arxiv:1003.3031 , mnras , in press maoz , d. , & gal - yam , a. 2004 , , 347 , 951 maughan , b. j. , jones , c. , forman , w. , & van speybroeck , l. 2008 , , 174 , 117 mazzali , p. a. , rpke , f. k. , benetti , s. , & hillebrandt , w. 2007 , science , 315 , 825 mennekens , n. , vanbeveren , d. , de greve , j. p. , & de donder , e. 2010 , arxiv:1003.2491 , , in press nelemans , g. , & tout , c. a. 2005 , , 356 , 753 porter , s. c. , raychaudhury , s. , pimbblet , k. a. , & drinkwater , m. j. 2008 , , 388 , 1152 poznanski , d. , et al .
2007 , , 382 , 1169 pratt , g. w. , croston , j. h. , arnaud , m. , boehringer , h. 2009 , , 498 , 361 pritchet , c. j. , howell , d. a. , & sullivan , m. 2008 , , 683 , l25 raskin , c. , scannapieco , e. , rhoads , j. , & della valle , m. 2009 , , 707 , 74 reiss 2000 , phd thesis , university of washington renzini , a. 1997 , , 488 , 35 renzini , a. , ciotti , l. , dercole , a. , & pellegrini , s. 1993 , , 419 , 52 rosati , p. , stanford , s. a. , eisenhardt , p. r. , elston , r. , spinrad , h. , stern , d. , & dey , a. 1999 , , 118 , 76 rosati , p. , et al .
2004 , , 127 , 230 ruiter , a. j. , belczynski , k. , & fryer , c. 2009 , , 699 , 2026 saintonge , a. , tran , k .- v .
h. , & holden , b. p. 2008
, , 685 , l113 sadat , r. , blanchard , a. , guiderdoni , b. , & silk , j. 1998 , , 331 , l69 salpeter , e. e. 1955 , , 121 , 161 scannapieco , e. , & bildsten , l. 2005 , , 629 , l85 sharon , k. , gal - yam , a. , maoz , d. , filippenko , a. v. , & guhathakurta , p. 2007
, , 660 , 1165 sharon , k. , et al .
2010 , apj , 718 , 876 sim , s. a. , et al .
2010 , , 714 , l52 stanford , s. a. , holden , b. , rosati , p. , eisenhardt , p. r. , stern , d. , squires , g. , & spinrad , h. 2002 , , 123 , 619 stanford , s. a. , et al .
2005 , , 634 , l129 stanford , s. a. , et al .
2006 , , 646 , l13 strolger , l .- g .
, et al .
2004 , , 613 , 200 sullivan , m. , et al .
2006 , , 648 , 868 tojeiro , r. , wilkins , s. , heavens , a. f. , panter , b. , & jimenez , r. 2009 , , 185 , 1 totani , t. , morokuma , t. , oda , t. , doi , m. , & yasuda , n. 2008 , , 60 , 1327 van kerkwijk , m.h . , chang , p. , justham , s. 2010 , apjl , submitted , arxiv:1006.4391 wang , b. , li , x .- d . , & han , z .- w .
2010 , , 401 , 2729 webbink , r. f. 1984 , , 277 , 355 whelan , j. , & iben , i. j. 1973 , , 186 , 1007 woosley , s. e. 2007 , nature physics , 3 , 832 yasuda , n. , & fukugita , m. 2010 , , 139 , 39 york , d. g. , et al .
2000 , , 120 , 1579 yungelson , l. r. , & livio , m. 2000 , , 528 , 108 | knowledge of the supernova ( sn ) delay time distribution ( dtd ) the sn rate versus time that would follow a hypothetical brief burst of star formation can shed light on sn progenitors and physics , as well as on the timescales of chemical enrichment in different environments .
we compile recent measurements of the type - ia sn ( sn ia ) rate in galaxy clusters at redshifts from @xmath0 out to @xmath1 , just 2 gyr after cluster star formation at @xmath2 .
we review the plausible range for the observed total iron - to - stellar mass ratio in clusters , based on the latest data and analyses , and use it to constrain the time - integrated number of sn ia events in clusters . with these data ,
we recover the dtd of sne ia in cluster environments .
the dtd is sharply peaked at the shortest time - delay interval we probe , @xmath3 gyr , with a low tail out to delays of @xmath4 gyr , and is remarkably consistent with several recent dtd reconstructions based on different methods , applied to different environments .
we test dtd models from the literature , requiring that they simultaneously reproduce the observed cluster sn rates and the observed iron - to - stellar mass ratios .
a parametrized power - law dtd of the form @xmath5 from @xmath6 myr to a hubble time , can satisfy both constraints .
shallower power laws , such as @xmath7 can not , assuming a single dtd , and a single star - formation burst ( either brief or extended ) at high @xmath8 .
this implies 5085% of sne ia explode within 1 gyr of star formation .
dtds from double - degenerate ( dd ) models , which generically have @xmath9 shapes over a wide range of timescales , match the data , but only if their predictions are scaled up by factors of @xmath10 .
single degenerate ( sd ) dtds always give poor fits to the data , due to a lack of delayed sne and overall low numbers of sne .
the observations can also be reproduced with a combination of two sn ia populations a prompt sd population of sne ia that explodes within a few gyr of star formation , and produces about 60 percent of the iron mass in clusters , and a dd population that contributes the events seen at @xmath11 an alternative scenario of a single , prompt , sn ia population , but a composite star - formation history in clusters , consisting of a burst at high @xmath8 , followed by a constant star - formation rate , can reproduce the sn rates , but is at odds with direct measurements of star formation in clusters at @xmath12 .
our results support the existence of a dd progenitor channel for sne ia , if the overall predicted numbers can be suitably increased . |
ductile fracture of metals commonly occurs through the nucleation , growth and coalescence of microscopic voids .
@xcite much can be learned about the ductile fracture through the study of these voids .
a particularly interesting case is the dynamic fracture of ductile metals , @xcite in which the strain rates are so high that processes such as diffusion operating on relatively long time scales may be neglected , while inertial effects become relatively important .
void growth is driven by the need to relax tensile stress that builds up in the system , and to minimize the associated elastic energy .
the material around a void deforms plastically in order to accommodate the void growth . naturally , the plastic deformation results from a local shear stress , which may arise from the applied stress , but it also may arise from the stress field of the void even if the applied stress is hydrostatic .
so the expectation is that the evolution of the plastic zone , and hence the growth of the void , is influenced by the degree of stress triaxiality ; _ i.e. _ , the ratio of the mean ( hydrostatic ) stress to the shear stress .
it is this relationship that we study here by varying the triaxiality of the loading . in particular , we conduct simulations in which one , two or three directions of the system are expanded , producing a state of uniaxial , biaxial or triaxial strain , respectively .
variation in the triaxiality of the strain causes variation in the triaxiality of the stress state , where it should be noted that uniaxial ( biaxial ) strain does not imply pure uniaxial ( biaxial ) stress . besides dynamic crack propagation experiments
, dynamic fracture can be measured for instance in shock physics or spallation experiments , to which the simulations performed here are compared .
various techniques are employed to generate the shock waves : hopkinson bar , gas gun , high explosives , and laser ablation . with hopkinson bar
the strain - rates @xmath2 usually are of the order @xmath3/sec , in gas gun of @xmath4/sec , with high - explosives even higher strain - rates can be produced , and with lasers strain - rates exceeding @xmath0/sec are attained . in a gas gun for instance the fracture results from essentially one - dimensional shock loading .
two compressive shock waves are generated by the impact of a flier on a metal target , propagate away from each other , reflect from opposite free surfaces becoming tensile release waves and finally come into coincidence again .
if the combined tensile stress exceeds the rupture strength of the material , the metal fails , after some incubation time , producing a fracture surface . in strong shocks
, a scab of material may spall from the back side of the target and fly off .
spallation experiments @xcite for single and polycrystal copper report spall strength values of @xmath5 gpa at strain - rates @xmath6/sec , and scaling between the spall strength and strain - rate , @xmath7 . in this study of dynamic fracture in ductile metals at high strain - rates ( @xmath0
@xmath1/sec ) we have concentrated on void growth starting from a single crystal copper lattice containing an infinitely weakly bound inclusion or a pre - existing nanoscale void .
the lattice is initially free of other defects .
we have focused on the effect of stress - triaxiality on void growth . in some fracture experiments , for example in necking and cup - cone fracture , @xcite the uniaxial strain produces a stress state that transitions rapidly to triaxial state due to the plastic flow during the course of loading .
it is during the triaxial phase that void growth and failure take place . because of the connection with shock experiments ,
the stress triaxiality study done here is carried out using strain control , and it is the strain that is varied from uniaxial to biaxial to triaxial .
much of the damage modeling of metals has been carried out at mean - field or continuum level based on constitutive theories .
the continuum models concentrate especially on two areas : macroscopic crack growth phenomenon @xcite and studies of porosity , _
i.e. _ behavior of an array of voids , at sub - grain level during loading .
@xcite in the latter area for example the locus of yield surfaces in stress space has been studied , which is related to the question of the effect of stress - triaxiality studied here . of the void growth studies especially the gurson model @xcite
is commonly used to model cavitation ( the development of porosity ) at the sub - grid level in what is termed as damage modeling .
these continuum calculations often assume that the matrix material , where the voids are embedded , is elastically rigid and plastically incompressible , and the dilation of the void - matrix aggregate is completely due to the void growth . of particular interest , and relevance in terms of this study
, is a single crystal plasticity study of void growth .
@xcite the calculations are typically done by determining approximate solutions for integrals of incremental equations of virtual work using the finite element method .
continuum modeling has been used to study some of the phenomena addressed in this paper such as the effect of triaxiality on void growth and void shape changes .
@xcite the validity of the approximate solutions of incrementals limits the strain - rates to be rather low compared to the strain - rates used in this study . in order to characterize the void growth not only with macroscopic quantities and at the continuum level , but to investigate what happens at the atomistic level , we have employed molecular dynamics ( md ) simulations .
md simulations enable us to see what the effects are on the void surface at the single atom level , when it grows while the total system yields .
this article presents work , which is an extension to the work done earlier by some of this article s authors of void growth in a single crystal copper with hydrostatic loading , void nucleation and growth in single and polycrystalline copper .
@xcite molecular dynamics simulations of void growth in single crystal copper have also been conducted by other groups for slab geometries of interest in the semiconductor chip metalization problem .
@xcite the effectively two - dimensional , thin film systems are in contrast to three dimensional bulk systems studied here . in some cases md simulations of plasticity and
crack propagation have been as large as billion atoms .
@xcite this article is organized as follows .
it starts in section [ method ] with an overview of the md method used and the simulations which have been carried out in this study .
exploration of the results of the simulations starts in section [ stressstrain ] by the study of the mean or hydrostatic stress versus strain as well as the deviatoric part of the stress tensor , von mises stress , which is used to measure the shear stress , and stress - triaxiality versus strain for all the simulated strain - rates and modes of expansion .
section [ plasticstrain ] concentrates on the macroscopic plastic quantities , such as mean and equivalent plastic strain , plastic work and its relation to the temperature .
the evolution of the void in terms of its growth and shape changes is studied in section [ voidevol ] .
section [ summary ] summarizes the results and compares different measured quantities with each other concentrating on one of the simulations , uniaxial strain with strain - rate @xmath8/sec .
the article is concluded with discussions of the results and suggestions for future studies in section [ discussion ] .
in this atomistic - level study of void growth , the simulations have been done using empirical embedded - atom ( eam ) potentials in classical molecular dynamics @xcite following the scheme developed earlier .
@xcite the copper eam potential we have used is due to oh and johnson . @xcite the system , in which the simulations are done , is a three - dimensional single - crystal face - centered - cubic ( fcc ) lattice in a cubic box with @xmath9 faces .
periodic boundary conditions are used in all the three directions so that there are no free boundaries in the system apart from the void .
equivalently the system can be imagined to consist of an infinite periodic array of voids .
note that periodic boundary conditions have also been used in continuum models of void growth , but in the continuum modeling of void growth in isotropic materials the calculations are done in a reduced cell , which exhibits one quarter of the box in two dimensions , and one eighth in three dimensions , and the behavior of other areas are derived from the symmetries .
we use the full cubic box because the cubic symmetry present in the continuum is broken in md at finite temperature , and processes such as dislocation nucleation at the void surface would be over - constrained in a reduced simulation box . in the simulations ,
the system is brought to thermal equilibrium at room temperature , @xmath10 k , with a commonly used thermostat @xcite and at ambient pressure , @xmath11 mpa , keeping the volume constant .
after that a spherical void is cut in the middle of the system , later the thermostat is turned off , and the dilational strain is applied uniformly with a constant strain - rate @xmath2 .
the removal of the atoms in the spherical region may be considered to simulate the instantaneous separation of the matrix material from an infinitely weakly bound inclusion .
the uniform expansion in these strain - controlled simulations is applied through rescaling the coordinates as in the parrinello - rahman method .
@xcite technically the three cartesian coordinates of the atoms are rescaled to the unit - box , each coordinate @xmath12 . when calculating the forces and velocities , as well as updating the new positions of the atoms , the unit - box is multiplied by a diagonal scaling matrix @xmath13 , where @xmath14 s are the simulation box s side lengths , to compute the true positions of the atoms , @xmath15 this scaling matrix @xmath16 is updated each time - step , when the load is applied , by multiplying the initial matrix @xmath17 with the sum of the unit matrix and the strain matrix @xmath18 , @xmath19 for our purposes the strain - rate matrix @xmath20 is always diagonal , since neither rotation nor simple shear type strains are studied . in the triaxial case
all the terms in the diagonal are equal ; in the uniaxial there is a single non - zero term ; and in the biaxial case two of the three diagonal terms differ from zero and are equal .
prior to expansion the system is cubic , its scaling matrix @xmath17 is diagonal , and all the terms are equal and correspond to the equilibrium size at ambient pressure .
hence the scaling matrix @xmath16 remains diagonal throughout the simulation , and the strain in each case is in a @xmath21 direction . in fracture and plasticity simulations the first quantity to consider is the stress - strain behavior . with the strain as an input parameter , here we have to measure the stress .
in this study of the stress - triaxiality we are interested in both mean and shear stresses . therefore the whole stress tensor @xmath22 is needed .
the stress tensor ( the negative of the pressure ) can be calculated atomistically on each time - step using the virial formula : @xcite @xmath23 the first term in the stress tensor is the kinetic contribution of atoms denoted with @xmath24 and having masses @xmath25 and momenta @xmath26 .
the second term , a microscopic _ virial _ potential stress , consists of sums of interatomic forces @xmath27 of atom pairs @xmath28 with corresponding distances @xmath29 .
it should be noted , that here and in the rest of the article @xmath24 and @xmath30 denote the atoms , and @xmath31 and @xmath32 the cartesian coordinates .
note that the thermal stress is included , although in practice in these simulations it contributes less than 1 gpa , less than 10% of the yield stress value , and never dominates the changes in stress . typically in the simulations carried out here , the cube consists of 60 fcc unit cells in each direction , giving 864 000 atoms .
the equilibrium side - length of such a copper system is @xmath33 nm at room temperature and ambient pressure . the radius of the spherical void cut from the system , unless otherwise noted , is 0.1 of the side - length of the box ; thus @xmath34 nm . after the void is cut , there are 860 396 atoms in the system . the relatively inexpensive potential used enables us to do extensive simulations in time .
a single time - step takes typically about 40 sec of cpu - time in a system with 864 000 atoms in a linux workstation with intel xeon 1700mhz processor .
the longest calculation required 835 050 time - steps corresponding to 5.6 nanoseconds .
the time - step was @xmath35 femtoseconds . as mentioned earlier , in order to study the effects of the stress - triaxiality and different modes of expansion on the void - growth , we have applied three different types of expansion , namely uniaxial , biaxial , and triaxial .
the strain - rates used for each of the three modes of expansion are @xmath36/sec , @xmath37/sec , @xmath38/sec , @xmath8/sec , and @xmath0/sec . for the lowest strain - rate ,
the md code was parallelized in order to take advantage of massively parallel computers .
the parallelization was done using a spatial domain - decomposition , and was shown to scale nearly linearly up to 128 processors .
the parallel code was used in the case with 835 050 time - steps mentioned above , for example . for comparison in the elastic regime
, we have also performed simulations without a void in all three modes of expansion .
these simulations have been used to determine the bulk , elastic stress - strain response of the eam copper and hence the elastic constants . without a void ,
the system is not so strain - rate and system size dependent , at least up to the point of failure , so the so - called `` no void '' simulations have been performed with a smaller system size , 45 fcc - cells in each direction ( 364 500 atoms ) and at the single strain - rate @xmath39/sec . a uniaxial study of the 60@xmath40 system size , but with a smaller initial void radius of @xmath41 nm , was carried out with the strain - rate @xmath42/sec in order to study the void - size dependence . in this case
the system with the void contains 863 543 atoms .
it should be mentioned , too , that all of the intermediate strain - rate simulations ( @xmath43/sec and @xmath38/sec ) expansion were not started from equilibrium conditions at @xmath44 mpa , but from systems expanded previously at the strain - rate @xmath39/sec .
these simulations have been restarted well before yielding , when the system s behavior is rate independent , and relaxed for 2000 time - steps , or 13.4 picoseconds , without expansion before continuing the expansion at the intermediate strain rates .
the energy is conserved during the relaxation in md simulations .
these restarts have been accomplished at strain values @xmath45 % , @xmath46 % , and @xmath47 % in uniaxial , biaxial , and triaxial cases , respectively .
let us begin to explore the results of the md simulations by looking at the stress - strain curves . figure [ fig1 ] shows these curves for each of the modes of expansion at all the strain - rates computed .
the data from `` no void '' cases are also plotted .
the mean or hydrostatic stress , @xmath48 is plotted to indicate the principal impetus for void growth .
note that the strain is the _ engineering strain _
, defined as the expanded system size divided by the original system size minus one . in the uniaxial and biaxial cases the strains are the principal strain values @xmath49 in the direction of the strain , such as @xmath50 , @xmath51 in the uniaxial case , and @xmath52 , @xmath53 in the biaxial case .
hence the strain is the value of a non - zero diagonal term of @xmath54 in eq .
( [ eq_update ] ) .
the mean strains @xmath55 are 1/3 and 2/3 of the plotted uniaxial and biaxial strains , respectively .
thus the total volumetric strain - rates are not the same in the different expansion modes . in the triaxial cases
the plotted and the mean strain values are the same .
the shape of the stress - strain curves do not differ much depending on the modes of expansion in these cases , at least when plotted versus mean strain .
independent of the strain - rate , and whether with or without a void , the stress - strain curves lie essentially on top of each other during elastic expansion , _
i.e. _ , the initial smooth behavior when the system is still recoverable and has not deformed plastically .
the stress - strain curve starts to deviate from the trend of the elastic behavior at a specific , `` critical '' point which we call here a _
yield point_. in other quantities we measure a change in behavior happens at a specific point , too , and as we shall see later the critical or yield points mostly coincide with each other , _ i.e. _ , their strain values are approximately the same independent of from which quantity we derived it .
the same point is also the one when the void starts to grow , which is the primary mechanism for plasticity in this study .
here we define numerically the yield point of the cases with void by comparing their stress - strain curves with the reference `` no void '' curve , which behaves elastically beyond the yield points of the other cases [ cf .
[ fig1](a ) inset ] .
ultimately the no void case does fail by homogeneous nucleation of voids , and this is the reason for the drop in the mean stress .
there is a small offset between cases with a void compared to the case without a void due to the elastic relaxation of the void .
the value of the stress at the yield point in the cases with void is lower with lower strain - rate , and thus the strain to yield is also lower . in each of the modes of expansion , the stress at the yield point for the strain - rate @xmath56/sec is close to the value to which the higher strain - rates converge .
of course , at much lower strain rates the physics changes , new mechanisms become active , so this value need not hold for arbitrarily low strain rates . however , it is noteworthy that the stress at the yield point is not scaling with strain , contrary to the experimental finding for the spall strength explained in section [ intro ] .
overshooting , the phenomenon that the maximum stress is much higher than the stress at the yield point , is evident here for the higher strain rates .
the scaling of the spall strength versus the strain - rate @xcite with an exponent 0.2 is reproduced here when one compares the maximum stress values instead of the stresses at the yield point for strain - rates @xmath57/sec , @xmath37/sec , and @xmath1/sec , since then the exponent is @xmath58 , lowest for the uniaxial case and highest for the triaxial case . on the other hand the stress value at the yield point , which is at the same time the maximum stress , when @xmath56/sec is very close to the value of 6 - 8 gpa the spall strength scaling predicted from the lower strain - rates mentioned in section [ intro ] .
it should be noted also , that since we are limited to finite , fairly small , system sizes , at late stages of the stress - strain curves , at the failure , the data is not realistic anymore .
the reason is that at the plastic part of the stress - strain behavior when the void grows , it also emits dislocations , and in a finite system with periodic boundaries , when the dislocations have traveled long enough , they propagate through the boundaries and reenter from the other side . in the picture where we have a periodic array of voids in an infinite system
this means that the voids are so close to each other that they start to interact . in reality voids
are never arranged in a perfect cubic lattice structure and in symmetric positions with respect to each other , and thus the interactions of the voids in the simulations with their periodic images are just an unphysical finite size effect . in the shear stress or
more precisely in the deviatoric part of the stress tensor @xmath59 plotted in fig .
[ fig2 ] , a much bigger difference is seen between the modes of expansion than in the mean stress . for the deviatoric part of the stress
@xmath59 we use _ von mises stress _ : @xmath60^{1/2 } , \label{eq_misesstress}\ ] ] where @xmath61 is the second invariant of the stress deviator @xmath62 .
@xcite thus von mises stress reads : @xmath63^{1/2 } \\ & = & \left [ \frac{1}{2 } \sum_{\alpha > \beta } ( \sigma_{\alpha \alpha } - \sigma_{\beta \beta})^2 + 3 \sum_{\alpha > \beta } \sigma_{\alpha \beta}^2 \right]^{1/2}. \end{array } \label{eq_misesstress2}\ ] ] while the mean stress at the yield point gets a value of about @xmath64 gpa when loaded with strain - rate @xmath65/sec in each of the three modes of expansion , von mises stress has a value of @xmath66 gpa and 0.7 gpa in the uniaxial and biaxial cases , respectively . in the triaxial case
it should be zero by symmetry , and the difference from zero , representing symmetry - breaking effects , is small .
thus the loading differences between the modes of expansion are quantified in von mises stress .
after the onset of plasticity or the void growth , von mises stress gets a value of about @xmath67 gpa , 0.2 gpa , and 0.1 gpa , in the uniaxial , biaxial , and triaxial cases , respectively , independent of the strain - rate , but with significant fluctuations in this regime . in this period
also dislocations move under the action of the shear stress until the stress has dropped to the point that it is no longer sufficient to move a dislocation through the forest of dislocations .
the final value of the shear stress corresponds to the flow stress , and they are close to the tensile strength values of copper , 200 - 400 mpa , quoted in the literature .
@xcite although von mises stress captures differences between the loading modes quite well , an even better quantity to study is the ratio between hydrostatic and shear stresses , the _ stress - triaxiality _
@xmath68 which has been plotted in fig .
[ fig3 ] . in the uniaxial case
the stress - triaxiality starts from the value @xmath69 and slowly decreases linearly to a value @xmath70 until the onset of rapid growth at the yield point .
after the rapid increase the stress - triaxiality saturates at @xmath71 .
the stress - triaxiality in the biaxial case starts with a much larger value than in the uniaxial case .
it begins at @xmath72 and increases linearly to a value @xmath73 at the yield point where it grows rapidly to a value of @xmath74 .
we have noted the correspondence of the final von mises stress and the flow stress above . similarly the spall strength provides an experimental measure of the mean stress that can be supported in void growth .
thus the stress - triaxiality values can be compared with the ratio of the spall and the tensile strength of copper .
previously quoted literature values for them are 6 - 8 gpa and 200 - 400 mpa , respectively , giving for their ratio values between 15 and 40 , and thus comparable with the stress - triaxiality values here .
the comparison is not fully rigorous , but it provides an indication of how reasonable the final stress triaxiality values are in terms of experiment . since the stress - triaxiality is the mean stress divided by von mises stress , which is equal to zero in the triaxial case until the yield point and very small even after that , the stress - triaxiality is diverging and therefore not plotted here in that case . the stress - triaxiality values in the uniaxial and biaxial cases at the elastic part of the simulation
is compared here also with the values one get from the elasticity theory : @xcite @xmath75 @xmath76 in the uniaxial case and @xmath77 in the biaxial case .
the literature values for the elastic constants of copper are @xmath78 168 gpa and @xmath79 121 gpa .
@xcite thus @xmath80 and @xmath81 in the uniaxial and biaxial cases , respectively , which compare quite well with the simulations presented here . when the elastic constants are derived from the stress - strain curves as @xmath82 , they are close to the actual experimental values : @xmath83 162 gpa , @xmath84 121
gpa and @xmath83 168 gpa , @xmath84 124 gpa in the uniaxial and biaxial cases , respectively . [
cols="<,^,^,^,^,^,^,^",options="header " , ] the critical mean stress , von mises stress , and the stress - triaxiality values where their behaviors start to deviate compared to the elastic ones , or what we call yield points , are summarized in table [ tab1 ] for the strain - rates @xmath85/sec , @xmath86/sec , @xmath8 , and @xmath0/sec of the principal strains . in the case of the highest strain - rate @xmath87/sec ,
the shapes of the stress - strain curves are so much rounded due to over - shooting , that there is no clear point , where the stress - strain curve deviates from the elastic behavior , and thus our definition of the yield point is no longer suitable . in comparing the mean strain values @xmath55 ( as in the table ) at the onset of plasticity for a particular strain rate ,
one finds that the uniaxial expansion always starts to yield at the least strain , and the hydrostatic expansion , at the greatest strain .
there are two effects that contribute to the increase in the plastic threshold as the triaxiality increases .
first , the shear component of the applied stress contributes to the resolved shear stress and lowers the threshold for heterogeneous nucleation of dislocations at the void surface .
@xcite and second , the volumetric strain - rate is lowest in the uniaxial case and the highest in the hydrostatic case .
strain - rate hardening then leads to an increase in the stress value at the onset of plasticity as the triaxiality increases .
the difference between the critical strain values when defined as when a behavior deviates from the elastic behavior is nearly negligible and thus independent of whether one uses the criterion from the hydrostatic stress , von mises stress or stress - triaxiality curves .
the differences reflect mainly the difficulties in defining the point what we call the yield point .
however , we will see later that if the mean stress and von mises stress start to deviate from the elastic behavior with the same ratio as they have during elastic expansion , the stress - triaxiality may deviate a bit later than the other quantities .
we shall see later , too , that the onset of plasticity defined from these quantities is very close to where the void starts to grow .
after sufficient expansion , the system yields and the mean stress is observed to drop with respect to the elastic response .
then as the simulation box continues to expand , the stress remains roughly constant until the precipitous drop at final failure . in the region of increasing strain but roughly constant mean stress ,
most of the strain is in the form of plastic strain , a macroscopic measure of the plastic , permanent and irrecoverable deformations in the system . in this section
we study the macroscopic quantities of plasticity such as mean and equivalent plastic strains as well as the plastic work , and in the next section in more detail what are the actual plastic deformations visible in the void .
the concomitant dislocations related to void s shape and volume changes are studied elsewhere .
@xcite in deriving the plastic strain here it is assumed that the tetragonal symmetry is approximately preserved and thus the off - diagonal terms of the stress tensor are negligible . following the literature we separate the strain increment @xmath88 into elastic and plastic parts .
@xcite thus by definition the plastic strain increment becomes @xmath89 where @xmath90 is the total increment of the strain .
below we use @xmath2 instead of @xmath91 since @xmath92 can be divided from both sides of eq .
( [ eq_plastic ] ) .
the total strain increment is an input parameter in these strain - controlled simulations .
it is given by the strain - rate matrix @xmath20 .
the compliance @xmath93 relating the elastic strain increment to the stress increment is derived from the stress - strain curves in the elastic region by : @xmath94 the stress matrix @xmath22 is calculated each time - step using eq .
( [ eq_stresstensor ] ) .
the elastic compliance tensor @xmath95 is retrieved from the elastic part of the stress - strain curve of the cases without the void as follows .
due to the nonlinearity of a stress - strain curve we have not only retrieved the slope of it , which would give @xmath96 , where @xmath97 is the bulk modulus , but fitted a fourth order polynomial to the strain versus stress curve , whose derivative gives us @xmath98 .
this is done separately for the uniaxial , biaxial , and triaxial no - void cases , and the respective curves are used for the cases with the void .
it should be mentioned , too , that in the derivation of the bulk modulus the mean total logarithmic strain is used instead of the engineering principal strain used in the plots of this article .
similarly the term @xmath99 is derived using a fourth order polynomial in the mean strain versus von mises stress curve giving @xmath100 .
note , that when @xmath101 is derived from the plot using mean strain there are prefactors 1/3 and 2/3 for @xmath102 in the uniaxial and biaxial cases , respectively .
using formulas which relate @xmath103 and @xmath104 to @xmath105 and @xmath106 in cubic crystals , @xcite and the correspondence between elastic constants and moduli we get for @xmath103 and @xmath104 @xmath107 using the @xmath108 as @xmath109 and @xmath110 as @xmath111 due to the symmetry and neglecting off - diagonal terms , which are small compared to the diagonal ones , we get all the necessary terms for @xmath112 , and thus @xmath113 from eq .
( [ eq_eepsilon ] ) .
note , that since the @xmath114 is used as a parameter instead of @xmath22 , the von mises stress must be mapped with the mean stress when finding the corresponding @xmath101 .
this was done again by fitting the von mises vs. mean stress curves with fourth order polynomials . subtracting the elastic strain from the total strain as in eq .
( [ eq_plastic ] ) we get the mean plastic strain increment @xmath115 which time - integral is plotted in fig . [ fig4 ] for all the loading modes and strain - rates . in these plots
one sees that after the yield point the mean plastic strain first increases roughly exponentially , although the region is too small to be definitive , and thenceforth roughly linearly .
note that the mean plastic strain is not the equivalent plastic strain commonly used in plasticity , which will be defined below , but a measure of the porosity .
this will be studied in the next section , where the mean plastic strain will be compared with the growth of the volume of the void .
we turn now to the quantification of the dislocation flow , conventionally computed at the continuum level as the second invariant of the deviatoric plastic strain , the _ equivalent plastic strain_. typically in the case of tetragonal total strain , the equivalent plastic strain rate would be calculated as : @xmath116 the equivalent plastic strain is calculated in turn as @xmath117 in practice this formula for the equivalent plastic strain is problematic in md for several reasons .
first , the time and length scales in md are much shorter than those assumed in continuum formulations of plasticity .
the time scale is a problem because dislocation flow becomes partially reversible at short enough time scales .
thermal fluctuations cause reversible oscillations of dislocations and fluctuations in the local elastic strain . to the contrary , the integrand in eq .
( [ eq_equivalent_t ] ) is positive definite , as appropriate for plastic deformation that is cumulative even when reversed . in practice
, the application of eq.([eq_equivalent_t ] ) in md gives a result dominated by the fluctuations for small time increments ; in fact , in our attempt to apply the formula to the md deformation every 10 time steps , the contribution of the fluctuations was 22 times as large as the applied total mean strain ( these values are obtained from the biaxial case with strain - rate @xmath43/sec ) . the formula must be modified to be insensitive to thermal fluctuations .
second , the formula for the equivalent plastic strain assumes isotropic plasticity in the following sense . in isotropic plasticity
, the plastic flow is driven by the shear stress quantified by the von mises stress .
the equivalent plastic strain is conjugate to the von mises stress , and therefore takes on a particular signficance in the theory .
implicit is the assumption , for example , that slip systems that experience the same shear stress will exhibit the same plastic strain .
this assumption is violated in md for two reasons . once again
, the thermal fluctuations may cause the initiation of flow on one glide system before that on a symmetrically related system .
this effect is observed in our md simulations .
typically the symmetry is restored after a brief period , but because the plastic strain is cumulative , the symmetry breaking fluctuation is never eliminated from the plastic strain ( [ eq_equivalent_t ] ) .
second , a more mundane reason the assumption of isotropy fails is that the single crystal systems are anisotropic , both because of the specific glide planes involved and because of the elastic constants , especially in copper . it may be possible to rectify these problems while retaining the basic formulation of the equivalent plastic strain , for example through a suitable multi - resolution calculation of the integral ( [ eq_equivalent_t ] ) .
we have made several attempts at a new formulation , but we were not able to develop a satisfactory algorithm , providing a meaningful measure of the plastic strain on md time scales based on the equivalent plastic strain integral ( [ eq_equivalent_t ] ) .
we found that we could eliminate the anomalies due to fluctuations or the anisotropy , but not both simultaneously in a robust manner .
we have therefore turned to a different quantification of the plastic strain .
certainly , the full deviatoric plastic strain tensor is a measure of the plastic flow , conjugate to the deviator stress .
its rate of increase is given by the traceless part of eq .
( [ eq_plastic ] ) .
typically , the rate would be integrated in a cumulative fashion , but we will not do so .
the nature of our simulations is such that at the continuum level plastic flow is only expected in one direction , so any sign reversal may be attributed to fluctuations .
we then calculate @xmath118 where the plastic strain rate is given by eq .
( [ eq_plastic ] ) .
we emphasize again that no absolute value is taken , so fluctuations cancel . then in order to have a scalar quantification of the plastic strain , we compute the @xmath119 invariant , normalized as the equivalent plastic strain would be : @xmath120 we must stress that this quantity is not equal to the equivalent plastic strain commonly used in plasticity , except in the extraordinary case of monotonic isotropic plasticity .
it is not conjugate to the von mises stress , for example , in our md simulations .
nevertheless , it is a useful qualitative measure of the degree of plasticity , and it allows us to compare the plastic response as the system is loaded in different ways and we call it here _ equivalent plastic strain _ for simplicity . the evolution of the equivalent plastic strain during uniaxial and biaxial expansion is plotted in fig . [ fig5 ] . in the triaxial case
it is essentially zero , as expected by symmetry , and therefore it has not been plotted . in practice ,
the stress during triaxial expansion has only a negligibly small fluctuating shear component , so the calculated elastic shear strains are very small , too .
since neither the box strain nor the elastic strain has an appreciable shear component , the equivalent plastic strain is found to be zero .
now , once the tensors for both the stress and the plastic strain are derived , ( actually only the diagonal terms of the plastic strain are needed ) , the plastic work can be calculated , @xmath121 ( see fig .
[ fig6 ] ) .
it should be compared with the temperature from the same simulations , fig .
note , that in these simulations , when the dilational strain is applied , the thermostat is turned off and thus the temperature is allowed to change .
first the system cools in the elastic regime due to adiabatic cooling on expansion , but when plastic deformations begin , work is done in the system resulting in heating .
we find that the increase in plastic work does not match exactly with the temperature . in principle
we expect several effects to contribute to this difference : the surface energy of the void , the defect formation energies for dislocations and point defects , further adiabatic cooling , and any error in calculating the elastic energy or strain from the stress . using the best data available to bound the contributions from surface , defect and adiabatic cooling energies
, we find that there remains an energy deficit that we attribute to an error in the calculated elastic energy .
the error comes from the use of the average stress despite stress inhomogeneity in the system due to the void : in the plastic work the product of plastic strain and stress is calculated with averaged quantities , while in the temperature the product is calculated at level of each atom and averaged afterward .
we now consider the volume and shape evolution of the void . during the md simulations undergoing expansion , the surface of the void
is determined by finding individual atoms that belong to the surface .
this is done by creating a fine two - dimensional mesh , in which each mesh point corresponds to spherical angular coordinates @xmath122 .
an atom is found to represent the surface at each point of the mesh , with some atoms representing multiple mesh points . in particular , taking the origin to be the center of the void , within the solid angle associated with each mesh point , the atom that is closest to the origin is defined to be the surface atom at that mesh point .
there are , however , some uncertainties related to this method .
if the mesh is too dense with its size diverging it can capture almost all the atoms in the system .
on the other hand if it is too sparse , it may neglect some surface atoms , especially when void is anisotropic , non - spherical , and has some sharp edges in it .
therefore we introduced a width to each of the atoms by drawing a circle around it that implies a width @xmath123 to the angles , so that one atom can occupy several mesh points in a fine mesh .
we have typically 75 - 100 points for each angular coordinate , giving a total of 5625 - 10000 mesh points . in the surface of the void
there are typically few thousand atoms .
besides introducing the width to the atoms , we also select atoms based on their radial distance : if an atom has much larger radial distance @xmath124 compared to its neighbors it is neglected in order not to capture atoms that do not belong to the surface .
once the surface atoms are identified , the surface is tessellated using a generalization of the delaunay triangulation method .
@xcite the delaunay triangulation is an optimal triangulation of a collection of points in our case atoms on the plane .
it is optimal roughly in the sense that the aspect ratio of the triangles is as near to unity as possible ; more precisely , the delaunay theorem guarantees that there is a unique ( up to degeneracy ) triangulation such that if any triangle in the triangulation is circumscribed by a circle , none of the other points will be in the interior of the circle . the delaunay theorem , as formulated ,
does not apply to points on a curved surface .
in fact , there appears to be a topological obstruction to the existence of a unique , optimal triangulation on a closed surface when the euler character is non - zero .
nevertheless , it is possible to extend the delaunay triangulation algorithm to achieve a locally optimal triangulation almost everywhere .
the approach we have taken is to project the points patchwise onto flat surfaces . in particular ,
stereographic projections are used to project the upper and lower hemispheres separately onto planes .
cylindrical coordinates are used to project the equatorial region to a cylinder .
the delaunay algorithm is used to triangulate each of these projected regions .
the patches overlap at latitudes of @xmath125 where the projections are not too distorted .
the three patches are sewn together using a simple advancing front triangulation at the boundaries . using this triangulation ,
the volume of the void can be calculated precisely by summing up the volumes of the tetrahedra with one apex at the center of the void and the opposing face on the void surface .
as we shall see below , the void shape evolves to be far from spherical .
therefore the approximation of the void surface by triangles captures the shape better than just assuming it to be spherical and using only the solid angles and radial distances of a sphere , when calculating the volume of the void .
an advantage of this method is that if some atom , which should be taken into account , is missing from the surface of the void , its position is filled with the triangles created by its neighboring atoms , and thus the `` hole '' is well approximated by its neighbors . in fig .
[ fig8 ] the porosity or the ratio between the volume of the void and the total volume of the system , @xmath126 is plotted for a fraction of the simulations of the strain - rates @xmath36/sec , @xmath37/sec , @xmath38/sec , and @xmath8/sec .
it should be noted that in order to get information about the positions of the atoms for the strain values of the interest ( _ i.e. _ , close to and after the onset of plasticity ) the calculations were restarted from already expanded system .
after the restart the expansion was applied with the same strain - rate as earlier but now to the already expanded system , thus the strain - rate was increased by a few percent compared to the original [ since @xmath127 in eq .
( [ eq_update ] ) was the restart value ] .
therefore these simulations are not precisely comparable with the ones plotted in figs .
[ fig1]-[fig7 ] , where the continuum quantities are shown . in these plots , as for the mean plastic strain , in most cases first the void grows exponentially and then ( if the calculation has been carried out that long ) it changes to a linear growth , see especially fig .
[ fig8](b ) and the strain - rates @xmath37/sec and @xmath128/sec there .
the cross - over to the linear growth happens at the same point as the rate of decrease of the mean stress slows .
although it is beyond the scope of this article to go into the analysis , the reduction in the growth rate coincides with the point at which the dislocation density along the shortest distance between the void and its periodic image ( at the apices of the faceted void , cf.section [ voidshape ] ) reaches saturation , and the nature of the dislocation activity changes dramatically .
this can be interpreted as a finite size effect as the void approaches the boundary of the simulation box or as a start of the coalescence process of the void with its own periodic image .
the void - void interactions and the coalescence process of two voids in a less restrictive geometry will be presented elsewhere .
@xcite the shapes of the porosity curves as a whole can be compared with the mean plastic strain plotted in fig .
although the volume of the void is not calculated throughout the whole simulation , one sees easily , that there is correspondence between mean plastic strain and the volume of the void .
there is of course an offset at the elastic part of the simulations , since mean plastic strain equals zero then , but the initial volume of the void is finite .
the correspondence will be revisited and studied more carefully in section [ summary ] .
however , it can be concluded already here that the macroscopic quantity mean plastic strain captures the microscopic behavior of the void growth very well .
effects such as the excess volume associated with defects are negligible .
this also means that the matrix material is plastically incompressible , the dilation comes from the void growth , and thus it is consistent with the gurson type of continuum models .
@xcite let us now look at the shape evolution of the void in more detail . in fig .
[ fig9 ] snapshots of the void are shown from uniaxial expansion at the strain - rate @xmath43/sec .
there are several interesting aspects in the snapshots . in the first two snapshots at strains @xmath129 and @xmath130 when the system still behaves elastically , the void is expanded in the @xmath131-direction , which is the direction of the strain . however , after that the void makes a rapid shape change and becomes more extended in the transverse @xmath132 and @xmath133-directions , _ i.e. _ , the strain - free directions , than @xmath131-direction .
this prolate - to - oblate transition may be counterintuitive , but the behavior has been observed previously in continuum calculations,@xcite and it has been related to the appearance of shallow dimples in the fractography studies of ductile fracture surfaces in low triaxiality conditions .
see also studies of non - spherical voids .
@xcite for example , budiansky _ et al .
_ , ref .
, investigated void shape change in a non - linear viscous plastic flow model .
they explained the oblate growth of voids under uniaxial loading as due to a non - linear amplification of the shear stress on the surface of the void , with the maximal void growth rate at the locations of maximal von mises stress : the equator .
their analysis does not apply directly to our simulations since they neglect elasticity , and the non - linear viscous solid model they have used is not expected to be a precise description of the plastic flow early in our md simulation when the prolate - to - oblate transition takes place .
furthermore , it is not clear from our simulations what value should be assigned to the strain - rate exponent that controls the non - linearity in the model of budiansky _
et al_. , although a large value is reasonable . despite these differences , the same localization of plastic flow to the equator of the void and the transition to an oblate shape
does appear in both our simulations and the viscous solid model of budiansky _
et al_. following some additional expansion beyond the transition , the void begins to become faceted , as visible in the last three snapshots .
it should be mentioned that the anisotropy visible in this uniaxial case is less pronounced in the biaxial case .
the cases with the hydrostatic loading are the most isotropic and the octahedral shape , somewhat visible in fig .
[ fig9](f ) , becomes more pronounced .
@xcite the octahedral shape has been seen in spallation experiments in the fcc metal aluminum,@xcite and also in experiments on the equilibrium void shape of another material , silicon , too , where it has been used to calculate anisotropic surface energies through an inverse wulff construction .
@xcite in void growth associated with dynamic fracture in copper , several effects contribute to the faceting : the low surface energy and high ad - atom energy of the \{111 } surfaces common to fcc metals , the high anisotropy of the copper elastic constants ( a=3.21 ) , and the \{111 } dislocation glide systems .
these effects are analyzed in detail elsewhere .
@xcite in order to characterize the shape change of the void not only qualitatively and visually , but also quantitatively , multipole moments of the void shape are calculated . to the best of our knowledge , this is the first time that multipole moments have been used to characterize surface shape .
they are a powerful way to quantify the evolution of the complex surface containing thousands of atoms , and they are suitable for use in continuum models and experimental void characterization as well .
using spherical harmonics , @xmath134 expressed as polynomials of cartesian coordinates , in contrast to more commonly used trigonometric forms , @xcite we are able to define different multipole moments of the void based on its surface atoms : @xmath135 where the mean square radius @xmath136 .
this is in contrast to the volume integral more commonly used when calculating multipole moments .
the axial index of the moment ranges @xmath137 , and for each @xmath138 except @xmath139 , @xmath140 has both real and imaginary parts .
here we concentrate on @xmath141 , and 4 .
only the positive moments of @xmath138 are calculated , since the negative ones are related by @xmath142 in all 24 different terms are calculated .
the polynomial forms used here of the moments are listed in the appendix .
the moments @xmath140 are not rotationally invariant , but depend on the way the coordinates @xmath131 , @xmath132 , and @xmath133 are chosen .
the set of @xmath143 moments at fixed @xmath14 form an irreducible representation of the so(3 ) rotation group , and are mixed by rotations according to the usual transformation rules .
they may be combined into a single rotationally invariant combination for each @xmath14 according to @xmath144^{1/2 } , \label{eq_ql}\ ] ] see _ e.g. _ ref . .
their use drastically reduces the amount of data to be shown .
only the positive @xmath138 s are needed for @xmath145 due to the square of the norm of @xmath140 and eq .
( [ eq_rel ] ) .
technically the calculation of the multipole moments has been done using the information about the shape of the void obtained from the surface triangulation procedure explained earlier . as in the calculation of the void volume ,
some refinements have been introduced to reduce the uncertainty in the values of the moments that arises from single atoms near the threshold for inclusion as surface atoms .
these borderline atoms can appear intermittently on the void surface during the growth , and the tessellation is used to minimize their effect on the moments . in calculating the volume of the void the triangulation gave one face of the tetrahedra that acted as small volume elements for the total volume . here
the triangulation is used to weight the atoms by the amount of solid angle associated with each surface atom .
in particular , each triangle in the tessalation contributes one third of its projected solid angle to each of the three atoms that make up its vertices , where the solid angle of a triangle is computed using the formula @xmath146 , where @xmath147 and @xmath148 for @xmath149 ( mod 3 ) and where @xmath150 is the unit vector in the direction of the @xmath151 vertex of the triangle . @xcite the weight of each atom is the sum of these solid angle contributions .
this reduces the sensitivity of moments to the atomic discretization of the surface , since evanescent atoms that occasionally appear and disappear from the fluctuating surface only make a small , local change to the value of @xmath152 .
it may be of interest to note that in the course of the development of these surface multipole moments , several different variations on the definition of the moments were tried .
the definition presented here ( [ eq_qlm ] ) produced substantially less noise ( up to a factor of 5 less noise ) than the other definitions we tried , even though they all showed the same trends in the shape evolution . using these weights for the atoms and after normalizing each atom s @xmath153 coordinate by its distance @xmath154 from the center of the void all the terms in eqs .
( [ eq_ql1])([eq_ql4 ] ) , and eq . ( [ eq_ql ] ) , are calculated .
the center of the void is defined to be the point where the three components of @xmath155 , as given by eq .
( [ eq_ql1 ] ) , are zero . due to space limitations , only a fraction of the multipole moment data is shown here . in fig .
[ fig10](a ) the quadrupole moments @xmath156 for all the positive @xmath138 of the void are shown in the uniaxial case for the strain - rate @xmath43/sec .
this is the same simulation as the snapshots in fig .
indeed , the quadrupole moments are able to represent numerically the shape changes one sees in the snapshots . in figs .
[ fig9](a ) and ( b ) at strains @xmath129 and @xmath130 the void is elongated to the direction of the load , which is visible as @xmath157 between strains @xmath158 [ fig .
[ fig9](c ) ] and @xmath159 [ fig .
[ fig9](d ) ] the void is extended more transverse to the direction of the strain , thus @xmath160 , and later it starts to becomes more of octahedral shape and the absolute value of @xmath161 saturates .
figures [ fig10](b)(d ) show the rotational invariant multipole moments @xmath145 , eq .
( [ eq_ql ] ) , in cases with uniaxial , biaxial , and triaxial loading , respectively .
each of the cases has strain - rate @xmath43/sec . in the plots it is clear that the behavior that the quadrupole moment has first a non - zero value and then makes a rapid dip but returns back to a non - zero value due to the transverse elongation is the strongest in the uniaxial case . on the other hand
the octahedral shape measured by the @xmath162 is more pronounced in the biaxial and triaxial cases than in the uniaxial case as explained qualitatively earlier .
hence we find that the multipole moments introduce a good method to measure the shape changes of the void .
the non - zero values for @xmath163 , as well as @xmath161 in other cases than uniaxial , indicate that the void is not ( cubically ) symmetric in these simulations .
it should be mentioned that these first four moments are enough to characterize the shapes of the void and the higher moments contain little relevant information .
this was checked by creating a three dimensional surface based on the moment values and drawing it in the same figure with the actual positions of the surface atoms using a standard commercial visualization program .
the surfaces overlapped very well .
based on the data shown earlier in this article for the shape and volume changes of the void as well as the stress - strain behavior and the stress - triaxiality , it is evident that the uniaxial loading raises many interesting aspects to be studied in more detail .
therefore we now concentrate on the uniaxial case when summarizing how the void evolves and how the evolution is related to the stress - triaxiality as the system is expanded . by plotting most of the measured values together in one figure for the uniaxial simulation at the strain - rate @xmath43/sec , it is possible to compare the evolution sequence and investigate causality , see fig .
[ fig11](a ) . for clarity ,
we have chosen not to plot many quantities in the figure , _
e.g. _ plastic strain , plastic work , and temperature .
however , their concomitant behaviors are included in the verbal explanation below and shown in previous figures .
the data shown in this figure are from the restarted simulation ( see the explanation near fig . [ fig8 ] ) , as are the data in figs .
[ fig8]-[fig10 ] .
the mean stress and stress - triaxiality data are from that simulation , too , and thus are different from the data shown in figs .
[ fig1](a ) and [ fig3](a ) . in any case
, the overall behavior stays the same as well as the other details as the system size , _ etc .
_ the evolution of the void and the system s stress - strain behavior can be divided in three or even four different regions .
the first region is when the system expands elastically .
the mean and von mises stresses increase smoothly , nearly linearly , and the stress - triaxiality stays nearly constant . through the elastic region
the void volume fraction remains nearly constant , too .
it is not exactly constant , since due to the free surface of the void , the elastic expansion is a bit greater at the surface of the void compared to the total system .
trivially the mean and equivalent plastic strains as well as the plastic work are equal to zero in the elastic region , and temperature decreases in the system .
the quadrupole moment has a non - zero value because of the elongation in the direction of the strain .
the second region begins at what we call the yield point , _ i.e. _ , the onset of rapid growth of the void facilitated by plastic deformation .
heterogeneous nucleation of dislocations at the void surface is the primary mechanism for plasticity in the simulation , and thus it is at this point that the measured quantities start to deviate from their elastic behavior .
the mean stress begins to plateau here , but fluctuating somewhat .
the early departure from elastic behavior prior to the plateau is much less pronounced .
the change in the void shape begins just at the point the mean stress deviates from elastic behavior : @xmath164 goes rapidly from the positive value acquired during elastic expansion to a negative value , _
i.e. _ , from a prolate shape ( elongated in the direction of the strain ) to an oblate shape ( expanded in the transverse directions ) .
@xmath161 , on the other hand , drops from a positive value , almost reaching zero at the prolate - to - oblate transition point ( @xmath165% ) and rising even larger value after that [ in fact , the oblate shape is somewhat more pronounced than the earlier prolate shape , seen as a larger absolute value of @xmath164 in fig .
[ fig10](a ) ] .
when @xmath161 starts to drop @xmath162 starts to rise . then after the prolate - to - oblate transition point , @xmath162 begins to saturate . at a strain of @xmath166 5.55%
, @xmath161 is 1.5 times as large as its value at the end of the elastic phase ( @xmath166 5.25% ) .
mean plastic strain , equivalent plastic strain , plastic work , and temperature increase together with porosity .
their increase starts immediately at the yield point , _ i.e. _ , when the mean stress first deviates from the elastic behavior .
a bit later than the plastic strain , equivalent plastic strain , plastic work , and temperature , the stress triaxiality increases simultaneously with the first substantial drop in the von mises stress .
the fact that the increase in stress - triaxiality follows later is dependent on how the ratio between mean stress and von mises stress develops , as discussed earlier . in fig .
[ fig3](a ) and in the data reported in table [ tab1 ] the stress - triaxiality started to increase simultaneously with the mean and von mises stresses deviating from the elastic behavior .
the increase of stress - triaxiality is caused by von mises stress plummeting in contrast to nearly constant mean stress .
the drop in von mises stress follows from the flow of dislocations nucleated at the void and from the relaxation of the shear stress of the system due to the flow .
the third region is when the void fraction , mean plastic strain , equivalent plastic strain , plastic work , and temperature switch from rapid increase to a linear growth or even saturate .
subsequently the increase of the stress - triaxiality slows down and plateaus .
the value at the plateau is related in continuum models to the ratio of the mean stress threshold for void growth to the flow stress . at
the plateau von mises stress saturates at a small value , close to the tensile strength , and the shape of the void starts to become more of octahedral shape although having a non - zero quadrupole moment , too .
hence at the second and third regions the mean stress is nearly constant , but von mises stress and the stress - triaxiality changes .
a conclusion might be that once the threshold for void growth is reached , the population of dislocations rises sufficiently to relax the shear stress quite effectively and it drops to a low value ( the flow stress ) ; the mean stress , on the other hand , plateaus since it is relaxed by void growth and requires that the stress at the void surface be sufficiently high to continue to nucleate dislocations .
the fourth region is the failure , when the system breaks , and it is not studied here . in order to see if the rapid changes studied above are due to the smallness of the size of the void we have done one additional simulation with a system in which the initial void radius is half that in the other simulations ; otherwise , the system size is the same , see fig .
[ fig11](b ) . for the small void simulation ,
uniaxial strain at a strain - rate @xmath43/sec has been used .
here the difference is that the quadrupole moment suffers stronger fluctuation due to the smallness of the void , where each of the surface atom contribute more to its value and therefore is even more sensitive to the selection criterion of surface atoms , and also the shape changes are harder to determine based on the @xmath164 behavior .
the other main difference is that the growth of the void is linear all the time .
also the changes in stress - triaxiality in fact advances the porosity when saturating , and the mean stress does not fluctuate but drops more rapidly ( this can be compared with the case without the void , where the mean stress drops abruptly ) . we have also compared the mean plastic strain and the void volume fraction calculations . in continuum
solid mechanics , it is assumed that solid materials are plastically incompressible .
any local dilation , as indicated by a change in the mean strain , is attributed either to elastic dilation or to a change in the porosity of the material , where the porosity is equated to the void volume fraction .
the porosity @xmath167 and the mean plastic strain @xmath168 are then related according to the equation @xcite @xmath169 where the dots denote time derivatives .
integration with respect to time , porosity from @xmath170 to @xmath167 , and mean plastic strain from zero to @xmath168 , gives @xmath171 where @xmath170 is the initial porosity .
it is interesting to check whether this relationship holds for the md simulation , where other effects such as excess volume associated with dislocation cores , vacancies or other defects could require corrections . in comparing the porosity inferred from the mean plastic strain and
that calculated directly as the void fraction , the agreement is very good .
the trends are in excellent agreement , but there is a small discrepancy between the curves , so that the porosity from the mean plastic strain is overestimated .
we believe that the discrepancy arises because of the void surface . in calculating the void fraction ,
we have defined the void surface to pass through the center of the surface atoms .
however , the properties of the surface atoms are distinct from the bulk atoms .
therefore , there is some ambiguity in where the surface should be placed , and a small uniform shift @xmath172 of the surface radially into the bulk is enough to account for the discrepancy . in fig .
[ fig12 ] we have plotted the comparison of the porosity from the mean plastic strain , eq .
( [ eq_porosity ] ) , and from the void fraction , eq .
( [ eq_void_frac ] ) , using a constant radius increase @xmath173 , where @xmath174 is the lattice constant for the void volume calculation .
the correction for the void size , @xmath175 , is a fit parameter and it varies for different strain - rates and slightly for different loading modes , but is always positive and of the order of the lattice constant , @xmath174
. it should be noted , too , that by taylor expanding eq .
( [ eq_porosity ] ) and discarding the higher order terms , it becomes @xmath176 indicating a linear correspondence between @xmath167 and @xmath168 as long as the void fraction is small .
in this article void growth in copper has been studied in a high range of strain - rates at the atomistic level .
the model has been designed to simulate the growth of a void nucleating from a very weakly bound inclusion during strain - controlled dynamic fracture . in order to see the effect of various modes of expansion and the related stress - triaxiality ,
three different modes have been applied , namely uniaxial , biaxial , and triaxial . the molecular dynamics method developed here
has been shown to be efficient enough to explore the different loading conditions and strain - rates varying over four orders of magnitude .
a uniform expansive loading of a system with periodic boundary conditions has been implemented using a well defined scaling matrix method . for the longest calculations ,
the md method was parallelized successfully .
the macroscopic observables mean stress , von mises stress , stress - triaxiality , mean plastic strain , equivalent plastic strain , plastic work , and temperature have been calculated and compared with the microscopic quantities measured at the atomistic - level , such as the volume of the void and its shape change .
a method to describe the shape changes in the void is introduced and employed , namely calculations of the multipole moments of the void based on spherical harmonics in polynomial , not trigonometric , form . when calculating the volume of the void with an unknown shape or defining solid surface for the multipole moment calculation a useful method , namely optimal triangular tessellation , has been introduced .
this method has been extended from the usual planar case to non - planar objects such as the surface of the void .
when the different measured quantities are compared with each other during an md simulation in uniaxial expansion , it is found that at early stages of plasticity von mises stress , and thus also stress - triaxiality , plays a more significant role to the void growth and its shape change than expected . on the other hand ,
most of the macroscopic plastic quantities as mean and equivalent plastic strain as well as plastic work and temperature , seem to be more dependent on the simultaneous saturation of the mean stress .
these calculations show a counter - intuitive behavior , observed previously in continuum void growth modeling , @xcite that a prolate - to - oblate transition occurs .
when the system starts to yield , the expansion of the void switches from its original elastic extension in the direction of the load to transverse plastic expansion .
the yield stress values for the lowest strain - rates @xmath0/sec are in reasonable agreement with the experimentally measured spall strength .
@xcite the fact that mean plastic strain can be mapped to the growth of the void is consistent with continuum models .
@xcite this study leaves many open questions .
for instance related to the void growth are the dislocations , which occur when the system yields . since the fcc crystal studied here is perfect apart from the void , the dislocations form from void s surface .
they are also responsible for its growth by carrying material away .
thus the characterization of plasticity surrounding a growing void at the level of dislocations should be investigated , too , especially with respect to the stress - triaxiality .
these investigations are underway .
@xcite their results are beyond the scope of this article , other than to mention that the identification of the yield point in this article does indeed correspond to the point of initial nucleation of dislocations .
another topic that is beyond the scope of this article and needs further investigation , but is closely related to the studies here , is the quantitative connection between the shear stress , and thus the mode of the loading , and the onset of the void growth and the resulting change in the stress state .
other areas where this study can be extended are different materials including different lattice structures such as body - centered cubic ( bcc ) lattices ; in the uniaxial case other orientations of the lattice as @xmath177 and @xmath178 ; continuously changing stress - triaxiality in order to create the full yield surface to the stress space similarly as in gurson type of continuum studies ; @xcite to include grain boundaries , defects , pre - existing dislocations , several voids , etc .
this work was performed under the auspices of the u.s .
of energy by the university of california , lawrence livermore national laboratory , under contract number w-7405-eng-48 .
the authors would like to thank dr .
richard becker bringing to our attention refs . and that find the prolate - oblate transition in continuum modeling .
the 24 different polynomial terms of the multipole moments used in this study are listed below .
@xcite here the conventional notation is used , so that the principal axis of the coordinates is the @xmath133-direction . when these formulas are used in interpreting the void shape evolution , the principal axis is the direction with uniaxial loading , _ i.e. _ , the @xmath131-coordinate in the article
. similarly the load free directions @xmath132 and @xmath133 correspond to @xmath131 and @xmath132 below . | the effect of stress - triaxiality on growth of a void in a three dimensional single - crystal face - centered - cubic ( fcc ) lattice has been studied .
molecular dynamics ( md ) simulations using an embedded - atom ( eam ) potential for copper have been performed at room temperature and using strain controlling with high strain rates ranging from @xmath0/sec to @xmath1/sec .
strain - rates of these magnitudes can be studied experimentally , _
e.g. _ using shock waves induced by laser ablation .
void growth has been simulated in three different conditions , namely uniaxial , biaxial , and triaxial expansion .
the response of the system in the three cases have been compared in terms of the void growth rate , the detailed void shape evolution , and the stress - strain behavior including the development of plastic strain .
also macroscopic observables as plastic work and porosity have been computed from the atomistic level . the stress thresholds for void growth are found to be comparable with spall strength values determined by dynamic fracture experiments .
the conventional macroscopic assumption that the mean plastic strain results from the growth of the void is validated .
the evolution of the system in the uniaxial case is found to exhibit four different regimes : elastic expansion ; plastic yielding , when the mean stress is nearly constant , but the stress - triaxiality increases rapidly together with exponential growth of the void ; saturation of the stress - triaxiality ; and finally the failure . |
the kaczmarz method is an iterative projection algorithm for solving linear systems of equations @xcite . due to its simplicity
, the kaczmarz method has found numerous applications including image reconstruction , distributed computation and signal processing to name a few @xcite , see @xcite for more applications .
the kaczmarz method has also been rediscovered in the field of image reconstruction and called art ( algebraic reconstruction technique ) @xcite , see also @xcite for additional references .
it has been also applied to more general settings , see ( * ? ? ?
* table 1 ) and @xcite for non - linear versions of the kaczmarz method .
let @xmath0 and @xmath1 . throughout the paper all vectors
are assumed to be column vectors .
the kaczmarz method operates as follows : initially , it starts with an arbitrary vector @xmath2 . in each iteration
, the kaczmarz method goes through the rows of @xmath3 in a cyclic manner . ] and for each selected row , say @xmath4-th row @xmath5 , it orthogonally projects the current estimate vector onto the affine hyperplane defined by the @xmath4-th constraint of @xmath6 , i.e. , @xmath7 where @xmath8 is the euclidean inner product . more precisely , assuming that the @xmath9-th row has been selected at @xmath10-th iteration , then the @xmath11-th estimate vector @xmath12 is inductively defined by @xmath13 where @xmath14 are the so - called relaxation parameters and
@xmath15 denotes the euclidean norm .
the original kaczmarz method corresponds to @xmath16 for all @xmath17 and all other setting of @xmath18 s are usually referred as the _ relaxed kaczmarz method _ in the literature @xcite .
kaczmarz proved that this process converges to the unique solution for square non - singular matrices @xcite , but without any attempt to bound the rate of convergence .
bounds on the rate of convergence of the kaczmarz method are given in @xcite , @xcite and ( * ? ? ?
* theorem 4.4 , p.120 ) .
in addition , an error analysis of the kaczmarz method under the finite precision model of computation is given in @xcite .
nevertheless , the kaczmarz method converges even if the linear system @xmath6 is overdetermined ( @xmath19 ) and has no solution . in this case and provided that @xmath3 has full column rank , the kaczmarz method converges to the least squares estimate .
this was first observed by whitney and meany @xcite who proved that the relaxed kaczmarz method converges provided that the relaxation parameters are within @xmath20 $ ] and @xmath21 , see also ( * ? ? ?
* theorem 1 ) , @xcite and @xcite for additional references . in the literature
there was empirical evidence that selecting the rows non - uniformly at random may be more effective than selecting the rows via kaczmarz s cyclic manner @xcite .
towards explaining such an empirical evidence , strohmer and vershynin proposed a simple randomized variant of the kaczmarz method that has exponential convergence _ in expectation _ @xcite assuming that the linear system is solvable ; see also @xcite for extensions to linear constraints . a randomized iterative algorithm that computes a sequence of random vectors @xmath22
is said to _ converge in expectation _ to a vector @xmath23 if and only if @xmath24 as @xmath25 , where the expectation is taken over the random choices of the algorithm .
soon after @xcite , needell analyzed the behavior of the randomized kaczmarz method for the case of full column rank linear systems that do not have any solution @xcite .
namely , needell proved that the randomized kaczmarz estimate vector is ( in the limit ) within a fixed distance from the least squares solution and also that this distance is proportional to the distance of @xmath26 from the column space of @xmath3 .
in other words , needell proved that the randomized kaczmarz method is effective for least squares problems whose least squares error is negligible .
in this paper we present a randomized iterative least squares solver ( algorithm [ alg : rek ] ) that converges in expectation to the minimum euclidean norm solution of @xmath27 the proposed algorithm is based on @xcite and inspired by @xcite .
more precisely the proposed algorithm can be thought of as a randomized variant of popa s extended kaczmarz method @xcite , therefore we named it as _ randomized extended kaczmarz_. [ [ organization - of - the - paper ] ] organization of the paper + + + + + + + + + + + + + + + + + + + + + + + + + in section [ sec : related ] , we briefly discuss related work on the design of deterministic and randomized algorithms for solving least squares problems . in section [ sec : back ] , we present a randomized iterative algorithm for projecting a vector onto a subspace ( represented as the column space of a given matrix ) which may be of independent interest .
in addition , we discuss the convergence properties of the randomized kaczmarz algorithm for solvable systems ( section [ sec : rk ] ) and recall its analysis for non - solvable systems ( section [ sec : noisyrk ] ) . in section [ sec : result ] , we present and analyze the randomized extended kaczmarz algorithm . finally , in section [ sec : impl ] we provide a numerical evaluation of the proposed algorithm .
in this section we give a brief discussion on least squares solvers including deterministic direct and iterative algorithms together with recently proposed randomized algorithms . for a detailed discussion on deterministic methods ,
the reader is referred to @xcite .
in addition , we place our contribution in context with prior work . [ [ deterministic - algorithms ] ] deterministic algorithms + + + + + + + + + + + + + + + + + + + + + + + + in the literature , several methods have been proposed for solving least squares problems of the form . here
we briefly describe a representative sample of such methods including the use of qr factorization with pivoting , the use of the singular value decomposition ( svd ) and iterative methods such as krylov subspace methods applied on the normal equations @xcite .
lapack provides robust implementations of the first two methods ; dgelsy uses qr factorization with pivoting and dgelsd uses the singular value decomposition @xcite . for the iterative methods ,
lsqr is equivalent to applying the conjugate gradient method on the normal equations @xcite and it is a robust and numerically stable method . [ [ randomized - algorithms ] ] randomized algorithms + + + + + + + + + + + + + + + + + + + + + to the best of our knowledge , most randomized algorithms proposed in the theoretical computer science literature for approximately solving least squares are mainly based on the following generic two step procedure : first randomly ( and efficiently ) project the linear system into sufficiently many dimensions , and second return the solution of the down - sampled linear system as an approximation to the original optimal solution @xcite , see also @xcite .
concentration of measure arguments imply that the optimal solution of the down - sampled system is close to the optimal solution of the original system .
the accuracy of the approximate solution using this approach depends on the sample size and to achieve relative accuracy @xmath28 , the sample size should depend inverse polynomially on @xmath28 .
this makes these approaches unsuitable for the high - precision regime of error that is considered here .
a different approach is the so called randomized preconditioning method , see @xcite .
the authors of @xcite implemented blendenpik , a high - precision least squares solver .
blendenpik consists of two steps . in the first step ,
the input matrix is randomly projected and an effective preconditioning matrix is extracted from the projected matrix . in the second step ,
an iterative least squares solver such as the lsqr algorithm of paige and saunders @xcite is applied on the preconditioned system .
blendenpik is effective for overdetermined and underdetermined problems .
a parallel iterative least squares solver based on normal random projections called lsrn was recently implemented by meng , saunders and mahoney @xcite .
lsrn consists of two phases . in the first preconditioning phase
, the original system is projected using random normal projection from which a preconditioner is extracted . in the second step ,
an iterative method such as lsqr or the chebyshev semi - iterative method @xcite is applied on the preconditioned system .
this approach is also effective for over - determined and under - determined least squares problems assuming the existence of a parallel computational environment . in section [ sec : impl ]
, we compare the randomized extended kaczmarz algorithm against dgelsy , dgelsd , blendenpik .
lsrn @xcite did not perform well under a setup in which no parallelization is allowed , so we do not include lsrn s performance .
the numerical evaluation of section [ sec : impl ] indicates that the randomized extended kaczmarz is effective on the case of sparse , well - conditioned and strongly rectangular ( both overdetermined and underdetermined ) least squares problems , see figure [ fig : sparse ] .
moreover , the randomized extended kaczmarz algorithm has also comparable performance with lapack s routine for the dense random input matrices , see figure [ fig : dense ] ( notice that the proposed algorithm almost matches blendenpik s performance for the underdetermined case , see figure [ fig : denseunder ] ) . on the other hand , a preconditioned version of the proposed algorithm
does not perform well under the case of ill - conditioned matrices , see figure [ fig : cond ] .
[ [ preliminaries - and - notation ] ] preliminaries and notation + + + + + + + + + + + + + + + + + + + + + + + + + + for an integer @xmath29 , let @xmath30:=\{1,\ldots , m\}$ ] . throughout the paper all vectors
are assumed to be column vectors .
we denote the rows and columns of @xmath3 by @xmath31 and @xmath32 , respectively ( both viewed as column vectors ) .
@xmath33 denotes the column space of @xmath3 , i.e. , @xmath34 and @xmath35 denotes the orthogonal complement of @xmath33 . given any @xmath1 , we can uniquely write it as @xmath36 , where @xmath37 is the projection of @xmath26 onto @xmath33 . @xmath38 and @xmath39 denotes the frobenius norm and spectral norm , respectively . let @xmath40 be the non - zero singular values of @xmath3 .
we will usually refer to @xmath41 and @xmath42 as @xmath43 and @xmath44 , respectively .
the moore - pensore pseudo - inverse of @xmath3 is denoted by @xmath45 @xcite .
recall that @xmath46 . for any non - zero real matrix @xmath3
, we define @xmath47 related to this is the scaled square condition number introduced by demmel in @xcite , see also @xcite .
it is easy to check that the above parameter @xmath48 is related with the condition number of @xmath3 , @xmath49 , via the inequalities : @xmath50 .
we denote by @xmath51 the number of non - zero entries of its argument matrix .
we define the _ average row sparsity _ and _ average column sparsity _ of @xmath3 by @xmath52 and @xmath53 , respectively , as follows : @xmath54 where @xmath55 for every @xmath56}$ ] and @xmath57 for every @xmath58}$ ]
. the following fact will be used extensively in the paper .
[ fact : xls ] let @xmath3 be any non - zero real @xmath59 matrix and @xmath1
. denote by @xmath60 . then @xmath61 .
we frequently use the inequality @xmath62 for every @xmath63 .
we conclude this section by collecting a few basic facts from probability theory that will be frequently used . for any random variable @xmath64 ,
we denote its expectation by @xmath65 $ ] or @xmath66 .
if @xmath64 is a non - negative random variable , markov s inequality states that @xmath67 $ ] .
let @xmath64 and @xmath68 be two random variables , then @xmath69=\operatorname{\mathbb{e}}[x ] + \operatorname{\mathbb{e}}[y]$ ] .
we will refer to this fact as _ linearity of expectation_. let @xmath70 be a set of events defined over some probability space holding with probabilities @xmath71 respectively , then @xmath72 .
we refer to this fact as _
union bound_. initialize @xmath73 pick @xmath74 $ ] with probability @xmath75 $ ] set @xmath76 output @xmath77 in this section we present a randomized iterative algorithm ( algorithm [ alg : randop ] ) that , given any vector @xmath1 and a linear subspace of @xmath78 represented as the column space of a given matrix @xmath3 , approximately computes the orthogonal projection of @xmath26 onto the column space of @xmath3 ( denoted by @xmath37 , @xmath79 ) , see @xcite for a different approach .
algorithm [ alg : randop ] is iterative .
initially , it starts with @xmath80 . at the @xmath10-th iteration ,
the algorithm randomly selects a column @xmath81 of @xmath3 for some @xmath82 , and updates @xmath83 by projecting it onto the orthogonal complement of the space of @xmath81 .
the claim is that randomly selecting the columns of @xmath3 with probability proportional to their square norms implies that the algorithm converges to @xmath84 in expectation .
after @xmath85 iterations , the algorithm outputs @xmath77 and by orthogonality @xmath86 serves as an approximation for @xmath37 .
the next theorem bounds the expected rate of convergence for algorithm [ alg : randop ] .
[ thm : randop ] let @xmath0 , @xmath1 and @xmath87 be the input to algorithm [ alg : randop ] .
fix any integer @xmath88 . in exact arithmetic , after @xmath10 iterations of algorithm [ alg : randop ] it holds that @xmath89 moreover , each iteration of algorithm [ alg : randop ] requires in expectation ( over the random choices of the algorithm ) at most @xmath90 arithmetic operations .
a suggestion for a stopping criterion for algorithm [ alg : randop ] is to regularly check : @xmath91 for some given accuracy @xmath92 .
it is easy to see that whenever this criterion is satisfied , it holds that @xmath93 , i.e. , @xmath94 .
we devote the rest of this subsection to prove theorem [ thm : randop ] .
define @xmath95 for every @xmath96 $ ] .
observe that @xmath97 , i.e. , @xmath98 is a projector matrix .
let @xmath64 be a random variable over @xmath99 that picks index @xmath82 with probability @xmath100 .
it is clear that @xmath101 = { \mathbf{i}}_m - { { { \ensuremath{\mathsf{a } } } } } { { { \ensuremath{\mathsf{a } } } } } ^\top /{\ensuremath{\left\| { { { \ensuremath{\mathsf{a } } } } } \right\|_{\text{\rm f}}}}^2 $ ] .
later we will make use of the following fact .
[ lem : technical ] for every vector @xmath102 in the column space of @xmath3 , it holds @xmath103 .
define @xmath104 for every @xmath17 .
a direct calculation implies that @xmath105 indeed , @xmath106 using the definitions of @xmath107 , @xmath83 , @xmath108 and the fact that @xmath109 for any @xmath110}$ ] .
moreover , it is easy to see that for every @xmath111 @xmath107 is in the column space of @xmath3 , since @xmath112 , @xmath113 and in addition @xmath114 is a projector matrix for every @xmath115 $ ] .
let @xmath116 be a sequence of independent and identically distributed random variables distributed as @xmath64 .
for ease of notation , we denote by @xmath117 = \operatorname{\mathbb{e}}_{x_k } [ \cdot\ |\ x_1 , x_2 , \ldots , x_{k-1}]$ ] , i.e. , the conditional expectation conditioned on the first @xmath118 iteration of the algorithm .
it follows that @xmath119 { \ensuremath{{\mathbf e}}}^{(k-1 ) } } \right\rangle } \\ & \leq { \ensuremath{\left\|{\ensuremath{{\mathbf e}}}^{(k-1)}\right\|_2 } } { \ensuremath{\left\| \left({\mathbf{i}}_m - \frac { { { { \ensuremath{\mathsf{a } } } } } { { { \ensuremath{\mathsf{a } } } } } ^\top } { { \ensuremath{\left\| { { { \ensuremath{\mathsf{a } } } } } \right\|_{\text{\rm f}}}}^2}\right ) { \ensuremath{{\mathbf e}}}^{(k-1 ) } \right\|_2 } } \ \leq \ \left(1 - \frac{\sigma^2_{\min}}{{\ensuremath{\left\| { { { \ensuremath{\mathsf{a } } } } } \right\|_{\text{\rm f}}}}^2}\right ) { \ensuremath{\left\|{\ensuremath{{\mathbf e}}}^{(k-1)}\right\|_2}}^2\end{aligned}\ ] ] where we used linearity of expectation , the fact that @xmath120 is a projector matrix , cauchy - schwarz inequality and fact [ lem : technical ] . repeating the same argument @xmath121 times
we get that @xmath122 note that @xmath123 to conclude .
step @xmath124 can be rewritten as @xmath125 . at every iteration ,
the inner product and the update from @xmath83 to @xmath126 require at most @xmath127 operations for some @xmath110}$ ] ; hence in expectation each iteration requires at most @xmath128 operations .
set @xmath129 to be any vector in the row space of @xmath3 pick @xmath130 $ ] with probability @xmath131 $ ] set @xmath132 output @xmath133 strohmer and vershynin proposed the following randomized variant of kaczmarz algorithm ( algorithm [ alg : randomized ] ) , see @xcite for more details .
the following theorem is a restatement of the main result of @xcite without imposing the full column rank assumption .
[ thm : rk : consistent ] let @xmath0 , @xmath1 and @xmath87 be the input to algorithm [ alg : randomized ] .
assume that @xmath6 has a solution and denote @xmath134 .
in exact arithmetic , algorithm [ alg : randomized ] converges to @xmath135 in expectation : @xmath136 the above theorem has been proved in @xcite for the case of full column rank .
also , the rate of expected convergence in @xcite is @xmath137 where @xmath138 .
notice that if @xmath139 , then @xmath140 is infinite whereas @xmath48 is bounded .
we devote the rest of this subsection to prove theorem [ thm : rk : consistent ] following @xcite . the proof is based on the following two elementary lemmas which both appeared in @xcite . however , in our setting , the second lemma is not identical to that in @xcite .
we deferred their proofs to the appendix .
[ lem : ortho ] assume that @xmath6 has a solution and use the notation of algorithm [ alg : randomized ] , then @xmath141 is perpendicular to @xmath142 for any @xmath17 .
in particular , in exact arithmetic it holds that @xmath143 .
the above lemma provides a formula for the error at each iteration . ideally , we seek to minimize the error at each iteration which is equivalent to maximizing @xmath144 over the choice of the row projections of the algorithm .
the next lemma suggests that by randomly picking the rows of @xmath3 reduces the error in expectation .
[ lem : avg ] assume that @xmath145 has a solution .
let @xmath146 be a random variable over @xmath30 $ ] with distribution @xmath147 and assume that @xmath148 is a vector in the row space of @xmath3 .
if @xmath149 ( in exact arithmetic ) , then @xmath150 theorem [ thm : rk : consistent ] follows by iterating lemma [ lem : avg ] , we get that @xmath151 the analysis of strohmer and vershynin is based on the restrictive assumption that the linear system has a solution .
needell made a step further and analyzed the more general setting in which the linear system does not have any solution and @xmath3 has full column rank @xcite . in this
setting , it turns out that the randomized kaczmarz algorithm computes an estimate vector that is within a fixed distance from the solution ; the distance is proportional to the norm of the `` noise vector '' multiplied by @xmath48 @xcite .
the following theorem is a restatement of the main result in @xcite with two modifications : the full column rank assumption on the input matrix is dropped and the additive term @xmath152 of theorem @xmath153 in @xcite is improved to @xmath154 .
the only technical difference here from @xcite is that the full column rank assumption is not necessary , so we defer the proof to the appendix for completeness .
[ thm : rk : inconsistent ] assume that the system @xmath155 has a solution for some @xmath156 .
denote by @xmath157 .
let @xmath158 denote the @xmath10-th iterate of the randomized kaczmarz algorithm applied to the linear system @xmath6 with @xmath159 for any fixed @xmath160 , i.e. , run algorithm [ alg : randomized ] with input @xmath161 .
in exact arithmetic , it follows that @xmath162 in particular , @xmath163
given any least squares problem , theorem [ thm : rk : inconsistent ] with @xmath164 tells us that the randomized kaczmarz algorithm works well for least square problems whose least squares error is very close to zero , i.e. , @xmath165 .
roughly speaking , in this case the randomized kaczmarz algorithm approaches the minimum @xmath166-norm least squares solution up to an additive error that depends on the distance between @xmath26 and the column space of @xmath3 . in the present paper ,
the main observation is that it is possible to efficiently reduce the norm of the `` noisy '' part of @xmath26 , @xmath84 ( using algorithm [ alg : randop ] ) and then apply the randomized kaczmarz algorithm on a new linear system whose right hand side vector is now arbitrarily close to the column space of @xmath3 , i.e. , @xmath167 .
this idea together with the observation that the least squares solution of the latter linear system is equal ( in the limit ) to the least squares solution of the original system ( see fact [ fact : xls ] ) implies a randomized algorithm for solving least squares .
next we present the randomized extended kaczmarz algorithm which is a specific combination of the randomized orthogonal projection algorithm together with the randomized kaczmarz algorithm .
initialize @xmath168 and @xmath73 pick @xmath130 $ ] with probability @xmath131 $ ] pick @xmath74 $ ] with probability @xmath75 $ ] set @xmath169 set @xmath170 [ alg : stopping ] check every @xmath171 iterations and terminate if it holds : @xmath172 output @xmath148 we describe a randomized algorithm that converges in expectation to the minimum @xmath166-norm solution vector @xmath135 ( algorithm [ alg : rek ] ) .
the proposed algorithm consists of two components . the first component consisting of steps @xmath124 and @xmath173 is responsible to implicitly maintain an approximation to @xmath37 formed by @xmath174 .
the second component , consisting of steps 4 and 7 , applies the randomized kaczmarz algorithm with input @xmath3 and the current approximation @xmath175 of @xmath37 , i.e. , applies the randomized kaczmarz on the system @xmath176 .
since @xmath177 converges to @xmath37 , @xmath148 will eventually converge to the minimum euclidean norm solution of @xmath178 which equals to @xmath179 ( see fact [ fact : xls ] ) .
the stopping criterion of step [ alg : stopping ] was decided based on the following analysis .
assume that the termination criteria are met for some @xmath88 .
let @xmath180 for some @xmath181 ( which holds by the definition of @xmath83 ) .
then , @xmath182 by re - arranging terms and using the second part of the termination criterion , it follows that @xmath183 .
now , @xmath184 where we used the triangle inequality , the first part of the termination rule together with @xmath185 and the above discussion .
now , since @xmath186 , it follows that @xmath187 equation demonstrates that the forward error of rek after termination is bounded . the following theorem bounds the expected rate of convergence of algorithm [ alg : rek ] .
[ thm : rek ] after @xmath87 iterations , in exact arithmetic , algorithm [ alg : rek ] with input @xmath3 ( possibly rank - deficient ) and @xmath26 computes a vector @xmath133 such that @xmath188 for the sake of notation , set @xmath189 and denote by @xmath190 : = \operatorname{\mathbb{e}}[\cdot \ |\ i_0,j_0 , i_1,j_1,\ldots , i_k , j_k]$ ] , i.e. , the conditional expectation with respect to the first @xmath10 iterations of algorithm [ alg : rek ] .
observe that steps @xmath124 and @xmath173 are independent from steps @xmath191 and @xmath192 of algorithm [ alg : rek ] , so theorem [ thm : randop ] implies that for every @xmath193 @xmath194 fix a parameter @xmath195 .
after the @xmath196-th iteration of algorithm [ alg : rek ] , it follows from theorem [ thm : rk : inconsistent ] ( inequality ) that @xmath197 indeed , the randomized kaczmarz algorithm is executed with input @xmath198 and current estimate vector @xmath199 .
set @xmath200 and @xmath201 in theorem [ thm : rk : inconsistent ] and recall that @xmath202 .
now , averaging the above inequality over the random variables @xmath203 and using linearity of expectation , it holds that @xmath204 simplifying the right hand side using the fact that @xmath205 , it follows @xmath206 moreover , observe that for every @xmath193 @xmath207 now for any @xmath88 , similar considerations as ineq .
implies that @xmath208 to derive the last inequality , consider two cases .
if @xmath85 is even , set @xmath209 , otherwise set @xmath210 . in both cases , @xmath211 .
in this section , we discuss the running time complexity of the randomized extended kaczmarz ( algorithm [ alg : rek ] ) .
recall that rek is a las - vegas randomized algorithm , i.e. , the algorithm always outputs an `` approximately correct '' least squares estimate ( satisfying ) but its runnning time is a random variable . given any fixed accuracy parameter @xmath92 and any fixed failure probability @xmath212 we bound the number of iterations required by the algorithm to terminate with probability at least @xmath213 .
[ lem : runtime ] fix an accuracy parameter @xmath214 and failure probability @xmath212 . in exact arithmetic , algorithm [ alg : rek ] terminates after at most @xmath215 iterations with probability at least @xmath213 .
denote @xmath216 for notational convenience .
it suffices to prove that with probability at least @xmath213 the conditions of step [ alg : stopping ] of algorithm [ alg : rek ] are met . instead of proving this
, we will show that : 1 . with probability at least @xmath217 : @xmath218 .
2 . with probability at least @xmath217 : @xmath219 .
later we prove that items ( 1 ) and ( 2 ) imply the lemma .
first we prove item ( 1 ) . by the definition of the algorithm , @xmath220 the first equality follows since @xmath221 ,
the second inequality is markov s inequality , the third inequality follows by theorem [ thm : randop ] , and the last inequality since @xmath222 .
now , we prove item ( 2 ) : @xmath223 the first inequality is markov s inequality , the second inequality follows by theorem [ thm : rek ] , and the last inequality follows provided that @xmath224 a union bound on the complement of the above two events ( item ( 1 ) and ( 2 ) ) implies that both events happen with probability at least @xmath213 .
now we show that conditioning on items ( 1 ) and ( 2 ) , it follows that rek terminates after @xmath225 iterations , i.e. , @xmath226 we start with the first condition .
first , using triangle inequality and item 2 , it follows that @xmath227 now , @xmath228 where the first inequality is triangle inequality , the second inequality follows by item @xmath229 and @xmath230 , the third and forth inequality follows by item @xmath231 and the fifth inequality holds by inequality and the last inequality follows since @xmath232 .
the second condition follows since @xmath233 the first equation follows by orthogonality , the second inequality assuming item ( 2 ) , the third inequality follows since @xmath230 , the forth inequality follows by and the final inequality since @xmath234 .
lemma [ lem : runtime ] bounds the number of iterations with probability at least @xmath213 , next we bound the total number of arithmetic operations in worst case ( eqn . ) and in expectation ( eqn . ) .
let s calculate the computational cost of rek in terms of floating - point operations ( flops ) per iteration .
for the sake of simplicity , we ignore the additional ( negligible ) computational overhead required to perform the sampling operations ( see section [ sec : impl ] for more details ) and checking for convergence .
each iteration of algorithm [ alg : rek ] requires four level-1 blas operations ( two _ ddot _ operations of size @xmath29 and @xmath235 , respectively , and two _ daxpy _ operations of size @xmath235 and @xmath29 , respectively ) and additional four flops . in total ,
@xmath236 flops per iteration
. therefore by lemma [ lem : runtime ] , with probability at least @xmath213 , rek requires at most @xmath237 arithmetic operations ( using that @xmath238 ) .
next , we bound the _ expected running time _ of rek for achieving the above guarantees for any fixed @xmath28 and @xmath239 .
obviously , the expected running time is at most the quantity in .
however , as we will see shortly the expected running time is proportional to @xmath240 instead of @xmath241 . exploiting the ( possible ) sparsity of @xmath3
, we first show that each iteration of algorithm [ alg : rek ] requires at most @xmath242 operations in expectation . for simplicity of presentation
, we assume that we have stored @xmath3 in compressed column sparse format _ and _ compressed row sparse format @xcite .
indeed , fix any @xmath243}$ ] and @xmath110}$ ] at some iteration @xmath10 of algorithm [ alg : rek ] . since @xmath3 is both stored in compressed column and compressed sparse format , steps @xmath192 and step @xmath244 can be implemented in @xmath245 and 5@xmath246 , respectively . by the linearity of expectation and the definitions of @xmath53 and @xmath52 ,
the expected running time after @xmath225 iterations is at most @xmath247 .
it holds that ( recall that @xmath248 ) @xmath249 } } { \ensuremath{\left\| { { { { \ensuremath{\mathsf{a } } } } } _ { ( j)}}\right\|_2}}^2}{\sigma_{\min}^2 } \ln \left(\frac{32(1 + 2{\ensuremath{\kappa^2\left ( { { { \ensuremath{\mathsf{a } } } } } \right)}})}{\delta{\varepsilon}^2}\right)\\ & \leq 2{\ensuremath{\mathrm{\textbf{\footnotesize nnz}}\left ( { { { \ensuremath{\mathsf{a } } } } } \right)}}{\ensuremath{\kappa^2\left ( { { { \ensuremath{\mathsf{a } } } } } \right ) } } \ln \left(\frac{32(1 + 2{\ensuremath{\kappa^2\left ( { { { \ensuremath{\mathsf{a } } } } } \right)}})}{\delta{\varepsilon}^2}\right)\end{aligned}\ ] ] using the definition of @xmath53 and @xmath225 in the first equality and the fact that @xmath250 } } { \ensuremath{\left\| { { { { \ensuremath{\mathsf{a } } } } } _ { ( j)}}\right\|_2}}^2 \leq \sigma_{\max}^2 $ ] and @xmath251 in the first and second inequality . a similar argument shows that @xmath252 using the inequality @xmath253 } } { \ensuremath{\left\| { { { { \ensuremath{\mathsf{a } } } } } ^{(i)}}\right\|_2}}^2 \leq \sigma_{\max}^2 $ ] . hence by lemma [ lem : runtime ] , with probability at least @xmath213 , the expected number of arithmetic operations of rek is at most @xmath254 in other words , the expected running time analysis is much tighter than the worst case displayed in equation and is proportional to @xmath240 times the square condition number of @xmath3 as advertised in the abstract .
the proposed algorithm has been entirely implemented in c. we provide three implementation of algorithm [ alg : rek ] : rek - c , rek - blas and rek - blas - precond .
rek - c corresponds to a direct translation of algorithm [ alg : rek ] to c code .
rek - blas is an implementation of rek with two additional technical features .
first , rek - blas uses level-1 blas routines for all operations of algorithm [ alg : rek ] and secondly rek - blas additionally stores explicitly the transpose of @xmath3 for more efficiently memory access of both the rows and columns of @xmath3 using blas .
rek - blas - precond is an implementation of rek - blas that additionally supports upper triangular preconditioning ; we used blendenpik s preconditioning code to ensure a fair comparison with blendenpik ( see section [ sec : exp ] ) . in the implementations of rek - c and rek - blas we check for convergence
every @xmath255 iterations .
moreover , all implementations include efficient code that handles sparse input matrices using the compressed column ( and row ) sparse matrix format @xcite .
[ [ sampling - from - non - uniform - distributions ] ] sampling from non - uniform distributions + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the sampling operations of algorithm [ alg : rek ] ( steps 4 and 5 ) are implemented using the so - called `` alias method '' for generating samples from any given discrete distribution @xcite .
the alias method , assuming access to a uniform random variable on @xmath256 $ ] in constant time and linear time preprocessing , generates one sample of the given distribution in constant time @xcite .
we use an implementation of w. d. smith that is described in @xcite and c s _ drand48 _ ( ) to get uniform samples from @xmath256 $ ] .
we report our experimental results in this section .
we compared the randomized extended kaczmarz ( rek - c , rek - blas , rek - blas - precond ) algorithm to lapack s dgelsy and dgelsd least squares solvers , blendenpick ( version 1.3 , @xcite ) and matlab s backslash operator .
lsrn @xcite did not perform well under a setup in which no parallelization is allowed as the one used here , so we do not include lsrn s performance .
dgelsy uses qr factorization with pivoting and dgelsd uses the singular value decomposition .
we use matlab with version 7.9.0.529 ( r2009b ) .
in addition , we use matlab s included blas and lapack packages and we call lapack s functions from matlab using matlab s cmex technology which allows us to measure only lapack s elapsed time .
we should highlight that matlab is used as a scripting language and no matlab - related overheads have been taken under consideration .
blendenpik requires the fftw library ; we used fftw-3.3.3 . to match the accuracy of lapack s direct solvers , we fixed @xmath28 in algorithm [ alg : rek ] to be 10e-14 .
moreover , during our experiments we ensured that the residual error of all the competing algorithms were about of the same order of magnitude .
we used a pentium(r ) dual - core e5300 ( 2.60ghz ) equipped with 5 gb of ram and compiled our source code using gcc-4.7.2 under linux operating system .
all running times displayed below are measured using the _ ftime _ linux system call by taking the average of the running time of 10 independent executions .
we experimented our algorithm under three different distributions of random input matrices ( sparse , dense and ill - conditioned ) under the setting of strongly rectangular settings of least squares instances .
in all cases we normalized the column norms of the input matrices to unity and generate the right hand side vector @xmath26 having gaussian entries of variance one .
[ [ sparse - least - squares ] ] sparse least squares + + + + + + + + + + + + + + + + + + + + we tested our algorithm in the overdetermined setting of random sparse @xmath59 matrices with @xmath257 and @xmath258 and density @xmath259 .
we also tested rek - blas on the underdetermined case where @xmath260 and @xmath261 . in both cases ,
the density of the sparse matrices was set to @xmath259 ( for even sparser matrices rek - blas performed even better compared to all other mentioned methods ) .
to generate these sparse matrix ensembles , we used matlab s _ sprandn _ function with variance one .
the results are depicted in figure [ fig : sparse ] .
both plots demonstrate that rek - blas is superior on both the underdetermined ( figure [ fig : sparseunder ] ) and overdetermined case ( figure [ fig : sparseover ] ) .
it is interesting that rek - blas performs well in the underdetermined case .
[ [ dense - and - well - conditioned - least - squares ] ] dense and well - conditioned least squares + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + in this scenario , we used random overdetermined dense @xmath59 matrices with much more rows than columns , i.e. , we set @xmath262 and @xmath263 .
we also tested rek - blas on the underdetermined case where @xmath264 and @xmath265 .
we generated this set of matrices using matlab s _ randn _ function with variance ten .
we depicted the results in figure [ fig : denseover ] . in the overdetermined case ( figure [ fig : denseover ] ) ,
rek - blas is marginally superior compared to lapack s routines whereas rek - c ( as a naive implementation of algorithm [ alg : rek ] ) is inferior .
blendepik is the winner in this case .
interestingly , rek - blas almost matches the performance of blendenpik in the underdetermined case , see figure [ fig : denseunder ] .
[ [ dense - and - ill - conditioned - least - squares ] ] dense and ill - conditioned least squares + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + finally , we tested all algorithms under a particular case of random ill - conditioned dense matrices with @xmath262 and @xmath266 .
namely , we used higham s _ randsvd _ function for generating these matrices @xcite .
more precisely , we set the condition number of these matrices to be @xmath267 ; set the top singular value to one and the rest to 10e-6 .
the results are displayed in figure [ fig : cond ] .
unfortunately , in the ill - conditioned setting rek - blas - precond is inferior compared to lapack s routines and blendepik .
we also verified the results of @xcite that blendepik is superior compared to lapack s least squares solvers in this setting .
we would like to thank the anonymous reviewers for their invaluable comments on an earlier draft of the present manuscript .
the first author would like to thank haim avron for his technical support on several issues regarding blendenpik and philip a. knight for sharing his unpublished manuscript @xcite .
dmms11 e. anderson , z. bai , j. dongarra , a. greenbaum , a. mckenney , j. du croz , s. hammerling , j. demmel , c. bischof , and d. sorensen . .
in _ proceedings of the 1990 acm / ieee conference on supercomputing _ , supercomputing 90 , pages 211 .
ieee computer society press , 1990 .
h. avron , p. maymounkov , and s. toledo . .
, 32(3):12171236 , 2010 .
r. ansorge . .
33(34):367375 , september 1984 .
r. barrett , m. berry , t. f. chan , j. demmel , j. donato , j. dongarra , v. eijkhout , r. pozo , c. romine , and h. van der vorst . .
software , environments , tools . society for industrial and applied mathematics , 1987 .
a. bjrck . .
society for industrial and applied mathematics , 1996 .
y. censor , p. eggermont , and d. gordon . .
, 41:8392 , 1983 . y. censor .
, 23(4):444466 , 1981 .
e. s. coakley , v. rokhlin , and m. tygert . .
, 33(2):849868 , 2011 . k. l. clarkson and d. p. woodruff . .
in _ proceedings of the symposium on theory of computing ( stoc ) _ , pages 205214 , 2009 .
k. l. clarkson and d. p. woodruff . .
available at arxiv:1207.6365 , july 2012 .
y. censor and s. a. zenios . .
numerical mathematics and scientific computation series .
oxford university press , 1997 . j. w. demmel . .
, 50(182):pp . 449480 , 1988 .
p. drineas , m. w. mahoney , and s. muthukrishnan . .
in _ proceedings of the acm - siam symposium on discrete algorithms ( soda ) _ , pages 11271136 , 2006 .
p. drineas , m. w. mahoney , s. muthukrishnan , and t. sarls . . , 117(2):219249 ,
feb 2011 .
h. g. feichtinger , c. cenker , m. mayer , h. steier , and t. strohmer . .
pages 299310 , 1992 . n. m. freris and a. zouzias . . ,
a. galntai . .
advances in mathematics .
springer , 2003 .
r. gordon , r. bender , and g. t. herman . .
, 29(3):471 481 , 1970 . g. h. golub and c. f. van loan . . ,
third edition , 1996 .
g. h. golub and r. s. varga . .
, 3:157168 , 1961 .
g. t. herman . .
omputer science and applied mathematics .
new york etc .
: academic press ( a subsidiary of harcourt brace jovanovich , publishers ) .
xiv , 1980 .
n. j. higham . .
technical report , ithaca , ny , usa , 1989 . nicholas j. higham . .
society for industrial and applied mathematics , philadelphia , pa , usa , first edition , 1996 .
g. t. herman and l. b. meyer . .
, 12(3):600609 , 1993 . m. hanke and wilhelm n. .
, 130(0):83 98 , 1990 . s. kaczmarz . . , 35:355357 , 1937 .
p. a. knight . .
, manchester university , 1993 .
p. a. knight . .
unpublished manuscript , may 1996 .
d. leventhal and a. s. lewis . .
, 35(3):641654 , 2010 . s. f. mccormick . .
, 23:371385 , 1975 .
x. meng , m. a. saunders , and m. w. mahoney . .
available at http://arxiv.org/abs/1109.5981 , sept 2011 .
a. magen and a. zouzias . .
in _ proceedings of the acm - siam symposium on discrete algorithms ( soda ) _ , pages 14221436 , 2011 . f. natterer . .
society for industrial and applied mathematics , 2001 .
n. h. nguyen , t. t. do , and t. d. tran . .
in _ proceedings of the symposium on theory of computing ( stoc ) _ , pages 215224 , 2009 . d. needell . .
, 50(2):395403 , 2010 .
c. popa . .
, 6:5164 , 1999 .
c.c . paige and m.a .
saunders . .
, 8(1):4371 , 1982 .
v. rokhlin and m. tygert . .
, 105(36):1321213218 , 2008 .
y. saad . .
society for industrial and applied mathematics , 2nd edition , 2003 .
t. sarls . .
in _ proceedings of the symposium on foundations of computer science ( focs ) _ , pages 143152 , 2006 .
w. d. smith . .
technical report , princeton , nj , usa , 2002 .
t. strohmer and r. vershynin . .
, 15(1):262278 , 2009 .
k. tanabe . .
, 17:203214 , 1971 . c. tompkins . .
in _ proc .
2nd symposium of linear programming _ , pages 425448 , washington , dc , 1955 .
m. d. vose . .
, 17(9):972975 , september 1991 .
a. j. walker . . ,
3(3):253256 , september 1977 .
t. whitney and r. meany . .
, 4(1):109118 , 1967 .
we present the proof of known facts from previous works for completeness .
( of lemma [ lem : ortho ] ) it suffices to show that @xmath268 . for notational convenience ,
let @xmath269 for every @xmath58}$ ] .
assume that @xmath270 for some arbitrary @xmath271 $ ] .
then , @xmath272 using the definition of @xmath12 , and the fact that @xmath273 since @xmath135 is a solution to @xmath145 .
now , by the definition of @xmath274 , @xmath275 .
( of lemma [ lem : avg ] ) in light of lemma [ lem : ortho ] , it suffices to show that @xmath276 . by the definition of @xmath12
, it follows @xmath277 \ = \
\operatorname{\mathbb{e}}_{z } \frac{{\left\langle { \xls - { \mathbf{x}}^{(k ) } } , \ { { { { { \ensuremath{\mathsf{a } } } } } ^{(z ) } } } \right\rangle}^2}{{\ensuremath{\left\| { { { { \ensuremath{\mathsf{a } } } } } ^{(z)}}\right\|_2}}^2 } \\ & = \sum_{i=1}^{m } \frac{{\left\langle { \xls - { \mathbf{x}}^{(k)}},\ { { { { { \ensuremath{\mathsf{a } } } } } ^{(i ) } } } \right\rangle}^2}{{\ensuremath{\left\| { { { \ensuremath{\mathsf{a } } } } } \right\|_{\text{\rm f}}}}^2 } = \frac{{\ensuremath{\left\| { { { \ensuremath{\mathsf{a } } } } } ( \xls - { \mathbf{x}}^{(k)})\right\|_2}}^2}{{\ensuremath{\left\| { { { \ensuremath{\mathsf{a } } } } } \right\|_{\text{\rm f}}}}^2}.\end{aligned}\ ] ] by hypothesis , @xmath148 is in the row space of @xmath3 for any @xmath10 when @xmath129 is ; in addition , the same is true for @xmath135 by the definition of pseudo - inverse @xcite .
therefore , @xmath278 .
( of theorem [ thm : rk : inconsistent ] ) as in @xcite , for any @xmath58}$ ] define the affine hyper - planes : @xmath279 assume for now that at the @xmath10-th iteration of the randomized kaczmarz algorithm applied on @xmath280 , the @xmath4-th row is selected .
note that @xmath158 is the projection of @xmath281 on @xmath282 by the definition of the randomized kaczmarz algorithm on input @xmath161 .
let us denote the projection of @xmath281 on @xmath283 by @xmath148 .
the two affine hyper - planes @xmath284 are _ parallel _ with common normal @xmath5 , so @xmath148 is the projection of @xmath158 on @xmath283 and the minimum distance between @xmath283 and @xmath282 equals @xmath285 .
in addition , @xmath286 since @xmath287 , therefore by orthogonality we get that @xmath288 since @xmath148 is the projection of @xmath281 onto @xmath283 ( that is to say , @xmath148 is a randomized kaczmarz step applied on input @xmath289 where the @xmath4-th row is selected on the @xmath10-th iteration ) and @xmath281 is in the row space of @xmath3 , lemma [ lem : avg ] tells us that @xmath290 note that for given selected row @xmath4 we have @xmath291 ; by the distribution of selecting the rows of @xmath3 we have that @xmath292 inequality follows by taking expectation on both sides of equation and bounding its resulting right hand side using equations and . applying inequality inductively , it follows that @xmath293 where we used that @xmath129 is in the row space of @xmath3 .
the latter sum is bounded above by @xmath294 . | we present a randomized iterative algorithm that exponentially converges in expectation to the minimum euclidean norm least squares solution of a given linear system of equations .
the expected number of arithmetic operations required to obtain an estimate of given accuracy is proportional to the square condition number of the system multiplied by the number of non - zeros entries of the input matrix .
the proposed algorithm is an extension of the randomized kaczmarz method that was analyzed by strohmer and vershynin . |
it follows from both extrapolating pulsar birthrates ( narayan & ostriker 1990 ) and from the number of supernovae required to account for the heavy element abundance ( arnett et al .
1989 ) that the galaxy is populated by about @xmath1 neutron stars ( nss ) .
however , only @xmath2 of this amount has been observed so far , mainly as radio pulsars .
thus , one can expect that there exists a vast population of nonpulsating nss which could be observed as faint sources of thermal radiation .
it has recently been claimed by walter , wolk , & neuhuser ( 1996 ; hereafter wwn96 ) that the bright , nonpulsating , soft @xmath3 x - ray source rxj1856.53754 is a viable candidate to be a member of this population .
wwn96 found that ( a ) a blackbody fit of the spectrum detected with the @xmath3 positional sensitive proportional counter ( pspc ) ( count rate @xmath4 s@xmath5 ) yields an apparent blackbody temperature of @xmath6 k , a hydrogen column density of @xmath7 @xmath8 , and an emitting area of @xmath9 km@xmath10 ; ( b ) other standard models do not give better fits ( see also neuhuser et al .
1996 for details ) ; ( c ) the @xmath3 high resolution imager ( hri ) count rate is @xmath11 s@xmath5 ; ( d ) there is no evidence for x - ray variability in this source ; and ( e ) the object is fainter than @xmath12 in the visual band .
no radio , uv , or gamma radiation have been detected .
the source is projected onto the r cra dark molecular cloud , at a distance of @xmath13 pc ( wang 1994 ) .
the observed x - ray extinction allows one to assume that the source is closer than the cloud .
the inferred temperature looks surprisingly high for this presumably very old ns ( see arguments in wwn96 ) , although it may be caused by enhanced accretion of interstellar matter if the ns relative motion is slow enough .
however , the real spectrum of the ns thermal emission may differ substantially from the planck spectrum due to the presence of an atmosphere on the ns surface ( e. g. , pavlov et al .
1995 ) , and the actual shape of the spectrum strongly depends on chemical composition of the emitting layers ( rajagopal & romani 1996 ; zavlin , pavlov , & shibanov 1996 , hereafter zps96 ) .
fitting observed spectra with the model atmosphere spectra may result in quite different effective temperature @xmath14 and ratio @xmath15 of the size of emitting region to the distance .
the chemical composition of the ns surface depends on various effects . due to strong gravitation at the ns surface ,
the elements there are separated in accordance with their atomic weights so that the lightest element resides at the very surface and forms the atmosphere , whereas heavier elements sink down .
thus , if an amount of hydrogen ( h ) is present in the outer layers ( e. g. , due to accretion of the interstellar matter ) , one can expect a pure h atmosphere .
if during the ns life time there were no substantial accretion of the h - rich matter , or the hydrogen were burnt or ejected from the surface , the atmosphere may be comprised of a heavier element . in principle
, it might be helium ( he ) or carbon , although the most natural choice seems to be iron ( fe ) formed in the core of the supernova progenitor star .
the h and he atmospheres emit spectra which are much harder in the wien tail than the blackbody ( bb ) spectrum because hotter deep layers are seen at high frequencies where the spectral opacity is lower . in this case ,
the best - fit atmosphere effective temperature is considerably lower than the bb temperature , whereas the @xmath15 ratio is larger than for the bb fit . on the contrary , in the case of the fe atmosphere
the overall shape of the spectrum is not so drastically different from that of the bb model , although local deviations may be quite substantial due to numerous spectral lines and photoionization edges .
the chemical composition of the ns surface could be determined by the analysis of its soft x - ray spectrum . however , spectral resolution and sensitivity of current soft x - ray detectors is not high enough to firmly detect spectral features in the most interesting range @xmath16 kev . on the other hand ,
this range is so short that the shape of the continuum can be fitted with quite different models .
thus , it is desirable to measure the ns flux over a broader range . since the ns thermal flux is intrinsically very low at medium and hard x - rays , and it is severely suppressed by the interstellar extinction at euv , the most perspective would be optical / uv observations ( @xmath17 ev ) .
although nss are extremely faint in this range ( typical expected magnitudes are @xmath18 at @xmath19 pc ) , broad - band photometry of these sources is quite feasible with the _ hubble space telescope _
( _ hst _ ) and the largest ground - based telescopes ( pavlov 1992 ) . so far , optical / uv _ thermal _ radiation has been apparently detected from only two active pulsars , psr b1929 + 10 and ( more questionable ) psr b0950 + 08 ( pavlov et al .
1996 ) ; optical radiation from younger , ` middle - aged ' pulsars geminga and psr b0656 + 14 is most likely of a non - thermal origin ( bignami et al . 1996
; pavlov et al .
the pulsars b1929 + 10 and b0950 + 08 have relatively strong magnetic fields ( @xmath20 g ) , which precludes analysis of the atmosphere chemical composition because no reliable opacities have been calculated for elements other than hydrogen for strongly magnetized dense plasmas . on the contrary , for very old nss , where the magnetic field is believed to have been decayed and rotation has slowed down so that the nss can not manifest themselves as active pulsars , the chemical composition can be analyzed with the use of non - magnetic atmosphere models . in the present paper
we fit the @xmath3 pspc spectra of the ns candidate rxj1856.53754 , which seems to be sufficiently old and does not show pulsar activity , with the h , he , and fe atmosphere models developed by zps96 .
we use these fits to calculate the ir through x - ray fluxes expected from this source , assuming that it is indeed an old , isolated ns with low magnetic field .
we show that the detected optical flux ( and the distance to the object ) would immediately allow us to distinguish between light - element ( h or he ) and heavy - element composition .
we show that for any assumptions on the chemical composition the object should be observable with the @xmath0 deep survey ( ds ) instrument in a relatively short exposure time .
the most dramatic effect of the chemical composition is expected at optical wavelengths , where the ns covered with hydrogen would be about @xmath21 magnitudes brighter than the ns with an iron surface .
we re - analized the pspc data using the exsas software ( zimmermann et al .
the source in the pspc image is @xmath22 off - axis , and its measured ( off - axis ) count rate is @xmath23 @xmath24 , which corresponds to the corrected ( on - axis ) count rate of @xmath25 s@xmath5 , in agreement with wwn96 .
we binned the observed count rate spectrum into 19 energy channels between approximately @xmath26 and @xmath27 kev ( no source counts were detected above 1 kev ) and fitted it with the h , he , and fe atmosphere models of zps96 , and with the bb spectrum .
the best - fit parameters ( effective hydrogen column density @xmath28 , unredshifted surface effective temperature @xmath14 , distance to the source @xmath29 ) , and reduced @xmath30 ( for 16 degrees of freedom ) are presented in table 1 .
we assumed standard ns mass @xmath31 and radius @xmath32 km ( which corresponds to the apparent radius @xmath33 km , where @xmath34^{1/2}$ ] ) .
although the bb temperature and @xmath28 are essentially the same as obtained by wwn96 ( notice that the unredshifted temperature @xmath14 is higher than the apparent temperature @xmath35 presented by wwn96 ) , the formal quality of our fit is much better ( minimum @xmath36 vs. @xmath37 ) , most likely because we chose another binning of the pspc count spectrum .
ccccccccc model & @xmath28 & @xmath14 & @xmath29 & @xmath30 & v & hri & ds & scanner + & ( @xmath38 @xmath8 ) & ( @xmath39 k ) & ( pc ) & & mag & ( s@xmath5 ) & ( s@xmath5 ) & ( s@xmath5 ) + h & @xmath40 & @xmath41 & @xmath42 & 0.94 & 21.8 & 0.60 & 0.046 & 0.045 + he & @xmath43 & @xmath44 & @xmath45 & 1.24 & 20.8 & 0.60 & 0.043 &
0.043 + fe & @xmath46 & @xmath47 & @xmath48 & 0.68 & 27.5 & 0.69 & 0.087 & 0.065 + blackbody & @xmath49 & @xmath50 & @xmath51 & 0.80 & 26.8 & 0.61 & 0.053 & 0.047 +
h magnetic & @xmath52 & @xmath53 & @xmath54 & 1.32 & 23.3 & 0.60 & 0.044 & 0.043 + since the spectra emitted by the h and he atmospheres are substantially harder , in the @xmath3 range , than the bb spectrum with the same effective temperature , we obtained quite different fitting parameters for the atmosphere model fits .
for example , @xmath55 and @xmath56 . on the other hand , the difference between the parameters obtained for the bb and fe atmosphere fits is not so dramatic because the fe atmosphere spectrum is close , on average , to the bb spectrum of the same temperature ( zps96 ) .
moreover , the fe model spectrum is even slightly softer than bb in the @xmath3 range due to strong bound - free absorption , which leads to higher @xmath14 and @xmath29 . despite the significant difference in the best - fit parameters (
particularly , @xmath14 and @xmath29 ) , the fits look very similar . from the viewpoint of formal statistics , all the four fits are acceptable , i. e. , one can not choose the most acceptable model without invoking additional arguments .
the projections of the three - dimensional confidence regions onto the @xmath14-@xmath28 plane , together with the lines of constant most probable distances , are shown in figure 1 for the four spectral models .
we see that the confidence regions for h and he models overlap with each other , being very distant from those for the fe and bb models . * fig
confidence contours for the helium , hydrogen , and iron atmosphere models , and the blackbody model , at 68% ( @xmath57 ) , 90% and 99% confidence levels .
nearly straight solid lines are the lines of constant most probable distance @xmath29 ( in pc ) .
the thick dashed lines correspond to the constant hri count rates ( figures near the curves , in s@xmath5 ) .
since the energy dependence of the hri and pspc effective areas are different , the hri observations of the same object can provide an additional opportunity to examine the models for its radiation .
we re - analyzed the hri image and obtained the source count rate @xmath58 s@xmath5 .
the hri count rates corresponding to the best pspc fit models are presented in table 1 , and the curves of constant hri count rates are shown in figure 1 .
we see that the model hri count rates exceed the measured one by @xmath59 for the h and he models , @xmath60 for the bb model , and @xmath61 for the fe model
. the systematic uncertainty of the hri effective area at low energies , @xmath62 kev , which give the main contribution to the hri count rate , may reach @xmath63 ( b. aschenbach , private communication ) . since the pspc response is also not free from systematic errors at low energies
, one can not exclude that the h and he atmosphere models are consistent with the hri data , whereas the fe model seems to give a too high hri count rate . to demonstrate how observations of this object in other spectral bands can be used to discriminate between the source models , we calculated the spectral fluxes , both attenuated and non - attenuated by the interstellar extinction , for the best pspc fit models over the broad photon energy range , @xmath64 ev ( fig . 2 ) .
to calculate the extinction at ir / optical / uv energies , we use the approximate relation between the hydrogen column density and the color excess : @xmath65 see savage & mathis ( 1979 ) . in the range of @xmath66
kev all the attenuated spectra are very similar to each other , especially for the h , he , and bb models , in accordance with approximately the same quality of the pspc fits . in the range of @xmath67 ev the fluxes
are unobservable for all the models because of the interstellar absorption .
they are observable in the high - energy part of the @xmath0 spectral range , @xmath68 ev , which considerably overlaps with the @xmath3 soft channels .
we evaluated expected count rates for the @xmath0 ds and @xmath0 scanners a / b with the lexan / b filter for different models ( see table 1 ) .
the count rates are not much different for the h , he , and bb models , @xmath69 s@xmath5 , whereas for the fe model the count rate is higher , particularly for the ds instrument .
a @xmath70 upper limit of 0.066 s@xmath5 , reported by wwn96 for a short ( 1,001 s ) scanner exposure , is too high to discriminate between the models ( although the fe atmosphere count rate almost coincides with the limit ) .
a longer , @xmath71 ksec , observation with the @xmath0 ds is needed to detect @xmath72 source counts and to draw a convincing conclusion on the source nature . *
fig . 2 . *
photon spectral fluxes for four models in the range of @xmath64 ev with and without allowance for interstellar absorption .
much more striking difference between the model fluxes is seen in the ir / optical / uv range ( 110 ev ) : for instance , the source flux is @xmath73 magnitudes brighter for the h and he models than for the fe model ( see table 1 ) .
the reason is that a @xmath74 times smaller distance ( at a given emitting area ) is needed for the h / he models to provide the same soft - x - ray flux at @xmath75 times lower effective temperature .
the spectra at the low energies follow the rayleigh - jeans law , @xmath76 , so that the flux for the h / he models is @xmath77 times higher than for the fe model .
since the effect of extinction is not very significant , the source is expected to be the brightest at uv frequencies .
the @xmath78 magnitude predicted for the h / he models is about 2022 , which is brighter than the limit of @xmath79 reported by wwn96 .
if this limit is correct , then the non - magnetic h / he models can be excluded .
one more striking feature of the h / he atmosphere models is a very small @xmath80 ratio for instance , for @xmath32 km the distance would be in a range of @xmath81 pc , at a 99% confidence level , if the ns is covered with hydrogen . for the expected local ns density , @xmath82 pc@xmath83 , the probability to find a ns at such
a distance is low ( e. g , the probability to find one ns at a distance of 5 pc from the sun is @xmath84 ) . in principle
, the ns radius may be larger than 10 km ; for example , @xmath85 km for a ns of the mass @xmath86 for a hard equation of state ( e. g. , shapiro & teukolsky 1983 ) . in this case , the distance would become larger by a factor of 1.5 .
the distances following from the h / he atmosphere models can be easily measured with the @xmath87 even for such a faint object . on the contrary , if we adopt the fe atmosphere model , the distance of @xmath88 pc would be even larger than for the bb fit , considerably exceeding 120 pc to the r cra molecular cloud .
this also seems unlikely because the total gas column density of the cloud in this region , @xmath89 @xmath8 , is substantially greater than the 99% upper limit of @xmath90 @xmath8 .
thus , at the present stage , the situation looks puzzling light element models give uncomfortably small distances and optical magnitudes brighter than the reported limit , whereas the fe model results in a distance larger than acceptable , if the ns is foreground to the r cra dark cloud .
this conclusion , however , is based on the assumption that the magnetic field at the ns surface does not exceed @xmath91 g ( otherwise the non - magnetic atmosphere models are inapplicable ) . although there have been many speculations about the field decay in isolated nss ( e. g. , itoh et al .
1995 , and references therein ) , this idea is not firmly supported by observations to date .
the main argument against the presence of high magnetic fields in rxj1856.53754 is that no significant pulsations have been found in the @xmath3 data ( becker 1996 ) .
if the lack of pulsations is confirmed , it still might be consistent with the presence of high magnetic fields if the rotation axis coincides with the magnetic axis or with the line - of - sight .
the former is not unexpected for an old ns ; the probability of the latter is very low .
fitting the observed spectra with the magnetic h atmosphere models may not be as definitive as with the nonmagnetic ones because we know nothing about the strength and geometry of the field which , in turn , affects the distribution of temperature over the ns surface . besides
, the available magnetic models may not be accurate at relatively low temperatures ( pavlov et al .
an example of the magnetic h atmosphere fit ( assuming uniform temperature and @xmath92 g ) is presented in table 1 . generally , magnetic fits result in higher effective temperatures , greater distances , @xmath93 pc , and fainter optical magnitudes , @xmath94 , than the nonmagnetic h / he atmosphere fits .
although the distances still look very small , such parameters do not contradict to the data available . the most reliable way to understand the nature of rxj1856.53754 is to conduct deep optical / uv observations of a relatively wide field around its x - ray position .
it should be noted that although the x - ray position of a star detected in the hri pointed observation is coincident , within @xmath95 , with its @xmath87 gsc position ( wwn96 ) , the latter may itself be in error by a few arcseconds ( e. g. , because of a large proper motion ) .
hence , the hri position of the ns may be less accurate than @xmath96 , as it was adopted by wwn96 .
in addition , if the distance to the ns is indeed as low as a few parsec , one should expect a large proper motion of the object .
for instance , if one adopts the ns tangential velocity of 300 km / s , a typical value for radio pulsars , and a distance of 10 pc , then the proper motion would be @xmath97 yr@xmath5 .
the proper motion might be limited by the relatively small mean offset , @xmath98 , between the @xmath99 ipc slew survey position and the hri position measured 14 years later .
this limit , however , is not very stringent because of the large uncertainty , @xmath100 at 90% confidence level , of the @xmath101 position : the resulting proper motion can be estimated as @xmath102 yr@xmath5 .
thus , one can not exclude that the object could travel @xmath103 during 1.4 years between the hri and optical observations , so that the reported limit of @xmath79 should be considered with caution , and a wider field should be carefully investigated before making final conclusions . to prove that a faint object in the field is indeed the ns ,
the color of the object must be examined nss of expected temperatures should be very blue . if a more accurate investigation of the field results in detection of this source with a @xmath78 magnitude of about @xmath104 , and a distance of @xmath105 pc , the conclusion would be quite unambiguous : the ns has an h or he atmosphere and a very low magnetic field .
if the magnitude is slightly fainter , about 24 , and the distance larger , @xmath106 pc , we would argue for the magnetic h atmosphere model and conclude that magnetic and rotation axes coincide with each other , or the rotation axis is co - aligned with the line of sight .
finally , if the magnitude is as faint as @xmath107 and/or the distance exceeds @xmath108 pc , then we conclude that the atmosphere consists of fe or another heavy element . in this case
we should also conclude that the ns is seen through a hole in the cloud r cra , unless the distance to the cloud was underestimated .
thus , optical observations of this object would enable one , for the first time , to probe chemical composition of surface layers of nss and to firmly evaluate their temperature .
obviously , the same approach , i. e. , combined analysis of the x - ray and optical / uv spectra with the use of the ns atmosphere models , can be applied to other similar objects .
we thank fred walter , werner becker , bernd aschenbach , stuart bowyer , and the referee for useful discussions .
the @xmath3 project is supported by the bmbf and the max - planck - society .
g. p is grateful to the max - planck - institut fr extraterrestrische physik for hospitality ; his work was partially supported by nasa grant nag5 - 2807 and intas grant 94 - 3834 .
v. z. acknowledges max - planck fellowship and the intas grant . | we show that an analysis of multiwavelength observations of isolated neutron stars based on neutron star atmosphere models can be used not only to evaluate the neutron star effective temperature , but also to determine chemical composition of its surface . to demonstrate how this new method can be applied to a specific object ,
we chose the old isolated neutron star candidate rxj1856.53754 , whose soft x - ray radiation has been studied recently by walter , wolk , & neuhuser ( 1996 ) .
we fitted the soft x - ray spectrum of this object with neutron star atmosphere models of different chemical compositions and used these fits to calculate the source spectrum over a broad wavelength range .
we showed , in particular , that the optical / uv flux expected from this object depends drastically on the composition of its surface .
in particular , the neutron star covered with hydrogen would be 56 magnitudes brighter than the neutron star with an iron surface .
the object should also be observable with @xmath0 ; the euv flux is expected to be almost twice higher for the iron surface than for the hydrogen one .
thus , multiwavelength observations of this object would enable one to examine , for the first time , chemical composition of the neutron star surface .
the method proposed can be applied to other nearby isolated neutron stars . |
the metal - insulator transition ( mit ) driven by electron interaction , the so - called mott transition , @xcite is one of the most challenging issues in modern solid state physics , especially because of its possible connections with other phenomena , like high - temperature superconductivity .
the prototypical model to study the mit is the one - band hubbard model , where the transition shows up if the ratio between the bandwidth @xmath0 and the on - site coulomb repulsion @xmath1 is varied .
@xcite in a pioneering work , brinkman and rice @xcite argued that the mit could be continuous , however , their approach led to a mott insulator where charge fluctuations are completely frozen .
recent developments , based on dynamical mean - field theory calculations , @xcite are able to give a more realistic description of the mit where , at zero temperature , a wide gap in the excitation spectrum opens immediately in the insulator .
however , within this approach , the role of the dimensionality is not taken into account , whereas its relevance comes out from recent experiments on si mosfet s @xcite and organic materials .
@xcite moreover , the situation can be radically different whenever the hamiltonian contains a true long - range coulomb interaction , as pointed out in the original mott argument .
@xcite in this work , we focus our attention on a correlated wave function ( wf ) , which is expected to closely describe the mit in two dimensional ( 2d ) systems with long - range coulomb interactions .
the important fact is that , in the strong - coupling regime , we can adiabatically connect our wf with the fully - projected one , usually considered to describe systems in the limit of infinite coulomb repulsion , e.g. , the so - called resonating valence - bond state .
@xcite moreover , the metallic phase has no quasiparticles defined , showing low - energy properties similar to the one - dimensional luttinger liquid .
the paper is organized as follow : in sec .
[ mapping ] we show the mapping between the quantum wave function and a classical model at finite temperature and in sec . [ results ] we show our numerical results and we draw the conclusions .
let us discuss how to construct a wf for correlated insulators . in general , starting from the ground state @xmath2 of a system with @xmath3 electrons , with energy @xmath4 , it is possible to construct simple variational states for the lowest - energy excitations .
for instance , in analogy with the feynman s construction for the liquid helium , @xcite the plasmon excitation with momentum @xmath5 is given by : @xmath6 where @xmath7 is the fourier transform of the local electron density .
its variational energy is @xmath8 where @xmath9 is the ground - state kinetic energy per particle and @xmath10 is its static charge structure factor .
when applied to an insulator , with gapped charge excitations , this implies that , for small momenta , @xmath11 .
this argument has very general consequences on the form of @xmath2 , that do not depend on the particular microscopic model .
to this purpose , let us denote an electronic configuration by the positions @xmath12 of the particles .
for all the operators @xmath13 that depend only on such positions , e.g. , the structure factor itself , the quantum average @xmath14 can be written in terms of the _ classical _ distribution @xmath15 , as @xmath16 since @xmath17 is a positive quantity , we can define an appropriate correspondence between the wf and an effective potential @xmath18 : latexmath:[\[\label{classical } that the potential @xmath18 is extensive , namely of order @xmath3 for typical configurations . in the limit of strong coulomb interactions ,
there are small charge fluctuations and , therefore , we can safely assume that only the two - body term is relevant and all multi - particle interactions are negligible .
this leads to the quadratic potential @xmath20 being @xmath21 the fourier transform of the local density of the configuration @xmath22 . to obtain the expected behavior @xmath23
the effective potential must diverge as @xmath24 here @xmath25 can be considered as the effective temperature of classical charges interacting through a potential @xmath26 . within this choice of @xmath27 ,
@xmath11 is generally valid and can be understood by considering @xmath7 as a complex continuous variable , so that the classical average of @xmath28 turns into a standard gaussian integral , yielding @xmath29 it should be noted that the fully - projected wave function with no charge fluctuations , and therefore @xmath30 for @xmath31 , is recovered when @xmath32 .
let us now consider a general one - band fermionic system in 2d , in which every site of a square lattice can be either empty , singly occupied , being the electron either with spin up or down , or doubly occupied . a true mott insulator , that does not break any lattice symmetry , can not be represented by a simple wf containing a single determinant , and , at this stage , it is useful to define a state that is simple enough and yet is compatible with the predicted form of @xmath27 in the insulating phase . a straightforward way to modify the effective potential determined by an uncorrelated determinant @xmath33 is obtained by taking into account an appropriate jastrow factor @xmath34 : latexmath:[\[\label{wf } @xmath33 is an electronic determinant that will be specified in the following and @xmath34 is a jastrow term that depends upon the electronic density : @xmath36,\ ] ] where @xmath37 is the jastrow potential , whose small-@xmath5 behavior is given by : @xmath38 } \sim \frac { \pi \beta}{q^2},\ ] ] @xmath39 fixing its strength . at
half filling , in one - dimensional electronic systems , we found @xcite that the singular jastrow @xmath40 always leads to an insulator , for any positive @xmath39 . in 2d
the situation is different and a much more interesting scenario is obtained , with a phase transition as a function of the correlation strength @xmath39 .
indeed , given the behavior of the jastrow @xmath41 , the potential @xmath18 of eq .
( [ classical ] ) turns out to be the one of the classical coulomb gas model ( cgm ) . in this approach ,
particles with charge @xmath42 , corresponding to empty ( @xmath43 ) and doubly ( @xmath44 ) occupied sites , interact through a coulomb potential in a neutral background , represented by singly occupied sites ( @xmath45 ) . in the half - filled case , there is an equal number of empty and doubly occupied sites , implying the charge neutrality of the cgm .
the fugacity @xmath46 of the cgm , that sets the average value of the charges , can be identified with the on - site gutzwiller term in the jastrow potential , i.e. , @xmath47 , where @xmath48 is the gutzwiller parameter .
( color online ) inverse of the dielectric function @xmath49 [ see eq .
( [ epsilon ] ) ] for the free - electron determinant . left panel : @xmath49 as a function of the effective temperature @xmath50 and for different sizes @xmath51 of the cluster .
the critical temperature of the classical coulomb gas model @xmath52 is marked with a dashed line for a comparison .
right panel : size scaling of @xmath49 for various @xmath39 .
, scaledwidth=43.0% ] the classical cgm in 2d is particularly interesting because it shows a kosterlitz - thouless ( kt ) transition at a finite temperature @xmath55 .
@xcite this transition is transparent from the classical dielectric function : @xmath56,\ ] ] where @xmath25 is the temperature of the classical model .
the charge structure factor is quadratic at small momenta , i.e. , @xmath57 , for all temperatures , but the coefficient @xmath58 changes discontinuously at @xmath55 . above @xmath55 ,
the cgm is in the plasma phase , i.e. , a metallic phase with infinite dielectric function , perfect screening , and exponential correlation functions . on the other hand , below @xmath55 ,
the cgm is in the confined phase , with a finite dielectric constant . in this phase
the charges are bound together forming dipoles , that , because of their residual interaction , induce power - law correlations . at the transition
, the inverse of the dielectric function has a finite jump , changing from zero , in the plasma phase , to a finite value , in the confined phase . a similar mapping between a quantum state and
a classical model has been emphasized also in the context of the fractional quantum hall effect : the laughlin wf can be related to a classical system with particles interacting through a logarithmic potential .
@xcite however , in this case , all the particles have the same charge , forming a one - component plasma , and by varying the strength of the potential , there is a first - order transition between an incompressible fluid and a wigner crystal .
@xcite the peculiarity of our approach is that , due to the mapping onto the two - component cgm , it is possible to connect continuously the plasma phase to the insulating one .
( color online ) equal - time density structure factor @xmath59 for the correlated wave function of eq .
( [ wf ] ) ( full squares ) , compared to the same quantity calculated within the gaussian approximation [ indicated as gsa and given by eq .
( [ enneq ] ) ] , ( full triangles ) for @xmath60 ( upper panel ) and @xmath61 ( lower panel , notice the different scale of the gsa data ) .
, scaledwidth=40.0% ] in the following , we show that , in analogy with the classical cgm , also in the case of fermionic systems at zero temperature , a kt - like transition is found by varying the correlation strength that can be tuned by the jastrow strength @xmath39 .
however , the existence of the fermionic part induces non - trivial properties for the two phases involved , that are not present in the classical problem .
for example the uncorrelated part of the wf may contribute to the expression of the effective temperature @xmath25 , as shown below .
whenever the square of the wf describes the plasma phase of the corresponding classical model , we can safely assume that the gaussian fluctuations are exact for small @xmath5 s and the classical temperature can be determined by imposing @xmath62 in eq .
( [ epsilon ] ) , namely @xmath63 . in the language of quantum states ,
the gaussian approximation leads to the well - known expression @xmath64 where @xmath65 is the charge structure factor of the uncorrelated determinant @xmath66 .
@xcite the previous form of @xmath59 allows us to identify the effective temperature as : @xmath67 where @xmath68 . in order to show the general validity of our approach , we consider the case of a free - electron determinant , obtained by occupying the lowest - energy states in the tight - binding model with dispersion @xmath69 and a gapless bcs state with a superconducting order parameter @xmath70
. in these cases @xmath71 and ,
therefore , the effective temperature in eq .
( [ teff ] ) is determined only by the jastrow coefficient , namely @xmath72 . in fig .
[ fig : jumpa ] , we report the inverse of the dielectric function for the free - electron determinant and different sizes @xmath51 of the system at half filling , i.e. , @xmath73 . in order to have closed - shell states for @xmath33
, we used 2d square lattices tilted by @xmath74 ( i.e. , with @xmath75 and @xmath76 odd ) and periodic boundary conditions . by increasing @xmath51 ,
the curves show a steeper and steeper shape in the vicinity of the critical temperature @xmath77 .
this result is further confirmed by the size scaling of @xmath49 , which clearly supports the existence of a finite jump in the thermodynamic limit : @xmath78 for @xmath79 , whereas @xmath80 for @xmath81 .
interestingly , @xmath77 depends slightly upon the choice of the uncorrelated determinant ( see for comparison fig . [
fig : jumpb ] for the gapless bcs state ) and is quite close to the cgm critical temperature @xmath82 .
these results give an important and transparent insight into the strong - coupling limit described by the fully - projected wf , @xcite that can be connected to our wf by letting @xmath83 , i.e. , @xmath32 .
indeed , in the confined phase for @xmath81 , the classical kt scaling equations of the cgm flow to fixed points with zero fugacity : this translates into the fact that the fully - projected state represents the fixed - point of the correlated wfs describing the 2d mott insulating phase .
therefore , in the confined regime , the ground - state properties are universal and represented by the ones of the fully - projected wf . in this respect ,
the total projection is not an unrealistic assumption and can accurately reproduce the low - energy physical properties of a strongly correlated system . on the other hand , for @xmath79
the classical kt scaling equations flow to strong coupling and are useful only close to the transition point .
( color online ) quasiparticle weight @xmath84 at @xmath85 for the gapless bcs state with @xmath53 and @xmath54 symmetry as a function of @xmath51 and for different jastrow strengths @xmath39 ( full circles ) .
the case of the fully - projected wave function ( empty circles ) is also reported for @xmath53 and @xmath86 .
, scaledwidth=40.0% ] ( color online ) the behavior of @xmath13 [ the exponent of the quasiparticle weight , see eq .
( [ powerlaw ] ) ] as a function of @xmath39 for bcs state ( full circles ) and the fermi gas ( fg ) determinant ( full squares ) .
the value of the fully - projected states are also reported ( arrows ) .
, scaledwidth=40.0% ] in fig .
[ fig : rpa ] we show that in the plasma phase the gaussian approximation , given by eq .
( [ enneq ] ) , is very accurate , not only for small @xmath5 s ( where it is exact ) but also for large momenta . in this case , the cusp singularity in @xmath59 for @xmath87 , related to the friedel oscillations , is not removed , even though @xmath11 at low momenta .
thus , for @xmath79 , the wf ( [ wf ] ) describes a `` coulomb metal '' , with @xmath11 at small @xmath5 s but with the sign of the fermi surface at large momenta . as shown previously in the limit of infinite coulomb repulsion @xcite or in the low - density regime , @xcite this wf has low - energy properties similar to one - dimensional luttinger liquid conductors , where the absence of a jump in the momentum distribution is replaced by a weaker singularity , yielding to @xmath88 and @xmath89 power - law density correlations .
it is important to emphasize that , in the quantum case , the power - law correlations come from the large momentum singularity , that are absent in the classical cgm .
@xcite indeed , in the quantum state , the subleading corrections in the classical potential of eq .
( [ classical ] ) are very important and can actually turn the cgm exponential correlations to power laws in the plasma phase , and _ vice - versa _ in the confined phase . on the other hand , in the confined phase
the gaussian approximation is not adequate both at small and large momenta , see fig .
[ fig : rpa ] . indeed , at small @xmath5 s , the coefficient of the quadratic term is not simply given by the gaussian approximation and , more importantly , the strong jastrow factor washes out completely the singularities of @xmath65 , leading to a smooth charge - structure factor , a genuine fingerprint of an insulating phase .
( color online ) quasiparticle weight @xmath84 at @xmath85 of the optimized _ paramagnetic _ wave function containing a jastrow factor applied to the fermi gas as a function of the interaction @xmath90 in the hubbard model , for three different sizes of the system .
inset : the number of double occupancies @xmath91 as a function of @xmath90 . , scaledwidth=40.0% ] ( color online ) optimized jastrow potential @xmath37 , multiplied by @xmath92 , for the hubbard model as a function of @xmath93 [ in the ( 1,1 ) direction ] for different sizes of the cluster and ratios @xmath90 .
the arrow indicates @xmath94 , the expected value of @xmath95 at the classical transition point .
, scaledwidth=40.0% ] in order to further characterize the two phases , we consider the quasiparticle weight @xmath96 where @xmath97 and @xmath98 are the wf with @xmath3 and @xmath99 particles , @xmath100 is the destruction operator of a particle of momentum @xmath101 and spin @xmath102 . in particular , the wave function with @xmath103 particles is constructed from @xmath97 : @xmath104 in a previous work , @xcite it was argued that the singular jastrow factor can induce non - fermi liquid properties , and in particular a vanishing @xmath84 at the fermi surface . in fig .
[ fig : zeta ] , we report @xmath84 for @xmath85 and for different jastrow strengths @xmath39 .
we find that the quasiparticle weight vanishes with a power - law behavior : @xmath105 both in the confined and in the plasma phase , with an exponent @xmath13 that depends upon @xmath39 and the type of the uncorrelated state . in the plasma phase
, @xmath13 varies continuously with the jastrow strength @xmath39 and there is no appreciable dependence on the uncorrelated determinant . on the other hand , in the confined phase , the exponent is constant , i.e. , @xmath106 for the bcs state and @xmath107 for the free - electron state , and equal to the value found for the fully projected wf , as shown in fig .
[ fig : theta ] .
it must be mentioned that , for the bcs state , @xmath13 does not depend upon the value of the superconducting order parameter @xmath108 ( see fig .
[ fig : zeta ] ) , indicating the universal properties of the confined phase .
our results show that it is possible to describe a continuous mit in 2d electronic systems with a jastrow correlated wf .
we characterized both the metallic region , with a zero - quasiparticle weight @xmath109 , and the insulator , which can be continuously connected to the totally projected gutzwiller wf .
of course , other scenarios are possible for the mit , e.g. , the one proposed in the infinite - dimensional limit .
indeed , whenever the metallic phase has @xmath110 with @xmath111 , the mit is not described by the functional form ( [ veff ] ) of the jastrow : by approaching the transition from the metallic phase , we enter directly into the confined phase with a quadratic charge structure factor at small momenta , i.e. , @xmath112 with a large finite @xmath113 at the critical point . in this case ,
in the metallic region , a less singular jastrow factor @xmath114 is expected , leading to a finite quasiparticle weight .
finally , we would like to comment about the possibility to stabilize the `` coulomb metal '' phase in a microscopic model .
for simplicity , let us consider the one - band hubbard model on the square lattice with nearest - neighbor hopping : @xmath115 where @xmath116 creates an electron with spin @xmath102 at the site @xmath117 , @xmath118 is the density operator at the site @xmath117 . in the following , we will consider a _
paramagnetic _ state , by taking a projected fermi gas wave function , as described above , and by minimizing the variational energy for the determination of the jastrow factor @xmath34 .
in particular , by using the method described in ref . , we are able to optimize all the independent jastrow parameters in the real space @xmath119 ( i.e. , the fourier transform of @xmath37 ) . within this approach , that neglects magnetic phases ,
we obtain a mit for @xmath120 . in the weak coupling regime , for @xmath121
, we obtain a fermi liquid with a finite quasiparticle weight , whereas at strong couplings , i.e. , for @xmath122 , we have an insulating phase with a vanishing @xmath84 , see fig .
[ fig : transition ] .
@xcite moreover , the calculation of the double occupancy @xmath91 clearly indicates that the transition is continuous and the insulating phase still possess finite charge fluctuations , see inset of fig .
[ fig : transition ] .
as discussed above , in the metallic region , we find that @xmath114 ( see fig . [ fig : vq ] ) .
unfortunately , as soon as we enter in the insulating phase , @xmath123 defines an effective @xmath39 which is larger than the critical value for the kt transition , and , therefore , no evidence for the `` coulomb metal '' is found .
@xcite indeed , we expect that the optimized jastrow factor @xmath37 , containing subleading corrections with respect to eq .
( [ veff ] ) , will define a critical @xmath39 very close to the value of the classical cgm , i.e. , @xmath124 .
therefore , in light of the results of fig .
[ fig : vq ] , the stabilization of the `` coulomb metal '' seems to be very unlikely : although there are large size effects around @xmath125 , we have a clear evidence that @xmath126 for @xmath121 and @xmath127 for @xmath122 ( see fig . [ fig : vq ] ) . on the other hand
, we can safely predict the occurrence of the novel kt - like scenario described above in 2d systems with long - range ( logarithmic ) interaction . in this case , the application of the gaussian approximation for small interaction and our ansatz for the insulating phase imply the presence of a transition of the type considered here .
it is remarkable that the proposed picture crucially depends on the long - range nature of the coulomb interaction , recalling mott s original idea . in this regard
, it should be mentioned that the original mott argument for a discontinuous metal - insulator transition , driven by the long - range coulomb interaction , can not be applied in 2d . in such case there
always exists a bound state for two opposite charges interacting with the screened coulomb potential , so that , according to this argument , no metallic phase with unbound charges is possible . however , this is clearly an artifact of such mean - field argument , since for small interaction the random - phase approximation leads to an anomalous metallic state ( see ref . ) .
we thank m. fabrizio , a. parola , and e. tosatti for the valuable and fruitful interaction during the accomplishment of this project .
we also thank c. castellani , n. nagaosa , t. senthil , and x .-
wen for interesting discussions .
this work has been supported by infm and miur ( cofin 2004 and cofin 2005 ) . | we show , by using a correlated jastrow wave function and a mapping onto a classical model , that the two - dimensional mott transition in a simple half - filled one - band model can be unconventional and very similar to the binding - unbinding kosterlitz - thouless transition of vortices and anti - vortices , here identified by empty and doubly occupied sites . within this framework ,
electrons strongly interact with collective plasmon excitations that induce anomalous critical properties on both sides of the transition .
in particular , the insulating phase is characterized by a singular power - law behavior in the photoemission spectrum , that can be continuously connected to the fully - projected insulating state , relevant to strongly correlated low - energy models . |
it is often argued that einstein gravity on a given three dimensional spacetime manifold @xmath0 is equivalent to a chern simons theory on the same manifold with the poincar group being the underlying gauge group @xcite .
in fact , writing the first order einstein hilbert action in the dreibein formulation of general relativity , one finds that this is , up to a total derivative , equal to the chern simons action .
the dreibein and the spin connection are then interpreted as the translational and rotational components of a poincar algebra valued gauge field .
the only difference between the two theories is that in einstein gravity the dreibein is restricted to be invertible , whereas no such restriction exists in chern simons theory . in this sense
, chern simons theory can be thought of as an extension of einstein gravity including singular metrics , or vice versa einstein gravity is considered as a restricted version of chern simons theory . the purpose of this article is to show that this equivalence holds at the level of field configurations only .
there is indeed a one - to - one relation between off shell as well as on shell field configurations of einstein gravity , and those field configurations of chern simons theory which have a non - vanishing dreibein determinant everywhere on @xmath0 .
however , there is no such relation at the level of _ gauge equivalence classes _ of physical states . after dividing out the gauge symmetries of the two theories ,
one ends up with two different reduced phase spaces .
both are finite dimensional manifolds , that is , both theories are _ topological _ in the sense that they have only finitely many physical degrees of freedom .
but there is no one - to - one correspondence between the reduced states of chern simons theory and those of einstein gravity .
the basic idea of the proof is the following .
assume that @xmath1 is the configuration space of chern simons theory , that is , the set of all poincar algebra valued one - forms on a given spacetime manifold @xmath0 .
they are typically subject to some boundary conditions , but these are not of any importance here .
assume further , that there is an action principle defined on @xmath1 , which provides a set of field equations that singles out a submanifold @xmath2 of _ physical states _ , or _ on shell _ field configurations .
this set of solutions @xmath3 can be identified with the _ physical phase space _ of chern simons theory .
the action defines a canonical symplectic two - form @xmath4 thereon .
it is highly degenerate , because only finitely many dimensions of the infinite dimensional phase space @xmath3 are actually physical degrees of freedom .
all other dimensions , namely the null directions of @xmath4 , are directions of gauge transformations . in general , two states @xmath5 are gauge equivalent , denoted by @xmath6 , if they lie on the same _ null orbit _ of @xmath4 , that is , if they can be joint by a curve in @xmath3 whose tangent vector lies in the kernel of the two - form @xmath4 everywhere along the curve .
this is the definition of a _ smoothly generated _ gauge symmetry . in the case of chern simons theory , the gauge symmetries are the local poincar transformations . depending on the topology of the spacetime manifold @xmath0 , some of them might not be smoothly generated .
these are called _ large _ local poincar transformations .
large gauge transformations are symmetries of the physical phase space that map those states onto each other , which are physically indistinguishable but do not lie on the same null orbit of the symplectic two - form .
it is thereby important to note that the existence of large gauge symmetries can _ not _ be inferred from the given action .
instead , it depends on the physical interpretation of the model whether two states related by a large transformation should be considered as gauge equivalent or not . in particular
, one has to decide which quantities can be measured and which can not .
depending on this choice , an equivalence class of physical states , in the following also called a _ gauge orbit _ , consist of one or several null orbits of @xmath4 .
if there are no large gauge symmetries , then each gauge orbit is a connected submanifold of @xmath3 and consists of exactly one null orbit of @xmath4 .
otherwise , the gauge orbits are disconnected submanifolds of @xmath3 consisting of more than one null orbit of @xmath4 . in any case
, the reduced phase space @xmath7 is defined to be the quotient space of the physical phase space modulo gauge symmetries .
provided that the gauge orbits on @xmath3 behave regularly enough , it is a manifold whose dimension is equal to the rank of @xmath4 . a unique , non - degenerate symplectic two - form @xmath8 can then be defined on @xmath9 , such that @xmath4 is its pullback on @xmath3 . by definition ,
each gauge invariant function on @xmath3 , usually called an _
observable _ , can be written as a function on @xmath9 and vice verse .
hence , the reduced phase space @xmath9 contains the full physically relevant information about the system . in other words , if einstein gravity is equivalent to chern simons theory , or a restricted version thereof , then the reduced phase space of einstein gravity should be equal to , or a subset of the reduced phase space of chern simons theory .
but this , as we are now going to see , is not the case .
let me briefly explain what goes wrong , before making things more explicit .
in the configuration space @xmath1 of chern simons theory , there is a _
submanifold of singular metrics _ , consisting of all field configurations with vanishing dreibein determinant at some point in @xmath0 . the configuration space @xmath10 of einstein gravity
is obtained from @xmath1 but taking away this submanifold .
the action and also the equations of motion remain the same , which implies that the same holds for the physical phase space @xmath11 of einstein gravity .
it is the subset of @xmath3 obtained by taking away the submanifold of singular metrics .
but now something crucial is happening to the gauge orbits .
first of all , we have @xmath12 , and therefore the null orbits of @xmath13 in einstein gravity are the same as the null orbits of @xmath4 in chern simons theory .
the problem is that they intersect with the submanifold of singular metrics .
this is illustrated in figure [ orb ] . as a consequence , when the submanifold is taken away , the null orbits fall apart into disconnected components .
for example , the states @xmath14 and @xmath15 , which are related by a smoothly generated gauge symmetry in chern simons theory , no longer lie on the same null orbit of @xmath13 , and thus they are not related by a smoothly generated gauge symmetry in einstein gravity . as a result , the reduced phase space @xmath16 of einstein gravity is different from that of chern simons theory , @xmath7 . ' '' '' now , one might argue that the equivalence of the two theories can be easily restored by considering the states @xmath14 and @xmath15 in einstein gravity as being related by a _ large _ gauge symmetry . each gauge orbit
then consists of several disconnected parts , namely exactly those that were formerly , in chern simons theory , a single null orbit . in einstein gravity
, we would then say that @xmath14 and @xmath15 , or @xmath17 and @xmath18 , are related by a large local poincar transformation . from the mathematical point of view , this would be perfectly consistent , and the resulting reduced phase spaces @xmath7 and @xmath16 would indeed be identical .
the only difference would be that a reduced state of einstein gravity has less representatives in @xmath11 than the corresponding reduced state in chern simons theory has in @xmath3 .
however , remember that the question whether large gauge symmetries exist was a physical question rather than a mathematical one .
so , what we should ask is whether the states @xmath14 and @xmath15 are physically indistinguishable .
unfortunately , this is not the case .
a simply example will show that declaring two such states as gauge equivalent in einstein gravity leads to a straight contradiction to our intuitive understanding of the very basic principles of general relativity . to express it more drastically ,
if nobody would know about the relation between einstein gravity and chern simons theory , then nobody would have ever suggested to declare the states @xmath14 and @xmath15 in figure [ orb ] to be gauge equivalent . or , phrased differently , given only the phase space @xmath11 of einstein gravity , why should we glue two _ a priori _ unrelated regions of this space together along the dashed line , in the way shown in the figure , especially if we do not know how to interpret the states _ on _ that line ? in any case , we have to conclude that the similarity between einstein gravity and chern simons theory can be quite useful , but when we speak about gauge symmetries and the reduced phase space , we have to decide which of the two theories we are dealing with . without making this decision ,
some authors run into yet another problem @xcite .
it arises if one tries to include too many features of both einstein gravity and chern simons theory into a kind of mixture of the two theories , which then has too many large gauge symmetries . in einstein gravity ,
an essential part of the gauge group are the diffeomorphisms of the underlying spacetime manifold @xmath0 . in the spirit of _ general relativity _ , one assumes that the individual points of @xmath0 do not have any particular physical meaning .
this implies that every transformation that interchanges them without destroying the differentiable structure of @xmath0 should be considered as a gauge symmetry . for a generic manifold @xmath0 ,
there are smoothly generated diffeomorphisms as well as large ones .
now , the orbits of the smoothly generated diffeomorphisms in phase space are included in the orbits of smoothly generated local poincar transformations .
this is because the infinitesimal generator of any diffeomorphisms can be written as a local poincar transformation with a special parameter .
however , it turns out that the large diffeomorphisms of @xmath0 have nothing to do with the local poincar transformations .
they act on the physical phase space in a completely different way .
for example , the state @xmath15 in the figure could be mapped onto @xmath17 by some large diffeomorphism , lying on a different null orbit of @xmath4 . in einstein gravity , including the large diffeomorphism as gauge symmetries does not course a problem .
what happens is that several null orbits of @xmath13 are united into a single gauge orbit .
however , the null orbits to the left and to the right of the dashed line in the figure are thereby identified in a different way , as compared to the way they are glued together by including the submanifold of singular metrics .
a problem arises if we now consider the large diffeomorphisms as gauge symmetries in chern simons theory , or if _ both _ the large poincar transformations _ and _ the large diffeomorphisms are considered as gauge symmetries in einstein gravity . in both cases ,
the states @xmath14 through @xmath18 in the figure all become gauge equivalent , and we can extend this sequence into both directions , as indicated by the arrows .
now , assume that the plane shown in the figure is actually a part of a cylinder , and after a finite number of steps we get back to this region , but we never again hit the null orbit we started from .
the gauge orbit then fills the plane densely , and the resulting reduced phase space is no longer a manifold .
it is in particular not hausdorff , because two such gauge orbits which both fill the plane densely can not be separated by open neighbourhoods . using the example
already mentioned above , we shall see that this is a quite typical situation .
it arises because one tries to keep too many features of both theories .
if one either excludes singular metrics , sticking to einstein gravity , or does not regard large diffeomorphisms as gauge symmetries , for which there is actually no motivation in chern simons theory , then this problem is absent .
let us now make things more explicit . in the following ,
we assume that @xmath0 is an orientable , three dimensional manifold with a trivial tangent bundle , equipped with a levi civita tensor @xmath19 . on @xmath0 ,
we introduce two @xmath20 valued one - forms @xmath21 and @xmath22 .
in chern simons theory , these are interpreted as the rotational and the translational components of a poincar algebra valued gauge field .
note that @xmath20 is the spinor representation of the three dimensional lorentz algebra , but it is also , as a vector space , isomorphic to minkowski space .
the poincar field strength also splits into an @xmath20 valued rotational component @xmath23,\ ] ] and an also @xmath20 valued translational component @xmath24 - [ { \bomega}_\nu,{\mathb{e}}_\mu ] .\ ] ] in einstein gravity , @xmath21 is interpreted as the spin connection and @xmath22 is the dreibein , defining the metric and the determinant , @xmath25 the latter is required to be different from zero everywhere on @xmath0 .
hence , the configuration space @xmath1 of chern simons theory is the set of all field configurations @xmath26 on @xmath0 , and the configuration space @xmath10 of einstein gravity is the set of all @xmath26 with @xmath27 . in both cases ,
they are subject to some additional restrictions such as smoothness and fall off conditions at infinity . a more detailed discussion , taking into account all such boundary conditions for a specific model , will be given in @xcite .
here , we only need to know that all these extra conditions can be chosen to be the same in both theories , such that @xmath10 is the subset of @xmath1 with the singular metrics excluded .
the vacuum einstein hilbert action can then be written as @xmath28 up to a total derivative , this coincides with the chern simons action @xcite .
the equations of motions are @xmath29 defining the physical phase space @xmath3 , respectively @xmath11 .
in chern simons theory , they state that the poincar field strength vanishes . in einstein gravity
they are interpreted as the _ torsion _ the _ curvature _ equation , stating that the dreibein is covariantly constant and that the metric is flat .
all we need to know about the symplectic form @xmath12 is that its null directions are the generators of the local symmetries of the action ( [ eh ] ) .
these are the _ local poincar transformations _ , @xmath30,\qquad \delta{\mathb{e}}_\mu = { \partial}_\mu { \bzeta}+ [ { \bomega}_\mu,{\bzeta } ] + [ { \mathb{e}}_\mu,{\blambda}],\ ] ] where @xmath31 is the generating parameter of a _ local lorentz rotation _ and
@xmath32 that of a _ local translation_. it is easy to see that the generators of spacetime diffeomorphisms are included in these transformations .
if we choose the parameters to be @xmath33 for some vector field @xmath34 on @xmath0 , then the transformation ( [ pnc ] ) becomes @xmath35 where @xmath36 denotes the lie derivative with respect to the vector field @xmath34 .
as the field strength , respectively the curvature and the torsion vanishes on the physical phase space , this is equal to the infinitesimal generator of a diffeomorphism of @xmath0 with generating vector field @xmath34 .
what we also need in the following are the finite versions of the gauge transformations .
a finite local lorentz rotation is given by @xmath37 where @xmath38 is a parameter field taking values in the spinor representation of the lorentz group @xmath39 .
at this point , however , we already have to be a little bit more careful .
not every transformation of the form ( [ lor ] ) can be written as a smooth deformation of the fields generated by ( [ pnc ] ) .
this is the case if and only if the map @xmath38 is _ contractible _ , that is , if it can be smoothly deformed into the trivial map @xmath40 .
otherwise the given transformation is a _
large _ local lorentz rotation .
so , already here we have to make a decision , namely whether to consider large local lorentz rotations as gauge symmetries or not .
both in einstein gravity as well as in chern simons theory , it is physically reasonable to include them .
doing this essentially means that the rotational component of the gauge field , respectively the spin connection , can only be measured by parallel transport of vectors or spinors along closed loops in @xmath0 .
one can show that this leads to the conclusion that all transformations of the form ( [ lor ] ) are gauge symmetries @xcite . but this is not the crucial point regarding the difference between einstein gravity and chern simons theory , and therefore it is actually not relevant how to deal with the large local lorentz rotations .
the more interesting gauge symmetries are the local translations .
a finite local translation is given by @xmath41,\ ] ] where @xmath42 is actually the same parameter field as @xmath43 in ( [ pnc ] ) , but now it is a finite parameter rather than an infinitesimal generator .
one can easily check that together with the local lorentz rotations ( [ lor ] ) , the local translation form a local poincar group .
it is also useful to note that there are no _ large _ local translations in chern simons theory .
every transformation of the form ( [ trn ] ) can be written as a smooth deformation by simply replacing @xmath44 by @xmath45 , with @xmath46 . for einstein gravity , things are a bit more evolved , as we shall see below . before coming to this ,
let us also write down the action of a diffeomorphism @xmath47 on the field configuration .
the one - forms are thereby replaced by their pullbacks , so that the finite version of ( [ lie ] ) reads @xmath48 from the relations between the infinitesimal generators of local poincar transformations ( [ pnc ] ) and diffeomorphisms ( [ lie ] ) , we know that the action of every _ smoothly generated _ diffeomorphism ( [ dif ] ) can be written as a combination of a local lorentz rotation ( [ lor ] ) and a local translation ( [ trn ] ) , both being smoothly generated .
this is valid in einstein gravity as well as in chern simons theory . in einstein gravity
, however , there is yet another relation between the infinitesimal generators which is not present in chern simons theory .
if the dreibein is invertible , then ( [ par ] ) provides a one - to - one relation between the generating parameter @xmath43 of local translations and the generating vector field @xmath34 of spacetime diffeomorphisms .
hence , in einstein gravity we can consider the spacetime diffeomorphisms instead of the local translations as the basic gauge symmetries .
regarding the smoothly generated ones , this is just a reparametrization .
however , it leads us to a different set of large transformations . to see that the situation is indeed as shown in figure [ orb ] ,
consider the action of a local translation ( [ trn ] ) in einstein gravity . given a field configuration and a parameter @xmath44 , there is no guarantee that the transformed dreibein is still invertible .
in fact , it is easy to show that for any given field configuration it is also always possible to find a parameter @xmath44 such that the transformed dreibein becomes singular .
so , the gauge orbits indeed intersect with the submanifold of singular metrics , as indicated in the figure .
moreover , given two states @xmath14 and @xmath15 with non - singular metric , and related by a local translation ( [ trn ] ) for some parameter @xmath44 , the simple trick to show that they can be smoothly transformed into each other does not work , because when we replace @xmath44 by @xmath45 , then some of the intermediate states might have a singular metric .
this does not yet prove that the gauge orbits fall apart into disconnected components , because the phase space is infinite dimensional , and therefore the two dimensional picture does not necessarily give the correct impression .
there might still be a null curve of @xmath13 that connects @xmath14 and @xmath15 without passing through the submanifold of singular metrics . a possible way to prove that such a path does not exist is to show that the states @xmath14 and @xmath15 are not related by a combination of a diffeomorphism and a local lorentz rotation .
they are then , in particular , not related by a smoothly generated diffeomorphism and a local lorentz rotation . on the other hand
, we know that in einstein gravity every smoothly generated gauge symmetry can be written as a combination of a smoothly generated diffeomorphism and a local lorentz rotations . if this is not the case for the two given states , it follows that they are not related by a smoothly generated gauge symmetry , and therefore they do not lie on a connected component of a gauge orbit . from this
we have to conclude that in einstein gravity some of the local translations are _ large _ , and we can ask the question whether these should be considered as gauge symmetries or not .
it is important to note that in chern simons theory this question never arises , because all local translations are smoothly generated .
hence , if we decide _ not _ to consider the large local translation as gauge symmetries of einstein gravity , and we shall shortly see that our intuitive understanding of general relativity forces us to do so , then the two theories are not equivalent
. we shall give an explicit example of two states @xmath14 and @xmath15 , related by a local translation , but representing two spacetimes which are obviously not diffeomorphic and therefore physically clearly distinguishable .
from this we infer that , first of all , the two states are not related by a smoothly generated gauge symmetry , and secondly , that we must not consider large translations as gauge symmetries of einstein gravity if we want to stick to its original physical interpretation .
to give the explicit example , let us first have a closer look at the solutions to the field equations ( [ eom ] ) .
it is well known that locally , within a simply connected region @xmath49 , they can be parametrized by two fields @xmath50 and @xmath51 , such that @xcite @xmath52 in chern simons theory , this means that locally the fields are _ pure gauge_. in fact , setting @xmath53 in ( [ lor ] ) and then @xmath54 in ( [ trn ] ) , we can always transform the fields into the trivial solution @xmath55 and @xmath56 .
it follows , in particular , that there are no local physical degrees of freedom . in einstein gravity , there is a restriction on @xmath57 , and the interpretation of the parameters is as follows .
the spacetime metric and the dreibein determinant can be written as @xmath58 the second equation states that the dreibein determinant is equal to the jacobian of @xmath57 .
if this is non - zero , then @xmath57 is locally one - to - one .
hence , in contrast to chern simons theory , @xmath57 is not completely arbitrary . according to the first equation
, it then provides an _ isometric embedding _ of the simply connected subset @xmath59 of spacetime into minkowski space .
the components of the vector @xmath57 are called _ minkowski coordinates _ on @xmath59 .
so , even though we can not transform to @xmath55 and @xmath56 , because then the metric is singular , we can still see that there are no local gravitational degrees of freedom , because the spacetime is always flat .
the action of the gauge symmetries on the fields @xmath60 and @xmath57 is as follows . under a local lorentz rotation with parameter @xmath38 , they transform as @xmath61 for a local translation with parameter @xmath42 , we find that @xmath62 and finally , a spacetime diffeomorphism @xmath47 acts as @xmath63 to specify a physical state completely , using the field @xmath60 and @xmath57 as parameters , we first have to fix a spacetime manifold @xmath0 , cover it by a set of simply connected regions , and define the fields @xmath60 and @xmath57 independently within in these regions .
let us do this for a special example . as a spacetime manifold , we choose the direct product @xmath64 of a real line , which is going to represent the time coordinate @xmath65 , with the space manifold @xmath66 shown in figure [ mfd ] .
it is a plane with four punctures , that is , four points are taken away .
it is not simply connected , but it can be covered by two simply connected , closed subsets @xmath67 , overlapping along a sequence of five lines @xmath68 , with @xmath69 , as indicated in the figure
. a physical state can then be parametrized by two pairs of fields @xmath70 and @xmath71 , where @xmath72 .
these fields are subject to a boundary condition on the overlap surfaces @xmath73 of @xmath74 and @xmath75 .
for the dreibein and the spin connection to be well defined , the right hand sides of ( [ sol ] ) must be the same in @xmath74 and @xmath75 .
this is the case if and only if the values of @xmath76 and @xmath77 in the overlap region are related by @xmath78 for some constants @xmath79 and @xmath80 .
the second equation states that @xmath81 is related to @xmath82 by a rigid poincar transformation , that is , an isometry of the embedding minkowski space , consisting of a translation @xmath83 and a lorentz rotation @xmath84 .
this implies that the spacetime metric induced by @xmath82 is the same as that induced by @xmath81 .
the first equation then ensures that not only the metric but also the dreibein and the spin connection are continuous on @xmath0 .
if the overlap region splits into several disconnected components , then there are independent constants for each part . in our case
, we have five independent _ transition functions _ @xmath84 and @xmath83 , with @xmath69 . using this
, we can formulate the following statement , which holds in chern simons theory , but not in einstein gravity . given two physical states , parametrized by fields @xmath60 and @xmath57 , respectively @xmath85 and @xmath86 , such that the transition functions @xmath84 and @xmath83 coincide , then the two states are gauge equivalent .
the proof is quite simple .
define a field @xmath87 . by assumption , this is a continuous field on @xmath0 , which follows from the first equation in ( [ cut ] ) .
inserted into ( [ emb - lor ] ) , it transforms @xmath60 into @xmath85 by a local lorentz rotation .
next , assume that @xmath88 and define @xmath89 .
according to the second equation in ( [ cut ] ) , this is again a continuous field on @xmath0 , and when we take this as a parameter of a local translation ( [ emb - trn ] ) , @xmath57 is transformed into @xmath86 . hence , given two states with coinciding transition functions , we can always find a combination of a local lorentz rotation and a local translation that maps the two states onto each other .
what remains to be done to prove the non - equivalence of einstein gravity and chern simons theory is to find two states of einstein gravity with coinciding transition functions , which are however physically distinguishable and therefore not gauge equivalent .
they are then related by a local translation . but this can not be a smoothly generated one because then the two states would also be related by a diffeomorphism .
hence , it must be a large local translation , and we have to conclude that the large local translations are not to be considered as gauge symmetries of einstein gravity .
to give two such states explicitly , we have to introduce some more notation .
let @xmath90 , with @xmath91 , be some orthonormal basis of @xmath20 , such that @xmath92 , where @xmath93 is the minkowski metric with signature @xmath94 and @xmath95 is the levi civita tensor with @xmath96 .
this implies , for example , that conjugation of a vector @xmath97 with a group element @xmath98 provides a spatial rotation of @xmath97 by @xmath99 .
this is all we need to know about the algebra of the gamma matrices . to define an explicit solution to the field equations , we first specify the following transition functions , @xmath100 where @xmath101 . these group elements represent spatial rotations by @xmath102 , @xmath103 , @xmath104 , @xmath105 , and @xmath106
then , we choose two arbitrary , smooth fields @xmath70 , such that the boundary conditions ( [ cut ] ) are satisfied .
their precise form does not matter .
they are all related by local lorentz rotations anyway .
all we need to know is that such fields exist , which can be easily verified .
finally , we have to define the fields @xmath77 .
we decompose them into time and space components , @xmath107 where @xmath108 maps the points @xmath109 onto points in the euclidean plane , considered as the spatial plane in minkowski space , which is spanned by @xmath110 .
if the functions @xmath111 are independent of @xmath65 , it follows from ( [ fex ] ) and ( [ emb ] ) that the resulting spacetime is static , and that the time coordinate @xmath65 is the physical time .
it is then sufficient to specify the spatial metric to fix the geometry of the whole spacetime manifold .
we choose the maps @xmath111 to be one - to - one , and such that their images @xmath112 are the subsets of the spatial plane shown in figure [ img](a ) .
it is , once again , not important to know what these maps precisely looks like .
in fact , the images determine them up to a diffeomorphism of @xmath66 , which can even be shown to be a smoothly generated one .
hence , we have the statement that given the transition functions @xmath84 and the images of @xmath111 , the state is fixed up to a local lorentz rotation and a smoothly generated diffeomorphisms .
what is not immediately obvious is that , given the images , we can choose @xmath111 such that the boundary conditions ( [ cut ] ) are satisfied .
the condition for this to be possible is that the edges of the upper and lower surface in the figure are mapped onto each other by isometries of the euclidean plane , consisting of a translation and a rotation by @xmath102 , @xmath103 , @xmath104 , @xmath105 , and @xmath106 , respectively . in particular
, we must have the following relations between the corner points @xmath113 and @xmath114 in the figure and the transition functions , @xmath115 for @xmath116 .
these relations are indeed satisfied , with the @xmath84 taken from ( [ uu - k ] ) and for some suitable chosen values of @xmath83 .
the first equation follows from evaluating ( [ cut ] ) at the beginning of the line @xmath68 , and the second from the same condition at the end of @xmath117 .
to see what the geometry of the spacetime manifold looks like , we have to cut out the surface in figure [ img](a ) , and glue them together along their edges .
note that they always fit together , because the corresponding edges of the two half spaces are mapped onto each other by isometries of the embedding minkowski space .
in particular , we always obtain a locally flat spacetime .
the resulting space manifold can be described as a cone with four tips , arranged in a row .
its total deficit angle is @xmath118 , and at each tip there is a deficit angle of @xmath119 .
the spacetime obtained by taking the direct product with a real line can be interpreted as a universe containing four point particles with mass @xmath120 , being at rest with respect to each other .
the mass is thereby measured in units of the planck mass @xmath121 @xcite .
it is now possible to construct a _
different _ universe , containing the same particles , which is not diffeomorphic to the first , but with all transition function being the same , so that the two states of einstein gravity are related by a large translation .
it is thereby sufficient to note that the transition functions are already determined by the corner points @xmath113 and @xmath114 in the figure , and the directions of the edges extending to infinity .
this follows from ( [ map ] ) , or simply from the fact that a poincar transformation , which is in this case just an isometry of the euclidean plane , is fixed if its action on two points is known .
if we , for example , choose the maps @xmath111 such that their images are the surfaces in figure [ img](b ) , then the transition functions remain the same , because the corner points and the directions of the edges extending to infinity are unchanged .
in fact , one can easily see that the rotations and translations needed to map the corresponding edges of the upper surface onto those of the lower surface as the same in figure [ img](a ) and ( b ) .
but if we now cut out the surfaces and glue them together along the edges , then we find that the particles are no longer arranged in a row . instead , they form a parallelogram .
it is immediately obvious that the two static spacetimes are not diffeomorphic , and therefore they are not gauge equivalent states of einstein gravity .
this completes the proof that there are states on the gauge orbits of chern simons theory @xmath14 and @xmath15 , as shown in figure [ orb ] , which do not lie on the same gauge orbit of einstein gravity , and it also explains why it is contradictory to the usual interpretation of general relativity to consider large translations as gauge symmetries of einstein gravity .
an important point in this consideration is that it does not depend on the boundary conditions imposed on the fields .
they usually play a very important role in the precise definition of the phase space , because they have to guarantee that the action is finite , and they also have some impact on what is a gauge symmetry and what is not @xcite .
for example , a transformation with a _ constant _ parameter @xmath122 or @xmath43 is typically not a gauge symmetry when certain fall off conditions are imposed on the fields .
however , all this has nothing to do with our simple counter example .
the equivalence of the two theories can not be restored by imposing boundary condition on the fields , neither at infinity nor in the neighbourhood of the punctures .
the crucial transformation , namely the large translation that takes us from ( a ) to ( b ) in figure [ img ] , takes place within a _ compact _ region of @xmath66 .
the fields @xmath111 can be chosen to be the same in ( a ) and ( b ) , say , outside the finite region shown in the figure , and by slightly deforming the images in the neighbourhoods of the corner points @xmath113 and @xmath114 , we can even achieve that the fields are unchanged in a finite neighbourhood of the particles . what remains to be seen
is that regarding the large diffeomorphisms as gauge symmetries of chern simons theory , or to consider both large diffeomorphisms and large translations as gauge symmetries of einstein gravity , makes the gauge orbits behave such that certain regions of the phase space are filled densely by a single orbit , so that the reduced phase space becomes non - hausdorff . to see this ,
observe that there are infinitely many inequivalent ways to draw the edges of the surfaces in figure [ img ] .
the only conditions are that the corners are connected in the correct ordering , and that the corresponding edges of the lower surface are mapped onto those of the upper surfaces by the given isometries ( [ cut ] ) .
but even if we , for example , fix the edges extending to infinity and the one between the second and the third corner , then there is still an infinite series of possibilities , two of which are shown .
the edge starting at @xmath123 , for example , can wind around @xmath124 any number of times before it gets to @xmath125 .
we can label the different states by this winding number .
all these states are then related to each other by local translations .
how do the resulting spacetimes look like ?
they always consist of four identical point particles being at rest .
two of them , number @xmath126 and @xmath127 , are at fixed positions , and the other two , number @xmath128 and @xmath129 , are located in between them , with varying angular orientation .
each time we increase the winding number , the inner particles are rotated by @xmath130 with respect to the outer ones . for the example given , we have @xmath131 , which means that increasing the winding number by six , or three if we consider the particles as being identical , leads us back to the same geometry .
a closer analysis shows that the two states are then related by a large diffeomorphism .
but what happens if the mass @xmath120 of the particles is an irrational multiple of @xmath132 ?
we can then make the same construction , but the possible angular orientations take countably many different values for the different winding numbers .
the possible locations of the inner particles with respect to the outer ones fill a circle densely . in other words ,
a single gauge orbit fills a certain region of the phase space densely , are there are different gauge orbits that do this for the same region .
these can not be separated by open neighbourhoods . as a result ,
the quotient space is not hausdorff . once again
, it should be emphasized that this is a consequence of the assumption that _ both _ the large local translations _ and _ the large diffeomorphisms are considered as gauge symmetries . if either of them is dropped , that is , if one sticks to the original definition of chern simons theory , which does not know anything about large diffeomorphism , or to einstein gravity , without singular metrics and without large local translations as gauge symmetries , then this problem is absent . in the former case ,
the states with the particles filling the circle densely are in fact not close to each other in phase space , and in the second case they are not gauge equivalent , so in neither case a single gauge orbit fills a region of phase space densely . | a simple example is given to show that the gauge equivalence classes of physical states in chern simons theory are not in one - to - one correspondence with those of einstein gravity in three spacetime dimensions .
the two theories are therefore not equivalent . it is shown that including singular metrics into general relativity has more , and in fact a quite counter - intuitive , impact on the theory than one naively expects .
mz - th/99 - 04 + gr - qc/9903040 * on the relation between 2 + 1 einstein gravity + and chern simons theory * * hans - jrgen matschull * + institut fr physik , johannes gutenberg - universitt + staudingerweg 7 , 55099 mainz , germany + e - mail : [email protected] |
liquid droplets on a solid surfaces pose several interesting theoretical and experimental challenges and has a direct importance in several industrial processes .
in general , the interaction of the liquid with the surface is characterized through the contact angle that a droplet , with known volume , makes with the surface . for a flat , chemically homogeneous surface , the equilibrium contact angle is easily obtained from an energy minimization procedure .
the result , known as young s law , includes the three surface energies ( solid - liquid , solid - vapor , liquid - vapor , @xmath0 , @xmath1 , @xmath2 , respectively ) with a `` weight factor '' for the last one , the cosine of the contact angle , @xmath3 .
the dynamical approach to equilibrium is also well understood , both for complete @xcite and partial @xcite wetting . in reality , the static contact angle , as well as its dynamics , is obviously influenced by the variations in the physicochemical or geometrical properties of the substrate on which the droplet spreads @xcite .
such surfaces are present in an enormous variety of industrial processes .
typical examples range from printing and coating to painting , as well as the creation of tailored superhydrophobic surfaces @xcite .
the static contact angles for surfaces with local variation can be obtained from variants of the original energy minimisation .
the classic cassie - baxter treatment for spatially varying surface energies , considers a simple spatial average , ie .
, @xmath4 @xcite .
likewise , wenzel law for surfaces with height variations introduces the ratio - always larger than unity - of the total surface to the projected surface as a correction factor @xcite .
clearly both cassie - baxter and wenzel treatments may not apply in practice , since a droplet spreading towards equilibrium will encounter several pinning states in which it can be trapped , thereby never reaching the theoretical result @xcite .
this is a much - studied issue both experimentally and theoretically , particularly for engineered surfaces .
approaches that have been tried range from hydrodynamic simulations of spreading to quasi - static considerations @xcite . in all cases , a crucial concept is the spreading power , or the difference of the actual contact angle from the static or equilibrium one .
here we analyze the consequences of surface heterogeneity on the spreading and final static state of liquid droplets .
our approach centers on the evolution of the droplet radius , as defined by the contact line that separates the wetted and non - wetted parts of the substrate .
the radii are then easily related to values of contact angle through volume conservation .
the droplet radius evolves under the influence of three factors : i ) the imbalance in surface tension , directly related to the spreading power , ii ) the quenched noise , induced by the locally varying surface properties , that introduces angular deformation in the radius , and iii ) contact line elasticity that tends to smoothen radius angular variations .
we use contact line pinning and depinning dynamics to follow the evolution of the droplet radius . in the initial stages of spreading
, surface tension imbalance controls the dynamics .
however , as the droplet spreads , disorder effects start to dominate and the contact line comes to a halt _ before _ an equilibrium state can be reached .
disorder , elasticity and spreading thus predict a _ pinning transition _ for the contact line , and thus a droplet _ pinning radius _ that is sample - dependent .
we show how this depends on the the independent parameters of the theory : disorder strength , disorder length - scale , droplet radius and volume , and the fluid properties such as surface energy and contact angle .
we further obtain the statistical distributions of the pinning radius .
small droplets are strongly effected by disorder , leading to results that are markedly different from what could be expected within a wenzel or cassie - baxter framework .
disorder effects are apparent in the case of droplets with a small contact angle and disorder variations that are correlated on a short length - scale with respect to the droplet radius .
the pinning predictions for the contact angle have simple experimental consequences : they can be tested by varying for a given type of surface the nominal contact angle ( by changing liquid for instance ) and by varying the droplet size .
the rest of this paper is structured as follows . in section [ intro ]
we outline the basic spreading phenomena as well as the approach to the pinning phase .
droplet spreading in the pinning phase is analyzed in details in section [ model ] , and section [ final ] finishes with conclusions .
we consider a liquid droplet with air / liquid interfacial tension @xmath2 , density @xmath5 and viscosity @xmath6 deposited on a macroscopically flat surface . when the size of the droplet is smaller than the capillary length @xmath7 , the droplet has the overall shape of a spherical cap with basal radius @xmath8 , the air / liquid interface joining the solid surface with contact angle @xmath9 . for a perfectly homogeneous surface
, the equilibrium contact angle @xmath10 can be obtained from young s law : @xmath11 related to the equilibrium radius @xmath12 via the drop volume @xmath13 : @xmath14 this relation reduces to @xmath15 for small contact angles @xmath16 .
equation [ eq : young ] has to be supplemented by a correction term involving the line tension @xmath17 , related to @xmath18 , the curvature of the contact line @xcite @xmath19 the actual value of the line tension remains a difficult quantity to measure , due to chemical or topographical heterogeneity of the substrate and the difficulty to perform experimental work with sub - micron sized droplets . in general , the value of the line tension can be estimated to be @xmath20 , thus representing a relevant correction only for droplets with nanometer range dimensions .
substrate disorder induces variations in the contact line and hence in the local values of the contact angle and radius of the droplet .
if @xmath21 is the local coordinate along the contact line , the local droplet radius is @xmath22 , where @xmath23 is the spatially averaged radius .
variations along the contact line are connected to variations of the local contact angle @xmath24 , since the overall volume of the drop is conserved .
to first order in the radius variations , the variations in the contact angle are @xcite @xmath25 correction terms coming from the overall curvature of the droplet can be neglected for small variations of the contact line . upon deposition on a surface ,
a droplet will tend to its equilibrium shape through hydrodynamical spreading , under the influence of uncompensated young s force @xmath26 and disorder variations @xmath27
@xcite @xmath28 where @xmath29 is a numerical factor arising from finite slip at the contact line @xcite . for small values of the contact angle @xmath16 and to first order in contact line deviations ( using eq .
[ ca_first_order ] ) ) , the uncompensated young force , corrected with a term arising from the dissipation at the contact line , @xcite @xmath30 is decomposed into an overall term independent of contact line variations and a non - local term @xmath31 where the spreading power @xmath32 is balanced by the elastic restoring force @xmath33 for simplicity , we consider a chemically disordered substrate , which contributes a quenched random force .
again , in the limit @xmath16 and to first order in the contact line variations , @xmath34 where @xmath35 the noise is generally delta - correlated over two microscopic scales @xmath36 , @xmath37 , parallel and perpendicular to the direction of spreading , dependent on the nature of the substrate .
it includes the chemical contrast ( via the variation of the surface energies , @xmath1 and @xmath0 , where the indices denote the solid - vapor and solid - liquid cases , respectively ) and may also include surface roughness through local surface tilts @xcite . in what follows ,
we consider the slowly evolving dynamics of the drop . this can be achieved experimentally by slowly depositing a droplet on the substrate , thereby avoiding initial impaction effects .
after that , the viscous contact line dynamics dictates the ( slow ) evolution of the droplet @xcite .
the analysis does not apply to volatile liquids and the ohnesorge number @xmath38 is an appropriate parameter to estimate how viscous the liquid must be in order to prevent inertial as well as front instability ( coming from evaporation ) effects to occur @xcite .
we only consider microscopic disorder such that the length scales @xmath36 and @xmath37 are the smallest length scales of the problem .
the droplet then globally keeps a spherical shape , with disorder inducing variations in the contact line itself . on a disordered surface ,
the equilibrium contact angle of a droplet may be markedly different from the prediction of eq .
( [ eq : young ] ) , depending on the type or disorder . for example
, bi - modal chemical disorder , where a fraction @xmath39 of the surface has surface tension @xmath40 , is usually analyzed in terms of cassie - baxter equation @xmath41 } = \cos \theta_{eq } - f \frac{\delta \gamma}{\gamma } , \label{eq : cb}\ ] ] which is essentially a weighted average of the equilibrium contact angle on each surface . on the other hand , topographical disorder ( surface roughness )
is encompassed within wenzel s result @xmath42 where the relative area @xmath43 is the ratio of the total surface to the projected surface under the drop ( see @xcite for recent similar analyses ) .
it is interesting to note that , through wenzel s analysis , roughness increases the wetting or non - wetting tendencies originally present in the problem , ie .
, if @xmath44 , @xmath45 until @xmath46 at which point a wetting film should be formed .
both approaches however neglect that spreading is a dynamical phenomenon and that a droplet may become pinned in several configurations before reaching equilibrium .
spatial scales and correlations of the disorder are also not included . during spreading
, the contact line roughens due to the quenched disorder of the substrate , a phenomenon analyzed extensively in the literature @xcite , also in the context of the hysteresis for advancing and receding contact lines @xcite .
when disorder becomes relevant the interface dynamics enters a critical regime where pinning occurs , and the interface propagates through avalanches .
this regime is characterized by a force @xmath47 , at which an interface becomes pinned . around @xmath47 ,
the contact line develops local self - affine fluctuations ( roughness ) @xmath48 where @xmath49 is the length of the interface and @xmath50 the roughness exponent specific to contact line motion @xcite .
the propagation of the interface by avalanche is characterized by a series a correlation lengths and critical exponents . during the avalanche , a portion of the avalanche , with lateral size @xmath51 , moves by a distance @xmath52 .
the duration @xmath53 of the interface is characterized by the dynamical exponent @xmath54 .
the lateral extent of the avalanches is a correlation length related to the driving force @xmath55 and the pinning force @xmath47 by the critical exponent @xmath56 : @xmath57 .
finally , the velocity of the interface scales as @xmath58 .
critical scaling implies that @xmath59 and , for non - local elastic interface , @xmath60 .
again , for the specific case of contact line motion @xmath61 @xcite . for any finite system , @xmath47 and thus the corresponding critical angle @xmath62 have a finite - size correction and in particular a sample ( and disorder ) -dependent actual critical value , with an universal probability distribution .
this distribution is characterized by its width which decays with the interfacial length and depends only on the strength of the disorder , measured by the prefactor of the disorder two - point correlation function @xcite .
although a complete solution of roughening requires additional non - linear terms , eq .
( [ eq : dyn0 ] ) , together with typical scaling arguments @xcite , already provides much information . balancing the elastic restoring force ( eq .
( [ elastic_restoring_force ] ) ) against disorder ( eq . ( [ disorder_force ] ) ) yields a length scale @xmath63 where @xmath64 describes the pinning strength of disorder . for length scales
@xmath65 , the elastic restoring force dominates while disorder dominates for @xmath66 .
the contact angle @xmath67 at which pinning first becomes relevant is then obtained by balancing the spreading power ( eq .
( [ spreading_power ] ) ) against either the elastic or pinning force at the length scale @xmath68 .
to first order in @xmath69 , @xmath70 through volume conservation ( using @xmath15 ) , this translates to a pinning radius @xmath71 the volume of the droplet thus plays a crucial role .
large droplets will enter the pinning regime already close to equilibrium while , for smaller droplets , the ratio @xmath72 can be much smaller than 1 , even more so for strong wetting @xmath73 .
we now consider spreading on a disordered substrate .
the case of a contracting droplet ( for equilibrium contact angles in excess of @xmath74 ) is an easy extension not reported here .
the droplet has initial basal radius @xmath75 and increases its radius through hydrodynamical spreading until @xmath76 .
then , disorder becomes relevant and spreading continues if the spreading power ( cf . , eq .
( [ spreading_power ] ) ) at position @xmath8 , @xmath77 , can overcome the combined pinning potential coming from the disordered substrate and the elastic restoring force , denoted @xmath78 . at the first pinning radius , the spreading power @xmath79 where the last form is in the limit @xmath80 with @xmath81 .
@xmath82 corresponds to a contact angle value @xmath67 . in the similar limit
, the elastic and disorder - induced forces can be written @xmath83 an analysis of spreading to a radius past @xmath76 necessitates the knowledge of the pinning force distribution resulting from eq .
( [ eq_driving ] ) .
extensive numerical simulations of a driven interface subjected to the combined elastic / disorder force ( eq .
( [ eq : linearpot ] ) ) have shown that the probability @xmath84 of propagating past a radius @xmath49 derives from a distribution of pinning forces for an interface of length @xmath49 , @xmath85 @xcite @xmath86 the distribution follows a scaling form @xmath87 \label{eq : fdist}\ ] ] with a scaling function @xmath88 independent on the details of the disorder @xcite .
close to pinning , @xmath89 , the scaling function has a power law behavior , @xmath90 with @xmath91 . in this critical regime ,
the likelihood @xmath92 that the droplet radius will at least be @xmath8 is obtained from a probabilistic argument , an approach also used for indentation crack propagation @xcite .
the motion of the droplet ( see fig . [ fig1 ] ) consists of a succession of steps over independent configurations of the pinning disorder - ie .
the distance covered is divided into uncorrelated increments .
the size of the steps can be inferred from the critical dynamics of the contact line motion .
after step @xmath93 , during the sequence of jumps , the pinned contact line explores the combined potential ( elastic plus disorder ) over a distance @xmath94 , finding a configuration that tend to minimize its energy . upon depinning ( ie .
, step @xmath95 ) , the contact line moves , by a distance @xmath94 @xcite , into a new configuration . at this point , both the contact line configuration as well as the combined potential are completely uncorrelated with the previous one .
the appropriate step size during the sequence of events is thus @xmath94 , the roughness of the contact line .
a droplet will thus reach a given @xmath23 if it has passed through all the previous pinning zones without being stopped .
the probability of such a chain of events is @xmath96 where @xmath97 is the probability that the spreading power at radius @xmath49 is larger than the pinning spreading power @xmath78 .
time is not explicitly included in this argument . in the continuous limit ,
( [ eq : product ] ) can be rewritten @xmath98,\ ] ] where the zone size @xmath99 is related to the droplet radius and the correlation lengths of the disorder , @xmath100 . for a droplet size large as compared to the scale of heterogeneities , effective contact angle values will always remain in the immediate vicinity of @xmath67 and the power - law behavior of @xmath101 at the origin can be used to approximate @xmath102 . which leads to @xmath103 where the prefactor @xmath104 .
depends only on the material parameters through the correlation lengths of the disorder @xmath105 and @xmath106 . the use of the linearized form for the spreading power , eq .
( [ eq : sp ] ) and a change of variable @xmath107 finally yields : @xmath108 with @xmath109 .
the integral in eq .
( [ eq : dist ] ) exhibits an universal form which only depends on @xmath110 , a disorder scale parameter , and @xmath76 , related to @xmath12 and the strength of the disorder through eq .
( [ eq : rp ] ) .
the size of the droplet is thus implicitly present in eq .
( [ eq : dist ] ) . to analyze the results , it is convenient to set @xmath111 .
the relevant dimensionless ratios are then @xmath72 , a measure of the influence of disorder strength , droplet volume and equilibrium wetting properties , and @xmath112 , which relates the droplet typical size to the spatial structure of the disorder .
these ratios are made apparent from the limit @xmath113 of eq .
( [ eq : dist ] ) which reads @xmath114 the probability for the droplet to reach a given @xmath8 thus decays quickly close to @xmath76 .
this decay is sharper for drops that are large compared to the disorder scale ( @xmath115 ) than for smaller drops .
this tendency to cluster around @xmath76 drastically increases as the ratio @xmath116 , which occurs for weak disorder or very large drops . the importance of the ratio @xmath117 is clearly shown in figure [ fig2 ] , where the probability to reach a radius @xmath8 as calculated from eq .
( [ eq : dist ] ) is shown for various ratios @xmath117 and @xmath72 . for values
@xmath118 , this probability drops sharply and the drop remains essentially pinned at a radius @xmath76 .
it is only for relatively small values of this ratio that the probability to reach a radius larger than @xmath76 increases significantly . in other words , occasionally for small droplets the spreading can get closer to the equilibrium radius .
the effect of disorder on the values of contact angles at pinning can then be elaborated by comparing the predicted value to @xmath10 using volume conservation , eq .
( [ eq : volume ] ) .
figure [ fig3 ] shows the average contact angles calculated from the theory for two values of @xmath10 . at large values of the ratio @xmath112
, the final value of the contact angle is essentially determined by the value of @xmath76 , ie .
, strong disorder , characterized by a small ratio @xmath72 leads to a larger apparent contact angle .
in such a case , it is only for a small value of the ratio @xmath112 that the apparent contact angle can be quantitatively close to the expected @xmath10 . in figure [ fig4 ]
we show for three different disorder ( @xmath119 ) values how @xmath72 itself scales for various contact angles and a fixed @xmath51 taken to be 10 micrometers .
the values of @xmath119 are taken to have representative values ; note that the parameter measures the relative variation of the surface energy due to chemical disorder or due to roughness .
out of a variety of cases , we depict nine representative ones to show the trends .
figure [ fig4]a shows the actual droplet volumes at hand for each contact angle and disorder , while figure [ fig4]b shows the ratios of the pinning radius to the equilibrium one .
the trends themselves are obvious ( larger contact angles lessen the effect of pinning , while stronger disorder works to the other direction ) , and the final prediction is then to be computed similarly to figure [ fig3 ] for all the cases , separately .
it is thus interesting to note that the wetting properties of the surface , present through @xmath10 also enter explicitly the problem through the values of @xmath76 and @xmath12 .
this is in contrast with wenzel law or cassie law which relate the apparent contact angles to the equilibrium contact angle through a set of constants independent of the nature of the surface or of the liquid . for naturally wetting surfaces ( @xmath120 )
, wenzel law predicts that the apparent contact angle is _ larger _ than the equilibrium contact angle
. this however neglects the fact that spreading is a dynamical process and that pinning of the contact line impedes the droplet from reaching an equilibrium state .
the temporal aspects of the radial pinning process can be obtained from the relation between the interface velocity and the driving force in the critical regime .
hydrodynamical spreading of the droplet occurs until the droplet reaches the radius @xmath76 , at time @xmath121 . at this point , the spreading velocity @xmath122
after this point , motion proceeds by avalanches , with a velocity @xmath123 where again @xmath124 .
the time needed to move across a given shell of thickness @xmath125 is simply @xmath126 and the total time to pinning is obtained from summing the successive contributions of each shells . in the continuous limit , @xmath127 which is calculated to @xmath128\ ] ] and
can be reduced to @xmath129 in the limit of strong pinning ( @xmath130 ) .
this result , averaged over the distribution of pinning radius @xmath92 is illustrated in fig .
[ fig5 ] and shows that that the time to reach a given @xmath131 increases as a function of @xmath72 , particularly for @xmath8 close to @xmath12 .
to summarize , we have presented a statistical physics theory of droplet asymptotic contact angles on heterogeneous surfaces .
this allows us to identify the important quantities , such as the first pinning radius @xmath76 and the ratios to the disorder scale and the equilibrium radius , @xmath72 and @xmath117 .
the theory presented here applies to droplets with typical lengths above the nanometer scale , such that line tension effects are irrelevant .
it is also important that radius variations are smaller than the radius of the droplet , for it to keep a compact circular shape on average .
this last requirement implies that the disorder length scale @xmath132 .
the contact line dynamics model loses its validity in certain cases - one example is when long - range microscopic forces make it invalid , or when the presence of features such as corrugations make it so that the coarse - grained surface tension indirectly assumed is not present .
our results predict a dependence of the average contact angle on the volume of the droplet . at constant disorder strength
, we expect small droplets to exhibit a markedly larger contact angle at pinning than larger droplets , even more so when the droplet is large with respect to the spatial scale of the disorder .
the predictions of the theory can be easily tested through repeated spreading experiments using droplets of different volumes on the same substrate , so that the first pinning radius @xmath76 ( cf . , eq .
( [ eq : rp ] ) ) only depends on the drop volume .
it is also possible to test the theory using different liquids on a given surface .
large variations of the apparent contact angle due to the final stage of spreading are expected for small droplets .
many of the consequences of the theory of elastic manifolds are in contrast to static , energy - minimization based results as the wenzel or cassie - baxter laws .
further theoretical developments include extending the theory presented in this paper to receding radii on hydrophobic surfaces and to develop similar arguments for a structured surface on which gas phase pockets can develop .
it is also clear that the theory can be tested through large scale numerical simulations of the relevant hydrodynamics equations .
extending the probabilistic argument to the finite - temperature very - long time creep motion regime , relevant for contact lines as elastic manifolds , is also possible .
we acknowledge the support of the academy of finland ( via the center of excellence -program ) and tekes finnano program via the silsurf -project .
ma would like to thank for the hospitality of prof .
e. frey and the arnold sommerfeld center at the ludwig - maximilians - universitt , munich , germany . | the contact angle of a fluid droplet on an heterogeneous surface is analysed using the statistical dynamics of the spreading contact line .
the statistical properties of the final droplet radius and contact angle are obtained through applications of depinning transitions of contact lines with non - local elasticity and features of pinning - depinning dynamics .
such properties not only depend on disorder strength and surface details , but also on the droplet volume and disorder correlation length
. deviations from wenzel or cassie / baxter behaviour are particularly apparent in the case of small droplet volumes and small contact angles . |
the boltzmann - gibbs formalism of statistical mechanics is heavily based on the equivalence of ensembles , which for many systems holds in the thermodynamic limit .
it is not adequate to describe the thermodynamic behaviour of small systems . in particular
, phase transitions can only occur in the thermodynamic limit of the canonical ensemble .
recently @xcite , the use of the microcanonical ensemble has been advocated to describe phase transitions in small systems .
the present letter supports this approach , but argues that one of its premises should be slightly modified .
indeed , it is tradition to use boltzmann s entropy @xmath0 with @xmath1 a small but fixed energy and with @xmath2 the density of states , as the obvious definition of thermodynamic entropy of isolated systems .
alternatively ( see @xcite and the references cited there ) , the density of states @xmath3 in ( [ bent ] ) can be replaced by the integrated density of states @xmath4 .
the proposal of the present letter is to replace it by a related quantity @xmath5 , solution of the non - linear differential equation @xmath6 the resulting entropy is denoted @xmath7 @xmath8 a conceptual argument supporting this modification is given below .
practical consequences for macroscopic systems are neglegible .
but for small systems the consequences are important .
according to boltzmann the equilibrium state of an isolated system is the macrostate that can be realized in the largest number of ways @xcite .
all accessible states have the same probability .
this can be taken as the definition of the microcanonical ensemble .
the notion of entropy enters when two or more isolated systems are compared .
the density of states in the combined system is the product of density of states of each of the composing systems . on the other hand
, thermodynamic entropy must be the sum of the entropies of the subsystems .
the standard argument is then that entropy necessarily is proportional to the logarithm of the density of states .
this results in boltzmann s entropy .
problem with this reasoning is that the compound system still consists of two unrelated systems , e.g. physically separated by a wall .
removal of this wall would affect entropy .
if the wall is not removed then the compound system is in equilibrium , but this is the equilibrium of a microcanonical ensemble constraint to the requirement that the different subsystems , labeled with index @xmath9 , have fixed energies @xmath10 .
not all possible ways of distributing the total energy @xmath11 over the subsystems are realized .
hence it is not in equilibrium in the unconstraint ensemble .
the usual way to overcome this problem is by means of equivalence of ensembles .
however , this works only for macroscopic systems , if it works at all .
indeed , the essence of the microcanonical approach to phase transitions in finite systems is precisely that the equivalence of ensembles can break down in presence of phase transitions .
when this happens then the microcanonical ensemble is the one that decribes nature , not the canonical ensemble .
several conclusions can be drawn from this discussion . in the first place it shows that the standard argument for associating boltzmann s entropy with the microcanonical ensemble is not justified in case of small systems .
it also shows that the possibility of phase separation should be taken into account because it modifies the accessible part of phase space . a different type of reasoning is now followed .
it starts from the assumption that it is possible to compare isolated systems and that it is meaningful to state that they all have the same microcanonical temperature .
note that energy is the control parameter of the microcanonical ensemble , not temperature .
therefore the argument is restricted to simple systems for which energy and temperature are uniquely related .
initially , put all the isolated subsystems in their ground state .
next distribute small amounts of energy d@xmath12 to the system in such a way that all degrees of freedom of the system are treated in the same manner . to this purpose ,
introduce an effective number of degrees of freedom @xmath13 which may depend on the energy @xmath10 of subsystem @xmath9 .
then the amount of energy received by subsystem @xmath9 is proportional to @xmath13 .
let us postulate that this procedure of distributing energy over the subsystems keeps them all at the same temperature @xmath14 .
for each of them is then @xmath15 with @xmath16 the heat capacity of the total system .
the obvious solution of this equation is @xmath17 integration of @xmath18 then gives temperature @xmath19 as a function of energy @xmath10 .
let us next try to relate the effective number of degrees of freedom @xmath20 to the density of states @xmath2 .
a suitable definition is @xmath21 for a system with hamiltonian @xmath22 the density of states equals @xmath23 in case of the harmonic oscillator is @xmath24 , which makes ( [ degfree ] ) badly defined .
however , if @xmath25 , with @xmath26 constant , then @xmath27 follows .
hence , taking the limit @xmath28 one concludes that for a constant density of states one has @xmath29 , as expected .
this implies @xmath30 , which is the canonical result . in case of the ideal gas
is @xmath31 .
this implies @xmath32 and hence @xmath33 which is again the canonical result .
one concludes that ( [ degfree ] ) gives physical results in both cases .
note that one has @xmath34 if the system is macroscopic then the constant 1 in this expression may be omitted , in which case integration can be done easily .
the result is that in that case boltzmann s entropy is the thermodynamic entropy , satisfying the thermodynamic relation @xmath35 in the general case boltzmann s entropy is not the thermodynamic entropy .
the proposal is to replace it by @xmath7 with @xmath5 chosen in such a way that ( [ thermtemp ] ) is satisfied and that the heat capacity @xmath16 equals @xmath36 , with @xmath20 given by ( [ ms ] ) .
this implies that @xmath5 must satisfy @xmath37 in combination with ( [ degfree ] ) this gives ( [ difeq ] ) .
let us investigate some properties of @xmath7 .
a short calculation yields @xmath38 hence the condition for @xmath7 to be concave is that @xmath39 . on the other hand ,
it is clear from ( [ ms ] ) that the condition for boltzmann s entropy to be concave is @xmath40 .
one concludes that some of the systems that are unstable w.r.t .
boltzmann s entropy are still stable when analysed with @xmath7 . in particular ,
negative heat capacities , often discussed in the context of finite systems @xcite , are less likely to be negative when using @xmath7 . of practical importance
is that ( [ difeq ] ) can be solved in a systematic manner . introduce a function @xmath41 the function @xmath42 is defined in a similar way .
then ( [ difeq ] ) can be written as @xmath43 by integration of @xmath42 one obtains the inverse of @xmath44 .
inverting and integrating again one obtains @xmath7 . finally , note that the two entropies @xmath45 and @xmath7 coincide only when the density of states is an exponential function @xmath46 . in that case the effective number of degrees of freedom is infinite
of interest is the situation where @xmath20 is constant .
e.g , if @xmath25 , then one has @xmath47 .
this case includes the harmonic oscillator as well as the ideal gas .
however , there is a larger class of systems with constant @xmath20 .
note that from ( [ degfree ] ) follows @xmath48 solutions of this equation are of the form @xmath49 , involving the deformed exponential functions @xcite @xmath50_+^{1/(1-q)}\end{aligned}\ ] ] ( the notation @xmath51_+=\max\{0,x\}$ ] is used ) .
the corresponding @xmath52-entropy ( determined up to a constant term which is fixed by choosing an arbitrary positive energy @xmath1 ) is @xmath53 here , @xmath54 is the ground state energy of the system .
temperature @xmath19 , calculated using ( [ thermtemp ] ) , satisfies @xmath55 one concludes that any microcanonical system described by a density of states of the form @xmath49 satisfies the equipartition law .
an interesting example of a finite system with phase transition is the pendulum with hamiltonian @xmath56 for low energy @xmath57 the motion is oscillatory . at large energy @xmath58
it rotates in one of the two possible directions .
a correct understanding of the thermostatistics of this model is highly relevant because rotational motion of ions and molecules , or part of molecules , occurs frequently in all states of matter .
see e.g. @xcite .
the density of states @xmath2 can be written as @xmath59 with @xmath1 a constant , inserted for dimensional reasons , and with @xmath60 given by @xmath61 see fig .
note that a factor @xmath62 has been inserted in case @xmath63 to take into account that the pendulum is rotating either clockwise or anti - clockwise .
the density of states @xmath2 diverges at @xmath64 .
it is log - convex on both intervals @xmath57 and @xmath65 , rather than log - concave .
hence , boltzmann s entropy can not be used to analyse this model .
the effective number of degrees of freedom @xmath20 has the value 1/2 at large values of @xmath12 , as expected .
however , it does not start with the value 1 at @xmath66 , but with @xmath67 .
this is due to anharmonicity of the oscillatory motion .
the heat capacity @xmath68 has a shape typical for a second order phase transition at @xmath69 .
see fig .
arguments have been given why boltzmann s entropy is not suited to discuss phase transitions in small systems and should be slightly modified .
the proposed replacement is given by ( [ thent ] ) .
it involves a function @xmath5 , related to the density of states @xmath2 and obtained by solving ( [ difeq ] ) .
it is based on the notion of effective number of degrees of freedom @xmath20 and the principle that , if two systems with equal @xmath20 are heated up in the same manner , then they have at all times the same temperature .
crucial point is relation ( [ degfree ] ) between @xmath20 and the density of states @xmath2 . omitting the constant term @xmath70 in this expression leads to boltzmann s entropy .
hence , if the number of degrees of freedom is macroscopic then the proposed entropy ( [ thent ] ) coincides with boltzmann s entropy .
but for small systems one can expect important differences . to illustrate this point the classical pendulum has been discussed . in the present approach
it exhibits a second order phase transition between vibrations and rotations .
it can not be treated using boltzmann s entropy because the density of states is piecewise convex instead of concave . in the approach of @xcite it undergoes a first order transition because heat capacity is negative in a range of energies ( this calculation has not been included but can be done straightforwardly ) .
negative heat capacities have been discussed at many places in the literature , in both theoretical and experimental papers .
it is out of scope of the present letter to reanalyse all those applications .
but it is clear that this has to be done .
the main omission of the present letter is a discussion of the quantum mechanical microcanonical ensemble . in the quantum case
the density of states @xmath71 counts the number of eigenstates with energy in the range from @xmath12 to @xmath72 , for some fixed @xmath73 .
it is a highly singular function of energy @xmath12 .
this complicates the calculation of the function @xmath5 , and hence , of the modified boltzmann entropy @xmath7 .
the only obvious solution to this problem is to replace the density of states by a continuous function .
this is of course only meaningful when the eigenvalues of the hamiltonian are densely spread over some relevant interval of energies .
a more fundamental solution is lacking and requires further investigation .
uffink j. , `` boltzmann s work in statistical physics '' , the stanford encyclopedia of philosophy ( winter 2004 edition ) , edward n. zalta ( ed . ) , forthcoming url = http://plato.stanford.edu / archives / win2004/entries / statphys - boltzmann/. gobet f. , farizon b. , gaillard m.j .
, buchet j.p . , carr m. , scheier p. , mrk t.d . , phys .
89 * , 183403 ( 2002 ) .
chabot m. , wohrer k. , phys .
93 * , 039301 ( 2004 ) .
gobet f. , farizon b. , gaillard m.j .
, buchet j.p . , carr m. , scheier p. , mrk t.d . , phys .
lett . * 93 * , 039302 ( 2004 ) . | boltzmann s entropy is slightly modified to make it suitable for discussing phase transitions in finite systems . as an example
it is shown that the pendulum undergoes a second order phase transition when passing from a vibrational to a rotating state . |
single electron control in semiconductor quantum dots would be valuable for spintronics , quantum information , and spin qubits@xcite .
adiabatic time evolution of degenerate eigenstates of a quantum system provides a means for controlling individual electrons since mixing between degenerate levels generates a matrix berry phase@xcite .
it has been shown that universal quantum computation is possible by means of non - abelian unitary operations@xcite . recently
several proposals on how to generate matrix berry phases have been made in semiconductor systems@xcite . in one of these proposals
it has been suggested that a _ single electron _ in a n - type semiconductor quantum dot with spin - orbit terms can have matrix berry phases@xcite .
all the discrete energy levels of quantum dots possess a double degeneracy because the rashba and/or dresselhaus spin orbit coupling terms have time reversal symmetry@xcite .
in ref.@xcite it was suggested that the adiabatic transformation can be performed _ electrically _ by changing the confinement potentials of the quantum dot .
it was shown that matrix berry phases can be produced only when the parity symmetry of two - dimensional harmonic potentials is broken .
a degenerate pair of states also exists in quantum rings@xcite .
the relevant symmetry of this system is the combined operation of time reversal and large gauge transformations .
however , it is often not easy to populate a quantum dot or ring with just one electron and usually there are several electrons in it@xcite .
it is not certain that many body groundstates are doubly degenerate in the presence of time reversal symmetry .
moreover , it is unclear how the effects of exchange and correlation many body physics change the non - abelian vector potentials .
these issues are investigated in the present paper .
before we give a summary of the main points of the present paper let us give a brief introduction of the basic ideas behind matrix berry phases .
consider a degenerate pair of states @xmath0 and @xmath1 , which may be _ single particle or many body states_. if the system is in a superposition state @xmath2 at time t=0 an adiabatic evolution of the parameters @xmath3 can transform this state into another state @xmath4 after some time @xmath5 . here
the orthonormal basis states @xmath6 are the instantaneous eigenstates of the hamiltonian . for a cyclic change with the period @xmath7 , represented by a closed contour @xmath8 in the parameter space , the states @xmath9 and @xmath10 return to the initial states @xmath11 and @xmath12 , but the coefficients @xmath13 and @xmath14 may not return the initial values . in such a case a @xmath15 matrix berry phase (
non - abelian berry phase ) @xmath16 is generated @xmath17 the expansion coefficients @xmath18 and @xmath19 satisfy the time - dependent schr@xmath20dinger equation @xmath21 the matrix elements @xmath22 are given by @xmath23 , where the sum over @xmath24 in @xmath22 is meant to be the sum over @xmath3 .
the non - abelian vector potentials @xmath25 are @xmath15 matrices @xmath26 note that the wavefunctions @xmath27 and @xmath28 are _ degenerate_. the role of the non - abelian vector potentials is to keep the energy degeneracy intact while the system parameters change .
our investigation shows that when even number of electrons are present the many body groundstate may not be doubly degenerate .
however , when odd number of electrons are present the groundstate may be doubly degenerate . in the simplest possible approximation the antisymmetry of many body wavefunctions can be ignored .
then the degenerate groundstates for three electrons are given by @xmath29 and @xmath30 , where @xmath31 are single electron wavefunctions , see fig.[fig : single ] . in this case
one can show that the solutions of the time dependent schroedinger equation , eq.([eq : time_schrod ] ) , have the form @xmath32 the function @xmath33 depends on the single electron non - abelian vector potentials of the _ first _ shell , and the functions @xmath34 and @xmath35 depend on the single electron non - abelian vector potentials of the _ second _ shell , see fig.[fig : single ] ( the function @xmath33 is periodic with the period @xmath7 ) . the probability to find an electron in the state @xmath36 is @xmath37 .
so these probabilities are _ independent _ of the phase factor @xmath38 , i.e. , of the non - abelian vector potentials of the first shell . in the next simplest possible approximation
the antisymmetry of many body wavefunctions can be included . in this case
the doubly degenerate many body groundstates may be taken to be two slater determinant wavefunctions , @xmath39 and @xmath40 , that are time reversed states of each other , see fig.[fig : many ] .
it will be shown below that the antisymmetry of these slater determinant wavefunctions does _ not _ change the results obtained above for the matrix berry phase .
in addition to the antisymmetry one can include correlation effects .
this can be done by including four basis vectors @xmath39 , @xmath40 , @xmath41 and @xmath42 , see fig.[fig : many ] . in this case
the groundstates are given by linear combinations of these four states and are thus correlated .
it can be shown that the non - abelian gauge potentials for a correlated electron system are related to non - abelian gauge potentials for single electrons , provided that the definition of single electron non - abelian gauge potentials is extended to @xmath43 when @xmath44 and @xmath45 belong to _ different _ single electron energy shells .
we will call these _ inter shell non - abelian gauge potentials_. in sec .
ii we describe the hamiltonian of the system in detail , and in sec .
iii we discuss why matrix berry phases can be produced only when the parity symmetry of two - dimensional harmonic potentials is broken .
a computational scheme to include systematically many electron physics is explained in sec.iv .
the absence of a matrix berry phase for the two - electron case is shown in sec .
v. in sec.vi we show that a matrix berry phase is present in the three - electron case .
correlation effects are discussed in vii and conclusions are given in sec .
an electric field @xmath46 is along the z - axis and electrons are confined in a triangular potential @xmath47 . when the the width of the quantum well along the z - axis is sufficiently _ small
_ we may include only the lowest subband state along the z - axis .
we denote this wavefunction by @xmath48 . from the expectation value @xmath49
we estimate the characteristic length scale along the z - axis : @xmath50 .
the hamiltonian in the absence of the spin orbit coupling is @xmath51 we take the two - dimensional potential to be @xmath52 .
the strengths of the harmonic potentials are denoted by @xmath53 and @xmath54 .
they may be varied by _ changing gate potentials _ of the quantum dot system . the characteristic lengths scales along x- and y - axis are @xmath55 .
the potential @xmath56 is chosen in such a way that the parity symmetry of electron wavefunctions is broken .
we can choose it to be @xmath57 .
it represents an electric field along the y - axis and its strength @xmath58 may be _ varied electrically_. in a periodic crystal potential of a semiconductor the spin orbit interaction has two contributions .
the rashba spin orbit term@xcite is @xmath59 here @xmath60 are pauli spin matrices and @xmath61 are momentum operators ( @xmath62 and similarly with @xmath63 . ) .
the constant @xmath64 _ depends on the external electric field _
@xmath46 applied along the z - axis .
the dresselhaus spin orbit term@xcite is @xmath65 there is another term of the form @xmath66 in the dresselhaus spin orbit term but it vanishes since the expectation value @xmath67 for the first subband along z - axis .
the constant @xmath68 represents breaking of inversion symmetry by the crystal in zinc blende structures .
the total single electron hamiltonian of semiconductor quantum dot is @xmath69 .
when several electrons are present in the dot the total hamiltonian consists of single particle hamiltonians and two - particle interaction terms @xmath70 , where @xmath71 and @xmath72 with @xmath73 .
( @xmath74 is the single particle hamiltonian for the ith electron and @xmath75 is the background dielectric constant ) .
the total many electron hamiltonian is invariant under time reversal symmetry .
the eigenenergies of the single electron hamiltonian is depicted in fig.[fig : single ] .
each energy is doubly degenerate .
an eigenstate of the hamiltonian @xmath76 may be expanded as a linear combination of the eigenstates of @xmath77 .
a lowest energy state can be written as @xmath78 in the basis states @xmath79 the quantum number @xmath80 and @xmath81 label the harmonic oscillator levels along the x - axis ( y - axis ) and the component of electron spin .
the subband wavefunction @xmath48 is suppressed in the notation @xmath82 .
the wavefunction @xmath83 can also be written as a column vector @xmath84 , where @xmath85 and @xmath86 .
the expansion coefficients satisfy a matrix equation @xmath87 the single electron wavefunctions used in this section are self - consistent hartree - fock single electron states if @xmath76 is replaced by the hf single electron hamiltonian . in the absence of the zeeman term
the single particle hamiltonian is invariant under time reversal symmetry : @xmath88 and @xmath89 .
the time reversal operator is @xmath90 , where the operator @xmath8 stands for complex conjugation .
the time reversed state of @xmath91 is @xmath92 this can also be written as a column vector @xmath93 .
note that @xmath94 .
these two states are degenerate and orthonormal .
we have suppressed the bloch wavefunction of the conduction band in applying the time reversal operator since it is unaffected by the operator @xmath7 .
our wavefunctions are all effective mass wavefunctions and only the conduction band bloch wavefunction at @xmath95 is relevant . even in the presence of
doubly degenerate states a matrix berry phase may not always exist .
it was shown numerically that only distorted two - dimensional harmonic potentials breaking the parity symmetry generate matrix berry phases@xcite . here
reasons for this effect for the single electron case is given in detail . in this argument
only one electron is assumed to be present in the dot , @xmath96 , and @xmath97 .
the adiabatic parameters are @xmath98 and @xmath99 .
let us consider @xmath100 as a perturbation and apply a perturbation theory to @xmath101 @xmath102 where @xmath103 .
the wavefunction of @xmath104 has even parity .
the matrix element @xmath105 couples @xmath104 only to odd parity states @xmath106 .
therefore the wavefunction corresponding to the second term in eq.([eq : symm ] ) has odd parity with spin down .
the wavefunction of the third term in eq.([eq : symm ] ) has even parity with spin up .
this perturbation series indicates that the exact lowest energy state has the form @xmath107 , where @xmath108 and @xmath109 even and odd functions . the time reversed wavefunction is @xmath110 . for this particular pair of degenerate
states the off - diagonal elements of non - abelian vector potentials are zero : @xmath111 here @xmath112 with @xmath113 such that the two - dimensional harmonic wavefunctions @xmath114 are even functions .
the function @xmath115 with @xmath116 are such that @xmath117 are odd functions ( note that one can show @xmath118 )
. since the off - diagonal non - abelian vector potentials are zero the matrix berry phase is absent . in what follows a distorted two - dimensional harmonic potential
is assumed .
the adiabatic parameters are @xmath64 and @xmath58 .
a computational scheme that includes systematically many electron physics is explained here .
we find an efficient computational scheme of computing eigenstates and eigenvalues of the many electron hamiltonian of a semiconductor quantum dots with an arbitrary shape of the confinement potential .
we follow the following steps .
single electron hamiltonian is first solved in the presence of spin - orbit terms .
this can be done by following the steps given in sec.iii .
the number of single electron basis states @xmath119 can be truncated to a reasonable number by comparing the relative magnitude of the single particle energy spacing and the strength of the coulomb interaction . using _ these _ single electron eigenstates slater determinant states @xmath27 and their time reserved states
@xmath120 are formed .
@xmath120 is obtained from @xmath27 by replacing the single electron wavefunctions with their time reversed states : @xmath121 .
it will be inefficient to use two dimensional harmonic potential wavefunction @xmath122 to build these slater determinant states .
the following relation hold for two arbitrary states @xmath123 and @xmath124 : @xmath125 . from this relation
it follows that the hamiltonian matrix elements have the following properties : @xmath126 , @xmath127 , and @xmath128 . when there are odd number of electrons the basis vectors of many electron hamiltonian are chosen to include time reversed pairs : @xmath129 , see fig .
[ fig : many ] .
the eigenstates are doubly degenerate in the odd electron case .
let us show this for the three electron case using the following slater determinant wavefunctions as the basis vectors in the many body hilbert space @xmath130 .
these states are depicted in fig.[fig : many ] .
the wavefunctions of the lowest energy degenerate pair are @xmath131 and @xmath132 .
clearly @xmath133 .
note @xmath134 and @xmath135 are a time reversed pair , orthogonal , and satisfies @xmath136 .
the two eigenvalues of @xmath137 are each doubly degenerate . when there are even number of electrons the basis vectors of the many electron hamiltonian are chosen as @xmath138 . in this scheme
the groundstate is non degenerate .
this can be shown shown , for example , in an approximation that includes five many body basis vectors in the presence of two electrons , see fig .
[ fig : many2 ] . the groundstate can then be written as a linear combination @xmath139 .
its time reversed state is @xmath140 .
clearly these two states are _ not _ orthogonal .
since @xmath141 and @xmath142 one finds that the energy expectation values are non degenerate , @xmath143 , i.e. , time reversed states are non degenerate .
the matrix berry phase is thus absent . in the simplest possible approximation
the groundstate @xmath144 of two electron system is given by the single slater determinant made of the single particles states @xmath145 .
in this case the initial state should return to itself after the adiabatic cycle .
it is instructive to verify this explicitly by transforming each single particles states .
the initial state is @xmath146 the transformed two electron slater determinant is @xmath147 , which is @xmath148 . since the transformed single electron states are time reversed states of each other @xmath149 .
so the factor in the transformed wavefunction is @xmath150 .
now consider odd number of electrons , for example , three electrons in a dot . in the simplest possible approximation with fermionic antisymmetry
two many body basis states , @xmath39 and @xmath40 , can be included .
these are slater determinant states and are depicted in fig.[fig : many ] .
this approximation is well justified when the typical single electron energy spacing is larger than the characteristic coulomb energy .
clearly @xmath39 and @xmath40 are orthogonal @xmath151 .
the hamiltonian matrix is @xmath152 the off - diagonal matrix elements of the hamiltonian matrix is zero since @xmath153 , which follows from @xmath125 .
it is instructive to show this explicitly by evaluating the matrix element @xmath154+[\langle 13 |v|14 \rangle-\langle 13 |v|4 1 \rangle].\nonumber\\\end{aligned}\ ] ] here the first and third terms are direct coulomb matrix elements while the second and fourth terms are their exchange counter parts .
the direct coulomb matrix elements are zero since @xmath155 .
so only their exchange counter parts remain , which are given by @xmath156 and @xmath157 one can show @xmath158 this is because @xmath159 which follows from @xmath160 , @xmath161 , and @xmath162 note that the second and third terms are equal except for the minus sign .
the off - diagonal elements are thus zero , and @xmath163 this implies that @xmath39 and @xmath40 are eigenstates and are degenerate in energy . during the adiabatic change the wavefunction of three electrons evolves as @xmath164 .
the coefficients @xmath18 and @xmath19 satisfy the schroedinger equation eq.([eq : time_schrod ] ) governed by the non - abelian vector potentials @xmath165 it can be shown using eqs.([eq : diag ] ) and ( [ eq : matrixover ] ) that these can be written in terms of the non abelian vector potentials for single particle states @xmath166 where the non - abelian vector potentials involving the lowest energy single electron states are @xmath167 and involving the second lowest single electron states are @xmath168 the solution of the time dependent schroedinger equation has the form @xmath169 .
the function @xmath170 depends on the electrons in the _ first _ shell .
it satisfies @xmath171 , where @xmath172 .
the solution of this differential equation is a phase factor @xmath173 note that @xmath174 appears in the diagonal elements of the non abelian vector potentials @xmath175 and @xmath176 .
the functions @xmath34 and @xmath35 are the solutions of the time dependent schroedinger equation @xmath177 with the following non - abelian gauge potentials .
@xmath178 this is precisely the non abelian vector potentials for the _ second _ energy shell . the probability to find an electron in the state @xmath36 is @xmath37 .
so these probabilities are _ independent _ of the non - abelian vector potentials of the first energy shell .
in addition it should be noted that the antisymmetry of the many electron wavefunction does _ not _ change the value of the matrix berry phase obtained from the wavefunctions without the antisymmetry , see eq.([eq : berry ] ) .
this is the main conclusion of this section .
correlation effects can be taken into account by writing many body eigenstates as a linear combination of single slater determinant wavefunctions @xmath179 .
these wavefunctions arranged in the increasing order of the total kinetic energy @xmath180 .
note that @xmath181 for all i and j. if the groundstate is doubly degenerate one of the degenerate groundstates can be written as @xmath182 then its time reversed state has the form latexmath:[\[\begin{aligned } since @xmath7 is an antiunitary operator .
note that @xmath184 are ordered as follows @xmath185 .
the states @xmath186 and @xmath187 are eigenstates of the truncated hamiltonian with the dimension @xmath188 .
( note that as @xmath189 the energy @xmath46 will approach the exact eigenvalue . ) .
some of the slater determinant states @xmath190 are depicted in fig.[fig : many ] .
clearly @xmath186 and @xmath187 are orthogonal .
the matrix element between them is zero : @xmath191=0 .
these two states are degenerate since @xmath192 .
the non - abelian vector potentials are @xmath193 in terms of many body basis states they can be written as @xmath194 in the three electron case with the basis vectors given in fig.[fig : many ] the diagonal matrix elements @xmath195 and @xmath196 are given by @xmath197 similarly one can find off - diagonal matrix elements @xmath198 these results follow from eq.([eq : matrixover ] ) .
usually the non abelian vector potentials are defined between two degenerate states , see eq.([eq : vector_pot ] ) .
when electron correlations are important the matrix elements between two single electron states belonging to _ different _ energy shells are relevant ( see the last four equations in eq.([eq : result ] ) ) . we will call these inter shell non - abelian vector potentials .
physically the mixing between the second and third shells is facilitated by them .
note that @xmath199 .
we have developed a scheme of computing eigenstates and eigenvalues of the many electron hamiltonian of a semiconductor quantum dot with an arbitrary shape of the confinement potential .
the basis vectors of this many electron hamiltonian are chosen to include time reversed pairs of slater determinant states . in this scheme
the groundstate is non degenerate when an even number of electrons are present while the eigenstates are doubly degenerate when an odd number of electrons are present . using this computational scheme we have investigated matrix berry phases when several electrons are present in a semiconductor quantum dot with spin - orbit coupling terms .
when an odd number of electrons is present a matrix berry phase exists .
the effect of the antisymmetry of many electron wavefunctions does not affect the value of the matrix berry phase .
electron correlation effects can be treated by including the inter shell coupling . in this case
the many body non - abelian vector potentials can be written as a sum of those of the single electron non - abelian vector potentials , provided that the inter shell non - abelian vector potentials , @xmath200 with @xmath45 and @xmath201 non degenerate , are introduced to represent the shell coupling .
it would be interesting to investigate the role of these inter shell non - abelian vector potentials in parabolic dots .
optical dipole transitions@xcite can be used to detect matrix berry phases even when several electrons interact strongly with each other .
this is because the dipole oscillator strength will depend on the matrix berry phase@xcite .
however , our results in sec .
v indicate the matrix berry phases will be independent of many body physics when the energy level spacing between the shells is larger than the characteristic coulomb energy . in self
assembled quantum dots with wetting layers this condition is satisfied@xcite . in this case
only the intra shell non - abelian vector potentials , @xmath200 with @xmath45 and @xmath201 degenerate , are relevant , and the value of the matrix berry phase can be calculated as in ref.@xcite .
we leave it as a future study to investigate numerical values of the matrix berry phases for strongly interacting electrons . in ii - vi semiconductors the rashba term
is expected to be larger than the the dresselhaus coupling and in iii - v semiconductors the opposite is true@xcite .
the typical value of the energy scale associated with the rashba constant depends on the electric field applied along the z - axis and the semiconductor material : it is of order @xmath202 , where the length scale @xmath203 is the lateral dimension of the quantum dot .
the second adiabatic parameter @xmath204 represents the strength of the distortion potential @xmath205 : the expectation value of the distortion potential is @xmath206 .
its magnitude is of order @xmath207 , depending on the electric field applied along the y - axis . in self - assembled inas quantum dots@xcite
the level spacing can be larger than the characteristic coulomb energy , and our investigation shows that in such a case the many problem essentially reduces to a single particle problem ( the energy @xmath208 associated with the length scale @xmath209 is typically of order 20 - 40mev in self - assembled quantum dots ) . in this regime , as we explained above , the exchange effect disappears in matrix berry phases and it is only the last singly occupied electron that matters .
the presence of the other electrons in the inert filled shells will just modify the confinement potential .
thus the matrix berry phase can be calculated following the method outlined in ref[11 ] , and below we give some typical values of the matrix berry phases .
we take an elliptic adiabatic path given by @xmath210 . for the parameters @xmath211 , @xmath212 , @xmath213 , and @xmath214 we find @xmath215 , respectively , for different values of the semiaxis of the ellipse @xmath216 ( the initial state is @xmath217 ) .
note that the parameter for the semiaxis @xmath218 can be varied experimentally either by changing the rashba field @xmath219 or the size of the dot through @xmath208 .
we have verified that two different values of the frequency @xmath220 give the same values of @xmath221 as long as the other parameters are unchanged , confirming that the matrix berry phase is a geometric phase@xcite .
this should be tested in any experimental investigation of matrix berry phases .
experimental data of single quantum dots would be less complicated than data of an ensemble of dots with various sizes of dots , and can thus be compared directly with our theoretical results .
let @xmath222 consists of @xmath223 and let @xmath1 consists of @xmath224 : the diagonal elements of the non - abelian gauge potential between the slater determinant wavefunctions are @xmath227 let us find the off - diagonal non - abelian vector potential elements @xmath228 between the two slater determinant wavefunctions @xmath229 and @xmath230 .
when @xmath231 is replaced with @xmath232 the state @xmath222 transforms into @xmath1 .
we find @xmath233 .
this result may be easily generalized to @xmath234 particle slater determinant wavefunctions .
in general the non - abelian vector potential element between two _ different _ slater determinant wavefunctions , @xmath190 and @xmath235 , is non zero when @xmath190 can be transformed into @xmath235 by replacing one single electron wavefunction in @xmath190 with another wavefunction in @xmath235 .
if these single electron wavefunctions are denoted by @xmath236 with @xmath237 then we find @xmath238 if more than two single electron wavefunctions in @xmath190 must be exchanged with those in @xmath235 to transform @xmath190 into @xmath235 then @xmath239 .
this work was supported by grant no . r01 - 2005 - 000 - 10352 - 0 from the basic research program of the korea science and engineering foundation and by quantum functional semiconductor research center ( qsrc ) at dongguk university of the korea science and engineering foundation .
in addition this work was supported by the second brain 21 project .
d. d. awschalom , d. loss , and n. samarth , semiconductor spintronics and quantum computation ( springer , berlin , 2002 ) .
e. i. rashba and al .
l. efros , phys .
lett . * 91 * , 126405 ( 2003 ) .
, edited by a. shapere and f. wilczek ( world scientific , singapore , 1989 ) .
f. wilczek and a zee , phys . rev .
lett . * 52 * , 2111 ( 1984 ) .
p. zanardi and m. rasetti , phys
* 264 * , 94 ( 1999 ) .
j. pachos , p. zanardi and m. rasetti , phys .
a * 61 * , 010305(r ) ( 2000 ) ; j. pachos and p. zanardi , int . j. mod
b * 15 * , 1257 ( 2001 ) .
b. a. bernevig and s .- c .
zhang , phys .
b * 71 * , 035303 ( 2005 ) .
serebrennikov , phys .
b * 70 * , 064422 ( 2004 ) .
p. solinas , p. zanardi , n. zanghi , and f. rossi , phys .
rev.a * 67 * , 062315 ( 2003 ) .
d. parodi , m.sassetti , p. solinas , p. zanardi , and n. zanghi , phys .
rev.a * 73 * , 052304 ( 2006 ) . s .-
eric yang and n.y .
hwang , phys .
b * 73 * , 125330 ( 2006 ) .
m. valin - rodriquez , a. puente , and l. serra , phys .
b * 69 * , 085306 ( 2004 ) . s .-
eric yang , phys .
b * 74 * , 075315 ( 2006 ) .
l. p. kouvenhoven , c. m. marcus , p. l. mceuen , s. tarucha , r. m. westervelt , and n. s. wingreen , in nato asi conference proceedings , section 5.1 , edited by l. p. kouvenhoven , g. schn , and l. l. sohn ( kluwer , dordrecht , 1977 ) . s. d. lee , s. j. kim , y. b. cho , j. b. choi , sooa park , s .- r .
eric yang , s. j. lee and t. h. zyung , appl .
89 , 023111 ( 2006 ) .
rashba , physica e , * 34 * , 31 ( 2006 ) .
g. dresselhaus , phys . rev . * 100 * , 580 ( 1955 ) .
d. heitmann and j. p. kotthaus , phys .
today * 56 * ( 6 ) , 56 ( 1993 ) ; the effect of many electron physics on dipole absorption spectra are investigated in parabolic dots in the presence of spin orbit terms in the following papers : p. lucignano , b. jouault , a. tagliacozzo , and b. l. altshuler , phys .
b * 71 * , 121310(r ) ( 2005 ) ; p. pietilinen and t. chakraborty , phys . rev .
b * 73 * , 155315 ( 2006 ) . to detect matrix berry phases it is desirable to investigate similar effects in non - parabolic dots with parity breaking terms .
p. m. petroff , a. lorke , and a. imamoglu , phys .
today * 54 * ( 5 ) , 46 ( 2001 ) . | adiabatic time evolution of degenerate eigenstates of a quantum system provides a means for controlling electronic states since mixing between degenerate levels generates a matrix berry phase . in the presence of spin - orbit coupling in n - type semiconductor quantum dots
the electron hamiltonian is invariant under time reversal operation and the many body groundstate may be doubly degenerate . this double degeneracy can generate non - abelian vector potentials when odd number of electrons are present .
we find that the antisymmetry of many electron wavefunction has no effect on the matrix berry phase .
we have derived equations that allow one to investigate the effect of electron correlations by expressing the non - abelian vector potentials for many electron system in terms of single electron non - abelian vector potentials . |
the standard algorithm for square lattice dla@xcite is as follows : 1 .
let the initial cluster consist of a single seed particle at the origin of the lattice .
2 . launch a new particle on a launching circle of radius @xmath18 that contains the current cluster .
in other words , generate an angle @xmath19 from the uniform distribution on @xmath20 , and set @xmath21 and @xmath22 .
3 . move the particle east , west , north , or south with equal probability .
if the particle is adjacent to the cluster , add it to the cluster , and go back to step 2 .
if the particle has diffused outside the killing circle of radius @xmath23 , discard it , and go back to step 2 .
go back to step 3 . the launching radius is typically taken to be @xmath24 where @xmath25 is the radius of the bounding circle circumscribing the cluster.@xcite the killing radius may be as small@xcite as @xmath26 or as large@xcite as @xmath27 . obviously , dla is a stochastic process involving random numbers
. measurements of observables ( such as @xmath11 ) are subject to random error , which can not be eliminated , but can be reduced by averaging over many simulations , or by self - averaging as part of going to larger system sizes .
however , one should be wary of systematic errors .
these can not be removed by any amount of statistical averaging .
moreover , emergent phenomena such as the self - organized critical behavior of dla may be strongly affected by any bias inadvertently introduced by the algorithm .
the original algorithm suffers from two potential sources of systematic error : 1 .
the launching circle only passes through a few lattice points .
when launching a new particle , we must snap its coordinates to the grid , introducing roundoff error . for a particle accreting onto a cluster of linear size @xmath28
one may worry that the errors may be as large as @xmath29 .
2 . for a 2d random walk ,
even if a particle has wandered outside the killing circle , there is a @xmath30 probability that it will eventually re - enter the launching circle .
the particle is more likely to enter at the near side of the circle than at the far side . by removing the particle from the killing circle and re - launching it from a uniform distribution on the launching circle ,
the algorithm introduces a bias that may affect results such as @xmath11 .
even if the killing radius is @xmath28 times the cluster radius , the errors in the return probabilities may still be as large as @xmath29 .
* snap - to - grid error due to launching circle : * let us first address the first source of systematic error .
suppose we launch a particle on a launching circle of radius @xmath18 and snap its coordinates to the grid as described earlier . the probability distribution of the point @xmath31 is @xmath32 suppose we generate many particles from this distribution and let them diffuse via brownian random walks .
what is the steady - state concentration of particles within the launching circle ? how far does it deviate from a uniform distribution ?
this brownian problem maps to an electrostatics problem .
the source distribution maps to a charge distribution @xmath33 , and the steady - state distribution of particles maps to the electric potential @xmath34 where @xmath35 is the green function of the square lattice poisson equation such that @xmath36 because of the long - range logarithmic divergence in 2d , @xmath35 contains an infinite additive constant .
thus we define the regularized green function @xmath37 .
the quantities @xmath38 are related to the resistances between two points on a square lattice of resistors,@xcite and as described in appendix [ secgreenfunction ] , it can be calculated to machine precision for any @xmath39 and @xmath40 .
compare the potential with that at a reference point @xmath41 , which might as well be the origin @xmath42 : @xmath43 we calculate the charges @xmath44 numerically according to eq . , and we perform a fast 2d convolution with @xmath38 to obtain @xmath45 .
figures [ circlecharges ] , [ circlepotentials ] , and [ circlepotentialsalongpath ] show the charges and potentials for a launching circle of radius 20 .
the ring of charge is distorted by snapping to the grid , leading to potential fluctuations on the scale of @xmath46 . *
eliminating bias using a fuzzy launching annulus : * it is useful to take some insights from computer graphics .
early raster displays rendered oblique lines and circles with jagged edges .
modern displays eliminate this problem using antialiasing.@xcite in the context of dla , one might hope that launching bias might be reduced by `` antialiasing '' the launching circle .
the launching bias comes from the high - order fourier components in @xmath47 that can not be represented on the grid .
perhaps if we thicken the launching circle into an annulus and smear out its inner and outer boundaries , the resulting probability distribution will be smoother , and quantization error will be reduced .
we will show that this is indeed true .
suppose we pick a radius from the probability distribution @xmath48 where @xmath49 where @xmath50 is the bessel @xmath51 function .
this distribution corresponds to to a normalized kaiser - bessel window function@xcite on the interval @xmath52 $ ] .
we choose the kaiser - bessel window because it has very small spectral leakage beyond the central lobe , and because the distribution is easy to evaluate and sample compared to the optimal dolph - chebyshev window.@xcite we also pick an angle @xmath19 from the uniform distribution on @xmath20 , and set @xmath21 and @xmath22 .
the probability distribution of the point @xmath31 is then @xmath53 we calculate @xmath54 and @xmath55 as before .
figures [ annuluscharges ] , [ annuluspotentials ] , and [ annuluspotentialsalongpath ] show the charges and potentials for a launching annulus with blurred edges ( @xmath56 ) .
we see that the potential in the interior of the annulus ( @xmath57 ) is uniform to within @xmath58 . by increasing the inner radius and the thickness of the annulus , and by adjusting the parameter @xmath8
, the potential can be made even more uniform . for an annulus with @xmath59 , @xmath60 , and @xmath61
, we find that @xmath62 for all @xmath57 . for practical purposes
this means that the launching bias has been eliminated . in our dla simulations
we use a launching annulus with @xmath63 , @xmath64 , and @xmath61 , where @xmath25 is the bounding radius of the cluster .
since the cluster is well within the interior of the annulus , launching bias is negligible .
we sample @xmath39 from @xmath65 using rejection sampling@xcite with a gaussian envelope function .
( in rejection sampling one must keep retrying until a move is accepted , unlike in the metropolis algorithm .
a better term would be `` retrial sampling '' . )
the second source of systematic error is the killing - and - relaunching .
we will eliminate this error by using an enormous killing circle of radius @xmath66 .
what is the catch in using large launching and killing circles ? recall that a particle executing brownian motion has a r.m.s .
displacement that grows very slowly with time , @xmath67 .
if a particle is launched @xmath68 sites away from the cluster , it will take at least @xmath69 timesteps before the particle has an appreciable probability of encountering the cluster !
thus , in order to make a dla simulation feasible , we must find a way to accelerate the diffusion process that is , we must `` fast forward '' through the random walk .
we do this using `` first passage theory . ''
* electrostatic analogue of the first - passage problem : * suppose a particle starts at @xmath70 and executes a random walk until it encounters a `` marked '' site @xmath71 where @xmath72 .
we wish to find the probability distribution @xmath73 of the final position of the particle .
this is known as the first - passage position , that is , the position at which an infinite random walk first passes through a marked site .
this brownian problem maps to the following electrostatics problem.@xcite suppose a charge is placed at @xmath70 on a square lattice , and that the sites @xmath71 for @xmath72 are held at zero potential .
solve the discrete poisson equation : @xmath74 the charges @xmath75 give the desired first - passage probabilities .
in other words , we are to find the charge distribution induced on a grounded conducting object by an external point charge . rather than solving the discrete differential equation for the potential @xmath45 in all space , it is better to solve the discrete integral equation for the charges @xmath76 on the surface of the conductor .
this takes the form of @xmath77 simultaneous equations in @xmath77 variables .
formally , we have @xmath78 where @xmath79 and @xmath35 is the green function of the square lattice poisson equation such that @xmath36 the long - range logarithmic divergence in 2d poses two additional complications .
first , @xmath35 contains an infinite constant .
we regularize this by defining the resistance green function @xmath37 .
as described in appendix [ secgreenfunction ] , @xmath38 can be calculated to machine precision for any @xmath39 and @xmath40 .
take eq . and subtract the potential at a reference point @xmath41 , which might as well be @xmath80 .
this gives @xmath81 second , in order for the potential to be well defined , the total charge in the entire system must be zero : @xmath82 although eq . and contain @xmath83 equations , they form a linear system of rank @xmath77 , and so there is a unique solution for the @xmath77 charges , @xmath84 . *
walk - to - line algorithm : * we now apply this formalism to the situation shown in fig .
[ walktoline ] . we will call this the walk - to - line ( wtl ) problem on a square lattice .
it is analogous to the walk - to - plane algorithm@xcite for 3d continuum brownian diffusion .
suppose a charge is located at @xmath85 and every site on the @xmath39-axis @xmath86 is a conducting site held at zero potential .
the potential everywhere in the upper half - plane is given by the method of images as @xmath87 .
the electric field along the bond @xmath88@xmath86 is @xmath89 . mapping this back to the brownian problem
, we see that if a particle is launched at @xmath85 , the probability distribution of first passage to the @xmath39-axis is @xmath90 there are an infinite number of probabilities corresponding to all integer @xmath39 .
we sample this distribution using rejection sampling.@xcite in order to use rejection sampling , we need an envelope function that resembles the target distribution ( the probability distribution that we wish to sample from ) . to obtain a suitable envelope function , consider the continuum version of the walk - to - line problem .
suppose a particle starts at @xmath85 and executes continuum brownian motion until it encounters the line @xmath91 .
the electrostatic analog is a unit charge at @xmath85 and a grounded conducting plane at @xmath91 .
solve poisson s equation with dirichlet boundary conditions using the method of images .
the potential is @xmath92 , and so the charge density on the line is @xmath93 .
thus the first - passage probability distribution is @xmath94 .
the cumulative distribution function ( cdf ) is @xmath95 .
therefore , we can generate a sample from @xmath96 using the inverse cdf method as @xmath97 , where @xmath98 is a random number from the uniform distribution on @xmath99 .
now , since the target distribution is discrete , consider generating a sample using @xmath100 this corresponds to the discrete envelope distribution @xmath101 the rejection sampling algorithm is as follows :
1 . find a number @xmath102 such that @xmath103 for all @xmath39 .
for our purposes we can choose @xmath104 .
2 . draw a random integer @xmath39 from the envelope distribution @xmath105 ( gray symbols in fig .
[ retrialsampling ] ) .
3 . generate a uniform random number @xmath106 .
if @xmath107 , return @xmath39 . otherwise , go back to step 2 .
the returned value of @xmath39 is a sample from the target distribution @xmath108 .
* application to square lattice dla : * suppose the diffusing particle is just outside the bounding rectangle of the cluster ( see fig . [ walktoboundingbox ] ) .
apply the walk - to - line algorithm to return the particle to an infinite horizontal or vertical line parallel to the bounding rectangle.@xcite a few iterations of this procedure usually suffice to return the particle to the bounding rectangle .
now suppose the particle is a large distance @xmath109 away from the cluster .
then , eq . implies that the walk - to - line algorithm is roughly equivalent to multiplying @xmath109 by a random number @xmath110 drawn from a lorentzian distribution : @xmath111 thus , the logarithm of the radius is incremented by a random number @xmath112 drawn from a sech distribution : @xmath113 the variance of the sech distribution is @xmath114 therefore , @xmath115 executes a random walk with a step variance of @xmath116 .
after @xmath117 iterations of the walk - to - line algorithm , the accumulated variance is @xmath118 , and one might expect @xmath109 to have increased or decreased by @xmath119 .
if a walker somehow finds itself at a distance @xmath68 from a cluster of linear size @xmath120 , after 54 iterations of walk - to - line , there is an appreciable probability that it will either have returned to the cluster ( of radius @xmath120 ) or that it will have drifted outside the killing circle ( of radius @xmath121 ) . in summary ,
if an errant particle finds itself a distance @xmath68 away from the cluster , the original witten - sander algorithm would take about @xmath69 steps to return it to the cluster , whereas the walk - to - line algorithm would take about @xmath122 iterations .
although this is not perfect , it is certainly a great improvement .
now let us consider another brownian motion problem .
suppose a random walker begins at point @xmath70 within a square with corners @xmath42 and @xmath123 , as in fig .
[ walkouttosquare ] . what is the probability that the walker makes first passage through the square at position @xmath31 ?
the corresponding electrostatic situation is a point charge at the center of a grounded conducting square .
we wish to solve the discrete poisson equation with dirichlet boundary conditions on a square : @xmath124 this is a linear system involving a @xmath125 sparse matrix with integer coefficients .
brute force matrix algebra gives the solutions @xmath45 as rational numbers .
for example , for @xmath126 the solutions are @xmath127 a better approach is to separate variables in cartesians and superpose eigenfunctions to obtain the green function .
expand the charge distribution @xmath128 and the potential @xmath45 in fourier sine series , and connect them via the discrete poisson equation : @xmath129 we implement eq . using the fast 2d discrete sine transform ( dst ) .
this allows us to compute @xmath45 for all @xmath39 and @xmath40 in @xmath130 time .
this is faster than evaluating the double sums @xmath131 which takes @xmath132 time . for @xmath133
the dst method takes a long time and accumulates roundoff errors of the order of @xmath0 .
then it becomes preferable to calculate @xmath45 as a madelung sum involving an infinite series of positive and negative image charges , @xmath134 for speed and accuracy , split @xmath38 into a @xmath135 part plus a correction due lattice anisotropy ( eq . ) .
this gives @xmath45 as the solution to the continuum problem ( in terms of the jacobi cn function ) plus a lattice correction , which is best evaluated by grouping the charges into quadrupoles and truncating the sum appropriately .
having found @xmath45 , we can find the electric field and hence the charge distribution on the boundary , @xmath136 .
thus the first - passage probabilities @xmath137 are given simply by the first row of the @xmath45 matrix , such as that in eq . .
geometry for walk - out - to - square algorithm .
a particle starts at @xmath70 ( black disk ) and executes a random walk on the square lattice .
we wish to find the probability distribution of the first intersection of this walk with a surrounding square of side @xmath138 ( gray disks ) . ]
we have tabulated the first - passage probabilities @xmath139 from the center of a square of side @xmath138 to every point @xmath86 on the lower edge , for @xmath140 .
see fig .
[ walkouttosquareprobs ] .
these distributions can easily be sampled using precalculated walker alias tables.@xcite first - passage probabilities @xmath139 from the center of a square of side @xmath138 to every point @xmath86 on the lower edge , for @xmath141 .
horizontal and vertical axes are scaled to show that @xmath142 approaches a universal function of @xmath143 as @xmath144 .
the universal function is the charge distribution inside a grounded conducting square induced by a charge at its center . ] * application to square lattice dla : * we store the dla cluster as an array of 1 s and 0 s of dimensions @xmath145 , say .
we also maintain a hierarchy of coarse - grained representations of dimensions @xmath146 , representing @xmath147 blocks , for block sizes @xmath148 .
every time a particle is added to the cluster , we mark the corresponding block at each level of the hierarchy as occupied .
the total amount of memory required is @xmath149 .
thus , all the coarse arrays together require only @xmath150 extra memory .
suppose the diffusing particle is within the bounding box of the cluster , so that the walk - to - line algorithm is inapplicable .
we can revert to moving the particle one step at a time .
however , if the particle is in a cavity of radius @xmath151 , it will take of the order of @xmath152 steps for the particle to find its way to an occupied site . from fig .
[ agg4-l65536-cluster ] we can see that a particle in the corner of the bounding box would spend @xmath153 timesteps wandering around in a cavity of linear size @xmath154 .
fortunately , the walk - to - square algorithm allows us to speed up the diffusion process .
we start at the coarsest level of the hierarchy ( @xmath155 blocks ) and examine the @xmath156 array of blocks around the diffusing particle s block .
if any one of these nine blocks is occupied , we proceed to the next finer level of the hierarchy .
we repeat this until we get to a level where all nine blocks are empty .
this means that we can move the particle by @xmath157 units in the @xmath39 direction and @xmath157 units in the @xmath40 direction without contacting the cluster .
therefore we can apply the walk - to - square algorithm for a square of side @xmath158 centered on the diffusing particle .
many authors have used hierarchical representations and variable stepsizes.@xcite meakin et al.@xcite used a combination of off - lattice jumps and on - lattice steps .
al.@xcite used a lookup table calculated using laplace s equation to compute the first passage to the square in a manner accurate to better than @xmath15 .
however , to the best of our knowledge , our walk - to - square algorithm , which is based on exact green functions , is the first _ unbiased _ variable - stepsize algorithm for _ lattice _ dla .
having described all the ingredients , we now give a summary of our dla simulation algorithm , omitting optimization details : 1 .
set up a seed cluster .
2 . launch a new walker on a fuzzy annulus of inner radius @xmath159 and outer radius @xmath160 , where @xmath25 is the bounding radius of the cluster ( see fig .
[ annuluscharges ] ) .
3 . if the walker lies outside the bounding rectangle of the cluster , use the walk - to - line algorithm to move the walker to an infinite line on the near side of the bounding box ( see fig .
[ walktoline ] ) .
if this takes the walker outside the killing circle of radius @xmath161 , discard the walker and go back to step 2 .
+ + if instead the walker lies inside the bounding rectangle , start at the coarsest level of the hierarchy , and proceed to finer and finer levels until one finds a scale at which all eight neighboring blocks contain no sticky sites .
then , apply the walk - out - to - square algorithm to move the walker to an edge of a square contained within the empty region ( see fig .
[ walkouttosquare ] ) .
4 . if the walker is now at a sticky site , freeze it ( i.e. , add it to the cluster and mark its neighbors as new sticky sites ) and go to step 2 . otherwise , go to step 3 . in this algorithm
all probability distributions are sampled with errors less than @xmath0 .
thus systematic error is practically eliminated , leaving statistical error and finite cluster size as the only sources of error .
we have implemented this algorithm in c++ ( see supplementary material ) .
we use 64-bit floating - point arithmetic , and we represent @xmath31 coordinates by 64-bit integers to allow the killing circle radius to be @xmath161 . we assume nearest - neighbor diffusion , such that the diffusing particle only moves horizontally or vertically ( fig .
[ rulefordiffusion ] ) .
we try various aggregation rules ( figs .
[ agg2 ] , [ agg3 ] , [ agg4 ] , and [ agg8 ] ) ; square lattice dla studies in the literature typically use the 4-neighbor rule , where sticky sites are horizontally or vertically adjacent to a cluster site . at each level of the hierarchy we store the pattern of frozen / sticky sites using a bit array . for simplicity we do not distinguish frozen sites from sticky sites . in this work
we have used a built - in system random number generator , which is a non - linear additive feedback generator with a period of approximately @xmath162 .
we have not tested the effect of different random number generators .
* cluster shape : * figures [ agg2-l65536-cluster ] , [ agg3-l65536-cluster ] , [ agg4-l65536-cluster ] , and [ agg8-l65536-cluster ] show square lattice dla clusters grown using different aggregation rules . the 2-neighbor and
3-neighbor aggregation rules clearly manifest themselves by producing @xmath163-shaped and @xmath164-shaped clusters .
the 4-neighbor aggregation rule produces faster growth along horizontal and vertical directions , whereas the 8-neighbor aggregation rule produces faster growth along diagonal directions ; both these rules lead to a @xmath165 asymmetry in the angular mass distribution .
we have tried starting with seed clusters of various shapes ( @xmath166 , @xmath167 , @xmath168 , @xmath169 , @xmath170 ) .
regardless of the shape of the seed cluster , the growing cluster evolves toward a shape determined by the aggregation rule .
thus the asymptotic shape of the cluster is governed by the aggregation rule ( the way in which particles stick together),@xcite and not by the diffusion rule , nor by the seed cluster shape .
various authors have reported that a `` square deposition habit '' leads to @xmath171 and @xmath167 cluster shapes,@xcite whereas a `` diagonal deposition habit '' leads to @xmath168 shapes.@xcite for off - lattice dla , it was found that the ratio of the principal radii of gyration tends to unity for large clusters@xcite i.e. , if a cluster happens to start off with an elliptical shape , it evolves towards a circular shape .
our results agree with these statements .
* fractal dimension : * figure [ rvsm ] shows the radius of gyration @xmath5 as a function of the cluster mass @xmath6 , during the growth of a single cluster , for various aggregation rules . the data are roughly consistent with a power law @xmath172 somewhere between @xmath16 , which is the exponent for 2d continuum dla,@xcite and @xmath17 , which is kesten s upper bound for 2d dla .
figure [ dlnrdlnm ] shows @xmath173 estimated from ratios between successive @xmath174 data points . for 2- and 3-neighbor aggregation rules
, @xmath175 is close to @xmath14 for large @xmath6 . for 4- and 8-neighbor
aggregation rules @xmath175 appears to be close to @xmath176 for moderate @xmath6 , but for very large @xmath6 it appears possible that @xmath175 is increasing towards @xmath14 .
meakin et al.@xcite reported that @xmath8 evolves from @xmath176 towards @xmath177 during the growth of the cluster for square lattice dla .
menshutin et al.@xcite also reported that @xmath178 for a variant of dla in which particles diffuse via continuum 2d brownian motion but aggregate onto lattice sites using an `` antenna '' rule .
our results are consistent with these statements .
we present an improved algorithm for 2d lattice dla that reduces systematic errors in probabilistic sampling below @xmath0 . we build clusters of @xmath1 particles on lattices of size @xmath2 .
we verify that the anisotropy of the aggregation process leads to anisotropy of the cluster shape , so that the radius - of - gyration exponent evolves from @xmath16 towards @xmath17 .
our unbiased dla algorithm can be generalized to triangular lattices , cubic lattices , and other lattices .
there are analytic expressions for triangular lattice green functions,@xcite and green functions on 3d lattices can be calculated numerically.@xcite we are grateful to william schwalm for helpful discussions .
in this appendix we consider the green function @xmath179 and the regularized green function @xmath180 * recursion relations : * by symmetry it can be shown that @xmath181 and @xmath182 . using complex variable techniques it can be shown that@xcite @xmath183 where @xmath184 are the harmonic numbers .
@xmath38 also satisfies the discrete poisson equation @xmath185 in principle , eqs . and allow one to compute @xmath38 for all @xmath39 and @xmath40 .
however , this procedure is unstable to roundoff error if implemented numerically .
thus , we implement the recursion relations using symbolic algebra , and use extra - precision arithmetic to convert the results to floating - point numbers .
the first few @xmath38 are shown in table [ tabresistancegreenfunction].@xcite @xmath186 * series approximation at large distances : * for large values of @xmath39 and @xmath40 , the behavior of eq .
is dominated by small @xmath187 and @xmath188 .
expand the denominator in powers of @xmath189 and @xmath190 : @xmath191 expand the reciprocal in powers of @xmath192 : @xmath193 take eq . and extend the domain of integration to the entire @xmath194 plane .
let @xmath195 and @xmath196
. then @xmath197 \end{aligned}\ ] ] now let us derive an identity for the fourier transform of a 2d power law function , @xmath198 this gives @xmath199 the constant @xmath200 is infinite , but the other terms are finite . thus the regularized green function has the form @xmath201 where @xmath202 is a finite constant . by matching this to the inverse power series for the harmonic numbers , eq .
, it can be shown that @xmath203 where @xmath204 is the euler - mascheroni constant.@xcite for @xmath205 , the @xmath206 terms in eq .
have absolute value smaller than @xmath207 .
therefore , truncating the series at the @xmath208 term allows us to evaluate @xmath38 to machine precision for all @xmath209 . in our dla simulations we obtain @xmath38 by table lookup for @xmath210 and @xmath211 and using the series otherwise
in this appendix we discuss other approaches to bias - free lattice dla , which are less efficient than the algorithm we have presented .
* solution of laplace s equation for an arbitrary cluster : * for a particle launched at infinity , the first - passage probabilities to the sticky sites ( sites adjacent to cluster sites ) can be computed exactly by solving laplace s equation .
one can then add a particle at a position picked directly according to these probabilities .
however , for a cluster of @xmath77 sites , solving laplace s equation is a dense linear algebra problem taking @xmath212 time .
even if the sherman - morrison formula@xcite is used to update the inverse matrix incrementally , the problem still takes @xmath213 time for every particle that is added to the cluster .
this is prohibitively slow . * walk - in - to - circle methods : * for 2d or 3d continuum dla , it is easy to return a particle to the bounding circle or sphere of the cluster .
the electrostatic problem is easily solved by the method of images , and the return probability distribution can be evaluated and sampled analytically .
this is exploited in a killing - free algorithm for continuum dla@xcite where particles that escape from the launching circle are immediately returned to the launching circle . for lattice dla ,
however , circular or spherical boundaries do not fit naturally on the lattice . snapping to
the grid leads to large errors , as we have shown . it may be possible to reduce these errors by returning the particle to a fuzzy annulus
; we have not investigated this completely .
suppose a particle starts at @xmath70 ( black disk ) and executes a random walk on the square lattice until it reaches a point @xmath31 on a square of side @xmath138 ( gray disks ) .
the first - passage probabilities can be found using brute - force matrix computation , which takes @xmath214 time , but there is no simple analytic formula for general @xmath138 . ] * walk - in - to - square methods : * what if we wish to return a particle to a bounding rectangle , square , or cube ?
this requires finding the charge distribution on a conductor induced by an _ exterior _ point charge . whereas interior electrostatics problems are amenable to a variety of methods ( images , separation of variables , and conformal mapping ) , exterior electrostatics problems are notoriously difficult even in the continuum .
the most accurate results for the capacitance of a cube have been obtained by mapping the electrostatic problem back to a random walk problem and using monte carlo techniques!@xcite it is probably futile to search for a simple analytic formula for the lattice problem ( see fig .
[ walkintosquare ] ) .
the walk - in - to - square problem can be solved numerically , but this takes @xmath214 time , where @xmath138 is the perimeter of the boundary . in comparison , the iterated walk - to - line method may take several hundred iterations to return the walker to the bounding box of the cluster , but this number of iterations is independent of cluster size .
thus we prefer the walk - to - line method . | we identify sources of systematic error in traditional simulations of the witten - sander model of diffusion - limited aggregation ( dla ) on a square lattice .
we present an algorithm that reduces these biases to below @xmath0 .
we grow clusters of @xmath1 particles on @xmath2 lattices .
we verify that lattice dla clusters inevitably grow into anisotropic shapes , dictated by the anisotropy of the aggregation process .
we verify that the fractal dimension evolves from the continuum dla value , @xmath3 , for small disk - shaped clusters , towards kesten s bound of @xmath4 for highly anisotropic clusters with long protruding arms .
diffusion - limited aggregation ( dla ) is one of the most important models in nonequilibrium statistical physics , exhibiting self - organized criticality@xcite and complex pattern formation . in the dla process ,
one begins with a cluster ( `` seed crystal '' ) immersed in a very dilute solution of particles ( `` molecules '' ) .
each particle wanders around according to brownian motion until it encounters the cluster , at which point it `` freezes '' and becomes part of the cluster .
it is more likely that a diffusing particle will stick to a protrusion on the cluster than to a depression .
thus dla has a natural instability resulting in pattern formation ; protrusions grow quickly and spawn other protrusions , forming a treelike pattern somewhat like frost on a window pane .
dla and its variants have been used to model a whole host of nonequilibrium phenomena , including viscous fingering ( pattern formation when one fluid is injected into a viscous fluid ) , electrodeposition,@xcite dielectric breakdown , and surface poisoning in ion - beam microscopy.@xcite during the course of a dla simulation , the radius of gyration of the cluster , @xmath5 , and the cluster mass , @xmath6 , are usually recorded .
these data can usually be fit to a power law @xmath7 , where @xmath8 is known as the radius - of - gyration exponent .
then one has @xmath9 where @xmath10 is the fractal dimension of the dla cluster @xmath11 ( insofar as the fractal dimension can be defined for an inhomogeneous finite object ) . in two dimensions ( 2d ) ,
@xmath12 ; this is one of the few rigorous results@xcite on dla .
other results have been obtained using mean - field theories@xcite and iterated conformal maps,@xcite but these are uncontrolled approximations without small parameters.@xcite the bulk of the literature involves numerical simulations .
early papers@xcite reported similar values ( @xmath13 ) for 2d continuum , square lattice , and triangular lattice dla clusters .
later papers@xcite claimed that square lattice dla clusters evolve from a roughly circular shape for small clusters towards diamond and cross shapes for larger clusters , and that @xmath8 approaches @xmath14 for very large clusters .
some numerical data suggested that dla exhibits multiscaling,@xcite but later work suggested that multiscaling is a finite - size effect that is not intrinsic to dla.@xcite in monte carlo simulations of equilibrium systems , such as ising models , it is essential to construct a markov chain with the correct invariant distribution ( e.g. , by ensuring detailed balance ) .
any bias in the simulation results in sampling the wrong probability distribution , giving wrong answers for thermodynamic quantities and critical exponents .
non - equilibrium situations such as dla deserve the same amount of care .
however , dla studies to date have used approximations with errors potentially as large as @xmath15 .
the results were justified by noting that varying the severity of the approximation did not noticeably affect the results ; nevertheless one may still be concerned that the approximations led to subtle effects that went unnoticed .
in this paper we present an algorithm for square lattice dla where probability distributions are sampled with accuracies better than @xmath0 .
we verify that dla on a lattice produces anisotropic clusters , and that the anisotropy originates from the aggregation process rather than the diffusion process .
we confirm that small circular clusters have a radius - of - gyration exponent @xmath16 , but as they mature into anisotropic shapes with arms extending outwards , the exponent tends towards @xmath17 . |
in the hierarchical paradigm of large scale structure ( lss ) formation , galaxy clusters are the largest object ever to virialize . with masses in excess of @xmath3
, they are located at the nodes of the cosmic web , where they accrete matter from the surrounding voids and through large scale filaments . due to their vast size ,
galaxy clusters resemble island universes seen at great distance , providing a powerful cosmological probe and a unique astrophysical laboratory .
the gas accreted by a cluster is thought to abruptly heat and slow down in a strong virial shock wave surrounding the cluster .
such collisionless shocks are thought , by analogy with supernova remnant shocks , to accelerate charged particles to @xmath4 energies , where they compton - scatter cosmic - microwave background ( cmb ) photons up to the @xmath0-ray band@xcite .
consequently , one expects to find @xmath0-ray rings around clusters@xcite , as indicated by cosmological simulations@xcite , which suggest an elliptic morphology elongated towards the large - scale filaments feeding the cluster@xcite .
such rings are also expected in hard x - rays@xcite , and should coincide with a synchrotron radio ring@xcite and with a cutoff in the thermal sunyaev - zeldovich ( sz ) signal@xcite .
however , no such virial shock signature was detected so far , although a stacking analysis of egret data around a sample of 447 rich clusters did show a @xmath5 signal@xcite .
upper limit were imposed on the @xmath0-ray emission from clusters such as coma@xcite , but mostly focusing on the central parts of the cluster , well within the virial radius .
the very existence of cluster - scale virial shocks has thus remained unconfirmed .
the coma cluster ( abell 1656 ) is one of the richest nearby clusters . with mass @xmath6 , temperature @xmath7 , and richness class 2 ,
it lies only @xmath8 away@xcite , at a redshift @xmath9 .
its virial radius , @xmath10 , corresponds to an angular radius @xmath11 , where subscript @xmath12 refers to an enclosed density @xmath12 times above the critical value .
the cluster is somewhat elongated in the east - west direction , in coincidence with the western lss filament@xcite that connects it with the cluster abell 1367 ( see figure [ fig : sdss ] ) .
there is x - ray , optical , weak lensing , radio@xcite , and sz@xcite evidence that the cluster is accreting clumpy matter and experiencing weak shocks towards the filament well within the virial radius , at @xmath13 radii .
the veritas erenkov array has produced a @xmath14 diameter @xmath0-ray mosaic@xcite of coma , at energies @xmath15 .
we argue that the significance map ( figure [ fig : comaver ] ) shows extended @xmath0-ray emission away from the center , that appears as a thick ( @xmath16 ) elliptical ring with semi - minor axis @xmath17 , elongated along the east west direction , with semi - major to semi - minor axes ratio @xmath18 .
the ring is seen at a @xmath19 confidence level ( effectively a single - parameter fit , estimated using mock veritas noise maps ; see figures [ fig : comaverbin ] and [ fig : comavertilt ] and [ sec : veritas ] ) .
however , the data was taken in @xmath20 wobble mode@xcite , inappropriate for extended emission on larger scales ; correcting for the excessive background removal indicates a @xmath21 signal .
x - ray and galaxy maps suggest that the lss filament and the semi - major axis of the thermal gas distribution are slightly tilted southwest , at an angle @xmath22 .
indeed , the significance of the @xmath0-ray structure is maximized when thus tilted ; the best fit is obtained for @xmath23 and @xmath24 . if no assumption is made regarding @xmath25 , the ring significance weakens to @xmath26 ( @xmath27 corrected ) .
the size and morphology of the veritas ring resemble the predicted @xmath0-ray signature of a coma - like , @xmath28 and @xmath7 cluster in a @xmath29cdm simulation@xcite . the simulated map ( figure [ fig : comasim ] ) , oriented such that the major simulated lss filament lies to the west and normalized for a nominal @xmath30 wobble - mode veritas detection ( no remaining free parameters )
, correlates with the veritas map at the @xmath31 ( @xmath32 corrected ) confidence level .
the ring brightness agrees with predictions too , if the relativistic electron fraction @xmath33 of the thermal energy accretion rate @xmath34 ( normalized by the total thermal energy over a hubble time ) is a few percent .
the associated @xmath35 flux at @xmath36 energies can explain the hard x - ray signal observed in coma@xcite .
similar , @xmath37 values are inferred from the @xmath5 egret stacking analysis@xcite , and from lss hard x - ray sources@xcite . in order to examine the coincident radio signals , we spectrally decompose the wmap seven - year , full - sky maps@xcite into synchrotron and thermal sz maps . excluding the inner , @xmath38 part of the cluster , the veritas data correlates with the synchrotron map at a @xmath39 ( @xmath40 ) confidence level ( figure [ fig : compositeversyn ] ) , and anti - correlates with the sz signal at @xmath41 ( @xmath42 ) .
the radio signals are stronger in the western half of the ring , towards the lss filament , where the correlations strengthen up to @xmath43 ( @xmath44 ) with the synchrotron and @xmath45 ( @xmath46 ) with the sz .
these peripheral radio signals agree with predictions for the synchrotron emission from the @xmath0-ray emitting electrons , and for the sz decline with thermal pressure beyond the shock , provided that @xmath47 magnetic fields are found downstream , corresponding to @xmath48 magnetization .
bright spots along the ring suggest local enhancements , up to @xmath49 a few times its mean value . while the sz signal , extending in some parts out to @xmath50 , agrees with sz detections by wmap@xcite and planck@xcite , the peripheral synchrotron signal coincident with the @xmath0-rays and with the sz cutoff was not reported so far .
note that the @xmath0-ray ring and these radio features are much farther out than the @xmath51 weak shocks previously reported based on x - ray , radio@xcite , and sz@xcite data .
inside the @xmath52 ellipse , the veritas and synchrotron maps show a @xmath53 ( @xmath54 ) anti - correlation , mainly ( @xmath55 ( @xmath56 ) ) in the west , evident from the @xmath0-ray underluminous radio halo and relic ( figure [ fig : compositeversyn ] ) .
such a signature , predicted in secondary models@xcite where cosmic - ray ion collisions inject relativistic electrons in both strongly ( @xmath57 ) and weakly magnetized regions , which consequently appear radio bright/@xmath0-ray faint and vice versa , is unnatural in primary models where electrons are ( re)accelerated in turbulence@xcite or weak shocks .
secondary electrons can explain radio halos@xcite , relics@xcite and minihalos@xcite as arising from the same cosmic - ray distribution , if the latter is nearly homogeneous@xcite .
the extent and magnitude of the anti - correlation observed , if confirmed , support such secondary models , and favor an extended cosmic - ray distribution .
a future veritas observation deeper than the present @xmath58 hour exposure would decisively test our results , in particular if wobble - mode corrections are minimized . this would allow a more precise reconstruction of the virial shock and a measure of variations in the electron deposition rate along the shock front .
such data is valuable in the study of lss formation at low redshifts , mapping the feeding pattern of coma , probing the surrounding voids and filaments , and tracing the warm - hot intergalactic medium ( whim ) immediately behind the shock .
a detailed , more significant anti - correlation between @xmath0-ray and synchrotron maps at inner , @xmath59 radii would confirm the hadronic , secondary electron model for cluster radio sources , and provide a direct measure of magnetic variations in the icm .
a deeper @xmath0-ray observation would thus probe the cosmic - ray proton distribution out to large radii , address the origin of these protons , and place an upper limit on proton acceleration in the virial shock .
the veritas energy threshold , @xmath60 , is not far below the expected photon cutoff , @xmath61 .
observations at higher energies would become increasingly sensitive to this cutoff , thus probing the conditions at the shock .
interestingly , no evidence for extended emission was found in a @xmath62 observation of coma by the hess telescopes@xcite , with @xmath63 the sensitivity in the veritas mosaic , but the @xmath0-ray ring may be shallowly buried in the hess noise . calibrating analytic@xcite , and numerical@xcite estimates with the above @xmath64 and @xmath65 values , yields @xmath66 inverse - compton and @xmath67 synchrotron cumulative backgrounds from cluster shocks . these constitute fair fractions of the @xmath0-ray and low - frequency radio signals observed far from the galactic plane , and dominate the respective diffuse extragalactic backgrounds@xcite .
the paper is organized as follows . in [ sec : veritas ] we show that an elongated @xmath0-ray ring around coma was observed by the veritas erenkov array . methods to correct for excessive wobble - mode background subtraction are discussed in general in [ sec : wobble ] , and specifically for cross - correlation analyses in [ sec : crosscorrelations ] .
contamination by galactic foregrounds is shown to be negligible in [ sec : galactic ] .
the @xmath0-ray ring is shown to agree with simulation - based predictions in [ sec : simulation ] .
significant correlations with both synchrotron and sz maps are presented in [ sec : radio ] . in [ sec : inversecompton ] we show that the @xmath0-ray signal agrees with predictions , if the shock deposits a few percent of the thermal energy per hubble time in relativistic electrons . in [ sec : radiotheory ] we show that the synchrotron and sz signals also agree with the predictions , if magnetic fields carry @xmath48 of the thermal energy .
a @xmath0-ray radio anti - correlation inward of the @xmath0-ray ring is interpreted in [ sec : inneremission ] as possible preliminary evidence for strong local magnetization and secondary electron injection at small radii .
the results are summarized and discussed in [ sec : discussion ] .
a simple @xmath68-model of the shock is provided in appendix [ sec : betamodel ] .
we adopt a concordance flat @xmath29cdm model with hubble constant @xmath69 , a baryon fraction @xmath70 , and a hydrogen mass fraction @xmath71 .
the plasma is approximated as an ideal gas with adiabatic index @xmath72 and mean particle mass @xmath73 , where @xmath74 is the proton mass .
colorbars are square - weighted .
veritas has produced a @xmath14 diameter @xmath0-ray mosaic@xcite of coma , at energies @xmath75 .
the significance map ( figure [ fig : comaver ] ) , taken in @xmath20 wobble mode@xcite with a @xmath76 integration radius , shows some extended @xmath0-ray emission away from the center .
the @xmath0-ray structure appears as an elliptical ring , elongated along the east - west direction , or as two parallel filaments lying symmetrically both north and south of the cluster .
the spectrum in the veritas band is probably flat , as a @xmath77 photon spectral index was used to optimize the gamma - hadron separation cuts@xcite . in [ sec : inversecompton ] we show that a flat spectrum with index @xmath78 ( equal energy per logarithmic energy interval ) is consistent with observations at lower energies .
denote the veritas significance map by @xmath79 .
the data is taken in wobble mode@xcite , where a pixel of significance @xmath79 means that in a beam of radius @xmath76 centered upon the pixel , there are @xmath80 counts in excess of the anticipated @xmath81 background events , where @xmath82 and the background is estimated based on a @xmath83 ring around the pixel . a @xmath84 analysis@xcite shows a background of about @xmath85 photon counts above @xmath86 in such a @xmath76 bin .
the background per pixel is denoted by @xmath87 , where @xmath88 is the number of pixels in the beam .
first we assume that the background is uniform , so the pixels are not correlated before the beam integration . the nominal significance of an extended source
may then be estimated as @xmath89 where @xmath90 and @xmath91 are respectively the total significance and the number of pixels enclosed . in order to quantify the @xmath0-ray structure and assess its significance
, we fit the data with a thick , elliptical ring model .
the center of the ellipse is chosen as the rosat x - ray peak . the median semi - major axis @xmath92 is taken along the east - west direction , as inferred from the rosat map and from the orientation of the sdss galaxy filament .
the thickness of the ring is fixed at @xmath93 along the semi - minor axis , because a thicker ring would exceed the @xmath94 limit imposed by the wobble - mode background subtraction bins ( figure [ fig : comaver ] ) , so @xmath95 exceeds @xmath96 at oblique angles . ] .
the ring is then defined by its median semi - minor axis @xmath97 and elongation ratio @xmath98 , taken as two free parameters . however , for sufficient elongation , @xmath99 , the results ( figures [ fig : comaverbin ] and [ fig : comavertilt ] ) depend weakly upon @xmath100 .
the ( effectively single - parameter ) fit then indicates an extended structure around @xmath17 , found at an @xmath101 nominal significance level .
the sdss ( figure [ fig : sdss ] ) and rosat maps suggest that the lss filament and the semi - major axis of the gas distribution are slightly tilted towards the southwest , at an angle @xmath22 .
indeed , the significance of the @xmath0-ray structure is maximized when tilted at precisely such angles ( figure [ fig : comavertilt ] ) .
the best fit is obtained for @xmath23 and @xmath24 . a more reliable way to estimate
the significance of the extended emission is to apply the same elliptical templates to a large sample ( @xmath102 ) of mock significance maps , containing random noise but no signal .
we prepare these maps by injecting poisson noise at the observed background level to the veritas mosaic template , integrating over the same @xmath76 beam , and converting to a significance map using either the mean background or a simulated @xmath20 wobble - mode background .
none of the resulting mock maps shows a signal as strong as the observed @xmath103 , so we extrapolate to the high significance levels assuming a gaussian distribution .
the resulting significance levels are shown as dashed curves in figure [ fig : comaverbin ] , for elliptical rings elongated in the east - west direction ( based on 1000 maps at a given @xmath104 ) .
if we make no a - priori assumption regarding the ring orientation @xmath25 , but rather choose in each mock map the value @xmath25 that maximizes its nominal ring significance , we obtain higher mock detection levels , shown as dot - dashed contours in the figure ( based on 200 maps , each scanned with @xmath105 tilt intervals ) .
the results depend weakly ( within the statistical fluctuations ) on @xmath104 .
the mock statistics indicate that the veritas ring presents at the @xmath19 confidence level if we assume that it is aligned with the lss filament , and at @xmath26 confidence if we make no a - priori assumption regarding @xmath25 .
these conservative estimates are based on the mean background , and are approximately independent of @xmath100 ; higher , @xmath106 significance levels are obtained when wobble - mode is used to estimate the mock background . as we show in [ sec : wobble ] below , the detection significance for such extended structures is actually higher , @xmath21 for east - west alignment , and @xmath107 for arbitrary @xmath25 .
the signal is locally on the order of @xmath108 of the noise , and stronger than the average galactic foreground at these energies@xcite ( see [ sec : inversecompton ] ) . in the coma region ,
the expected galactic foreground is relatively low , spatially smooth , and featureless on @xmath109 scales ; see [ sec : galactic ] .
wobble - background subtraction is highly efficient in removing smooth foregrounds ; a disadvantage is that it also removes much of the signal for extended sources . in wobble mode ,
an estimated background is subtracted based on a ring of radius @xmath110 around each pixel .
this method , appropriate for point sources , distorts and smears out features extended over scales comparable or larger than @xmath110 . as we show below
, analyses of the @xmath0-ray features , of a histogram of the @xmath0-ray pixels , and of wobble simulations , indicate that the wobble subtraction reduces the significance of the extended features observed by at least @xmath111 .
therefore , the corrected significance of a @xmath0-ray ring aligned with the lss filament is estimated as @xmath21 .
consider the wobble - induced correlation between @xmath0-ray pixels . due to the wobble background subtraction
, the significance of a pixel is anti - correlated with the pixels @xmath83 around it .
one implication is that sources that extend over scales much larger than @xmath110 are effectively erased by the background subtraction , except in an @xmath112 wide band in their periphery .
in such a band , and in sources that extend @xmath113 in at least one direction , the signal is diminished but not completely erased . for example , the estimated signal at the inner edge of a large linear source will be diminished by a factor of two , and the signal at the outer edge will have an equal but negative significance .
therefore , for sources that extend over @xmath114 in at least one direction , the actual significance may be @xmath115 times higher than estimated from the wobble - mode map . in order to quantify
the effect wobble - mode has on the signal , we examine the distribution of @xmath0-ray significance pixels , using the same @xmath76 integration beam applied to the map .
the significance histogram ( figure [ fig : ver_hist ] ) , taken with @xmath116 intervals , can be modeled@xcite as a single gaussian function , with adjusted coefficient of determination @xmath117 .
however , the histogram shows peak deviations and sharp cutoffs , indicating a more complicated underlying distribution .
indeed , a simple model combining two gaussian functions provides a much better fit to the data ( @xmath118 ; note that @xmath119 improves slowly near @xmath120 ) .
the best fit is obtained for gaussian means @xmath121 and @xmath122 , with corresponding standard deviations @xmath123 and @xmath124 .
the data were also modelled as the sum of two gaussian functions constrained to have the same standard deviation ; the resulting best fit is similar to the unconstrained sum : @xmath125 and @xmath126 , with standard deviation @xmath127 ( @xmath118 ) .
both fits suggest that a signal is present in the map , with pixels that are on average @xmath128 above the noise .
interestingly , a histogram of the pixels found only in the central , @xmath129 wide region of the map agrees much better with a single gaussian@xcite .
this indicates that the additional signal is found at radii @xmath130 , in agreement with the inferred ring .
the above arguments indicate that the local pixel significance should in fact be @xmath131 higher than it appears in the significance map ( figure [ fig : comaver ] ) .
this imposes a lower limit @xmath132 to the correction factor needed in order to estimate the significance of an extended structure or of a cross - correlation signal ( see [ sec : crosscorrelations ] ) with respect to the true background .
we conservatively use @xmath111 for extended emission exceeding the beam size .
another indication that such a correction is needed are the negative , @xmath133 significance regions , lying just outside the bright feature we identify as a ring .
assuming that the bright region is part of an extended , linear structure , this suggests that its local significance should be @xmath134 higher than it appears .
another way to assess the effect of wobble background subtraction is to apply it to a simulated cluster map .
we thus normalize the @xmath0-ray map of the simulated cluster derived in [ sec : simulation ] below , add homogeneous poisson noise , introduce a @xmath20 wobble - mode correction , and integrate over the @xmath76 beam . the normalization and background
are chosen such that the final map thus obtained shows @xmath135 bright spots over a background comparable to the veritas background . for such parameters , a map identically prepared but with mean background , rather than wobble background subtraction , shows @xmath136 bright spots , implying a true significance @xmath137 higher than inferred in wobble - mode
. a @xmath0-ray structure that extends over scales much greater than @xmath110 would be largely erased by the wobble background subtraction .
hence , the above estimates provide only a lower limit to the correction needed to determine the significance of such a structure .
next , consider a cross - correlation between the veritas map @xmath79 and some potential tracer of the virial shock , @xmath138 .
after binning the two maps onto some common grid , we compute the correlation as @xmath139 where the means @xmath140 , standard deviations @xmath141 , and number of pixels @xmath91 , pertain to the region being examined .
the factor @xmath88 accounts for the correlation between pixels within the @xmath76 beam , ensuring that the result is approximately independent of the grid resolution .
the significance of the correlation is determined by generating a large ( @xmath142 , converged ) sample of maps with randomly swapped pixels , and computing the corresponding significance distribution .
consider the effect wobble - mode background subtraction has on such cross - correlations .
as in the significance estimation discussed in [ sec : wobble ] , any existing correlation is diminished because extended bright regions experience excessive background subtraction . however , the effect here is more severe , because background regions lying @xmath143 away from the bright structure are affected as well , appearing excessively faint .
as such `` dark edges '' appear in the wobble - mode map but not in the tracers , the correlation is further diminished .
the effect of wobble - mode background subtraction on a cross - correlations signal can be crudely estimated by correlating mock tracers with simulated @xmath0-ray maps , prepared with ( _ e.g. , _ figure [ fig : comasimwobble ] ) and without ( _ e.g. , _ figure [ fig : comasim ] ) wobble background subtraction .
we generate mock tracers by adding random noise of various mean amplitude to the significance map of the simulated cluster , smoothed with a gaussian filter of standard deviation @xmath144 .
the results are well - fit by a wobble correction factor @xmath145 where @xmath146 is the nominal significance of the cross - correlation using the wobble - mode signal , @xmath147 is the true significance , and @xmath148 is the solid angle involved . for example , for the typical @xmath149 of wmap , the @xmath150 correlation found ( see [ sec : radio ] below ) between @xmath0-ray and synchrotron emission outside the inner @xmath52 ellipse bounding the @xmath0-ray structure from below yields a corrected @xmath151 confidence level .
as pointed out in [ sec : wobble ] , these corrections are only lower limits if the @xmath0-ray structure is more extended than the simulated ring used to derive eq .
( [ eq : corrwobble ] ) .
when averaged over large angular scales , the galactic foreground anticipated above @xmath86 based on fermi observations@xcite is lower by at least a factor of a few than the ring signal ( see [ sec : inversecompton ] ) .
moreover , galactic signals on @xmath152 scales are efficiently removed from the veritas map by the wobble - mode background correction .
next , we show that no significant galactic contamination on @xmath153 scales is expected in the veritas mosaic of coma .
the main @xmath0-ray galactic foregrounds at intermediate latitudes arise from _
( i)_inverse compton emission from cosmic - ray ( cr ) electrons ; _
( ii)_relativistic bremsstrahlung of cr electrons with the interstellar plasma ; and _ ( iii)_nucleon - nucleon scattering between cr ions and the interstellar medium .
the crs are unlikely to show significant structure on @xmath154 scales , as this would require @xmath155 magnetic fields at @xmath156 distances . moreover ,
synchrotron maps@xcite show relatively very little emission from the coma region .
therefore , the galactic foreground on small scales is dominated by electron bremsstrahlung and by @xmath157 decay from nucleon collisions , both approximately following the distribution of gas along the line of sight .
hence , we may estimate the galactic contamination remaining in the veritas map using gas tracers , such as the full - sky h@xmath158 @xcite and infrared dust emission@xcite maps ( figure [ fig : galactictracers ] ) .
both tracers indicate a very low level of galactic structure in the coma region , and towards the @xmath0-ray ring in particular , with only minor structure found mainly in the east .
we confirm this using a spectral linear regression of the seven - year wmap data , outlined in [ sec : radio ] .
both tracers show no positive correlation with the veritas map , ruling out any significant remaining galactic contamination .
note that synchrotron tracers should not be used here to estimate or eliminate the galactic foreground .
first , these tracers are inefficient as they follow the smooth cr electron distribution .
more importantly , they include an interesting synchrotron signal from the coma virial ring , as we argue in [ sec : radio ] .
the veritas signal can be assessed by comparing it to the @xmath0-ray signature of a simulated coma - like cluster in a @xmath29cdm simulation@xcite .
the @xmath0-ray signal was computed by injecting relativistic electrons at the strong shocks of the simulation , with a mach number - dependent power - law spectrum given by the fermi diffusive shock acceleration model .
these electrons were assumed to carry a fraction @xmath33 of the thermal post - shock energy ; see [ sec : inversecompton ] for details .
we choose the richest cluster in the @xmath159 simulation box , found@xcite to have a coma - like mass @xmath160 and temperature @xmath7 .
= 6.5 cm the simulated virial shock ( figure [ fig : comasim ] ) has an elliptic structure , elongated towards the main lss filament ( figure [ fig : sdss ] ) , with semi - minor axis @xmath17 and semi - major axis @xmath161 ( at the distance of coma ) ; the average @xmath162 spectral index is @xmath163 .
the shock elongation , the morphology of the surrounding filaments , and the enhanced accretion towards the main filament , suggest a prolate geometry aligned with the main filament feeding the cluster .
unrelaxed , merger clusters such as coma are indeed known to preferentially be prolate@xcite .
the center of the simulated cluster is determined by the peak density .
the simulated map is rotated about this point such that the main simulated lss filament coincides with the @xmath164 orientation of the sdss filament and the rosat ellipse . by coincidence , the agreement between the simulated filaments and the sdss map remains fairly good even far from the cluster , if the simulated map is inverted along the semi - major axis of the virial shock ( figure [ fig : comasim2 ] ) .
the simulated @xmath0-ray signal shows a positive correlation with the veritas map ( figure [ fig : comaversim ] ) , at the @xmath31 confidence level ( @xmath32 after correcting for the excessive wobble - mode background subtraction ; see [ sec : wobble ] ) .
the correlation remains strong , @xmath165 ( @xmath166 ) , if the maps are rotated by @xmath167 with respect to each other .
the extent of the virial shock may be somewhat overestimated by the adiabatic , smooth particle hydrodynamics ( sph ) simulation .
indeed , the correlation strengthens to @xmath168 ( @xmath169 ) if the simulated map is resized by a factor @xmath170 .
the mass accretion rate @xmath171 of the cluster , normalized by the mass - to - hubble time ratio in the dimensionless parameter @xmath172 , varies among clusters and fluctuates in time , and is found@xcite to be somewhat low when averaged over the simulated cluster , @xmath173 .
the @xmath0-ray map shown in figure [ fig : comasim ] corresponds to normalized average accretion rate @xmath174 , and an acceleration efficiency @xmath175 .
equivalently , we may examine the accretion rate of the total thermal energy @xmath176 , using the dimensionless parameter @xmath177 .
in the simulation ,
we find@xcite @xmath178 , although slightly larger , @xmath179 values arise in a simple isothermal sphere model@xcite .
in order to quantify the coincidence between @xmath0-ray and radio signals , we analyze the wmap seven - year , full - sky maps@xcite .
we use a simple spectral decomposition of the five wmap bands into synchrotron and sz maps , avoiding any spatial masking which may complicate the comparison to the veritas map .
the synchrotron spectrum is approximated as a pure power - law , with brightness @xmath180 and spectral index @xmath181 , typical of cluster radio sources ; the results are not sensitive to the precise spectrum for @xmath182 .
the sz signal is clear in the central parts of coma , but is difficult to spectrally separate from the cmb fluctuations at large radii@xcite . for simplicity
, we leave these cmb fluctuations in the sz map , thus somewhat diluting any correlations with the veritas map .
in general , the wmap data includes non - negligible foregrounds from galactic synchrotron , dust and free - free emission@xcite , which are strong towards the galactic plane and in principal should be removed . however , as discussed in [ sec : galactic ] , these signals are particularly weak in the coma region ( figure [ fig : galactictracers ] ) , are spatially too smooth ( in particular the synchrotron foreground ) to have a @xmath0-ray counterpart surviving the wobble - mode background removal , and their tracers do not correlate with the veritas map .
moreover , the extragalactic synchrotron signal can not be separated from the galactic @xmath183 synchrotron , @xmath184 dust , and @xmath185 free - free signals , using spectral regression alone , in particular considering the varying synchrotron spectral index in the cluster radio sources .
therefore , we leave these faint and probably smooth galactic contaminations in our synchrotron extragalactic tracer .
this only dilutes the correlations between this tracer and the veritas map , as the galactic signals show no positive correlations with the latter .
therefore , we fit the five wmap channels by @xmath186 + { { t_{\mbox{\scriptsize syn}}}}\left ( \frac{\nu}{\nu_0}\right)^{-3.2 } { \mbox { , } } \nonumber\end{aligned}\ ] ] where @xmath187 is the dimensionless frequency and @xmath188 is the comptonization parameter .
the low - frequency sz temperature shift @xmath189 and the synchrotron brightness temperature @xmath190 ( at the arbitrary frequency @xmath191 ) are used as fit parameters , giving the tracer maps ( figure [ fig : wmap_regression ] ) .
next , we compute the cross - correlation amplitudes of the veritas data with these @xmath192 and @xmath190 maps . on average , the veritas data show no correlation with the synchrotron map , and an insignificant , @xmath193 ( @xmath194 ) anti - correlation with the sz map ( _ i.e. _ a positive correlation with the @xmath188-parameter ) . here and below ,
results in parenthesis are the conservative confidence levels after correcting for the wobble - mode background subtraction , as described in
[ sec : crosscorrelations ] .
features seen in both sz and synchrotron maps are more pronounced in the western part of the cluster , possibly due to some eastern foreground ( _ e.g. , _ see figure [ fig : halpha ] ) or cmb fluctuation .
consequently , for @xmath195 ( no more than @xmath196 east of the cluster s center ) , the sz anti - correlation strengthens to @xmath194 ( @xmath197 ) . note that this signal does not arise from the spatial separation between the central sz decrement and the peripheral @xmath0-ray ring , as this would correspond to a positive correlation here .
more importantly , when we split the data into two regions , outside and inside the inner @xmath52 ellipse bounding the @xmath0-ray signal , correlations emerge between @xmath0-rays and both radio signals . outside this inner ellipse ,
i.e. along the @xmath0-ray ring , the veritas map correlates with the synchrotron map at the @xmath39 ( @xmath40 ) confidence level , and anti - correlates with the sz map at the @xmath41 ( @xmath42 ) level .
the signal is dominated by the western half of the ring , where these correlations strengthen to @xmath198 ( @xmath199 ) with the synchrotron and @xmath45 ( @xmath46 ) with the sz .
the results depend somewhat on the precise region examined .
for example , including the @xmath196 region east of the center , the synchrotron correlation strengthens to @xmath43 ( @xmath44 ) while the sz correlation remains unchanged . within the @xmath52 ellipse
, the veritas map does not correlate with the sz signal , but does show a @xmath53 ( @xmath54 ) anti - correlation with the synchrotron map , mainly in the western part , where it reaches @xmath55 ( @xmath56 ) . the inner anti - correlation between the veritas and synchrotron maps , evident in the @xmath0-ray underluminous radio halo and radio relic , and their outer positive correlation ,
are illustrated in figure [ fig : compositeversyn ] . in the periphery of the cluster
, the positive correlation between the veritas map and the synchrotron signal agrees with predictions for the coincident inverse - compton and synchrotron emission from the same relativistic , shock - accelerated electrons , as we show in [ sec : inversecompton ] and [ sec : radiotheory ] below .
the anti - correlation with the sz map agrees with the anticipated sz signal from the thermal gas , expected to cutoff sharply as the thermal pressure drops beyond the virial shock , as shown in
[ sec : radiotheory ] . in the inner parts of the cluster ,
an anti - correlation between @xmath0-ray and synchrotron radiation was anticipated in secondary electron models , where electrons are continuously injected into the icm in both strongly and weakly magnetized regions@xcite . here ,
electrons magnetized above ( below ) @xmath200 , where @xmath201 is the cmb energy density , lose most of their energy to synchrotron ( inverse - compton ) radiation in the radio ( @xmath0-ray ) band@xcite , leading to alternating radio- and @xmath0-ray - bright regions , as discussed in [ sec : inneremission ] .
cluster virial shocks last for long , @xmath203 hubble timescales , and are strong , in particular where cold gas accretes from the voids and the mach number squared is @xmath204 .
therefore , such shocks are thought to accelerate a flat , @xmath205 spectrum ( constant energy per logarithmic energy interval ) of relativistic electrons up to high , cooling - limited energies .
the dominant cooling process here is compton scattering off cmb photons , implying an electron energy cutoff@xcite @xmath206 where @xmath207 and @xmath208 are the temperature and magnetic field amplitude downstream of the shock . here , @xmath209 and @xmath210 are the electron charge and mass , @xmath211 is the thompson cross section , and @xmath212 is the sound velocity .
the photon energy @xmath213 which maximizes the veritas sensitivity corresponds@xcite to electrons with energy @xmath214 , below but not far from the estimated cutoff .
the main uncertainty in modelling the nonthermal emission from a virial shock stems from the presently poor understanding of particle acceleration and magnetization in collisionless shocks .
the inverse - compton emission from the shock can be estimated by assuming that a fraction @xmath215 of the post - shock thermal energy is deposited in relativistic electrons .
electrons with energies @xmath216 cool quickly , in a narrow shell behind the shock , so the logarithmic emissivity per unit surface area of the shock is @xmath217 where @xmath218 is the ( thermal ) electron number density , and @xmath219 is the downstream velocity .
the @xmath0-ray signal from a coma - like cluster can be computed using eq .
( [ eq : icemissivity ] ) in the framework of a cosmological simulation@xcite , without introducing additional free parameters . however , the comparison between simulated and observed clusters is complicated by a secondary uncertainty , involving the local accretion rate of the cluster .
this rate varies spatially along the shock front , fluctuates in time , and differs even among similar clusters as a function of their specific environment .
we parameterize the average accretion rate using the dimensionless variable @xmath220 where here we assumed a spherical shock of radius @xmath221 . in the @xmath68-model
discussed in appendix [ sec : betamodel ] , @xmath222 , and in the simulated cluster@xcite @xmath173 .
these are averaged quantities , with strong local fluctuations seen in simulations and expected in practice .
consider a planar cut parallel to the line of sight , along which the virial shock is approximately a ring of radius @xmath221 and of negligible thickness .
a beam that intersects the cut at distances @xmath223 from the center covers a shock segment of length @xmath224 .
therefore , at the near edge of a long cylindrical shock of angular radius @xmath225 , a beam of area @xmath84 would detect @xmath226 energy photons with number flux @xmath227 a similar result is expected in the prolate shock inferred in coma , where for the veritas map @xmath228 in the isothermal @xmath68-model , this may be written as @xmath229 . in order to compare this estimate with the @xmath0-ray structure seen in the veritas map , recall that a @xmath230 fluctuation corresponds to an excess of @xmath231 photons with @xmath15 in a beam of radius @xmath76 .
the map was produced by a @xmath232 hour exposure@xcite , with an @xmath233 effective area ( at @xmath234 , see @xcite ) .
a @xmath230 signal in the beam therefore corresponds to a photon flux @xmath235 the average signal inferred from the histogram ( figure [ fig : ver_hist ] ) , @xmath128 , is thus comparable to the predicted signal in eq .
( [ eq : jpredicted ] ) for standard parameters .
the agreement between the predicted ( eq .
[ eq : jpredicted ] ) and observed ( eq .
[ eq : iver ] ) @xmath0-ray flux , and the similar significance of the @xmath0-ray features in the observed ( figure [ fig : comaver ] ) and simulated ( figure [ fig : comasim ] ) maps , suggest that to within an uncertainty factor of a few , @xmath236 .
localized regions with @xmath237 detection , and possibly brighter emission masked by the wobble - mode background subtraction , suggest that at least locally the product @xmath238 can exceed its inferred average value by a factor of a few .
the total flux from the virial shock can be estimated by taking @xmath239 in eq .
( [ eq : jperbeam ] ) , and integrating over the solid angle enclosed by the shock . for a cylindrical ring of radius @xmath97 and length @xmath240
, this gives @xmath241 where @xmath242 and @xmath243 .
for the veritas band and @xmath244 , we use the isothermal @xmath68-model to obtain @xmath245 . calibrating the free parameters against the local @xmath230 veritas flux ( eq . [ eq : iver ] ) , this becomes @xmath246 .
extrapolating the veritas - calibrated signal to the egret band ( assuming a flat spectrum ) , yields @xmath247 , comparable to the @xmath248 egret upper limit on emission from coma@xcite .
the corresponding @xmath249 exceeds the @xmath250 upper limit imposed by fermi@xcite , but this limit was obtained using an aperture optimized for emission from the central , x - ray bright region of the cluster .
note that at such low energies , the galactic foreground is not negligible , and both instruments lack the high angular resolution of veritas essential in order to identify the ring and avoid foreground contamination .
the above analysis assumed a flat spectrum of accelerated electrons , leading to a flat ( @xmath78 ) photon spectrum . indeed , comparing the veritas measurement with the egret upper limit implies that the photon spectral index between @xmath251 and @xmath86 can not be much softer than @xmath252 .
as the veritas measurement probes only photons near @xmath86 , the above estimates strictly pertain to electrons in energies near @xmath253 . here
, we find a logarithmic electron injection efficiency @xmath254
the same electrons that scatter cmb photons up to the @xmath0-ray band also emit synchrotron radiation , as they gyrate in the shock amplified magnetic fields . the ratio between radio and @xmath0-ray emission from a strong virial shock is approximately given by the ratio between magnetic and cmb energy densities , @xmath256 in order to compute the synchrotron signal , some assumption must be made concerning the magnetic field behind the shock .
typically@xcite , one assumes that a fraction @xmath257 of the thermal energy is deposited in downstream magnetic fields , based on observations of supernova remnant shocks ; some magnetization is also needed to accelerate the @xmath0-ray emitting electrons . however
, this assumption was not tested so far near the virial shock , and the value of @xmath65 is not well constrained .
we may use the coincident @xmath0-ray and radio measurements to estimate the magnetic field at the shock .
the strongest signal is found in the lowest wmap frequency , @xmath258 , where bright spots along the shock show @xmath259 , corresponding to a brightness @xmath260 the same bright spots are detected by veritas at @xmath237 , corresponding to @xmath261 . equation ( [ eq : gamma_to_synch ] ) then yields @xmath262 , which according to the @xmath68-model corresponds to @xmath263 .
we may use the sz signal near the shock to test the above estimates .
for example , at @xmath258 one expects the ratio between the synchrotron emission and the sz decrement to be @xmath264 for an isothermal sphere distribution@xcite , and @xmath265 for a @xmath68-model ( see appendix [ sec : betamodel ] ) . inspecting the coincident synchrotron ( @xmath266 at @xmath267 ) and sz ( @xmath268 ) signals measured along the @xmath0-ray ring , we find that @xmath269 .
this supports the above estimates @xmath236 , and @xmath48 .
note that the parameter @xmath65 refers to the magnetic energy weighted over the relativistic electron distribution , and so is elevated by clumping effects@xcite .
within @xmath52 , we find an anti - correlation between @xmath0-rays and synchrotron emission .
this signal is not highly significant ( @xmath55 and @xmath56 , before and after correcting for wobble background removal ) , so deeper observations are needed to clarify its nature .
here we consider the implications of the anti - correlation assuming it is indeed real .
in such a case , as the synchrotron signal involved is dominated by the radio halo and radio relic , some of the veritas signal must arise from the inner icm , well inside the virial radius . in the radio band
, coma is a classical example of a merging cluster , harboring both an extended radio halo and a peripheral radio relic@xcite .
the @xmath270 mpc size of such radio halos implies an ongoing injection of relativistic electrons , but it is debated whether they arise from hadronic collisions involving cr ions ( secondary models@xcite ) or from in - situ ( re)acceleration in weak shocks or turbulence ( primary models@xcite ) .
secondary models can also explain radio relics@xcite and minihalos@xcite as arising from the same distribution of cr protons , if the latter are homogeneously distributed@xcite . here ,
radio - bright regions correspond to strong , @xmath271 magnetic fields .
the existence of such strongly magnetized regions is also being debated , in coma@xcite and in other clusters@xcite . in secondary models ,
electron injection is an ongoing process , proportional to the gas density but largely independent of shocks , turbulence , and other changes in the flow .
therefore , the sum of synchrotron and inverse - compton emission from the icm varies slowly across the cluster , whereas their ratio can show strong variations , proportional to @xmath272 , leading to a radio@xmath0-ray anti - correlation .
in contrast , in primary models , the electrons are injected only locally , in shocks or turbulence where the magnetic field is amplified , and cool long before the magnetic field can decay .
hence , a positive correlation is expected here between synchrotron and @xmath0-ray signals ; inverse - compton emission far from the radio sources is expected to be negligible . in both secondary and primary models ,
@xmath0-ray emission is expected also from @xmath157 decay , following cr ion collisions with the ambient nucleons .
this signal is expected to be smoother than the inverse - compton emission , and so more susceptible to wobble - mode background subtraction , in particular if the crs are distributed homogeneously , as inferred from hadronic radio models@xcite .
for example , in a @xmath68-model with homogeneous cr ions , wobble - mode removes @xmath273 of the signal in the central @xmath274 beam , which corresponds to the magnetized core , and @xmath275 of the signal beyond @xmath276 .
the @xmath157 signal , independent of the local magnetic field and stronger in the center , should positively correlate with the synchrotron map , which is dominated by the central radio halo .
it can not account for the anti - correlation presently observed .
therefore , the inner @xmath0-ray synchrotron anti - correlation , if confirmed , supports the notion of magnetized , @xmath57 regions in the icm , and favors a secondary electron injection model . here , the presence of a weak @xmath0-ray signal far from the center ( but inward of the virial shock ) supports a homogeneous cr proton distribution . in non - magnetized regions ,
the @xmath277 synchrotron signals from the radio halo and relic translate to an inverse - compton flux @xmath278 . by eq .
( [ eq : iver ] ) , this corresponds to an @xmath279 veritas significance per @xmath76 beam , so @xmath0-ray emission around the core can plausibly lead to the observed anti - correlation .
integrated within a @xmath280 ellipse , emission from secondary electrons corresponds to an @xmath281 nominal veritas signal , assuming homogeneous cr ions and a @xmath68-model gas distribution .
such a signal can be further tested in a deeper @xmath0-ray observation .
if wobble - mode corrections can be minimized , the ( @xmath282 times stronger@xcite ) @xmath157 signal may be measured as well .
we analyze the veritas mosaic ( figure [ fig : comaver ] ) of the coma cluster , and find a large - scale , extended @xmath0-ray feature around the cluster .
it is best described as an elongated ( @xmath99 ) , thick ( @xmath283 ) elliptical ring , with semi - minor axis @xmath17 ( figure [ fig : comaverbin ] ) , oriented approximately along the east - west direction ( @xmath284 ; figure [ fig : comavertilt ] ) , towards the lss filament connecting coma with abell 1367 ( figure [ fig : sdss ] ) .
the @xmath0-ray ring is seen at a nominal @xmath30 confidence level , but comparison with mock veritas noise maps implies a higher , @xmath19 confidence .
further correcting for excessive wobble - mode background subtraction , we conclude that the ring presents with a confidence level of at least @xmath285 .
if we do not assume a - priori that the ring is aligned with the lss filament , the significance becomes @xmath26 ( @xmath27 corrected ) .
our wobble - mode corrections , derived in [ sec : wobble ] and [ sec : crosscorrelations ] ( see figure [ fig : ver_hist ] , and eq .
[ eq : corrwobble ] ) , are quite conservative .
the size and morphology of the @xmath0-ray ring agree with predictions for the signature of the virial shock , produced as shock - accelerated electrons inverse compton - scatter cmb photons .
the veritas map thus correlates with a normalized simulated @xmath0-ray map of a coma - like cluster ( figure [ fig : comasim ] ) , once the main lss filaments are similarly oriented ( figure [ fig : comasim2 ] ) , at a @xmath31 ( @xmath32 ) confidence ( no free parameters ; see figure [ fig : comasimwobble ] ) .
the best fit is obtained for ellipticity @xmath23 , but we can not rule out a cylindrical shock parallel to the lss filament .
the brightness of the signal agrees with predictions as well , provided that the rate of relativistic electron energy deposition is @xmath286 of the thermal energy per hubble time ( within a factor of a few ) .
interestingly , a @xmath5 signal found@xcite by stacking the egret data around 447 rich clusters corresponds , when compared with source number counts computed analytically@xcite and using cosmological simulations@xcite , to an average @xmath287 , in agreement with the value we infer in coma .
inverse - compton emission from virial shocks can explain the hard x - ray signals observed in several clusters@xcite , provided that @xmath288 .
for example , @xmath289 was recently inferred from hard x - ray emission from a lss filament@xcite .
the veritas ring corresponds to a @xmath290 flux at @xmath36 energies ( assuming a flat spectrum ) , accounting for the @xmath291 hard x - ray signal observed@xcite .
high resolution observations in hard x - rays , in particular with nustar , should resolve the virial ring in coma , and detect virial shocks in many other clusters .
an analysis of the wmap seven - year data reveals a positive correlation between the @xmath0-ray ring and synchrotron emission ( figures [ fig : wmap_syn ] and [ fig : compositeversyn ] ) , at the @xmath292 ( @xmath293 ) confidence level , and an anti - correlation with the sz map ( figure [ fig : wmap_sz ] ) at @xmath41 ( @xmath42 ) .
the radio signals are clearer in the western half of the ring , towards the lss filament , where the correlations reach @xmath294 ( @xmath295 ) and @xmath45 ( @xmath46 ) , respectively .
they agree with predictions for the synchrotron emission from the @xmath0-ray emitting electrons and for the sz decline with thermal pressure beyond the shock , if the downstream magnetic field is @xmath262 , corresponding to @xmath263 magnetization ( within a factor of a few ) .
the sz signal we find , extending in some parts out to @xmath50 away from the center , agrees with the sz detections by wmap@xcite and planck@xcite .
however , the peripheral synchrotron signal coincident with the @xmath0-rays and with the sz cutoff was not reported so far . the @xmath0-ray ring and
these radio features are much farther out than the @xmath51 weak shocks seen in x - ray , radio@xcite , and sz@xcite data .
inward of the ring , the veritas map shows a @xmath53 ( @xmath54 ) anti - correlation with the synchrotron map , stronger in the western region where @xmath296 ( @xmath56 ) .
this is evident in the radio halo and radio relic , which appear underluminous in @xmath0-rays ( figure [ fig : compositeversyn ] ) .
such a signal was predicted in models where relativistic electrons are injected into the icm continuously , in both strongly and weakly magnetized regions which consequently appear radio bright/@xmath0-ray faint and vice versa .
while the present signal is weak and requires confirmation , it is consistent with analytic estimates .
the extent of the anti - correlation and its magnitude support secondary electron injection by collisions of cosmic - ray protons with the ambient nucleons , and favor a homogeneous cr distribution .
a veritas observation deeper than the present @xmath58 hour exposure would decisively test our results , in particular if wobble - mode corrections can be avoided or minimized , for example by using a larger , @xmath297 wobble radius .
this would allow for a more precise reconstruction of the virial shock structure and a determination of variations in the electron deposition rate along the shock front .
such data would be valuable in the study of large - scale structure formation at low redshift , mapping the feeding pattern of coma , probing the surrounding voids , the coma
abell 1367 filament , and other lss filaments , and tracing the warm - hot intergalactic medium ( whim ) immediately behind the shock .
such a deep observation would test the anti - correlation we find between @xmath0-ray and radio maps at small , @xmath298 radii .
a significant anti - correlation would confirm the hadronic , secondary electron model for cluster radio sources , and provide a direct measure of magnetic variations in the icm .
a deep observation would thus probe the cr proton distribution out to large radii , address the origin of these protons , and in particular place an upper limit on proton acceleration in the virial shock .
observations at lower energies , for example using the @xmath299 magic erenkov telescopes@xcite , should be able to utilize the higher photon flux to clearly identify the @xmath0-ray ring . at
yet lower energies , the fermi @xmath0-ray space telescope is marginally sensitive to the signal at @xmath300 , but it may be challenging to resolve the shock structure with the @xmath130 ( @xmath301 containment ) point spread function@xcite .
the veritas energy threshold , @xmath60 , is not far below the expected photon cutoff , @xmath302 .
observations at higher energies would become increasingly sensitive to this cutoff , thus probing the conditions at the shock and the nature of particle acceleration , although pair production off the infrared background also becomes important at such energies@xcite .
interestingly , no evidence for extended emission was found in a @xmath62 observation of coma by the hess telescopes@xcite . despite the higher energy threshold and the shorter , @xmath303 hour observation , it was only @xmath304 times less sensitive than the veritas observation ( assuming a flat spectrum ) , thanks to the larger effective area@xcite , @xmath305 .
the hess map shows no significant signal or correlation with the veritas data , but the @xmath0-ray ring may well be shallowly buried under the noise .
note that the optical depth of the hess photons due to pair production off the infrared background , estimated@xcite as @xmath306 , may not be negligible . finally , consider the contribution of cluster virial shocks to the extragalactic @xmath0-ray and radio backgrounds .
these components were computed analytically@xcite , and calibrated numerically using cosmological simulations@xcite . while the main model parameters @xmath238 and @xmath65 are likely to fluctuate within clusters and vary among different clusters , it is instructive to adopt the values inferred in coma as typical .
an accretion rate @xmath307 is intermediate between the analytic model@xcite and the low redshift result of numerical simulations@xcite .
normalizing all clusters by this value , we obtain a diffuse @xmath0-ray component @xmath308 , dominating the extragalactic background@xcite . in the radio ,
the revised parametrization yields a @xmath309 synchrotron signal , dominating the extragalactic low frequency radio background @xcite .
it is observable through @xmath310 fluctuations at multipoles @xmath311 with present interferometers such as lofar and evla .
we thank ido reiss and boaz katz for useful discussions .
the research of uk has received funding from the european union seventh framework programme ( fp7/2007 - 2013 ) under grant agreement n pcig09-ga-2011 - 293975 .
an analytic model for emission from the virial shock requires some assumptions specifying the gas distribution .
simple choices include an isothermal sphere@xcite or a isothermal @xmath68-model@xcite , the latter of which we discuss below . in our simple model
, it is assumed that _ ( i)_the electron number density is approximately given by @xmath312 out to the virial radius ; and _
( ii)_on average , the gas distribution is approximately isothermal and in hydrostatic equilibrium . in coma@xcite ,
the central density is approximately @xmath313 , the core radius is @xmath314 , the index @xmath315 , and the ( emission measure - weighted ) temperature is @xmath316 .
p. sreekumar , d. l. bertsch , b. l. dingus , j. a. esposito , c. e. fichtel , j. fierro , r. c. hartman , s. d. hunter , g. kanbach , d. a. kniffen , y. c. lin , h. a. mayer - hasselwander , j. r. mattox , p. f. michelson , c. von montigny , r. mukherjee , p. l. nolan , e. schneid , d. j. thompson , and t. d. willis . . ,
464:628 , june 1996 .
, a. p. t. arlen , t. aune , m. beilicke , w. benbow , a. bouvier , j. h. buckley , v. bugaev , k. byrum , a. cannon , a. cesarini , l. ciupik , e. collins - hughes , m. p.
connolly , w. cui , r. dickherber , j. dumm , a. falcone , s. federici , q. feng , j. p. finley , g. finnegan , l. fortson , a. furniss , n. galante , d. gall , s. godambe , s. griffin , j. grube , g. gyuk , j. holder , h. huan , g. hughes , t. b. humensky , a. imran , p. kaaret , n. karlsson , m. kertzman , y. khassen , d. kieda , h. krawczynski , f. krennrich , k. lee , a. s madhavan , g. maier , p. majumdar , s. mcarthur , a. mccann , p. moriarty , r. mukherjee , t. nelson , a. ofaolin de bhrithe , r. a. ong , m. orr , a. n. otte , n. park , j. s. perkins , m. pohl , h. prokoph , j. quinn , k. ragan , l. c. reyes , p. t. reynolds , e. roache , j. ruppel , d. b. saxon , m. schroedter , g. h. sembroski , c. skole , a. w. smith , i. telezhinsky , g. tei , m. theiling , s. thibadeau , k. tsurusaki , a. varlotta , m. vivier , s. p. wakely , j. e. ward , a. weinstein , r. welsing , d. a. williams , and b. zitzer . . ,
august 2012 .
e. komatsu , k. m. smith , j. dunkley , c. l. bennett , b. gold , g. hinshaw , n. jarosik , d. larson , m. r. nolta , l. page , d. n. spergel , m. halpern , r. s. hill , a. kogut , m. limon , s. s. meyer , n. odegard , g. s. tucker , j. l. weiland , e. wollack , and e. l. wright . . ,
192:18 , february 2011 .
f. aharonian , a. g. akhperjanian , g. anton , u. barres de almeida , a. r. bazer - bachi , y. becherini , b. behera , k. bernlhr , c. boisson , a. bochow , v. borrel , e. brion , j. brucker , p. brun , r. bhler , t. bulik , i. bsching , t. boutelier , p. m. chadwick , a. charbonnier , r. c. g. chaves , a. cheesebrough , l .- m .
chounet , a. c. clapson , g. coignet , m. dalton , m. k. daniel , i. d. davids , b. degrange , c. deil , h. j. dickinson , a. djannati - ata , w. domainko , l. oc .
drury , f. dubois , g. dubus , j. dyks , m. dyrda , k. egberts , d. emmanoulopoulos , p. espigat , c. farnier , f. feinstein , a. fiasson , a. frster , g. fontaine , m. fling , s. gabici , y. a. gallant , l. grard , b. giebels , j. f. glicenstein , b. glck , p. goret , d. ghring , d. hauser , m. hauser , s. heinz , g. heinzelmann , g. henri , g. hermann , j. a. hinton , a. hoffmann , w. hofmann , m. holleran , s. hoppe , d. horns , s. inoue , a. jacholkowska , o. c. de jager , c. jahn , i. jung , k. katarzyski , u. katz , s. kaufmann , e. kendziorra , m. kerschhaggl , d. khangulyan , b. khlifi , d. keogh , w. kluniak , t. kneiske , n. komin , k. kosack , g. lamanna , j .-
lenain , t. lohse , v. marandon , j. m. martin , o. martineau - huynh , a. marcowith , d. maurin , t. j. l. mccomb , m. c. medina , r. moderski , e. moulin , m. naumann - godo , m. de naurois , d. nedbal , d. nekrassov , j. niemiec , s. j. nolan , s. ohm , j .- f .
olive , e. de oa wilhelmi , k. j. orford , m. ostrowski , m. panter , m. paz arribas , g. pedaletti , g. pelletier , p .- o .
petrucci , s. pita , g. phlhofer , m. punch , a. quirrenbach , b. c. raubenheimer , m. raue , s. m. rayner , m. renaud , o. reimer , f. rieger , j. ripken , l. rob , s. rosier - lees , g. rowell , b. rudak , c. b. rulten , j. ruppel , v. sahakian , a. santangelo , r. schlickeiser , f. m. schck , r. schrder , u. schwanke , s. schwarzburg , s. schwemmer , a. shalchi , m. sikora , j. l. skilton , h. sol , d. spanglfoer , .
stawarz , r. steenkamp , c. stegmann , g. superina , a. szostek , p. h. tam , j .-
tavernet , r. terrier , o. tibolla , m. tluczykont , c. van eldik , g. vasileiadis , c. venter , l. venter , j. p. vialle , p. vincent , m. vivier , h. j. vlk , f. volpe , s. j. wagner , m. ward , a. a. zdziarski , and a. zech . . ,
502:437443 , august 2009 .
m. ackermann , m. ajello , w. b. atwood , l. baldini , j. ballet , g. barbiellini , d. bastieri , k. bechtol , r. bellazzini , b. berenji , r. d. blandford , e. d. bloom , e. bonamente , a. w. borgland , t. j. brandt , j. bregeon , m. brigida , p. bruel , r. buehler , s. buson , g. a. caliandro , r. a. cameron , p. a. caraveo , e. cavazzuti , c. cecchi , e. charles , a. chekhtman , j. chiang , s. ciprini , r. claus , j. cohen - tanugi , j. conrad , s. cutini , a. de angelis , f. de palma , c. d. dermer , s. w. digel , e. d. c. e. silva , p. s. drell , a. drlica - wagner , l. falletti , c. favuzzi , s. j. fegan , e. c. ferrara , w. b. focke , p. fortin , y. fukazawa , s. funk , p. fusco , d. gaggero , f. gargano , s. germani , n. giglietto , f. giordano , m. giroletti , t. glanzman , g. godfrey , j. e. grove , s. guiriec , m. gustafsson , d. hadasch , y. hanabata , a. k. harding , m. hayashida , e. hays , d. horan , x. hou , r. e. hughes , g. jhannesson , a. s. johnson , r. p. johnson , t. kamae , h. katagiri , j. kataoka , j. kndlseder , m. kuss , j. lande , l. latronico , s .- h .
lee , m. lemoine - goumard , f. longo , f. loparco , b. lott , m. n. lovellette , p. lubrano , m. n. mazziotta , j. e. mcenery , p. f. michelson , w. mitthumsiri , t. mizuno , c. monte , m. e. monzani , a. morselli , i. v. moskalenko , s. murgia , m. naumann - godo , j. p. norris , e. nuss , t. ohsugi , a. okumura , n. omodei , e. orlando , j. f. ormes , d. paneque , j. h. panetta , d. parent , m. pesce - rollins , m. pierbattista , f. piron , g. pivato , t. a. porter , s. rain , r. rando , m. razzano , s. razzaque , a. reimer , o. reimer , h. f .- w .
sadrozinski , c. sgr , e. j. siskind , g. spandre , p. spinelli , a. w. strong , d. j. suson , h. takahashi , t. tanaka , j. g. thayer , j. b. thayer , d. j. thompson , l. tibaldo , m. tinivella , d. f. torres , g. tosti , e. troja , t. l. usher , j. vandenbroucke , v. vasileiou , g. vianello , v. vitale , a. p. waite , p. wang , b. l. winer , k. s. wood , m. wood , z. yang , m. ziegler , and s. zimmer . . , 750:3 ,
may 2012 .
j. s. perkins . .
in f. a. aharonian , w. hofmann , and f. rieger , editors , _ american institute of physics conference series _
, volume 1085 of _ american institute of physics conference series _ , pages 569572 , december 2008 .
b. gold , n. odegard , j. l. weiland , r. s. hill , a. kogut , c. l. bennett , g. hinshaw , x. chen , j. dunkley , m. halpern , n. jarosik , e. komatsu , d. larson , m. limon , s. s. meyer , m. r. nolta , l. page , k. m. smith , d. n. spergel , g. s. tucker , e. wollack , and e. l. wright . .
, 192:15 , february 2011 . | galaxy clusters , the largest gravitationally bound objects in the universe , are thought to grow by accreting mass from their surroundings through large - scale virial shocks . due to electron acceleration in such a shock , it should appear as a @xmath0-ray , hard x - ray , and radio ring , elongated towards the large - scale filaments feeding the cluster , coincident with a cutoff in the thermal sunyaev - zeldovich ( sz ) signal .
however , no such signature was found so far , and the very existence of cluster virial shocks has remained a theory .
we report the discovery of a large , @xmath1 diameter @xmath0-ray ring around the coma cluster , elongated towards the large scale filament connecting coma and abell 1367 .
the @xmath0-ray ring correlates both with a synchrotron signal and with the sz cutoff .
the @xmath0-ray , hard - x - ray , and radio signatures agree with analytic and numerical predictions , if the shock deposits a few percent of the thermal energy in relativistic electrons over a hubble time , and @xmath2 of the energy in magnetic fields .
the implied inverse - compton and synchrotron cumulative emission from similar shocks dominates the diffuse extragalactic @xmath0-ray and low frequency radio backgrounds .
our results reveal the prolate structure of the hot gas in coma , the feeding pattern of the cluster , and properties of the surrounding large scale voids and filaments .
the anticipated detection of such shocks around other clusters would provide a powerful new cosmological probe .
01 0.2 in |
stoichiometric 50:50 ferh alloy has been studied extensively for many years .
fallot@xcite reported on the discovery of an af @xmath1 fm transition in this material more than half a century ago .
the unusual occurrence of a metamagnetic transition attracted considerable attention and many valuable information have been gathered over the years .
equiatomic ferh crystallizes in the cscl structure .
the magnetic transition happens abruptly as a function of temperature ( first order transition , @xmath3 @xmath4 350k ) .
the critical temperature is strongly dependent on the stoichiometry ( the transition exists only within @xmath5% deviation from the equiatomic composition ) as well as on the presence of transition metals ( tm ) impurities .
the transition is isostructural , but accompanied by a rather large lattice constant increase.@xcite the transition can also be reversed by application of external pressure.@xcite other physical quantities undergoing large variations at the transition are thermal capacity , electrical resistivity and entropy.@xcite magnetically , the antiferromagnetic phase has been identified via neutron scattering as of type ii ( af - ii ) , with nearest - neighbor fe sites ordered antiferromagnetically in the ( 001)-planes and ferromagnetically in ( 111)-planes.@xcite neutron data also provide evidence of an enhanced fe magnetic moment in this compound ( @xmath6 ) compared to bulk fe metal ( @xmath7 ) . according to electronic structure calculations , enhanced fe moments are present in both the af and fm phases .
the rh atoms are usually assumed to be unpolarized in the antiferromagnetic phase and to acquire @xmath8 in the ferromagnetic phase .
several models have been proposed to account for the properties of the metamagnetic transition , but the debate is still active .
early models , focusing on the conspicuous lattice expansion , attempted to justify the transition in terms of an underlying lattice instability .
these theories encounter serious difficulties in explaining the large entropy jump associated with the transition . on the other side ,
the large variation of thermal and electrical conductivities motivated the search for an electronic origin for the change in entropy .
this search generally led to theoretical models attributing a major role to modifications of the electronic structures ( i.e. change in density of states , dos ) in the af and fm phases .
recently , interest in the ferh system has revived .
its peculiar thermo - magnetic properties promise the realization of efficient thermally assisted magnetic recording media.@xcite with this new interest , the metamagnetic transition has been re - scrutinized using modern experimental techniques .
notably , time - resolved investigations have attempted to elucidate the role played by the spin - lattice dynamics in the transition.@xcite in spite of the richness of information on the properties of ferh , direct spectroscopic information on this system is conspicuously scarce , especially for ultra - thin films .
x - ray magnetic dichroism studies@xcite from polycrystalline samples focused on the magnetic properties without examining effects in the electronic structure , aside from the xmcd signal in the fm phase . in this work
, we report on a comprehensive photoemission study of ferh ultrathin films with close - to - equiatomic compositions .
epitaxial ferh(001 ) films are obtained by room - temperature co - deposition on a w(100 ) single crystal substrata , followed by mild annealing .
the atomic reordering process , induced by post - annealing , is accompanied by substantial modifications of the electronic structure , apparent in core levels as well as in the valence band spectra .
the appearance of a cubic leed pattern at this stage signals the formation of a well - defined , single crystalline alloy films .
the temperature dependence of the magnetic properties are monitored using magnetic linear dichroism ( mld ) in the fe 3@xmath2 core levels .
they reveal that the af @xmath1 fm metamagnetic transition , characteristic of bulk equiatomic alloy , is present also for these ultra - thin epitaxial films and in particular at their surfaces . in the ferromagnetic phase , spin - resolved valence band spectra display well polarized features which are consistent with _ first - principle _ calculations of the spin - polarized band structure of bulk ferromagnetic ferh . judging from the photoemission spectra
, the metamagnetic transition is accomplished without any major change in the electronic structure . in particular ,
large modifications of the @xmath9-derived dos can be excluded during the metamagnetic transition .
all experiments were performed at beamline u5ua at the national synchrotron light source ( nsls ) .
the beamline , equipped with a planar undulator and a high - resolution spherical - grating monochromator , is dedicated to spin - polarized angular - resolved photoemission ( sp - arpes ) studies.@xcite the incident light was linearly polarized in the horizontal plane .
a commercial hemispherical electron energy analyzer , coupled to a mini - mott polarimeter for spin analysis , was employed for these experiments .
the incoming photons , incident at 45@xmath10 from the sample normal , and the emitted ( detected ) electrons were in the horizontal plane .
the emission angle was along the sample surface normal .
the typical overall instrumental resolution was approximately 80 mev for spin integrated and 120 mev for spin - polarized photoemission spectra .
the sherman function of the spin - detector was estimated at 0.2 .
the ferh films were magnetized along the _ in - plane _ [ 010 ] ( vertical ) direction via a pulsed current passed through a magnetizing coil placed near the sample surface .
all magnetic measurements were performed in magnetic remanence .
ferh(001 ) films ( @xmath4 10ml ) were grown on a w(001 ) single crystal at rt by co - evaporation of fe and rh from two electron beam sources .
the evaporation rate was @xmath4 5 @xmath11/min .
the base pressure in the chamber was 5@xmath1210@xmath13 torr and rose to about 1@xmath1210@xmath14 torr during depositions .
after rt deposition , the leed was barely visible , consisting of large , diffused spots on a high background .
annealing to 600@xmath15 for approximately 2min restored a sharp cubic leed pattern ( see fig.[alloy - formation](a ) ) .
the pattern is in registry with the underlying w(001 ) substrate , indicating the formation of ordered films , exposing the ( 001 ) surface .
the annealing procedure has been limited because longer / higher annealing leads to substantial three dimensional island growth with partial uncovering of the substrate , as can be assessed by monitoring the w 4@xmath16 peaks .
= 54 ev ) and from an epitaxial film ( 10 ml ) of ferh(001 ) ( bottom ; @xmath17 = 57.2 ev ) .
( b ) photoemission spectra from shallow core levels ( fe 3@xmath2 and rh 4@xmath2 ) from equiatomic ferh films , before and after annealing .
the spectra from films of fe and rh metals are also shown for comparison .
( c ) valence band spectra corresponding to the spectra shown in ( b).,scaledwidth=47.0% ]
the formation of the ordered alloy is accompanied by considerable modifications in the electronic structure , easily detected in the photoemission spectra from core levels and the valence band ( see fig . [ alloy - formation](b ) and [ alloy - formation](c ) ) . at the 50:50 concentration
the fe 3@xmath2 signal is more intense than rh 4@xmath2 .
this is due to a combination of favorable fe photoemission cross sections at 130 ev ( @xmath18(fe@xmath19)/@xmath18(rh@xmath20 ) @xmath42 ) and to a rh signal spread over a wider energy range as a consequence of the larger spin - orbit coupling in the heavier element . in the _ as - grown _
films , the line - shape of the fe 3@xmath2 is similar to metallic fe . during the annealing process , the 3@xmath2 spectrum broadens and shifts toward higher binding energy .
these features represent the substitution of fe nearest - neighbors with rh atoms during the formation of the cscl cubic structure of ferh .
the annealing process does not change appreciably the rh - to - fe core level intensity ratio suggesting a negligible tendency to surface segregation , either of fe or rh .
a more quantitative assessment can not be safely obtained from our data due to the change in line shape of the fe @xmath21 and rh @xmath22 spectra in the alloy and in the pure metals and to their partial overlap .
the sharpening of all spectral features in the valence band spectrum upon annealing ( see fig .
[ alloy - formation](c ) ) can also be partially explained by the onset of long - range crystallographic order , with the consequent decrease of inelastic scattering events for the out - going electrons . in our case , however , a more substantial charge redistribution is taking place .
consistently with the formation of the ferh alloy at this stage , the form of the valence band is modified by the annealing process .
a new peak appears close to the fermi level and higher spectral intensity builds up at @xmath4 5 ev binding energy . at the relatively high photon energy used in these experiments ( @xmath17 = 170 ev ) , photoemission spectra represent , in first approximation , densities of states , mostly biased towards @xmath9 states .
the accumulation of intensity at high binding energy can be attributed to 3@xmath9 - 4@xmath9 majority spin hybridization during alloy formation .
indeed the majority @xmath9-bands largely overlap in fe and rh metals . by the same mechanism , the occupied fe minority spin states close to the fermi level
are pushed upward to become unoccupied , anti - bonding states .
the result is an enhancement of the fe magnetic moment , but also an increased separation between the fe and rh minority spin states .
largely unhybridized , rh minority spin states just below the fermi level therefore account for the sharp peak appearing in the spectra at approximately 0.5 ev binding energy . and rh 4@xmath2 core levels of an ferh film in the ferromagnetic phase .
inset shows a blow up of the rh @xmath22 region .
( b ) temperature dependence of the fe 3@xmath2 dichroism signal.,scaledwidth=47.0% ] the magnetic behavior of the ferh films can be monitored using magnetic linear dichroism in core - level photoemission.@xcite these spectra are shown in fig .
[ magnetism](a ) for ferh film in the ferromagnetic phase .
the two continuous lines ( black and gray ) correspond to the photoemission spectra measured with up and down magnetization directions .
the difference curve , the dichroism signal , is also shown .
clearly the fe dichroism signal is much larger than the rh one ( see fig . [
magnetism](a ) inset ) .
this is consistent with the large difference in local magnetic moments on fe ( @xmath43@xmath23 ) and rh ( @xmath41@xmath23 ) sites .
the similar shape of the difference curves of the two dichroisms ( within their binding energy ranges ) indicates the parallel alignment of the fe and rh moments .
the temperature dependence of the fe 3@xmath2 dichroism is shown in fig .
[ magnetism](b ) ( bottom ) in the range 100@xmath4600k .
we note that a pulsed magnetic field before each series of measurements as a function of temperature was applied .
it is clear that the @xmath24 transition is present in the ultrathin films prepared in this study .
the critical temperature of the metamagnetic transition , separating the antiferromagnetic state ( low temperature , no dichroism ) from the ferromagnetic state ( high temperature , finite dichroism ) , is @xmath4 50k above room temperature .
the main features of the transitions
e.g. the critical temperature and range of hysteretic behavior@xcite are remarkably close to the values found in bulk ferh alloys .
only the steepness of the transition increases in the single - crystalline films ; possibly an indication of the intrinsic first order nature of the transition .
it is relevant to recall that photoemission at the excitation energies used in our experiments probes at most the first few monolayers of material .
these results therefore demonstrate that the metamagnetic transition is present also at the ferh surface .
furthermore , considering the lattice strain usually associated with epitaxial thin films particularly at their surfaces and interfaces
these observations suggests that a lattice instability is a consequence rather than the driving force of the transition . and rh 4@xmath25 core level spectra from fe and rh metals and from equiatomic ferh alloy in the af ( solid line ) and fm ( open circles ) phases ; ( b ) similar to ( a ) but for fe 3@xmath2 and rh 4@xmath2 levels ; ( c ) normal emission valence band spectra taken close to the high - symmetry @xmath26 and x point.,scaledwidth=45.0% ] above the metamagnetic transition , further increasing the temperature , the magnetic signal diminishes . the curie temperature , above which the system is paramagnetic , is at approximately 530k
it is interesting to note that at @xmath4 430k the slope of the curie curve appears to change slightly , accelerating the decay of the magnetization .
this behavior , if further confirmed via a more direct probe of the magnetization , could indicate the disappearance of the local rh moment via a high- to low - spin state transition .
the existence of these two states for the rh atoms has been suggested in a recent model proposed to explain the ferh thermomagnetic behavior.@xcite and x points that are marked with arrows ( up / down : majority / minority ) .
the exchange splitting ( @xmath27 ) is estimated to be around 1.4 ev .
see text for details .
the clearest feature is perhaps the @xmath28 sharp minority state close to the fermi level ( @xmath4 0.2 ev binding energy ) in the spectra.,scaledwidth=44.0% ] the effect of the af@xmath1fm transition on the electronic structure is shown in fig .
[ transition ] .
the photoemission spectra from films in the antiferromagnetic state ( continuous lines , @xmath4100k ) and in the ferromagnetic state ( open circles , @xmath4380k ) are superimposed to facilitate the comparison .
the magnetic transition has remarkably little effect on all photoemission spectra ; core levels ( @xmath25-levels , top ; @xmath2-levels , middle ) as well as valence band ( bottom ) .
the small differences in the spectra are mostly attributable to thermal broadening . in fig .
[ transition](a ) , the 3@xmath25 and 4@xmath25 spectra of fe and rh metals are compared to the ferh one . the overall shift of these levels ( @xmath40.5 ev ) reflects an increased local electrostatic potential in the alloy .
furthermore , the distance between the main peak and the higher energy satellite of fe 3@xmath25 increases from 4.5 ev in the metal to 5.1 ev in the alloy . due to the absence of orbital contributions in the @xmath25-levels , the distance between the main peak and the higher energy satellite is proportional to the local spin moment on the fe site.@xcite this observation therefore implies a larger fe magnetic moment in the alloy .
the shape of the fe 3@xmath25 core level remains essentially unchanged during the metamagnetic transition .
a larger magnetic moment is therefore present in both phases of the alloy .
these observations agree with calculations of the electronic structure which give a fe magnetic moment of approximately 3 @xmath23 in ferh to be compared with 2.2 @xmath23 of fe metal .
unfortunately , the signal from the rh 4@xmath25 level is too weak to permit an unambiguous assessment on the presence of a high energy satellite , as would be expected during the formation of a rh magnetic moment in the fm state .
similarly , as in the case of the @xmath25-levels , the @xmath2-levels ( fig.[transition](b ) ) are also mostly unaffected by the metamagnetic transition .
the angular resolved photoemission spectra from the valence band are shown in fig .
[ transition](c ) .
the spectra are measured in normal emission ( i.e. @xmath29=0 ) from the ( 100 ) surface , and probe states along the bulk @xmath30 line .
the two photon energies shown sample widely spaced regions of the brillouin zone ( bz ) . assuming a free electron final state , 63 and 35
ev photon energies correspond to states close to the @xmath26-point and x - point , respectively .
similarly to the core levels , the valence band remains largely unchanged during the metamagnetic transition .
again the major effect is thermal broadening , easily visible at the fermi level .
this behavior is not limited to the two photon energies shown ; we have explored the photon energy range between 20 to 80 ev and found essentially the same result .
the spin - polarized electronic structure in the fm phase can be examined with spin - resolved photoemission .
the spin - resolved valence band spectra , measured under the same conditions as fig.[transition](c ) are shown in fig .
[ spin ] . these well polarized spectra
are quite complex , displaying several overlapping features in both minority and majority spin channels . as a first step in interpreting these spectra ,
it is useful a comparison with band structure calculations . in magnetic 3@xmath9 metals , however , electronic correlation have been shown to affect the photoemission spectra@xcite and a more refined analysis would need to take them into proper account .
the band structure caculation were performed using the wien2k package,@xcite with default settings and a lattice constant a = 2.985 .@xcite the band structure along the high symmetry @xmath30-line is displayed in fig .
[ band - structure ] .
the general form of the bands resembles a typical bcc elemental ferromagnet ( e.g. fe metal ) , doubly folded because of the two elements in ferh.@xcite this folding is indeed the reason why the photoemission spectra contains so many features .
figure [ band - structure ] also shows spin - resolved density of states ( dos ) which was calculated with a 21@xmath1221@xmath1221 mesh in the bz , with effective 286 _ k_-points in the irreducible bz , and plotted with 40 mev gaussian broadening .
the general shape of the dos and its magnitude compare favorably with previous calculations,@xcite with small alterations most likely related to the different computation methods used . in the experimental geometry
used , only initial states of @xmath31 or @xmath32 symmetry are dipole - allowed.@xcite their positions at the two high - symmetry points , @xmath26 and x , are marked with arrows ( up / down : majority / minority ) in fig .
although the presence of several broad and overlapping features do not allow for an unambiguous assignment of all features , there is generally a good correspondence between the experimental features and calculated bands .
the clearest feature is perhaps the @xmath28 sharp minority state close to the fermi level ( @xmath40.2 ev binding energy ) in the spectra at 35 ev , which is not present at 63 ev . along the @xmath30-line
the @xmath28 band disperses downward , being unoccupied at @xmath26 and occupied at x ( see fig.[band - structure ] ) .
the majority counterpart of this state is located at a binding energy of @xmath4 2.5 ev , giving a rather large exchange splitting ( @xmath27 = 2.3 ev , see fig . [ spin ] ) .
the 3@xmath9 - 4@xmath9 hybridization in this system results in an unusually variable exchange . besides states with high exchange splitting , other @xmath9 states have a much lower exchange .
an example of this are the minority and majority states located at 1.8 and 3.2 ev , respectively ; giving an exchange splitting of only 1.4 ev ( see fig.[spin ] ) .
it would be tempting to attribute states of high and low exchange to fe and rh states , respectively .
this picture however is oversimplified as can be concluded from the site - projected calculations shown in fig .
[ band - structure ] where bands are highlighted according to the weight on fe - sites or on rh - sites .
the main features emerging from fig . [ band - structure ] are that majority bands are roughly equi - distributed on the fe and rh sites while the minority bands are split into two groups : rh - like in the lower portion and fe - like close to the fermi level .
this behavior stems from the different degree of hybridization of majority and minority states mentioned above .
the main factor determining the charge distribution in the alloy is the energy location of the bands previous to the alloy formation rather than the spatial localization .
the majority bands of fe and rh , largely degenerate , strongly hybridize forming delocalized states equally distributed on the fe and rh sites .
the minority bands , hybridizing much less , remain localized , forming bonding states on the rh sites and anti - bonding states on the fe sites.@xcite it is therefore meaningless to attribute exchange pairs to either fe or rh .
as mentioned in the introduction , there is a vast literature on the origin of the metamagnetic transition in ferh .
it is therefore worth to discuss our results in the context of previously published work .
a primary feature of our experimental work is the investigation of epitaxial films .
the metamagnetic transition is accompanied by a relatively large ( in the range of @xmath4 0.51%)@xcite and isotropic lattice expansion . moreover ,
the transition temperature can be varied and the transition itself can even be suppressed by external pressure applied to ferh samples .
these observations have lead some researchers to link the transition to a lattice instability , somewhat along the lines of the invar phenomena in feni.@xcite it seems therefore appropriate to attempt ferh experiments on single crystalline samples ( epitaxial films ) consisting of only a few monolayers ( @xmath4 10 ml in the present films ) .
the ferh bulk lattice constant is 2.99 @xmath11 while the w lattice parameter is 3.15 @xmath11 .
the anticipated consequence of this 5% mismatch is the substantial lattice strain if pseudomorphic growth is assumed .
the fact that the magnetic properties of these ultrathin films are essentially the same as those of much thicker polycrystalline films , or even bulk ferh , is a strong indication that the lattice change may not be a decisive factor of the metamagnetic transition , but rather a consequence of it . the considerable change in the specific heat and resistivity at the metamagnetic transition had been noticed early on in the study of ferh.@xcite these measurements are challenging and often have to rely on comparison of samples with slightly different stoichiometry .
nevertheless , these experimental results suggest profound changes in the electronic structure .
redistribution of charge could account for large changes of the electrical and thermal properties of the material undergoing the metamagnetic transition.@xcite indeed , early electronic structure calculations displayed a large change of the density of states across the transition.@xcite however , core level , high - energy valence band ( xps ) , and low - energy valence band ( arpes ) photoemission spectra do not show any significant differences across the transition . although the relatively high temperature needed to reach the fm phase limits the resolution to about 200 mev , at least within this limit , large charge rearrangements can be excluded .
the direct sampling of the electronic structure using photoemission therefore substantiates and extends previous findings obtained with ellipsometry measurements@xcite and with positron annihilation techniques@xcite on bulk samples .
they also agree with modern first principle calculations of the electronic structure which find similar charge transfer from rh to fe in both magnetic phases as well as similar values for both majority and minority spin density of states at the fermi energy.@xcite
in conclusion , equiatomic ferh(100 ) films ( cscl structure , @xmath4 10 ml ) have been epitaxially grown on w(100 ) by room temperature co - deposition of fe and rh , followed by moderate annealing .
the formation of the ordered alloy is accompanied by profound modifications of the electronic structure , displayed in valence band and shallow core level photoemission spectra .
the epitaxial ultra - thin films and particularly their surfaces display magnetic properties similar to bulk ferh .
the temperature - induced @xmath33 transition is clearly detected via magnetic linear dichroism in photoemission .
surprisingly , the metamagnetic transition is accomplished without major changes in the electronic structure : valence band and core levels remain practically unaffected crossing the critical temperature . in the fm phase , the valence band spectra display well polarized features , in good agreement with spin - polarized band structure calculations .
the nsls , brookhaven national laboratory , is supported by the u.s .
doe , office of science , office of basic energy sciences , under contract de - ac02 - 98ch10886 .
the hysteretic behaviorare known to be affected by structural change ( a 1% unit cell volume expansion ) , including the stress state ( j. van driel , r. coehoorn , g. j. strijkers , e. bruck , and f. r. de boer , j. appl . phys . * 85 * , 1026 ( 1999 ) . ) and the sample s microstructural scale ( k. kang , a. r. moodenbaugh , and l. h. lewis , appl . phys . lett . * 90 * , 153112 ( 2007 ) . ) .
j. snchez - barriga , j. fink , v. boni , i. di marco , j. braun , j. minr , a. varykhalov , o. rader , v. bellini , f. manghi , h. ebert , m. i. katsnelson , a. i. lichtenstein , o. eriksson , w. eberhardt , and h. a. drr , phys .
lett . * 103 * , 267203 ( 2009 ) .
j. snchez - barriga , j. minr , j. braun , a. varykhalov , v. boni , i. di marco , o. rader , v. bellini , f. manghi , h. ebert , m. i. katsnelson , a. i. lichtenstein , o. eriksson , w. eberhardt , h. a. drr , and j. fink , phys .
b * 82 * , 104414 ( 2010 ) . p. blaha , k. schwarz , g. k. h. madsen , d. kvasnicka , and j. luitz , _ an augmented plane wave + local orbitals program for calculating crystal properties ( karlheinz schwarz , techn .
universitat wien , austria ) isbn 3 - 9501031 - 1 - 2 _ ( 2000 ) . the ferh cscl - structure can be thought as a simple bcc structure with every central atom substituted with the other element . in first approximation ,
the resulting band structure is therefore the one of a bcc crystal with an average electrostatic potential between fe and rh .
the doubling of the lattice periodicity , due to the different elements , additionally results in folding the bands in reciprocal space . | epitaxial ferh(100 ) films ( cscl structure , @xmath0 thick ) , prepared _ in - situ _ on a w(100 ) single crystal substrate , have been investigated via valence band and core level photoemission .
the presence of the temperature - induced , first - order , antiferromagnetic to ferromagnetic ( af@xmath1 fm ) transition in these films has been verified via linear dichroism in photoemission from the fe 3@xmath2 levels .
core level spectra indicate a large moment on the fe atom , practically unchanged in the fm and af phases . judging from the valence band spectra
, the metamagnetic transition takes place without substantial modification of the electronic structure . in the fm phase ,
the spin - resolved spectra compare satisfactorily to the calculated spin - polarized bulk band structure . |
all graphs considered in this paper are finite , undirected and simple .
let @xmath2 be a graph with vertex set @xmath3 and edge set @xmath4 . denote by @xmath5 the adjacency matrix of @xmath0 .
the eigenvalues of @xmath0 are those of @xmath5 .
an eigenvalue of a graph @xmath0 is called a main eigenvalue if it has an eigenvector the sum of whose entries is not equal to zero .
it is well known that a graph is regular if and only if it has exactly one main eigenvalue .
a long - standing problem posed by cvetkovic ( @xcite ) is that of how to characterize graphs with exactly @xmath6 main eigenvalues .
hagos @xcite gave an alternative characterization of graphs with exactly two main eigenvalues .
recently , hou and zhou @xcite characterized the trees with exactly two main eigenvalues .
a vertex of a graph @xmath0 is said to be pendant if it has degree one .
denote by @xmath7 and @xmath8 the cycle and path of order @xmath9 , respectively .
a connected graph is said to be tricyclic ( resp . , unicyclic and bicyclic ) , if @xmath10 ( resp . ,
@xmath11 and @xmath12 ) .
hou and tian @xcite showed that the graphs @xmath13 for some positive integers @xmath14 with @xmath15 , where @xmath13 is the graph obtained from @xmath16 by attaching @xmath17 pendant vertices to every vertex of @xmath16 , are the only connected unicyclic graphs with exactly two main eigenvalues .
@xcite and shi @xcite characterized all connected bicyclic graphs with exactly two main eigenvalues independently .
this paper will continue the line of this research and determine all connected tricyclic graphs with exactly two main eigenvalues . for any tricyclic graph @xmath0 , the base of @xmath0 , denoted by @xmath18 is the minimal tricyclic subgraph of @xmath0 .
clearly , @xmath18 is the unique tricyclic subgraph of @xmath0 containing no pendent vertex , and @xmath0 can be obtained from @xmath18 by attaching trees to some vertices of @xmath18 .
it follows from @xcite that there are 8 types of bases for tricyclic graphs , say , @xmath19 , which are depicted in fig . 1 . , title="fig : " ] +
in this section , we will present some notations and known results which will be used in the next section .
the reader is referred to @xcite for any undefined notation and terminology on graphs in this paper .
let @xmath0 be a graph . as usual
, we denote by @xmath20 and @xmath21 the degree of vertex @xmath22 and the set of all neighbors of @xmath22 in @xmath0 .
let @xmath23 a graph @xmath0 is called 2-walk @xmath1-linear if there exist unique rational numbers @xmath24 such that @xmath25 holds for every vertex @xmath26 . an internal path of @xmath0 is a walk @xmath27 such that the vertices @xmath28 are distinct , @xmath29 , and @xmath30 for @xmath31 .
an internal path is called an internal cycle if @xmath32 . if @xmath33 is a path or a cycle of @xmath0 , the length of @xmath33 , denoted by @xmath34 , is defined as the number of edges of @xmath33 .
( @xcite ) .
[ 222ab ] a graph @xmath0 has exactly two main eigenvalues if and only if @xmath0 is 2-walk @xmath1-linear .
( @xcite ) . [ 21 ] let @xmath0 be a 2-walk ( a , b)-linear graph
. then both @xmath35 and @xmath36 are integers .
( @xcite)[22 ] .
let @xmath0 be a 2-walk @xmath1-linear graph and @xmath37 be two vertices of @xmath0 with unequal degree @xmath38 , respectively .
then @xmath39 in the following , for convenience , we always assume that @xmath40 is the set of tricyclic graphs with exactly two main eigenvalues , @xmath41 is a pendent vertex of @xmath0 ( if exist ) . for each @xmath42 , let @xmath43 be the graph obtained from @xmath0 by deleting all pendent vertices . from the proof lemmas 3.1 - 3.7 in @xcite , we know that those lemmas are also hold for tricyclic graphs .
hence we have the following two lemmas .
[ 31 ] let @xmath44 and @xmath45 be an internal path of length at least @xmath46 in @xmath0
. then @xmath47 .
in particular , if @xmath48 , then there exists no path @xmath49 in @xmath0 such that @xmath50 and @xmath51 .
[ 3no ] let @xmath42 and @xmath52 .
then \(i ) @xmath53 ( see fig .
1 ) ; \(ii ) @xmath54 or @xmath55 ; \(iii ) if @xmath0 has at least one pendent vertex , then @xmath56 and @xmath57 ; \(iv ) for a cycle @xmath58 of @xmath0 with @xmath59 , @xmath60 , if @xmath0 has at least one pendent vertex , then there is an integer @xmath61 such that @xmath62 .
in this section , we will determine all tricyclic graphs with exactly two main eigenvalues . by lemma [ 222ab ] ,
it is sufficient to determine all 2-walk @xmath1-linear tricyclic graphs . * proof . * ( i ) by way of contradiction , assume that there is an integer @xmath79 such that @xmath80 . without loss of generality ,
suppose that @xmath81 is the smallest integer such that @xmath82 .
by lemma [ 3no ] ( ii ) , we may assume that @xmath83 and @xmath84 . hence @xmath85 .
applying with @xmath86 , we have @xmath87 this together with lemma [ 3no ] ( iii ) implies that @xmath88 by lemma [ 3no ] ( ii ) , we have @xmath89 for @xmath90 . thus @xmath91 . if @xmath94 , then @xmath95 .
it follows from that @xmath96 .
this together with lemma [ 3no ] ( iii ) implies that @xmath97 .
note that @xmath68
. then @xmath98 by . on the other hand , @xmath99
, so @xmath100 by lemma [ 3no ] ( ii ) .
thus by , @xmath101 , a contradiction .
\(iii ) by lemma [ 3no ] ( ii ) , we have @xmath113 . if @xmath114 , then @xmath115 for @xmath116 by ( i ) .
thus @xmath117 by lemma [ 3no ] ( iv ) .
applying with @xmath118 , we have @xmath119 note that @xmath120 or @xmath121 and @xmath122 .
it follows from that @xmath35 can not be an integer .
it contradicts lemma [ 21 ] .
hence @xmath68 .
therefore @xmath69 and @xmath123 .
\(iv ) by lemma [ 3no ] ( ii ) , @xmath113 .
if @xmath68 , then @xmath69 and @xmath128 by ( i ) and ( ii ) .
if @xmath114 , then @xmath115 for @xmath116 by ( i ) .
so @xmath129 by lemma [ 3no ] ( iv ) .
applying with @xmath118 , we have @xmath130 .
thus @xmath131 and @xmath132 by lemmas [ 21 ] and [ 3no ] ( iii ) .
suppose that @xmath109 .
then @xmath133 by ( ii ) .
applying with @xmath134 , we have @xmath73 .
it is a contradiction .
therefore @xmath69 .
@xmath135 * case 1*. @xmath144 . if @xmath0 has no pendent vertex , then @xmath145 ( see fig . 3 ) . by
, @xmath146 is 2-walk @xmath147-linear .
if @xmath0 has at least one pendent vertex , say @xmath41 , then @xmath148 since @xmath83 for @xmath149 by lemma [ p ] ( iii ) .
it follows from lemma [ 3no ] ( ii ) that @xmath150 . applying with @xmath151
, we have @xmath152 .
this contradicts lemma [ 3no ] ( iii ) . *
subcase 1*. @xmath0 has no pendent vertex
. then @xmath154 by and .
hence @xmath155 .
it implies that @xmath156 .
similarly , we have @xmath157 and @xmath158 . by lemma [ 31 ] , we have @xmath159 . without loss of generality ,
suppose that @xmath160 . if @xmath161 , then @xmath162 .
hence @xmath163 by . on the other hand , @xmath164
, so @xmath165 by , a contradiction .
hence @xmath166 .
similarly , we have @xmath167 .
therefore @xmath168 ( see fig . 3 ) .
by , @xmath169 is 2-walk @xmath170-linear . *
subcase 2*. @xmath0 has at least one pendent vertex . in this case , we show that there is no such graph with exactly two main eigenvalues .
since @xmath171 , we have @xmath73 by lemma [ p ] ( iii ) .
so @xmath83 for @xmath149 by lemma [ p ] ( iii ) and ( iv ) .
we claim that @xmath172 .
otherwise , let @xmath173
. then @xmath174 by and .
it follows from lemma [ 3no ] ( ii ) and the fact that @xmath175 that @xmath176 .
hence @xmath177 by . by , @xmath178 if @xmath179 , then @xmath180 .
this is impossible by lemma [ p ] ( i ) . if @xmath158 , then @xmath181
obviously , @xmath182 when @xmath183 and @xmath184 when @xmath185 by lemma [ p ] ( i ) .
hence @xmath186 , it contradicts the fact that @xmath187 .
therefore @xmath156 .
dually , we have @xmath158 . if @xmath191 .
applying with @xmath192 , we have @xmath127 .
. furthermore , @xmath194 by and .
thus @xmath195 .
it follows from lemma [ p ] ( i ) that @xmath196 for @xmath197 .
similarly , we have @xmath198 for @xmath199 and @xmath200 .
note that @xmath201 .
we have @xmath202 and @xmath203 by . on the other hand , @xmath204 by , a contradiction .
* proof . * if @xmath210 , then @xmath214 and so each cycle of @xmath43 has the length of @xmath63 by lemmas [ 31 ] and [ p ] ( iii ) . hence @xmath215 and @xmath83 for @xmath216 ( see fig . 2 ) , where @xmath217 . for convenience , we set @xmath218 and @xmath219 . by and ,
hence @xmath221 . next ,
. then @xmath232 . by and ,
so @xmath234 and @xmath235 .
similarly , @xmath236 .
so @xmath237 .
note that @xmath238 or @xmath239 , we have @xmath240 and @xmath241 .
therefore , @xmath242 ( see fig . 3 ) . by ,
@xmath243 is 2-walk ( 2,1)-linear . if @xmath254 , then @xmath255
. so @xmath256 by and . by lemma [ 3no ]
( ii ) , @xmath257 . without loss of generality ,
suppose that @xmath258 .
then @xmath259 for @xmath199 by lemma [ p ] ( i ) . for the vertex @xmath260 , @xmath261 by and .
so @xmath262 .
. hence @xmath166 by lemma [ p ] ( ii ) . therefore @xmath263
( see fig .
4 ) , where @xmath264 .
it is easy to see that any graph @xmath265 is 2-walk @xmath229-linear . if @xmath266 . in this case , we show that there is no such graph with exactly two main eigenvalues . applying with @xmath248 and @xmath267 , respectively .
we have @xmath73 and @xmath268 , respectively .
hence @xmath269 . by lemma [ 3no ] ( ii ) and
the fact that @xmath270 , we have @xmath271 .
we first show that @xmath232 .
otherwise , let @xmath277
. then @xmath206 by lemma [ 3no ] ( iii ) .
applying with @xmath267 , we have @xmath278 .
note that @xmath206 and @xmath279 or @xmath63 .
we have @xmath280 .
it contradicts lemma [ 3no ] ( ii ) .
hence @xmath232 .
it implies that @xmath281 .
next , applying with @xmath248 , we have @xmath127
. so @xmath131 and @xmath193 . for the vertex @xmath282 , @xmath283 by and
so @xmath284 .
thus @xmath285 for @xmath286 by lemma [ p ] ( i ) .
note that @xmath287 .
we have @xmath288 by lemma [ 3no ] ( ii ) .
hence @xmath285 for @xmath289 .
* proof . * if @xmath299 , then @xmath301 and so the cycle of @xmath43 has the length of @xmath63 by lemmas [ 31 ] and [ p ] . hence @xmath302 ( see fig . 2 ) , where @xmath303 . for convenience ,
we set @xmath304 and @xmath305 .
we first show that @xmath0 contains at least one pendent vertex . on the contrary ,
suppose that @xmath0 has no pendent vertex
. then @xmath306 by lemma [ 31 ] .
without loss of generality , suppose that @xmath307 . if @xmath225 , then @xmath308 and @xmath309 or @xmath310 by .
so @xmath311 .
on the other hand , @xmath312 by , a contradiction .
if @xmath281 , then @xmath313 . so @xmath314 and @xmath315 or @xmath316 by . on the other hand , @xmath317 by , a contradiction
therefore , @xmath0 has at least one pendent vertex .
we consider the following two cases : if @xmath318 , then @xmath320 by lemma [ p ] ( iv ) .
so @xmath321 by lemma [ 3no ] ( ii ) .
it follows from lemma [ p ] ( i ) that @xmath322 . without loss of generality ,
suppose that @xmath323 or @xmath63 .
then @xmath324 by and .
note that @xmath325 or @xmath63 by lemma [ 3no ] ( ii ) .
we have @xmath326 .
so @xmath327 by and .
thus @xmath284 .
it implies that @xmath328 .
therefore @xmath300 ( see fig . 3 ) . by
, @xmath329 is 2-walk @xmath330-linear .
* subcase 1*. @xmath331 . without loss of generality ,
suppose that @xmath332
. then @xmath333 and @xmath334 for @xmath335 by lemma [ p ] ( i ) .
note that @xmath336 .
we have @xmath337 by lemma [ 3no ] ( ii ) .
hence @xmath338 we next show that @xmath339 .
if @xmath340 , then @xmath341 , @xmath342 .
so @xmath343 . if @xmath344 , clearly , @xmath339 .
if @xmath345 , then @xmath339 by lemma [ p ] ( i ) .
therefore @xmath339 for all @xmath158 . *
subcase 2*. @xmath349 . without loss of generality ,
suppose that @xmath350 and @xmath231 or @xmath63 .
we claim that @xmath351 .
otherwise , let @xmath352
. then @xmath353 by and , which is impossible since @xmath354 .
hence @xmath351 .
similarly , we have @xmath355 . if @xmath357 . applying lemma [ 3no ] ( iv ) with @xmath358 , we have @xmath359 .
so @xmath191 .
applying with @xmath151 and @xmath248 , respectively .
we have @xmath360 and @xmath73 , respectively .
hence @xmath361 , a contradiction .
if @xmath362 , then @xmath363 by applying with @xmath248 .
it follows from lemma [ 3no ] ( iii ) that @xmath131 or @xmath239 . if @xmath131 , then @xmath364 .
hence @xmath365 by and .
this is impossible since @xmath191 by lemma [ 3no ] ( ii ) . if @xmath254 , then @xmath366 .
thus @xmath367 by and .
this is also impossible since @xmath191 or @xmath239 . *
case 2*. @xmath281 .
we show that in this case there is no such graph with exactly two main eigenvalues . by lemma [ p ] ( iii ) , we have @xmath73 and @xmath368 . applying with @xmath248 , we have @xmath369 .
so @xmath370 .
applying with @xmath371 , we have @xmath372 . hence @xmath373 . by lemma [ 3no ]
( ii ) , @xmath374 . if @xmath376 , then @xmath131
. so @xmath193 . by lemmas [ 3no ] ( ii ) and [ p ] ( i ) , we have @xmath377 and @xmath378
. hence @xmath379 by and .
it follows that @xmath380 .
thus @xmath381 by .
on the other hand , @xmath382 by , a contradiction .
@xmath135 * proof . *
if @xmath383 , then @xmath386 and so the cycle of @xmath43 has the length of @xmath63 by lemmas [ 31 ] and [ p ] ( iii ) .
hence @xmath387 ( see fig . 2 ) , where @xmath388 and @xmath389 . for convenience ,
we set @xmath390 and @xmath391 .
* case 1*. @xmath173
. then @xmath393 by .
we claim that @xmath394 .
otherwise , let @xmath395 .
then @xmath396 and @xmath397 or @xmath316 by .
so @xmath398 .
on the other hand , @xmath399 by , a contradiction .
hence @xmath394 . similarly , we have @xmath400 or @xmath63 .
if @xmath340 , then applying with @xmath401 and @xmath402 , respectively , we have @xmath73 and @xmath403 , respectively . a contradiction . if @xmath344 , then @xmath404 ( see fig
, @xmath405 is 2-walk @xmath229-linear . if @xmath231 , then @xmath408 by and .
so @xmath409 .
it implies that @xmath410 . hence @xmath411
otherwise , let @xmath295
. then @xmath412 by . on the other hand , @xmath177 by , a contradiction .
therefore @xmath411 and @xmath413 ( see fig . 3 ) . by
, @xmath414 is 2-walk @xmath415-linear . if @xmath428 , then @xmath254 by . note that @xmath73 .
we have @xmath429 and @xmath430 . by lemma [ p ] , @xmath431 for @xmath432 .
so @xmath433 by and .
thus @xmath434 .
similarly , we have @xmath435 . by and ,
this is impossible since @xmath437 or @xmath239 . if @xmath438 , then @xmath439 by . recall that @xmath73 , we have @xmath440 and @xmath441 . note that @xmath442 .
we have @xmath443 .
thus @xmath444 by and .
by lemma [ 3no ] ( ii ) , @xmath445 . hence @xmath446 .
therefore @xmath447 , which is impossible since @xmath448 .
applying with @xmath248 , we have @xmath452 .
so @xmath453 .
applying with @xmath267 , we have @xmath278 .
thus @xmath454 .
note that @xmath455 by lemma [ 3no ] ( ii ) .
we have @xmath284 or @xmath55 .
it follows that @xmath131 and @xmath193 . * case 1*. there is a vertex @xmath470 such that @xmath471 and @xmath472 . without loss of generality ,
let @xmath473 .
then @xmath474 for @xmath199 by lemma [ p ] ( i ) .
in particular , @xmath475 by lemma [ 3no ] ( iii ) . we first show that @xmath476 . applying lemma [ 3no ] ( iv ) with @xmath490 , we have @xmath491 and @xmath492 for some @xmath432 .
it follows from lemmas [ 3no ] ( ii ) and [ p ] ( i ) that @xmath493 for @xmath432 . applying with @xmath478 and @xmath151 , respectively .
we have @xmath73 and @xmath494 , respectively .
this together with lemma [ 3no ] ( ii ) and the fact that @xmath495 implies that @xmath496 . hence @xmath497
. therefore @xmath254 and @xmath255 . by and ,
@xmath498 . by lemma [ 3no ]
( ii ) , @xmath499 or @xmath239 . thus @xmath500
. hence @xmath501 by and , which is impossible since @xmath502 or @xmath239 .
* subcase 1*. @xmath508 . without loss of generality ,
suppose that @xmath509
. then @xmath477 by lemma [ p ] ( i ) .
note that @xmath175 .
we have @xmath510 by lemma [ 3no ] ( ii ) .
thus @xmath0 has at least one pendent vertex . applying with @xmath478 and @xmath151 , respectively .
we have @xmath73 and @xmath511 , respectively .
hence @xmath512 .
we claim that @xmath523 or @xmath63 .
otherwise , let @xmath524 .
without loss of generality , suppose that @xmath525 . then @xmath526 and @xmath527 .
note that @xmath528 . we have @xmath347 , a contradiction .
hence @xmath523 or @xmath63 .
we consider the following two cases : * subcase 2.1*. @xmath0 has no pendent vertex .
then @xmath161 .
otherwise , suppose that @xmath529 . then @xmath237 . by and ,
this is impossible since @xmath262 or @xmath63 .
hence @xmath161 .
if @xmath523 .
applying with @xmath531 and @xmath402 , respectively .
we have @xmath73 and @xmath403 , respectively . a contradiction .
if @xmath532 , also applying with @xmath531 and @xmath402 , respectively , we have @xmath533 and @xmath403 , respectively . also a contradiction . if @xmath537 .
applying with @xmath267 and note that @xmath522 , we have @xmath538 .
if @xmath539 , then @xmath540 is not an integer , which is impossible by lemma [ 21 ] . if @xmath541 , then @xmath254 and @xmath255 .
so @xmath542 by .
on the other hand , @xmath543 by , a contradiction . if @xmath417 .
applying with @xmath478 , we have @xmath544 . if @xmath539 , then @xmath545 . for the vertex @xmath461
, we have @xmath546 .
it follows that @xmath547 , which is impossible by lemma [ 3no ] ( ii ) . if @xmath548 , then @xmath363 .
note that @xmath549 . we have @xmath439 and @xmath550 by lemma [ 3no ] ( iii ) . by and ,
this is impossible since @xmath522 . *
case 1*. there is a vertex @xmath559 such that @xmath471 and @xmath472 . without loss of generality ,
suppose that @xmath560 . applying with @xmath561 , we have @xmath562 by lemmas [ 3no ] ( ii ) , @xmath563 or @xmath55 .
we consider the following three cases : we now show that @xmath565 for @xmath566 . if @xmath183 , then obviously , @xmath565 for @xmath566 .
if @xmath185 , then @xmath567 for @xmath568 and @xmath569 by lemma [ p ] ( i ) .
hence @xmath565 for @xmath566 .
applying lemma [ 3no ] ( iv ) with @xmath570 , we have @xmath571 for some @xmath572 .
it follows from lemmas [ 3no ] ( ii ) and [ p ] ( i ) that @xmath573 and @xmath574 for all @xmath572 .
thus @xmath575 .
note that @xmath73 .
we have @xmath576 .
so @xmath345 .
it follows from lemma [ p ] ( ii ) that @xmath509 .
similarly , we have @xmath577 and @xmath578 . * subcase 3*. @xmath586 . then @xmath183 and @xmath587 .
it follows from lemma [ 3no ] ( iii ) that @xmath131 and @xmath78 .
thus @xmath588 .
for the vertex @xmath461 , @xmath589 by and . note that @xmath590 .
we have @xmath591 .
so @xmath592 .
thus @xmath593 .
it implies that @xmath344 .
similarly , we have @xmath594
. therefore @xmath595 ( see fig . 3 ) . by
, @xmath596 is 2-walk @xmath330-linear . *
case 2*. for any vertex @xmath559 , @xmath597 or @xmath504 . by lemma
[ 31 ] , @xmath598 . without loss of generality ,
suppose that @xmath599
. then @xmath600 and @xmath601 . by and ,
thus @xmath603 .
it implies that @xmath604 .
similarly , we have @xmath605 and @xmath606 or @xmath46 .
hence @xmath607 or @xmath608 .
if @xmath0 has no pendent vertex , then @xmath609 for @xmath610 ( see fig . 3 ) . by , @xmath611 is 2-walk @xmath147-linear , @xmath612 is 2-walk @xmath613-linear , @xmath614 is 2-walk @xmath229-linear and @xmath615 is 2-walk @xmath230-linear . if @xmath0 has at least one pendent vertex @xmath41 , then @xmath148 or @xmath616 .
without loss of generality , suppose that @xmath617 . then @xmath618 .
applying with @xmath151 , we have @xmath619 . by lemma [ 3no ]
( ii ) , @xmath620 or @xmath55 .
it implies that @xmath536 .
this is impossible by lemma [ 3no ] ( iii ) .
if @xmath0 has no pendent vertex , then @xmath628 by lemma [ 31 ] . without loss of generality , suppose that @xmath629
. then @xmath183 or @xmath239 . by and ,
note that @xmath631 .
we have @xmath632 .
it implies that @xmath633 .
similarly , we have @xmath634 . if @xmath183 , then @xmath635 or @xmath63 by lemma [ 31 ] .
hence @xmath138 for @xmath636 ( see fig . 3 ) .
if @xmath166 , then @xmath637 or @xmath239 by lemma [ 31 ] . hence @xmath638 for @xmath639
( see fig .
, @xmath640 is 2-walk @xmath229-linear , @xmath641 and @xmath642 are 2-walk @xmath230-linear , @xmath643 is 2-walk @xmath644-linear , @xmath645 is 2-walk @xmath646-linear , @xmath647 and @xmath648 are 2-walk @xmath415-linear , @xmath649 is 2-walk @xmath170-linear . if @xmath652 . by way of contradiction , suppose that @xmath653 . by lemma [ 3no ] ( ii ) , we may assume that @xmath654 without loss of generality .
applying with @xmath655 , we have @xmath656 by lemma [ 3no ] ( ii ) , @xmath657 or @xmath55 . if @xmath658 or @xmath659 , then @xmath660 since @xmath534 .
so @xmath661 .
it implies that @xmath254 and @xmath255 . by and ,
so @xmath663 and @xmath542 by .
on the other hand , @xmath664 by , a contradiction . if @xmath665 , then @xmath666 or @xmath660 since @xmath534 . if @xmath667 , then @xmath206 and @xmath73 .
thus @xmath668 . by and ,
@xmath669 , which is impossible since @xmath670 or @xmath671 and @xmath668 .
if @xmath660 , then @xmath672 .
it implies that @xmath254 and @xmath366 . by and ,
note that @xmath670 or @xmath239 by lemma [ 3no ] ( ii ) , we have @xmath674 . thus @xmath675 by . on the other hand , @xmath676 by , a contradiction .
next , we show that @xmath677 . on the contrary ,
suppose that @xmath678 by lemma [ 3no ] ( ii )
. then @xmath679 .
note that @xmath680 . without loss of generality ,
suppose that @xmath573 .
by lemma [ 3no ] ( ii ) , we have @xmath355 or @xmath46 . if @xmath663 .
applying with @xmath478 , we have @xmath683 .
if @xmath684 , then @xmath576 by lemma [ p ] ( i ) .
note that @xmath206 .
we have @xmath685 .
it contradicts the fact @xmath57 .
thus @xmath344 .
hence @xmath686 and @xmath249 .
note that @xmath687 .
we have @xmath254 and @xmath255 . by and ,
so @xmath689 .
if @xmath235 , then @xmath690 . so @xmath691 .
it implies that @xmath394 and @xmath692 .
similarly , we get @xmath443 and @xmath202 . by and ,
note that @xmath694 or @xmath239 by lemma [ 3no ] ( ii ) .
it follows that @xmath695 or @xmath696 .
we may suppose that @xmath695 without loss of generality . then @xmath328 and @xmath697 for @xmath698 by lemma [ p ] ( i ) .
hence @xmath0 has no pendent vertex .
this is a contradiction . if @xmath699 , then @xmath375 or @xmath239 .
if @xmath375 , then @xmath700 . thus @xmath701 by .
this is impossible since @xmath702 by .
if @xmath703 , then @xmath704 by .
this is also impossible since @xmath705 by .
we claim that @xmath684 .
otherwise , suppose that @xmath344 . then @xmath710 and @xmath711 , which is impossible since @xmath712 or @xmath55 by lemma [ 3no ] ( ii ) .
. therefore @xmath509 by lemma [ p ] ( ii ) .
thus @xmath713 .
it follows from that @xmath714 . by lemma [ 3no ] ( iii ) , @xmath712 or @xmath55
* case 1*. there is a vertex @xmath744 with @xmath471 and @xmath472 . without loss of generality ,
suppose that @xmath745
. then @xmath746 and @xmath747 for @xmath748 by lemma [ p ] ( i ) .
in particular , @xmath749 by lemma [ 3no ] ( iii ) .
if @xmath759 , then @xmath73 . we claim that @xmath131 . otherwise , let @xmath760 . for the vertex @xmath282 , @xmath761 .
thus @xmath762 . by lemma [ 3no ]
( ii ) , @xmath763 or @xmath55 . note that @xmath760 . we have @xmath764 or @xmath765 or @xmath766 . without loss of generality , we consider the following two cases : if @xmath764 , then @xmath254 and @xmath340 . note that @xmath73 .
we have @xmath429 .
if @xmath767 , then @xmath768 by lemma [ p ] ( i ) .
if @xmath328 , then @xmath769
. so @xmath770 or @xmath239 by lemma [ 3no ] ( ii ) . on the other hand , @xmath771 by and .
so @xmath772 . by and ,
. thus @xmath774 , a contradiction . if @xmath0 has no pendent vertex , then @xmath628 by lemma [ 31 ] . if @xmath785 , then @xmath0 is regular .
it is well known that a graph is regular if and only if it has exactly one main eigenvalues .
thus @xmath786 .
without loss of generality , suppose that @xmath787 .
we consider the following two cases : if @xmath162 , then @xmath790 .
so @xmath791 and hence @xmath523 .
similarly , @xmath792 .
thus @xmath375 and @xmath793 . therefore @xmath794
( see fig .
, @xmath795 is 2-walk @xmath229-linear . with a similar argument
, we have : if @xmath0 has at least one pendent vertex @xmath41 , then @xmath805 for @xmath806 , where @xmath807 . if @xmath808 , then with a similar argument of case 1 , we have @xmath809 .
hence we may suppose that @xmath810 .
in particular , assume that @xmath811 without loss of generality . then @xmath537 and @xmath812 for @xmath813 , where @xmath807 . applying lemma with @xmath267
, we have@xmath814 note that @xmath815 or @xmath55 and @xmath206 .
we consider the following cases without loss of generality : if @xmath822 , then @xmath340 and @xmath488 . by lemmas [ 21 ] and [ 3no ] ( ii ) , we have @xmath254 and @xmath73 .
hence @xmath823 for @xmath824 . by and ,
thus @xmath826 .
similarly , @xmath827 .
it implies that @xmath828 .
thus @xmath773 .
hence @xmath774 . on the other hand , @xmath829 by lemma [ 3no ] ( ii ) and the fact that @xmath826 .
this is a contradiction . if @xmath832 , then @xmath833 and @xmath587 .
it follows from lemma [ 3no ] ( iii ) that @xmath254 and @xmath255 . by and ,
it implies that @xmath772 .
hence @xmath834 .
it follows that @xmath774 and @xmath793 .
similarly , we have @xmath532 . therefore @xmath835 ( see fig . 3 ) .
by , @xmath836 is 2-walk @xmath229-linear . if @xmath837 , then @xmath253 and @xmath130 .
thus @xmath439 and @xmath73 by lemma [ 3no ] ( iii ) .
hence @xmath838 . for the vertex @xmath839 , @xmath776 .
it implies that @xmath772 .
thus @xmath840 , which is impossible since @xmath778 by lemma [ 3no ] ( ii ) .
if @xmath842 , then @xmath833 and @xmath132 .
we claim that @xmath843 .
otherwise , let @xmath844 . by , @xmath845 .
note that @xmath846 by assumption .
we have @xmath847 . on the other hand , @xmath848 by , a contradiction .
hence @xmath843 .
similarly , we have @xmath443 .
thus @xmath849 ( see fig . 4 ) , where @xmath850 .
it is easy to see that any graph @xmath851 is 2-walk @xmath852-linear . | an eigenvalue of a graph @xmath0 is called a main eigenvalue if it has an eigenvector the sum of whose entries is not equal to zero . in this paper , all connected tricyclic graphs with
exactly two main eigenvalues are determined .
0.05 in * 2000 mathematics subject classification : * 05c50 * keywords : * main eigenvalues ; tricyclic graphs ; 2-walk @xmath1-linear graphs |
leptogenesis@xcite is a simple mechanism to generate baryon number asymmetry of the universe .
the idea is that a lepton asymmetry produced at a high temperature is converted to the baryon asymmetry through the sphaleron interactions@xcite which conserve @xmath1 number but break @xmath2 number .
a simplest version of the leptogenesis is based on the seesaw mechanism@xcite which can also explain the smallness of the neutrino masses by introducing heavy right - handed neutrinos ( rhn ) to the standard model . in the seesaw models , the cp - violating and out - of - equilibrium decay of the rhn can produce the lepton asymmetry .
the seesaw mechanism is also easy to be implemented in supersymmetric and/or grand unified theories ( guts ) which are most attractive candidates for the physics beyond the standard model .
in such a class of model , the successful leptogenesis is considered as a mechanism to generate the baryon asymmetry of the universe . however , in many models especially in gut models , the mass ( @xmath3 ) of the lightest rhn ( @xmath4 ) is too small to generate the observed baryon asymmetry by the @xmath4 decay and the asymmetry should be produced through another mechanism .
recently it is pointed out that flavor effects give significant contributions to the leptogenesis @xcite .
one of the interesting phenomena in the flavored leptogenesis is that the primordial lepton asymmetry generated by the decay of the second lightest rhn ( @xmath5 ) , of the inflaton or so can remain against the washout by the lightest one@xcite .
this is an interesting possibility to give enough baryon asymmetry even when the mass of the lightest rhn is too small .
in such a scenario , the study of the washout effect by the lightest rhn is very important . in this paper
, we study the detail of this flavored washout effect due to the lightest rhn and we point that there is a novel parameter region where an effect that is negligible in most cases plays an important role .
this effect is due to off - diagonal elements of the so - called a - matrix , and thus unique in the flavored leptogenesis .
most recently , the effects of the off - diagonal elements are studied numerically in the context that @xmath4 decay generates the lepton asymmetry@xcite . here
, we adapt the analysis on the washout effect to the case where the asymmetry produced by the @xmath5 decay dominates the baryon asymmetry of the universe , and show that a sizable lepton asymmetry can remain against the washout process in a different way from those studied in refs.@xcite . in the section 2 ,
we investigate the washout effect by the @xmath4 decay , assuming that there is primordial lepton asymmetry before the decay becomes relevant . in the section 3
, we will show examples that the primordial asymmetry is generated by the decay of the second lightest rhn .
the section 4 is devoted to summary and discussion .
in the seesaw mechanism , rhns are introduced to the standard model , @xmath6 with @xmath7 , @xmath8 , and @xmath9 being rhns , lepton doublets , and higgs doublet respectively . here
we take the basis where the yukawa matrix for charged leptons and the mass matrix of rhns are diagonalized . integrating out the heavy rhns and giving the vev to higgs ,
one can obtain the neutrino masses , @xmath10 , and pontecorvo - maki - nakagawa - sakata ( pmns ) mixing matrix@xcite , @xmath11 , as @xmath12 where @xmath13 is the higgs vev . supposing the reheating temperature after the inflation is enough high , the rhns are produced through the interaction with the doublet leptons and the higgs fields .
when the temperature decreases down to the mass of rhn , the production becomes inefficient and rhns decay away .
this out - of - equilibrium decay of the rhns generates @xmath1 asymmetry which is proportional to the cp violation in the decay , @xmath14 , defined as @xmath15 this asymmetry is converted to the baryon asymmetry through the electroweak sphaleron@xcite processes . in order to evaluate the baryon asymmetry of the universe , the boltzmann equation is used . in this analysis , for simplicity
, we omit the scattering effects which are considered to be subdominant .
the decays and inverse decays , @xmath16 , are considered with rate @xmath17 . with this simplification ,
the evolution of the asymmetry of @xmath18 after the decoupling of the second lightest rhn is described by the following set of boltzmann equations@xcite : @xmath19 } } , \label{boltzmanneqb - l}\end{aligned}\ ] ] where @xmath20 and @xmath21 .
the parameters @xmath22 and @xmath23 indicate the number density of the particle @xmath24 divided by the entropy density @xmath25 and its value in equilibrium respectively , and @xmath26 . the parameter @xmath27 is the total effective number of the degrees of freedom ( dof ) at the temperature around @xmath3 . with these definitions ,
we have @xmath28 where @xmath29 is the riemann s zeta function and @xmath30 is the modified bessel function . here , @xmath31 is ( not effective ) number of dof of the particle @xmath24 , for example @xmath32 and @xmath33 . after neglecting the finite temperature effects such as the thermal masses and running of the couplings ( for these effects , see ref.@xcite ) for simplicity , one can obtain @xmath34 and the hubble parameter @xmath35 as @xmath36 where @xmath37 is the total decay width of @xmath4 and @xmath38 is the planck scale .
now , let us define the `` washout mass parameter '' @xmath39 and `` equilibrium neutrino mass parameter '' @xmath40 as @xmath41 where @xmath42 . the partial decay width to a flavor @xmath43
is written as @xmath44 and the total decay width is given by the sum of them @xmath45 .
eventually , the boltzmann equations are written as@xmath46}}. \label{boltzmanneqb - l1}\end{aligned}\ ] ] the coefficient of @xmath47 is different from that in ref.@xcite by a factor @xmath48 which comes from the relative factor of the number density in the equilibrium of fermion / boson ( [ yeq ] ) .
notice that , however , in the derivation of the boltzmann equations ( [ boltzmanneqn ] ) and ( [ boltzmanneqb - l ] ) , an approximation @xmath49 is made . within this approximation
, the relative factor @xmath48 ( and the factor @xmath50 in eq.([mu2y ] ) ) disappears .
in fact the right hand side of eq.([boltzmanneqb - l ] ) is originally written in terms of not @xmath51 but the chemical potentials of the particle @xmath24 , @xmath52 . in literatures ,
these chemical potentials are replaced as eq.([boltzmanneqb - l ] ) using the relation between @xmath52 and @xmath51 with the approximation .
if one would not use this replacement , a factor 1/2 appears instead of 3/4 .
this difference of the factor does not affect the results a lot in many cases and the term of @xmath47 itself is often neglected . in our analysis
, the contribution can affect the result significantly . in the following we take basically the factor 3/4 for illustration .
it is straightforward to make analyses with a different factor .
an important point is that @xmath53 is invariant under the standard evolution of the universe and related to the present baryon asymmetry as @xmath54@xcite can be somewhat different , for instance @xmath55@xcite , depending on the timing of the freeze out of the electroweak sphaleron , but in any case , the value is approximately equal to @xmath56 .
] at the weak scale due to the electroweak sphaleron process .
thus , we define `` baryon asymmetry '' by multiplying the factor @xmath57 on @xmath53 even at a higher temperature .
because this value is proportional to the present baryon to photon ratio as @xmath58 where @xmath59 is the present total effective number of dof , successful leptogenesis scenario should predict the `` baryon asymmetry '' @xmath60 which comes from the observable @xmath61@xcite . as mentioned in the introduction , in some models ,
the cp violation @xmath62 is too small to produce enough lepton asymmetry . in this case
we can neglect the source term .
then , the evolution of @xmath53 is controlled by one equation as @xmath63 note that we assume that the asymmetries of the lepton doublet , @xmath64 , and of the higgs fields , @xmath65 , are much smaller than 1 and thus neglect the higher terms because it is proportional to the @xmath1 asymmetry , as shown below .
these relations are forced by the fast ( spectator ) processes , such as the sphaleron process , and depend on the temperature .
for example at a high temperature , only the interactions mediated by the gauge and the top yukawa coupling are in the thermal equilibrium , while at a lower temperature weaker interactions come in it .
for instance , let us concentrate on the range of the temperature where the interactions mediated by all the second and third generational yukawa couplings are in the equilibrium but the first generational ones are not .
this range is likely the one in which we are interested , namely @xmath66 . in this range ,
the weak sphaleron and the qcd sphaleron@xcite are considered to occur fast enough .
these fast interactions make the following relations hold among the chemical potentials : @xmath67 in addition to these relations , we impose the charge neutrality of the universe and assume the vanishing asymmetries for the right handed leptons ( and quarks if the qcd sphaleron is not considered ) of the first generation : @xmath68 notice that in the eq.([chemical - charge ] ) , the factor @xmath50 in front of @xmath69 comes from the relative factor in the relation between the asymmetry density and the chemical potential for massless particles : @xmath70 taking care of this factor 2 , we get similar relations among the asymmetries by replacing @xmath71 , @xmath72 and @xmath73 . by solving these relations
, we find the expression of @xmath74 and @xmath75 in terms of @xmath76 as @xmath77 with @xmath78 if we do not consider the qcd sphaleron and with @xmath79 if we take into account it . in the following , we examine only the latter case because there are no qualitative difference . from these expressions , we have the following boltzmann equation : @xmath80 with @xmath81 where the summation over @xmath82 should be understood . in order to analyse this equation ,
it is convenient to change the variable from @xmath83 to @xmath84 that satisfy @xmath85 so that @xmath86 the range of @xmath84 is from @xmath87 to @xmath88 .
the relation between them is shown in the fig.[fig : svsz ] .
as a function of @xmath83 . ]
this figure shows the washout occurs mostly in the temperature range @xmath89 . for comparison , the boltzmann equation for the usual one - flavor approximation , which is in reality valid only when the temperature is high enough that even the processes mediated by the tau yukawa coupling are out of equilibrium ,
is given as @xmath90 where @xmath91 and @xmath92 .
roughly speaking , the matrix @xmath93 is close to a diagonal one . and
thus , we can find an approximate solution by a perturbation with respect to rather small off - diagonal elements .
neglecting the off - diagonal elements , each initial asymmetry @xmath94 is exponentially washed out .
the evolution of the asymmetry is given by @xmath95 when the off - diagonal elements are switched on , the asymmetry follows @xmath96 up to the next leading order of the off - diagonal elements . inserting the leading order solution eq.([yi0th ] ) , we find @xmath97 this expression shows that even if the initial asymmetry of a certain flavor is zero , the asymmetry is generated from those of the others although it is suppressed by a small off - diagonal element of @xmath98 . for most cases ,
this order already gives good approximation for the final value of the total asymmetry . in this case
, one may expect the washout effect is small and thus the flavor effect can not play an important role .
however , even in this case , the final total asymmetry can be 2 times larger than the one - flavor approximation ( see ( 1 ) in fig.[fig : washout ] ) . in this case
, the summation @xmath104 is dominated by @xmath105 and is larger than @xmath40 .
it is called strong washout region .
as shown below , however , if we take account of the flavor effect the washout effect is drastically changed .
the effect depends strongly on the flavor structure of the initial asymmetry . in the following ,
let us take @xmath106 as a representative example for clarity .
it is straightforward to apply this analysis to the other cases .
we consider two typical sets of the initial asymmetries : * @xmath107 + in this case , the washout effect of @xmath108 is controlled by @xmath109 which is much smaller than @xmath110 , while @xmath111 are generated due to the small off - diagonal elements of @xmath112 , with the opposite sign .
because of the large @xmath113 , these generated @xmath111 are washed out strongly and can not become comparable with @xmath108 . thus , in terms of the perturbation , the leading order approximation is sufficient .
+ note that even in this case , the washout factor @xmath114 is quite different ( much larger ) than the one - flavor approximation , @xmath115 ( see ( 2a ) in fig.[fig : washout ] ) .
this is the case even when @xmath109 is not so much smaller than the others , due to the exponential washout factor@xcite . *
@xmath116 and/or @xmath117 + the asymmetry @xmath111 decreases rapidly , while the asymmetry @xmath108 produced due to the off - diagonal elements is washed out much more slowly .
this means that at some point @xmath108 becomes dominant .
once it becomes dominant , the following evolution is similar to the one in the case ( 2a ) .
thus , the washout factor is controlled basically by the small @xmath109 rather than @xmath110 or @xmath113 ( see ( 2b ) in fig.[fig : washout ] ) . interestingly in this case
, the sign of the total @xmath1 asymmetry changes through the washout . + note that the approximation at the nlo is quite bad for @xmath111 because the secondary conversion from @xmath108 , which is generated by the nnlo effect , is important .
nevertheless , the approximation for the total asymmetry is rather good because these are small as in the case ( 2a ) . in this case , all the asymmetries in each flavors are strongly washed out .
thus , it is hard that the observed value remains after the washout , as far as @xmath120 where the muon yukawa interaction is out of equilibrium , the asymmetry along with the direction in the flavor space that is orthogonal both to the direction determined by @xmath4 yukawa coupling and @xmath121 direction is free from the washout , even in this case@xcite . ] . from the above considerations , it is clear that the washout factor is basically controlled by the smallest washout mass parameter
this is in great contrast to the non - flavored case , where the factor is controlled basically by the largest washout mass parameter .
interestingly , this is also true even for the case that the initial asymmetry of the flavor with smallest washout mass parameter is tiny . for this case , the effect of the off - diagonal elements , which is usually negligible , is crucial .
cc asymmetries .
the horizontal line is @xmath122 , and the vertical line is @xmath123 . in the left figures ,
black ( solid ) , light blue ( broken ) and purple ( dot - dashed ) lines show the total @xmath1 asymmetry calculated by the full boltzmann equation ( [ boltzmanneqf ] ) , by the approximation formula ( [ yi1st ] ) and the one - flavor approximation ( [ boltzmanneqnf ] ) , respectively . in the right figures ,
red , green and blue lines respectively show the @xmath124 , @xmath125 and @xmath126 , and the solid and broken lines corresponds ( [ boltzmanneqf ] ) and ( [ yi1st ] ) , respectively .
, title="fig : " ] & asymmetries .
the horizontal line is @xmath122 , and the vertical line is @xmath123 . in the left figures , black ( solid ) , light blue ( broken ) and purple ( dot - dashed ) lines show the total @xmath1 asymmetry calculated by the full boltzmann equation ( [ boltzmanneqf ] ) , by the approximation formula ( [ yi1st ] ) and the one - flavor approximation ( [ boltzmanneqnf ] ) , respectively . in the right figures , red , green and blue lines respectively
show the @xmath124 , @xmath125 and @xmath126 , and the solid and broken lines corresponds ( [ boltzmanneqf ] ) and ( [ yi1st ] ) , respectively .
, title="fig : " ] + + asymmetries .
the horizontal line is @xmath122 , and the vertical line is @xmath123 . in the left figures , black ( solid ) , light blue ( broken ) and
purple ( dot - dashed ) lines show the total @xmath1 asymmetry calculated by the full boltzmann equation ( [ boltzmanneqf ] ) , by the approximation formula ( [ yi1st ] ) and the one - flavor approximation ( [ boltzmanneqnf ] ) , respectively . in the right figures , red , green and blue lines respectively show the @xmath124 , @xmath125 and @xmath126 , and the solid and broken lines corresponds ( [ boltzmanneqf ] ) and ( [ yi1st ] ) , respectively . , title="fig : " ] & asymmetries .
the horizontal line is @xmath122 , and the vertical line is @xmath123 . in the left figures , black ( solid ) , light blue ( broken ) and purple ( dot - dashed ) lines show the total @xmath1 asymmetry calculated by the full boltzmann equation ( [ boltzmanneqf ] ) , by the approximation formula ( [ yi1st ] ) and the one - flavor approximation ( [ boltzmanneqnf ] ) , respectively . in the right figures ,
red , green and blue lines respectively show the @xmath124 , @xmath125 and @xmath126 , and the solid and broken lines corresponds ( [ boltzmanneqf ] ) and ( [ yi1st ] ) , respectively .
, title="fig : " ] + + asymmetries .
the horizontal line is @xmath122 , and the vertical line is @xmath123 . in the left figures , black ( solid ) , light blue ( broken ) and purple ( dot - dashed ) lines show the total @xmath1 asymmetry calculated by the full boltzmann equation ( [ boltzmanneqf ] ) , by the approximation formula ( [ yi1st ] ) and the one - flavor approximation ( [ boltzmanneqnf ] ) , respectively . in the right figures , red , green and blue lines respectively
show the @xmath124 , @xmath125 and @xmath126 , and the solid and broken lines corresponds ( [ boltzmanneqf ] ) and ( [ yi1st ] ) , respectively .
, title="fig : " ] & asymmetries .
the horizontal line is @xmath122 , and the vertical line is @xmath123 . in the left figures , black ( solid ) , light blue ( broken ) and purple ( dot - dashed ) lines show the total @xmath1 asymmetry calculated by the full boltzmann equation ( [ boltzmanneqf ] ) , by the approximation formula ( [ yi1st ] ) and the one - flavor approximation ( [ boltzmanneqnf ] ) , respectively . in the right figures ,
red , green and blue lines respectively show the @xmath124 , @xmath125 and @xmath126 , and the solid and broken lines corresponds ( [ boltzmanneqf ] ) and ( [ yi1st ] ) , respectively .
, title="fig : " ] + + from eq .
( [ boltzmanneqf ] ) , one can obtain a coupled equations for @xmath127 as @xmath128\left(\frac{y_{\delta_{f'}}}{y_{\delta_{\tau}}}\right)\;.\ ] ] this couple of equations have fixed points in the space of @xmath129 , which is determined by solving the equations @xmath130\left(\frac{y_{\delta_{f'}}}{y_{\delta_{\tau}}}\right ) = 0\;.\ ] ] once the flow of the solution reaches close to a fixed point at @xmath131 , the set of ratios @xmath129 becomes invariant and the boltzmann equations can be rewritten as @xmath132 this means that the asymmetries of the all flavor are washed out with the universal washout mass parameter @xmath133 which corresponds to one of the eigenvalues of matrix @xmath134 .
there are three possible points in the case of three effective flavor numbers .
however two of them are unstable fixed points and only one point is attractive .
the attractive one most likely corresponds to the smallest eigenvalue which is smaller than the smallest @xmath135 . around the attractive fixed point ,
the asymmetry in the flavor with the smallest washout mass parameter , @xmath136 dominates the total asymmetry .
this can be understood as follows .
the asymmetry in the flavor with larger washout mass parameter , @xmath137 decrease more quickly as discussed in the case ( 2b ) in the sec .
[ solutions ] .
when @xmath137 becomes the order of @xmath138 , @xmath137 evolves similar to @xmath136 because the transportation from @xmath136 becomes to control the evolution .
then the washout of @xmath136 becomes weaker due to the transportation from @xmath137 .
when the initial condition is too far from the fixed point and @xmath139 , the washout term decouples from the system before the solution flows into the fixed point takes the value in the range of @xmath140 with respect to @xmath141 . ] .
one can see similar phenomena also in the case where only @xmath4 decay produces the asymmetries and they are washed out by the @xmath4 ( inverse ) decay . in fact
effects of off - diagonal elements of @xmath142 matrix in such a case are discussed in ref.@xcite and the asymmetry shown there behaves as discussed in the above .
in this section , we investigate the possibility that the initial asymmetries are generated via the second lightest rhn decay . for simplicity , we restrict ourselves to the case that the mass of the second lightest rhn is larger than @xmath143 .
this case is qualitatively discussed in ref.@xcite . for this mass range , the fast interactions that are in the equilibrium when the rhn decays are only the interactions mediated by the top yukawa coupling and qcd sphalerons .
this means that the fast interactions can not distinguish all the generations of the lepton doublets , and thus two liner combinations of the three doublets that do not interact with the rhn are never produced .
namely , only @xmath144 are produced .
then , the relevant boltzmann equations are for one flavor system , which is given by eqs.([boltzmanneqn1 ] ) and ( [ boltzmanneqb - l1 ] ) by replacing all the index @xmath145 to @xmath50 ( including @xmath146 ) and suppressing the flavor indexes . with the definitions given in the section [ boltzmanneq ] , we have @xmath147 the a factor for this case is calculated in a similar way to the discussion in the section [ boltzmanneq ] as @xmath148 it is possible , of course , that we solve these set of equations numerically to evaluate the @xmath1 asymmetry produced by the decay of the second lightest rhn . in this article , however , we use the following approximation formula proposed in ref.@xcite , which includes the effects of the scatterings , to evaluate the `` baryon asymmetry '' : @xmath149 with @xmath150 in any case , these equations are controlled by the parameters @xmath151 and @xmath152 which are determined by the mass spectrum of the rhns @xmath153 and the neutrino yukawa coupling @xmath154 . to be more concrete ,
they are given by sums of ( [ def;mtilde ] ) and @xmath155 here both the diagrams of the vertex correction and of the self - energy correction are implemented in each function as @xmath156 in this way , fixing @xmath153 and @xmath154 , all the parameters in the boltzmann equations are determined , and we can calculate the @xmath1 asymmetry @xmath157 just after the second lightest rhn decouples .
this asymmetry is washed out when the lightest rhn starts decaying . in this period ,
the @xmath121 and @xmath158 yukawa couplings enter into the equilibrium , and thus the fast interactions distinguish all the three flavors .
therefore , we should divide @xmath157 into @xmath159 , @xmath160 and @xmath108 , which follow the relation @xmath161 according to the probabilistic interpretation .
this set of asymmetries @xmath162 gives the initial condition of the analysis given in the section [ solutions ] .
the neutrino yukawa couplings @xmath163 should be related to the low energy neutrino parameters through the seesaw relation , eq.([seesawrelation ] ) . in order to represent the solution of this relation ,
we adopt the following famous parameterization@xcite , @xmath164 here @xmath11 is written as the product of a ckm - like mixing matrix @xmath165 which includes three mixing angles and one cp phase , we adopt the chau keung parametrization ( pdg parametrization ) @xcite . ] and a phase matrix with two majorana phases @xmath166 : @xmath167 and @xmath168 is a complex orthogonal matrix which can be decomposed as @xmath169 where @xmath170 are gell - mann matrices and @xmath171 are complex parameters . for simplicity , in this article , we use the following set of the parameters for the light neutrino sector as @xmath172 , @xmath173 for the pmns matrix , and majorana masses @xmath174 for the rhns . as representative examples ,
let us consider the following sets : @xmath175 for instance , for the example ( i ) , we find @xmath176 these show @xmath177 and @xmath178 are negligibly small .
thus , this is an example of the case ( 1 ) in the section [ solutions ] . using the approximation ( [ appformula ] )
, we see the `` baryon asymmetry '' generated by the @xmath5 decay is @xmath179 . when the temperature decrease to around @xmath3 , @xmath4 starts decaying , and the asymmetry is washed out in the way investigated in the last section . as mentioned above , in this period , the fast interactions distinguish all the flavor , and the asymmetry should be divided as @xmath180 .
after the washout , a total asymmetry @xmath181 remains . in a similar way ,
( i@xmath182ia ) and ( i@xmath182ib ) are examples of the cases ( 2a ) and ( 2b ) , respectively .
their results are listed in the table [ results ] . .
results of @xmath183 , @xmath14 , @xmath184 and @xmath185 for the three examples in ( [ examples ] ) . [ cols="^,^,^,^",options="header " , ] [ results ]
in this article , we investigate the washout effect due to the @xmath4 ( inverse ) decay , assuming non - vanishing initial lepton asymmetry and negligible lepton asymmetry production in @xmath4 decay .
we show that there is a novel parameter region in addition to those studied in refs.@xcite . there , off - diagonal elements of the @xmath142-matrix , which are often omitted , play a critical role .
this region is where some of @xmath186 is comparable to or smaller than @xmath40 , the others are larger than it , and the initial asymmetries on the flavors with small @xmath186 are tiny . in this case , if we would omit the off diagonal elements as usual , any initial asymmetries on the flavors with large @xmath186 were strongly washed out .
in fact , the off diagonal elements transform the asymmetries from those with large @xmath186 to those with small ones .
once transformed , such asymmetries are weakly washed out , and thus a sizable total asymmetry may survive . for completeness , we examine the possibility that the initial asymmetry is generated by the @xmath5 decay within the thermal leptogenesis scenario .
we show some concrete examples for each class discussed in the above analysis .
finally , let us make a comment on an ambiguity of the boltzmann equations , especially on the factor in front of @xmath47 in eq.([boltzmanneqb - l1 ] ) .
as briefly discussed below the equation , an approximation is used in the derivation of the boltzmann equations ( [ boltzmanneqn ] ) , ( [ boltzmanneqb - l ] ) , and it brings the ambiguity . because the contribution of this term ( @xmath47 ) is relatively small , as seen from ( [ cw / oqcds ] ) and ( [ cwqcds ] ) , this ambiguity does not affect results so much , @xmath187 . in the case
( 2b ) , however , the off diagonal element of @xmath142-matrix , which is of the same order as the @xmath47 contribution , is critical .
in addition , a cancellation occurs between @xmath47 contribution and that of @xmath188 in the off diagonal element .
thus , the result is largely changed due to the ambiguity . in fact , if we take a factor @xmath189 instead of @xmath48 as a possible choice , the final baryon asymmetry is reduced to @xmath190 with the same parameters as ( i@xmath182ib ) in ( [ examples ] ) .
thus , it is important to make a closer look on the boltzmann equations before discussing this novel effect quantitatively .
99 m. fukugita and t. yanagida , phys .
b*174 * , 45 ( 1986 ) .
v. a. kuzmin , v. a. rubakov and m. e. shaposhnikov , phys .
b * 155 * , 36 ( 1985 ) .
p. minkowski , phys . lett .
* b67 * , 421 ( 1977 ) ; + m. gell - mann , p. ramond , and r. slansky in _ supergravity _ , p. 315
, edited by f. nieuwenhuizen and d. friedman , north holland , amsterdam , 1979 ; + t. yanagida , proc . of the _ workshop on unified theories and the baryon number of the universe _ , edited by o. sawada and a. sugamoto , kek , japan 1979 ; + r. n. mohapatra , g. senjanovi , phys .
* 44 * 912 ( 1980 ) .
r. barbieri , p. creminelli , a. strumia and n. tetradis , nucl .
b * 575 * , 61 ( 2000 ) [ arxiv : hep - ph/9911315 ] .
h. b. nielsen and y. takanishi , nucl .
b * 636 * , 305 ( 2002 ) [ arxiv : hep - ph/0204027 ] .
o. vives , phys.rev .
d*73 * , 073006 ( 2006 ) [ arxiv : hep - ph/0512160 ] .
a. abada , s. davidson , f. x. josse - michaux , m. losada and a. riotto , jcap * 0604 * , 004 ( 2006 ) [ arxiv : hep - ph/0601083 ] .
e. nardi , y. nir , e. roulet and j. racker , jhep * 0601 * , 164 ( 2006 ) [ arxiv : hep - ph/0601084 ] .
g. engelhard , y. grossman , e. nardi and y. nir , hep - ph/0612187 .
f. x. josse - michaux and a. abada , arxiv : hep - ph/0703084 .
b. pontecorvo , zh . eksp .
( jetp ) * 33 * 549 ( 1957 ) ; _ ibid . _ * 34 * 247 ( 1958 ) ; _ ibid . _ * 53 * 1717 ( 1967 ) ; z. maki , m. nakagawa and s. sakata , prog .
* 28 * 870 ( 1962 ) .
p. di bari , nucl .
b * 727 * , 318 ( 2005 ) [ arxiv : hep - ph/0502082 ] .
s. blanchet and p. di bari , jcap * 0606 * , 023 ( 2006 ) [ arxiv : hep - ph/0603107 ] .
giudice , a. notari , m. raidal , a. riotto and a. strumia , nucl.phys .
d*685 * , 89 ( 2004 ) [ arxiv : hep - ph/0310123 ] .
m. laine and m. shaposhnikov , phys.rev . d*61
* , 117302 ( 2000 ) [ arxiv : hep - ph/9911473 ] .
jeffrey a. harvey and michael s. turner , phys.rev .
d*42*,3344 ( 1990 ) d. n. spergel _
et al . _ , astro - ph/0603449 ( submitted to apj ) l. bento , jcap * 0311 * , 002 ( 2003 ) ; g. d. moore , phys . lett .
b * 412 * , 359 ( 1997 ) ;
r. n. mohapatra and x. m. zhang , phys . rev .
d * 45 * , 2699 ( 1992 ) .
a. abada , s. davidson , a. ibarra , f .- x .
josse - michaux , m. losada and a. riotto , jhep * 0609 * , 010 ( 2006 ) [ arxiv : hep - ph/0605281 ] ; + s. pascoli , s.t .
petcov and a. riotto , hep - ph/0611338 .
j. a. casas and a. ibarra , nucl .
b * 618 * , 171 ( 2001 ) [ arxiv : hep - ph/0103065 ] .
l. l. chau and w. y. keung , phys .
* 53 * , 1802 ( 1984 ) . | we investigate a flavored washout effect due to the decay of the lightest right - handed neutrino , assuming that there is non - vanishing initial lepton asymmetry and the decay of the lightest right - handed neutrinos gives negligible contribution to the asymmetry .
we figure out general features of the washout effect .
it is shown that there is a novel parameter region where an effect that is negligible in most cases plays a critical role and a sizable lepton asymmetry can survive against the washout process even in a strong washout region . * a novel washout effect in the flavored leptogenesis * * tetsuo shindou*@xmath0 and * toshifumi yamashita*@xmath0 _ scuola internazionale superiore di studi avanzati , and + _ _ istituto nazionale di fisica nucleare , i-34014 trieste , italy + _ |
the action of @xmath0=4 four dimensional supersymmetric yang
mills theory was given for the first time in @xcite in the context of toroidal compactifications of the type i superstring .
the theory has the maximal amount of supersymmetry allowed for a rigid supersymmetric theory in four dimensions , namely the symmetry generated by sixteen real supercharges , and has been proved to be finite , possessing a vanishing @xmath2-function .
for this reason it has long been considered to be a rather trivial theory . however recent developments in the study of the correspondence with type iib superstring theory on anti de sitter space have shown that it is actually a very interesting quantum field theory , displaying very peculiar properties .
the field content of the theory , which is unique apart from the choice of the gauge group g , consists of six real scalars , four weyl spinors and one vector , which are all in the adjoint representation of the gauge group g. in the abelian case the theory is free , whereas in the non abelian case it has a moduli space of vacua parameterized by the vacuum expectation values ( vev s ) of the six real scalars . in the coulomb phase , reached giving non vanishing vev s to the scalars in the cartan subalgebra of g , the theory is again free , while the origin of the moduli space corresponds to a highly non - trivial superconformal field theory .
the model is classically invariant under the @xmath0=4 superconformal group and this property is supposed to be preserved at the quantum level @xcite .
furthermore it has a global su(4)@xmath3so(6 ) symmetry , which naturally emerges in the compactification from ten dimensions on a six - torus and is identified , in a four dimensional perspective , with the r - symmetry group of automorphisms of the @xmath0=4 superconformal algebra @xcite .
the @xmath0=4 super yang - mills theory is supposed to exactly realize a generalization of the electric - magnetic duality of montonen and olive @xcite ( s - duality ) , thanks to the presence in the spectrum of an infinite tower of stable bps dyonic states @xcite . at the perturbative level
the theory is finite up to three loops .
the @xmath2-function has been shown to be vanishing both in the component formulation @xcite and using superspace techniques @xcite .
in particular the formulation in terms of @xmath0=1 superfields has proved to be an extremely powerful tool in the perturbative analysis .
strong arguments have been proposed in order to extend the proof of the finiteness to all orders in perturbation theory @xcite and moreover general considerations from instanton calculus and duality arguments suggest that the same should hold at the non - perturbative level @xcite . following the proposal of maldacena @xcite of a correspondence relating ( super ) conformal field theories in @xmath4 dimensions and type iib superstring theory on @xmath4 + 1 anti de sitter space
there has been recently a renewal of interest in @xmath0=4 supersymmetric yang mills theory .
according to this conjectured duality both perturbative @xcite and non - perturbative @xcite contributions to correlation functions of gauge invariant composite operators in @xmath0=4 super yang mills theory should be related to type iib superstring amplitudes in ads@xmath5 .
work in this context is leading to very interesting results in the study of peculiar properties of @xmath0=4 yang
mills theory as a superconformal field theory .
the aim of this paper is to present a careful re - analysis of perturbation theory , which allows to point out various problems related to the choice of gauge .
both in the component field formulation and using @xmath0=1 superfields the gauge fixing procedure appears very subtle .
as we will discuss in detail later , in both cases one finds ultraviolet and/or infrared divergences in off - shell green functions . throughout the paper calculations
will be carried out in euclidean space .
two different formulations of the @xmath0=4 supersymmetric yang mills theory will be employed , one in terms of @xmath0=1 superfields and the other in component ( ` physical ' ) fields .
notations and conventions that will be used are those of @xcite .
the @xmath0=4 `` on - shell '' multiplet can be obtained by combining three @xmath0=1 chiral superfields and one @xmath0=1 vector superfield , so that the six real scalars of the model are assembled into three complex scalars which , together with three of the weyl spinors , form the chiral multiplets , the fourth spinor and the vector gives rise to the vector multiplet . in this description a su(3)@xmath6u(1 ) subgroup of the su(4 ) r - symmetry group is manifest . the component - field formulation that will be used
is directly related to that in @xmath0=1 superspace , actually it is obtained from the former by integrating over the grassmannian coordinates of @xmath0=1 superspace . in the component formulation
the propagators of the elementary fields are ultraviolet divergent in the wess
zumino ( wz ) gauge and these infinities are exactly cancelled when the contributions of the ` gauge - dependent ' fields , that are put to zero in the wz gauge , are taken into account .
the choice of the wz gauge , that is almost unavoidable in explicit computations , introduces divergences that require a wave function renormalization .
different problems emerge in the formulation of the theory in terms of @xmath0=1 superfields .
almost all of the calculations showing the vanishing of the quantum corrections to two- and three - point functions , that are presented in the literature , were performed in the supersymmetric generalization of the fermi feynman gauge . with a different choice of gauge
two- and three - point functions develop infrared singularities , leading to the result that the choice of the fermi feynman gauge is somehow privileged .
this conclusion was proposed for the first time in @xcite and then it was discussed in the case of the @xmath0=4 theory in @xcite ; however no possible explanation was proposed .
unlike those of the elementary fields , correlation functions of gauge invariant composite operators , that play a crucial rle in the correspondence with ads type iib supergravity / superstring theory , should not suffer from problems related to the gauge fixing .
we will also discuss , in the superfield formulation , the effect of introducing of a mass term for the ( anti ) chiral superfields , which breaks supersymmetry from @xmath0=4 down to @xmath0=1 .
it is shown that this deformation of the model does not modify the ultraviolet properties of the original theory .
this result was first proposed in @xcite , where it was proved that the inclusion of mass terms for the ( anti ) chiral superfields does not generate divergent corrections to the effective action .
we will argue that this statement can be reinforced , showing that , at least at one loop , no new divergences , not even corresponding to wave function renormalizations , appear as a consequence of the addition of the mass terms .
this result supports the claim put forward in @xcite , where the ` mass deformed ' @xmath0=4 theory was proposed as a supersymmetry - preserving regularization scheme for a class of @xmath0=1 theories .
the paper is organized as follows .
section [ wzproblems ] deals with the problems introduced by the choice of the wess
zumino gauge when the component formalism is used .
the following sections report calculations performed in the @xmath0=1 superfield formalism of two- , three- and four - point functions .
a discussion of the results is presented in the concluding section .
as already remarked the field content of @xmath0=4 super yang
mills theory can be obtained by coupling in a gauge invariant way one @xmath0=1 vector superfield @xmath7 and three chiral superfields @xmath8 , @xmath9 , all in the adjoint representation of the gauge group g , so that the colour index @xmath10 takes the values @xmath11 .
a su(3)@xmath6u(1 ) subgroup of the su(4 ) r - symmetry group is manifest and under this global symmetry the chiral superfields @xmath12 transform in the * 3 * , while the vector @xmath13 is a singlet .
the complete expression for @xmath13 is @xmath14 -i{\overline \theta}{\overline \theta } \theta \left [ \lambda(x ) +
\frac{i}{2}\sigma^{\mu}\partial_{\mu}{\overline \chi}(x ) \right]+\frac{1}{2 } \theta \theta { \overline \theta}{\overline \theta } \left [ d(x ) + \frac{1}{2 } \box c(x ) \right ] .
\label{vector}\end{aligned}\ ] ] the gauge transformations of the superfields take the form @xmath15 and @xmath16 where @xmath17 is a matrix - valued chiral superfield . for infinitesimal gauge transformations use of the becker
hausdorff s formula allows to write @xmath18 \ , , \label{vvariaz}\ ] ] where @xmath19 $ ] is the lie derivative and ( [ vvariaz ] ) is a compact form to be understood in terms of the power expansion of @xmath20 . by explicitly writing the component expansion of these relations
one can show that it is always possible to put to zero the lower components , @xmath21 , @xmath22 and @xmath23 , of @xmath13 by a suitable gauge transformation . as a result for the superfield @xmath13 one obtains @xmath24 this choice
is known as the wess
zumino gauge .
fixing the wess
zumino gauge still leaves with the ordinary non abelian gauge freedom on the remaining fields . in this section
we will discuss perturbation theory in components .
the one - loop correction to the propagators of elementary fields will be computed first in the wess - zumino gauge and then taking into account the contribution of the fields @xmath21 , @xmath22 and @xmath23 that are absent in this gauge .
we will show that the choice of the wz gauge introduces ultraviolet divergences in the propagators .
the standard formulation of the model is obtained by eliminating the auxiliary fields , @xmath25 from the chiral multiplet and @xmath26 from the vector multiplet .
this procedure in the wz gauge results in a polynomial action with a scalar potential containing a quadrilinear term for the scalars .
however in order to correctly deal with the lower components of the vector superfield @xmath13 , that are put to zero in the wess
zumino gauge , it is more convenient not to eliminate the auxiliary fields through their equations of motion , but rather keep them in the action : in the computation of green functions the corresponding @xmath27-space propagators are @xmath28-functions . the action is written in terms of the non abelian field strength superfield @xmath29 @xmath30 where @xmath31 \rule{0pt}{24pt}\end{displaystyle } \label{w1 - 2 } \end{array}\ ] ] and the terms @xmath32 , with @xmath33 , contain @xmath34 factors of @xmath13 and vanish in the wz gauge .
in euclidean space the action in the @xmath0=1 superfield formalism takes the form @xmath35 + \right .
\nonumber \\ & + & \left .
\phi_{i}^{\dagger } v \phi^{i } + \left [ \left ( \frac{1}{2 } w^{(1)\alpha } w^{(2)}_{\alpha } + \frac{1}{4 } w^{(2)\alpha } w^{(2)}_{\alpha } \right ) \delta ( { \overline \theta } ) + \nonumber \right .
\\ & + & \left .
\left ( \frac{1}{2 } { \overline w}^{(1)}_{\dot \alpha } { \overline w}^{(2)\dot \alpha } + \frac{1}{4 } { \overline w}^{(2)}_{\dot \alpha } { \overline w}^{(2)\dot \alpha } \right ) \delta(\theta ) \right ] + \frac{1}{2 } \phi_{i}^{\dagger } v^{2 } \phi^{i } + \ldots \right\ } \label{relevantaction } \ , , \end{aligned}\ ] ] where a standard gauge fixing term has been included , whereas no ghost term is displayed since it will not be relevant for the computations to be discussed in this section . in equation ( [ relevantaction ] )
dots denote terms of higher order in @xmath13 , which do not contribute to the green functions that will be considered and thus will be suppressed from now on . in the following calculations the fermi feynman gauge , @xmath36=1 , will be used . with
this choice one obtains @xmath37 + \left [ \left ( \frac{1}{2 } w^{(1)\alpha } w^{(2)}_{\alpha } + \frac{1}{4 } w^{(2)\alpha } w^{(2)}_{\alpha } \right ) \delta ( { \overline \theta } ) + \right .
\nonumber \\ & + & \hspace{-0.2 cm } \left .
\left ( \frac{1}{2 } { \overline w}^{(1)}_{\dot \alpha } { \overline w}^{(2)\dot \alpha } + \frac{1}{4 } { \overline w}^{(2)}_{\dot \alpha } { \overline w}^{(2)\dot \alpha } \right ) \delta(\theta ) \right ] + \left [ \phi_{i}^{\dagger } v \phi^{i } + \frac{1}{2 } \phi_{i}^{\dagger } v^{2 } \phi^{i } \right ] \right\ } \ , , \label{ffaction}\end{aligned}\ ] ] where from the definition ( [ w1 - 2 ] ) one gets @xmath38
\theta_{\beta } + \nonumber \\ & & + \theta \theta \sigma^{\mu}_{\alpha{\dot \alpha } } \partial_{\mu}{\overline \lambda}^{\dot \alpha } \ , .
\label{w1}\end{aligned}\ ] ] the explicit form of @xmath39 is not necessary for the moment ( see however equation ( [ w2nonwz ] ) at the end of this section ) .
expansion of the action ( [ ffaction ] ) in components using the complete expression of @xmath13 gives @xmath40 the free action @xmath41 comes from the terms @xmath42 and @xmath43 and reads @xmath44 + \left [ \frac{1}{2 } s^{\dagger}_{a}\box s^{a } + \right .
\nonumber \\ & & \hspace{-1.6 cm } \left .
-\frac{1}{2 } c^{a } \box d_{a } -\frac{1}{2 } d^{a}\box c_{a } - \frac{1}{2 } c^{a } \box^{2 } c_{a } + \frac{1}{2 } \chi^{a } \box \lambda_{a } + \frac{1}{2 } { \overline \chi}^{a}\box { \overline \lambda}_{a } + \frac{1}{2 } \lambda^{a}\box \chi_{a } + \right
\nonumber \\ & & \hspace{-1.6 cm } \left .
+ \frac{1}{2 } { \overline \lambda}^{a}\box { \overline \chi}_{a } + \frac{1}{2 } \chi^{a } \box \sigma^{\mu } ( \partial_{\mu}{\overline \chi}_{a } ) + \frac{1}{2 } { \overline \chi}^{a } \box { \overline \sigma}^{\mu } ( \partial_{\mu}\chi_{a } ) + \frac{1}{2 } ( \partial_{\mu}a^{a}_{\nu } ) ( \partial_{\nu}a_{a\mu } ) \right ] \right\ } .
\label{freeact}\end{aligned}\ ] ] note that the scalar field @xmath21 and the spinor @xmath22 have `` wrong '' physical dimension , resulting in non standard free propagators , as will be shown below .
the interaction part @xmath45 contains an infinite number of terms .
the propagators of the fermions @xmath46 and of the scalars @xmath47 belonging to the @xmath0=1 chiral multiplets will now be computed at one loop .
the terms that are relevant for these calculations come from the expansion of @xmath48 and @xmath49 in ( [ ffaction ] ) .
the latter generates tadpole type diagrams in the propagator @xmath50 , but not in @xmath51 .
the interaction part of the action to be considered is @xmath52 - \frac{1}{2 } \left [ \varepsilon^{ijk } \left ( \varphi^{a\dagger}_{i } \varphi^{b\dagger}_{j}f^{c\dagger}_{k } - \varphi_{i}^{a\dagger}{\overline\psi}^{b}_{j}{\overline\psi}^{c}_{k }
\right ) + \right .
\nonumber \\ & & \hspace{-1 cm } + \left . \left . \left .
\varepsilon_{ijk } \left ( \varphi^{ai}\varphi^{bj } f^{ck } - \varphi^{ai } \psi^{bj}\psi^{ck } \right ) \rule{0pt}{14pt } \right ] \right ) -\frac{g^{2}}{2 } f_{abe}f^{e}{}_{cd}\left ( \left [ \rule{0pt } { 18pt } -c^{b}\psi^{d } \sigma^{\mu}{\overline \psi}^{a } a^{c}_{\mu } + \right . \right .
\nonumber \\ & & \hspace{-1 cm } \left .
\left . + ( { \overline \chi}^{c}{\overline \psi}^{d})(\chi^{b } \psi^{a } ) + \frac{i}{2 } \left ( \partial_{\mu } \psi^{d } \sigma^{\mu } { \overline \psi}^{a } - \psi^{d } \sigma^{\mu } \partial_{\mu}{\overline \psi}^{a } \right ) c^{b}c^{c } \right ] + \varphi^{a\dagger } \left [ c^{b } \left ( d^{c}+ \rule{0pt}{17pt } \right .
\nonumber \\ & & \hspace{-1 cm } \left .
\left . \left .
+ \frac{1}{2 } \box c^{c } \right ) - \chi^{b } \left ( \lambda^{c } + \frac{i}{2 } \sigma^{\mu}\partial_{\mu}{\overline \chi}^{c } \right ) - { \overline \chi}^{b } \left ( { \overline \lambda}^{c } + \frac{i}{2 } \sigma^{\mu } \partial_{\mu } \chi^{c}\right ) + \frac{1}{2 } s^{b\dagger}s^{c } + \rule{0pt}{17pt } \right .
\nonumber \\ & & \hspace{-1 cm } \left.\left.\left .
-\frac{1}{2 } a^{bc}_{\mu}a^{c\mu}\right ] \varphi^{d } + \frac{i}{2 } c^{b } a^{c\mu } \left ( \varphi^{a\dagger } ( \partial_{\mu } \varphi^{d } ) - ( \partial_{\mu}\varphi^{a\dagger } ) \varphi^{d } \right ) + \frac{1}{4 } c^{b}c^{c}\left [ \varphi^{a\dagger}(\box \varphi^{d})+ \right . \right .
\nonumber \\ & & \hspace{-1 cm } \left .
+ ( \box \varphi^{a\dagger } ) \varphi^{d } \right ] + \frac{i}{2 } \chi^{b}\sigma^{\mu } { \overline \chi}^{c } \left [ \varphi^{a\dagger } ( \partial_{\mu } \varphi^{d } ) - ( \partial_{\mu}\varphi^{a\dagger } ) \varphi^{d } \right ] \rule{0pt}{18pt}\right ) \right\ } \ , .
\label{intaction}\end{aligned}\ ] ] the quadratic part of the action , @xmath41 , can be written in a more compact form introducing the notation @xmath53 so that @xmath54 + \right .
\nonumber \\
& & + \left .
\left [ \frac{1}{2 } s^{\dagger}_{a}\box s^{a } + \frac{1}{2 } a^{a}_{\mu}a_{a}^{\mu } + { \cal b}_{a}^{t } m { \cal b}^{a } + { \cal f}_{a}^{t } n { \cal f}^{a } \right ] \right\ } \ , , \label{compactfree}\end{aligned}\ ] ] where @xmath55 inverting the kinetic matrices @xmath56 and @xmath57 one gets the free propagators . from ( [ kineticmatrices ] )
it follows @xmath58 so that the free propagators are + phi+phi ( 90,50 ) + s+s ( 90,50 ) + cd ( 90,50 ) + dd ( 90,50 ) + f+f ( 90,50 ) + aa ( 90,50 ) + chilam ( 90,50 ) + bchiblam ( 90,50 ) + blamlam ( 90,50 ) bpsipsi ( 90,50 ) to summarize beyond the ordinary propagators for the physical fields , @xmath59 , @xmath60 , @xmath61 and @xmath62 , and those for the auxiliary fields , @xmath63 and @xmath26 , one further obtains the propagators @xmath64 , @xmath65 , @xmath66 and @xmath67 .
the latter are absent in the wess
zumino gauge .
the one - loop correction to the propagator of the fermions @xmath46 in the chiral multiplet is the simplest to compute . in the wz gauge
there are three contributions at the one loop level , that will be shown to lead to a logarithmically divergent result . from the action ( [ freeact ] ) , ( [ intaction ] ) , with
@xmath21=@xmath22=@xmath23=0 one obtains the following three diagrams psipsia ( 150,80 ) + psipsib ( 150,80 ) + psipsic ( 150,80 ) notice that the insertion of tadpoles such as tadpole ( 100,40 ) in a diagram gives a vanishing result because all the propagators are diagonal in colour space , so that the tadpole contains a factor @xmath68 .
the same is true also for diagrams in @xmath0=1 superspace that will be discussed in subsequent sections .
this divergence will now be shown to be a gauge artifact due to the choice of the wz gauge .
in fact if the fields @xmath21 , @xmath22 and @xmath23 are included two further contributions must be added , corresponding to the diagrams psipsid ( 150,80 ) + psipsie ( 150,80 ) the contribution of these two diagrams exactly cancels the divergence in equation ( [ onelooppsiwz ] ) giving a net vanishing one - loop result for the propagator .
in fact the computation of @xmath26 and @xmath76 gives @xmath77 \right\ } \
, , \end{aligned}\ ] ] @xmath78 in momentum space one finds @xmath79 \ , , \ ]
] @xmath80 in conclusion summing up all the terms gives @xmath81 so that the one - loop correction to the @xmath82 propagator without fixing the wess
zumino gauge is zero , as expected in @xmath0=4 super yang
mills theory .
the calculation of the propagator @xmath83 is more complicated because more diagrams are involved . in particular tadpole type graphs , coming from the term @xmath49 in the action ,
are present .
however the result is analogous to what was found for the @xmath84 propagator : in the wz gauge the one - loop correction is logarithmically ultraviolet - divergent , requiring a wave function renormalization , but when the contributions neglected in the wz gauge are taken into account the total one - loop result vanishes . the diagrams contributing to @xmath85 at the one - loop level in the wess zumino gauge are the following phiphia ( 150,80 ) + phiphib ( 150,80 ) + phiphic ( 150,80 ) + phiphid ( 150,80 ) + phiphie ( 150,80 ) + phiphif ( 150,80 ) the last three diagrams are all tadpole corrections since the free propagators for the auxiliary fields @xmath63 and @xmath26 are @xmath28-functions , so that the corresponding lines shrink to a point .
each one of these diagrams is quadratically divergent .
there is also a quadratically divergent contribution coming from the two diagrams @xmath86 and @xmath87 and the sum of all these terms exactly vanishes .
the total sum gives a final result that is logarithmically divergent . in momentum space
it is schematically of the form @xmath88 just like in the case of @xmath89 this divergence can be reabsorbed by a wave function renormalization for the field @xmath59 .
taking into account the contribution of the fields @xmath21 , @xmath22 and @xmath23 there are eight more diagrams to be calculated phiphig ( 150,80 ) + phiphih ( 150,80 ) + phiphil ( 150,80 ) + phiphim ( 150,80 ) phiphin ( 150,80 ) phiphip ( 150,80 ) phiphiq ( 150,80 ) + phiphir ( 150,80 ) single diagrams contain quadratically divergent terms that cancel in the sum leaving a total contribution that is again logarithmically divergent .
this correction is exactly what is needed to cancel the divergence in ( [ phioneloopwz ] ) . as a result
the complete one - loop correction to the @xmath85 propagator is actually zero .
notice that the situation induced by the choice of the wz gauge is a feature common to every supersymmetric gauge theory , since the contribution of the gauge - dependent fields @xmath21 , @xmath22 and @xmath23 is in general logarithmically divergent . however in theories with less supersymmetry a logarithmic wave function renormalization is in any case unavoidable , so that this effect is completely irrelevant . on the contrary it becomes important in the @xmath0=4 yang
mills theory and in general in finite theories .
of course the divergences encountered here are gauge artifacts and disappear in gauge invariant correlation functions .
examples of computations of correlators of gauge invariant operators , in which the choice of the wz gauge does not lead to this kind of problems , have been considered within the discussion of the correspondence with type iib superstring theory on ads space , @xcite .
the calculations we described seem to suggest the possibility of constructing improved feynman rules in which the effect of the gauge dependent fields , set to zero in the wz gauge , is dealt with by a suitable redefinition of the free propagators .
however this program proves extremely complicated when the propagators of fields in the vector multiplet or three- and more - point functions are considered .
the calculation of these correlation functions , already at the one - loop level , requires many other terms to be included in the action @xmath45 .
in particular terms coming from @xmath90 and @xmath91 must be considered .
the calculation of @xmath92 without the simplifications introduced by the wess
zumino gauge is rather lengthy and gives @xmath93 \theta_{\beta } + \left [ c^{b}\sigma^{\mu}_{\alpha{\dot\alpha } } \partial_{\mu}{\overline\lambda}^{c{\dot\alpha } } + \right .
\nonumber \\ & & \left .
-i\chi_{\alpha}^{b}d^{c}-\frac{1}{4 } ( \sigma^{\mu } { \overline\sigma}^{\nu})_{\alpha}{}^{\beta } \chi_{\beta}^{b } ( \partial_{\mu}a^{c}_{\nu}-\partial_{\nu } a^{c}_{\mu } ) - \frac{1}{2 } a_{\nu}^{b } ( \sigma^{\mu } { \overline\sigma}^{\nu})_{\alpha}{}^{\beta } \partial_{\mu } \chi_{\beta}^{c } + \right .
\nonumber \\ & & \left .
\left .
+ \frac{1}{2}\partial_{\mu}a^{b\mu}\chi_{\alpha}^{c } - \frac{i}{2 } \partial_{\nu}c^{b } ( \sigma^{\mu } { \overline\sigma}^{\nu})_{\alpha}{}^{\beta } \partial_{\mu } \chi_{\beta}^{c } - \frac{i}{2 } \sigma^{\mu}_{\alpha{\dot\alpha } } \partial_{\mu}{\overline\chi}^{{\dot\alpha}b}s^{c } + s^{b}\lambda^{c}_{\alpha}+ \right . \right .
\nonumber \\ & & \left .
-i a_{\mu}^{b } \sigma^{\mu}_{\alpha{\dot\alpha } } { \overline\lambda}^{c{\dot\alpha } } \right ] \theta\theta \rule{0pt}{18pt}\right\ } \label{w2nonwz}\end{aligned}\ ] ] substituting ( [ w2nonwz ] ) into @xmath45 results in a number of relevant interaction terms of the order of 100 , making this formulation totally impractical in explicit calculations . in conclusion perturbation theory in components almost unavoidably requires the wz gauge , but the latter introduces divergences in gauge dependent quantities . in the following sections the superfield formalism , that allows to avoid these problems , will be employed .
however we will show that new difficulties related to `` ordinary '' gauge fixing emerge .
the difficulties encountered in the previous section in the calculation of gauge dependent quantities , because of the divergences present in the wess zumino gauge , can be overcome using the superfield formulation .
because of the lack of a completely consistent @xmath0=4 formulation , we will employ the @xmath0=1 superfield formalism which has proved to be a powerful tool in the proof of the finiteness of the theory up to three loops . in this approach
there is no particular difficulty in working without fixing the wz gauge .
if one does not choose to work in the wz gauge the action is non polynomial , however a finite number of terms is relevant at each order in perturbation theory , so that only at very high order the choice of the wess
zumino gauge introduces significant simplifications .
the aim of this section is to show that there are other subtleties related to further fixing the gauge for the vector superfield , even if one does not work in the wz gauge . to be more general and for the purpose of studying the possibility of finding a supersymmetric regularization of a class of @xmath0=1 theories , following the proposal of @xcite
, we will consider a formulation in @xmath0=1 superspace of @xmath0=4 super yang
mills theory deformed with the addition of mass terms for the ( anti ) chiral superfields @xmath94 .
\label{massterm}\ ] ] the inclusion of this terms to the action breaks @xmath0=4 supersymmetry down to @xmath0=1 . in @xcite it was argued , by means of dimensional arguments , that this term should not affect the ultraviolet properties of the @xmath0=4 theory .
more precisely the statement of @xcite is that no divergences appear in gauge invariant quantities , so that no divergent contribution to the quantum effective action is generated perturbatively . as a result
the model obtained in this way would be an example of a finite @xmath0=1 theory .
the inverse construction , in which one deforms a @xmath0=1 model to @xmath0=4 super yang mills plus a mass term , has been proposed in @xcite as a regularization procedure preserving supersymmetry .
the calculations presented in this section will show that in the presence of the term ( [ massterm ] ) no divergence is generated , at the one - loop level , in the two- , three- and four - point green functions that are computed .
the results presented suggest the possibility of reinforcing the conclusions of @xcite ; namely the @xmath0=4 theory augmented with ( [ massterm ] ) appears to be finite in the sense that no divergences appear in the complete @xmath1-point irreducible green functions , at least at one loop .
in particular no wave function renormalization is required .
notice that this result is what is actually necessary for the consistency of the approach advocated in @xcite .
the complete action we will be using for our perturbative calculations is thus will be taken to be real for simplicity of notation . ]
@xmath95 and reads @xmath96 v_{a } + { \phi^{\dagger}}^{a}_{i } \phi_{a}^{i } - \frac{1}{2 } m \delta_{ij } \phi_{a}^{i } \phi^{ja } \delta({\overline \theta } ) + \right .
\nonumber \\ & & \hspace{-1 cm } -\frac{1}{2 } m \delta^{ij}{\phi^{\dagger}}^{a}_{i}{\phi^{\dagger}}_{ja } \delta(\theta)+\frac{i}{\sqrt{2 } } g f_{abc}{\phi^{\dagger}}^{a}_{i } v^{b } \phi^{ic}- \frac{g^{2}}{2 } { f_{ab}}^{e}f_{ecd } { \phi^{\dagger}}^{a}_{i } v^{b } v^{c } \phi^{id } + \nonumber \\ & & \hspace{-1 cm } - \frac{i}{16 \sqrt{2 } } g f_{abc } \left [ { \overline d}^{2 } \left ( d^{\alpha}v^{a } \right ) \right ] v^{b } \left ( d_{\alpha}v^{c } \right ) -\frac{1}{128 } g^{2 } f_{ab}{}^{e } f_{ecd } v^{a } \left ( d^{\alpha } v^{b } \right ) \left [ \left({\overline d}^{2 } v^{c } \right ) \left ( d_{\alpha}v^{d } \right ) \right ] \hspace{-0.15 cm } + \nonumber \\ & & \hspace{-1 cm } + \ldots - \frac{1}{3 ! } g f^{abc } \left [ \varepsilon_{ijk } \phi_{a}^{i } \phi_{b}^{j}\phi_{c}^{k } \delta({\overline \theta } ) + \varepsilon^{ijk } { \phi^{\dagger}}_{ia}{\phi^{\dagger}}_{jb } { \phi^{\dagger}}_{kc } \delta(\theta ) \right ] + \left ( { { \overline c}^{\prime}}_{a } c^{a } - { c^{\prime}}_{a } { \overline c}^{a } \right ) + \\ & & \hspace{-1 cm } \left . + \frac{i}{2\sqrt{2 } } g f_{abc } \left ( { c^{\prime}}^{a } + { { \overline c}^{\prime}}^{a } \right ) v^{b } \left ( c^{c } + { \overline c}^{c } \right ) - \frac{1}{8 } g^{2 } { f_{ab}}^{e}f_{ecd } \left ( { c^{\prime}}^{a } + { { \overline c}^{\prime}}^{a } \right ) v^{b } v^{c } \left ( c^{d } + { \overline c}^{d } \right ) + \ldots \rule{0pt}{18pt } \right\ } \ , , \nonumber \label{n1sfaction } \end{aligned}\ ] ] where dots stand for terms that are not relevant for the considerations of this paper .
notice that in the action ( [ n1sfaction ] ) a gauge fixing term corresponding to a family of gauges parameterized by @xmath97 has been introduced .
it will now be shown , by explicitly computing the propagators of both the chiral and the vector superfields , that the supersymmetric generalization of the fermi feynman gauge , corresponding to @xmath98 , is somehow privileged ( see also @xcite ) , since any other choice of the parameter @xmath36 leads to infrared divergences in green functions .
the propagator of the chiral superfield is the simplest green function to compute .
the calculation will be reported in detail in order to illustrate the superfield technique .
it follows from the form of the action ( [ n1sfaction ] ) that there are three diagrams contributing to the propagator @xmath99 at the one loop level .
the one - particle irreducible parts of these diagrams will be directly evaluated in momentum space using the improved super feynman rules of @xcite .
the convention employed is that all momenta are taken to be incoming .
the diagrams are the following sphiphia ( 150,80 ) + sphiphib ( 150,80 ) sphiphic ( 150,80 ) + where @xmath100 .
the notations for the internal propagators and the corresponding @xmath27-space expressions are sphi+phi ( 90,50 ) + svv ( 90,50 ) + @xmath101 and @xmath102 in the @xmath103 propagator are the projectors @xcite @xmath104 moreover in the presence of mass terms for @xmath105 and @xmath106 there are extra @xmath107 and @xmath108 propagators , that will enter the calculation of the vector superfield propagator at one loop sphiphi ( 90,50 ) + sphi+phi+ ( 90,50 ) using the above expressions for the free propagators and the rules of @xcite , with the vertices read from the action ( [ n1sfaction ] ) , the three contributions can be evaluated without too much effort .
for the fourier transform of @xmath109 one finds @xmath110 \left [ \left(1+\gamma(d^{2}_{1 } { \overline d}^{2}_{1}+ { \overline d}^{2}_{1}d^{2}_{1})\right ) \left(-\frac{\delta^{de}\delta(1,2)}{k^{2}}\right)\right ] \frac{i}{\sqrt{2 } } g f_{efb } \phi^{bj}(-p,\theta_{2},{\overline \theta}_{2 } ) \rule{0pt}{20pt}\right\ } , \label{atildephiphi}\end{aligned}\ ] ] where @xmath111 and the compact notation @xmath112 has been introduced .
the computation uses the properties of the grassmannian @xmath28-function , which imply for example @xcite @xmath113 and integrations by parts on the grassmannian variables in order to remove the @xmath26 and @xmath114 derivatives from one @xmath28 , so that one @xmath115 integration can be performed immediately . in this way
one obtains an expression that is local in @xmath115 as is expected from the @xmath0=1 non - renormalization theorem @xcite . from ( [ atildephiphi ] )
one obtains @xmath116 } \left\ { \phi^{\dagger}_{ai}(p,1 ) \cdot \right .
\\ & & \hspace{-1.7 cm } \left .
\cdot \phi^{bj}(-p,2 ) \left [ { \overline d}^{2}_{1 } d^{2}_{1 } \delta(1,2 ) \right]\left [ 1+\gamma ( d^{2}_{1}{\overline d}^{2}_{1}+ { \overline d}^{2}_{1}d^{2}_{1 } ) \delta(1,2 ) \right ] \right\ } = \tilde{a}_{1}(p ) + \tilde{a}_{2}(p ) \rule{0pt}{20pt } \ , .\end{aligned}\ ] ] the first term is trivially calculated using @xmath117 \delta(\theta-\theta^{\prime } ) = 16 \nonumber \\ & & \hspace{-1 cm } \int d^{2}\theta d^{2}{\overline\theta } \ , \left [ { \overline d}^{m } d^{n } \delta(\theta-\theta^{\prime } ) \right ] \delta(\theta-\theta^{\prime } ) = 0 \quad { \rm if } \quad ( m , n ) \neq ( 2,2 ) \label{thetaint}\end{aligned}\ ] ] and gives @xmath118 } \left\ { \phi^{\dagger}_{ai}(p,\theta,{\overline \theta } ) \phi^{bj}(-p,\theta,{\overline \theta } ) \right\ } \ , . \label{a1tildephiphi}\ ] ] in the computation of the second term one must use the ( anti ) commutators of covariant derivatives , which in particular imply @xcite @xmath119 then integration by parts gives @xmath120}{k^{4}[(p - k)^{2}+m^{2 } ] } \left\ { \phi^{\dagger}_{ai}(p,\theta,{\overline \theta } ) \phi^{bj}(-p,\theta,{\overline \theta } ) \right\ } \ , .\ ] ] in conclusion from the first diagram one obtains two contributions , the first proportional to @xmath121 and the second independent of it .
the second diagram gives one single @xmath121-independent contribution .
the feynman rules give @xmath122 \left [ \frac{\delta^{d}_{f}\delta^{l}_{n } \delta(1,2)}{k^{2 } + m^{2 } } \right ] \left(- \frac{1}{3 ! }
\right ) \varepsilon_{j}{}^{mn } f_{be}{}^{f } { \phi}^{bj}(-p,2 ) \right\ } \ , .\end{aligned}\ ] ] proceeding exactly as for @xmath123 one obtains @xmath124 } \left\ { \phi^{\dagger}_{ai}(p,\theta,{\overline \theta } ) \phi^{bj}(-p,\theta,{\overline \theta } ) \right\ } \ , . \label{btildephiphi}\ ] ] from the last diagram one gets @xmath125
\phi^{bj}(-p,1 ) \right\ } \ , .\end{aligned}\ ] ] the only non - vanishing contribution comes from the term in which the projection operators act on the @xmath28-function , since @xmath126 , so that @xmath127 from which one sees that the @xmath128 contribution is absent in the gauge @xmath98 ,
_ i.e. _ @xmath129 . putting the various corrections together
gives the following result .
the sum of the terms @xmath130 and @xmath131 , which does not depend on @xmath121 , is @xmath132 \cdot \\ & & \cdot \left\ { \frac{1}{k^{2 } \left [ ( p - k)^{2 } + m^{2 } \right ] } - \frac{1}{(k^{2 } + m^{2 } ) \left [ ( p - k)^{2 } + m^{2 } \right ] } \right\ } \ , .\end{aligned}\ ] ] this is finite and exactly vanishes for @xmath133 , _
i.e. _ in the limit in which the @xmath0=4 theory is recovered .
the sum of the terms @xmath134 and @xmath135 is proportional to @xmath121 and reads @xmath136 \cdot \\ & & \cdot \left\ { \left ( \frac{(p - k)^{2}}{k^{4 } \left [ ( p - k)^{2 } + m^{2 } \right ] } - \frac{1}{k^{4 } } \right ) + \left ( \frac{p^{2}}{k^{4 } \left [ ( p - k)^{2 } + m^{2 } \right ] } \right ) \right\ } \ , .\end{aligned}\ ] ] both terms in the last integral are infrared divergent . the first one is zero in the limit @xmath137 , while the second one gives an infrared divergence that survives in the @xmath0=4 theory , _
i.e. _ in the limit @xmath138 .
more explicitly putting @xmath139 \left [ i_{1}(p ) + i_{2}(p ) \right ] \
, , \label{totphiphi}\ ] ] one has @xmath140 } - \frac{1}{k^{4 } } \right\ } = \\ & & = - \int \frac{d^{4}k}{(2\pi)^{4 } } \left\ { \frac{m^{2}}{k^{4 } \left [ ( p - k)^{2 } + m^{2 } \right ] } \right\ } = \\ & & = - 2 m^{2 } \int_{0}^{1 } d\zeta \ , \zeta \int \frac{d^{4}k}{(2\pi)^{4}}\frac{1}{{\left [ k^{2 } + ( p^{2 } \zeta + m^{2})(1-\zeta ) \right]}^{3 } } = \\ & & = - \frac{2 m^{2}}{2(4\pi)^{2 } } \int_{0}^{1 } d\zeta \ , \zeta
\frac{1}{(p^{2}\zeta + m^{2})(1-\zeta ) } = \\ & & = \frac{m^{2}}{(4\pi)^{2 } } \left [ \frac{1}{(p^{2}+m^{2 } ) } \log \epsilon \ ; + \ ; \frac{m^{2}}{p^{2}(p^{2}+m^{2 } ) } \log \left ( \frac{p^{2}+m^{2}}{m^{2 } } \right ) \right ] \
, , \end{aligned}\ ] ] where a standard feynman parameterization has been used . analogously @xmath141}\right\ }
= \\ & & = 2p^{2 } \int_{0}^{1 } d\zeta \
, \zeta \int \frac{d^{4}k}{(2\pi)^{4 } } \frac{1}{{\left [ k^{2 } + ( p^{2 } \zeta + m^{2})(1-\zeta ) \right]}^{3 } } = \\ & & = \frac{p^{2}}{(4\pi)^{2 } } \int_{0}^{1 } d\zeta \ , \zeta \frac{1}{(p^{2}\zeta + m^{2})(1-\zeta ) } = \\ & & = -\frac{1}{(4\pi)^{2 } } \left [ \frac{p^{2}}{(p^{2}+m^{2 } ) } \log \epsilon \ ; + \ ; \frac{m^{2}}{(p^{2}+m^{2 } ) } \log \left ( \frac{p^{2}+m^{2}}{m^{2 } } \right ) \right ] \ , .\end{aligned}\ ] ] in the above expressions an infrared regulator @xmath142 has been introduced .
notice that the total correction ( [ totphiphi ] ) is exactly zero on - shell , _
i.e. _ for @xmath143=@xmath144 . to summarize the results , the propagator of the chiral superfields of the @xmath0=4 super yang mills theory in @xmath0=1 superspace is logarithmically infrared - divergent for any choice of the gauge parameter @xmath145 .
this divergence corresponds to a wave function renormalization for the superfields @xmath12 . in the fermi
feynman gauge @xmath36=1 the one - loop correction exactly vanishes . the one - loop calculation of the propagator of the vector superfield is much more complicated , because many more diagrams are involved producing a large number of contributions .
the final result is however completely analogous : in the fermi
feynman gauge the one - loop correction is zero , whereas off - shell infrared divergences arise for @xmath145 . in the presence of a mass term for the ( anti ) chiral superfields
no new divergences are generated . from a calculational viewpoint
the new feature with respect to the @xmath146 case is that there are also diagrams involving the ghosts .
there are two multiplets of ghosts , described by the chiral superfields @xmath21 and @xmath147 . the free propagators for these superfields will be denoted by scprime+c ( 90,50 ) sc+cprime ( 90,50 ) the ghosts are treated exactly like ordinary chiral superfields with the only difference that there is a minus sign associated with loops , because @xmath21 and @xmath147 are anticommuting fields @xcite the corrections to the @xmath103 propagator at the one - loop level are given by the following diagrams vva ( 150,80 ) + vvb ( 150,80 ) vvc ( 150,80 ) + vvd1 ( 150,80 ) + vvd2 ( 150,80 ) + vvd3 ( 150,80 ) vve1 ( 150,80 ) vve2 ( 150,80 ) + vvf ( 150,80 ) vvg ( 150,80 ) the contributions @xmath148 to @xmath76 are rather straightforward to evaluate much in the same way as the diagrams entering the @xmath149 propagator .
the last two graphs are more involved because the free propagator for the @xmath13 superfield is more complicated for generic values of the parameter @xmath36 .
moreover the cubic and quartic vertices @xmath150 v^{b } \left ( d_{\alpha}v^{c } \right ) \\ & & -\frac{1}{128 } g^{2 } f_{ab}{}^{e}f_{ecd } v^{a } \left ( d^{\alpha } v^{b }
\right ) \left [ \left({\overline d}^{2 } v^{c } \right ) \left ( d_{\alpha}v^{d } \right ) \right ] \ , , \end{aligned}\ ] ] lead to several terms corresponding to the many ways in which the covariant derivatives can act on the @xmath13 lines .
the @xmath151 vertex in particular vvv ( 90,80 ) gives rise to six different terms .
schematically the calculation goes as follows .
@xmath152 is a new contribution that appears because of the addition of the mass terms ( [ massterm ] ) ; it is not present in the @xmath0=4 theory , _
i.e. _ when @xmath153=0 .
it is useful to discuss separately the corrections coming from diagrams @xmath154 , @xmath155 and @xmath128 and those obtained from @xmath156 , @xmath157 , @xmath158 and @xmath159 , since the latter correspond to the one - loop contribution to the vector superfield propagator in the @xmath0=1 supersymmetric yang mills theory .
the first diagram , @xmath154 , is logarithmically divergent and reads @xmath160 } \left\ { v^{a}(p,\theta,{\overline\theta } ) v^{b}(-p,\theta,{\overline\theta } ) \right\ } , \label{vvmassdiverg}\ ] ] for the diagram @xmath155 application of the feynman rules gives rise to three different contributions , one quadratically divergent and two logarithmically divergent @xmath161 } \left\ { k^{2}v^{a } ( p,\theta,{\overline\theta})v^{b}(-p,\theta,{\overline\theta } ) \rule{0pt}{18pt } - \right .
\nonumber \\ & & \hspace{-1 cm } \left .
-\frac{i}{4 } p_{\mu}\sigma^{\mu}_{\alpha{\dot\alpha}}v^{a } ( p,\theta,{\overline\theta } ) \left [ ( { \overline d}^{\dot\alpha}d^{\alpha } ) v^{b}(-p,\theta,{\overline\theta } ) \right ] + \frac{1}{4 } v^{a}(p,\theta,{\overline\theta } ) \left[({\overline d}^{2}d^{2 } ) v^{b}(-p,\theta,{\overline\theta } ) \right ] \right\ } \ , . \label{bcorrvv}\end{aligned}\ ] ] the tadpole diagram @xmath128 gives a quadratically divergent result of the form @xmath162 putting the three corrections @xmath154 , @xmath155 and @xmath128 together gives a net result that is only logarithmically divergent @xmath163 } \cdot \nonumber \\ & & \hspace{-1.3 cm } \cdot \left\ { \frac{i}{4 } p_{\mu}\sigma^{\mu}_{\alpha{\dot\alpha}}v^{a } ( p,\theta,{\overline\theta } ) \left [ ( { \overline d}^{\dot\alpha}d^{\alpha } ) v^{b}(-p,\theta,{\overline\theta } ) \right ] + \frac{1}{16 } v^{a}(p,\theta,{\overline\theta } ) \left[({\overline d}^{2}d^{2 } ) v^{b}(-p,\theta,{\overline\theta } ) \right ] \right\ } \ , . \label{vvlog1}\end{aligned}\ ] ] notice that in particular the logarithmically divergent contribution proportional to @xmath144 exactly cancels out .
this is crucial because this correction would correspond to a mass renormalization for the vector superfield that is known to be excluded in any gauge theory as well as in supersymmetric theories in general . the diagrams @xmath156 and @xmath157 are completely analogous to the previous ones , with the only difference that the mass does not appear in the denominators and there is a minus sign associated with the loops .
their sum is logarithmically divergent and takes the form @xmath164 + \frac{1}{8 } v^{a}(p,\theta,{\overline\theta } ) \left[({\overline d}^{2}d^{2 } ) v^{b}(-p,\theta,{\overline\theta } ) \right ] \right\ } \ , .
\label{vvlog2 } \end{aligned}\ ] ] all of the above corrections are independent of the gauge parameter @xmath121=@xmath165 and must be summed to those coming from the last two diagrams .
@xmath158 exactly vanishes for any @xmath36 , so that only @xmath166 needs to be considered .
this diagram produces in principle 72 corrections because the feynman rules give rise to 18 terms ( distributing the covariant derivatives associated with the two vertices ) , each of which splits into 4 , since the free propagator itself contains two terms .
many of these contributions can be easily shown to vanish using the properties of the covariant derivatives .
in particular one uses @xmath167 which follows from @xmath168 and use of the ( anti ) commutation relations for the @xmath26 s .
it is useful to separate in the non - vanishing part terms proportional to @xmath121 and @xmath169 , from the @xmath121-independent terms .
the latter combine to give a logarithmically divergent correction that together with that of equation ( [ vvlog2 ] ) cancel the correction ( [ vvlog1 ] ) coming from the sum @xmath170 .
actually if @xmath171 this sum is finite and exactly vanishes at @xmath153=0 . as a result
the only non - vanishing corrections to the vector superfield propagator at the one loop level come from terms proportional to @xmath121 and to @xmath169 in @xmath172 .
the former contain an infrared divergent part of the form @xmath173 \right\ } \
, , \ ] ] but is ultraviolet finite .
furthermore there is a correction , finite both in the ultraviolet and in the infrared regions , proportional to @xmath169 , that reads @xmath174 \left [ ( d^{2}{\overline d}^{\dot\alpha } ) v^{b}(-p,\theta,{\overline\theta } ) \right ] \right\ } \ , .\end{aligned}\ ] ] in conclusion in the @xmath0=4 theory , _
i.e. _ when @xmath153=0 , the one - loop correction to the vector superfield propagator , just like that of the chiral superfield , is ultraviolet finite , but infrared singular unless the fermi feynman gauge , @xmath36=1 , is chosen , in which case it vanishes .
like in the case of the chiral superfield propagator the non - vanishing @xmath121-dependent corrections could be reabsorbed by a wave function renormalization of the superfield @xmath13 and are zero on - shell , _
i.e. _ for @xmath143=0 .
the proof of the finiteness of the theory in the presence of the mass terms ( [ massterm ] ) given in @xcite is based on naive power counting , which gives for the superficial degree of divergence , @xmath4 , of a diagram in @xmath0=1 superspace @xcite @xmath175 where @xmath76 is the number of external ( anti ) chiral lines and @xmath21 the number of @xmath176 or @xmath177 propagators . in @xcite
it is also argued that for corrections to the effective action involving only @xmath13 superfields the requirement of gauge invariance reduces the degree of divergence to @xmath178 however for the purposes of @xcite it appears crucial that no divergences , not even corresponding to a wave function renormalization , be present in complete @xmath1-point functions for any @xmath1 .
the computation of the two - point functions in this section has shown that this actually the case at one loop .
an argument for the generalization of this result to green functions with an arbitrary number of external @xmath13 lines will now be briefly sketched .
first note that diagrams involving only vector and ghost superfields are not modified by the inclusion of ( [ massterm ] ) , so that one must only consider graphs containing internal chiral lines .
the ultraviolet properties of diagrams that are only logarithmically divergent in the original @xmath0=4 theory are not modified by the presence of the mass in the propagators .
thus the contributions that we need to analyze are the quadratically divergent ones , which can acquire subleading logarithmic singularities , plus eventually new diagrams involving @xmath179 and @xmath177 propagators .
the relevant quadratic divergences come from tadpole diagrams vvvvc ( 160,110 ) the only new diagram containing @xmath179 and @xmath177 propagators that must be considered is vvvva ( 160,110 ) the only covariant derivatives that are associated with the vertices in these graphs come from the functional derivatives and must act on the internal ( anti ) chiral lines to give a non - vanishing result . as a consequence
the loop integral in the above diagrams is exactly the same as for the @xmath128 and @xmath154 corrections to the @xmath180 propagator . hence summing to the previous diagrams the contribution of vvvvb ( 160,110 ) leads to a net logarithmically divergent correction that is exactly the same as the one obtained in the original @xmath0=4 theory . like in the latter case
this divergence will be cancelled by the contributions coming from the other one - loop diagrams involving @xmath13 and ghost internal lines .
the argument given here reinforces the results of @xcite , at least at the one - loop level , and supports the proposal , put forward in @xcite , according to which the mass deformed @xmath0=4 model can be considered a consistent supersymmetry - preserving regularization for ( a class of ) @xmath0=1 theories . notice that for the previous discussion it is not necessary to consider equal masses for all the ( anti ) chiral superfields .
the same results can be proved giving different masses to the three superfields .
this can be easily understood since in each diagram considered in this section only one chiral / anti - chiral pair is involved , because the propagators are diagonal in ` flavor ' space and the vertices containing vector and ( anti ) chiral superfields couple @xmath181 and @xmath12 with the same index @xmath182 .
basically this means that the above discussed cancellations apply separately to the contributions of each ( anti ) chiral superfield . from the viewpoint of the dimensional analysis of @xcite having different masses
@xmath183 is irrelevant . as a consequence one can in particular give mass to only two of the ( anti ) chiral multiplets
this suggests the possibility of generalizing the approach of @xcite to the case of @xmath0=2 super yang
mills theories . a discussion of the effect of a @xmath0=2 mass term in @xmath0=4 supersymmetric yang
mills theory can be found in @xcite .
the computation of green functions with three and four external legs will now be considered . from now on the fermi feynman gauge will be assumed .
the three - point functions are expected to suffer from infrared problems of the kind encountered in the preceding section .
this issue will not be addressed here since the calculation with @xmath145 is quite involved , even at the one - loop level , because of the huge number of contributions .
four - point functions on the contrary should be infrared finite , because they are directly related to physical scattering amplitudes .
notice , however , that infrared divergences in the green function with four external @xmath13 lines were found in @xcite with @xmath145 . in that paper beyond the infrared singularity
, it was shown that the adimensionality of the superfield @xmath13 implies that it requires a non - linear renormalization , in the sense that the renormalized field @xmath184 will be a non - linear function of the bare field @xmath13
@xmath185 an example of three - point function will be briefly discussed here and then the more interesting case of a four - point function will be studied in greater detail .
the simplest three point function that one can consider corresponds to the correction to the vertex @xmath186 ( or @xmath187 ) , which is determined by one single diagram at the one loop level .
however , as a less trivial example of computation of three - point functions , the one - loop correction to the vertex phi+vphi ( 150,95 ) will be considered here .
the correction at the one - loop level comes from the following diagrams ( the notation for the propagators is the same as in the previous section ) phi+vphia ( 150,95 ) phi+vphib ( 150,95 ) phi+vphic ( 150,95 ) phi+vphid ( 150,95 ) phi+vphie ( 150,95 ) phi+vphif ( 150,95 ) in the presence of mass terms for the chiral and antichiral superfields there are two additional contributions phi+vphig ( 150,95 ) phi+vphih ( 150,95 ) the calculation of these diagrams is completely analogous to those that were presented in the previous section .
the last two diagrams are finite and vanish in the @xmath0=4 theory , _
i.e. _ at @xmath153=0 , as they are proportional to @xmath144 ; they will not be considered here .
the other contributions will be briefly studied in the limit @xmath153=0 .
the diagram @xmath188 is zero as a consequence of the contraction among the colour indices , so that there are five contributions to be calculated .
dimensional analysis gives a vanishing superficial degree of divergence @xmath4 , corresponding to a logarithmic divergence , for the green function under consideration .
single diagrams actually contain a logarithmically divergent term plus finite terms .
a straightforward but rather lengthy computation , based on elementary @xmath26-algebra , allows to prove that the divergent part of all the diagrams is of the form @xmath189 where @xmath190 is a constant .
moreover one finds a finite contribution of the form @xmath191 \right .
\cdot \\ & & \hspace{1 cm } \left .
\cdot \phi^{j}_{c}(q,\theta,{\overline \theta } ) + c_{2}\sigma^{\mu}_{\alpha{\dot\alpha}}(q+k)_{\mu } \left [ d^{\alpha}{\overline d}^{\dot\alpha}v_{b } ( -p - q,\theta,{\overline \theta } ) \right ] \phi^{j}_{c}(q,\theta,{\overline \theta } ) \right\ } \ , , \end{aligned}\ ] ] where @xmath192 and @xmath193 are numerical constants .
the sum of all the logarithmically divergent terms contained in the diagrams @xmath154 , @xmath155 , @xmath128 , @xmath166 and @xmath158 vanishes .
the residual finite part can be shown to be zero as well . in conclusion
the one - loop correction to the three - point function @xmath194 exactly vanishes in the fermi
feynman gauge .
the same result can be shown to hold for the other three - point functions .
the superfield formalism does not lead to significant simplifications in the calculation of three - point functions with respect to the same computation in components ( in the wess zumino gauge ! ) .
actually for the green function considered here the number of diagrams to be evaluated is approximately the same as in the component formulation .
however the power of superspace techniques becomes clear in the computation of four - point functions that will be considered in next subsection .
the computation of four - point functions in components is extremely complicated even at the one - loop level and in the wess
zumino gauge . in this case
the choice of the wz gauge should not lead to extra divergences because the complete four - point function must finally give the physical scattering amplitude . in the @xmath0=1 superfield formulation
the calculation of four - point functions , though rather lengthy , is much more simple . in this section
the computation of the one - particle irreducible one - loop correction to the green function @xmath195 in the fermi
feynman gauge will be presented .
there are several diagrams to be considered : each of them is free of ultraviolet divergences , as immediately follows from dimensional analysis .
moreover in the fermi
feynman gauge each single diagram is infrared finite . in conclusion one finds a finite and non - vanishing result .
the first subset of contributions corresponds to the diagram phi+phiphi+phia ( 150,110 ) together with those obtained from this one by crossing , for a total of four terms of this kind .
steps completely analogous to those entering the evaluation of the propagators allow one to find the following results .
@xmath196 where @xmath197 \phi^{c\dagger}_{k}(p+q - r,\theta,{\overline \theta } ) \phi_{d}^{l}(r,\theta,{\overline\theta } ) \phi_{b}^{j}(q,\theta , { \overline\theta } ) + \phi^{a\dagger}_{i}(p,\theta,{\overline \theta } ) \cdot \right .
\nonumber \\ & & \hspace{-2.5 cm } \left .
\cdot\left[(d^{2}{\overline d}^{2 } ) \phi^{c\dagger}_{k}(p+q - r,\theta,{\overline \theta})\right ] \phi_{d}^{l}(r,\theta,{\overline\theta } ) \phi_{b}^{j } ( q,\theta,{\overline\theta } ) + 2\left[(d^{2}{\overline d}_{\dot\alpha})\phi^{a\dagger}_{i}(p , \theta,{\overline \theta})\right ] \cdot \right
. \nonumber \\ & & \hspace{-2.5 cm } \left.\cdot\left[{\overline d}^{\dot\alpha } \phi^{c\dagger}(p+q - r,\theta,{\overline \theta})\right ] \phi_{d}^{l}(r,\theta,{\overline\theta } ) \phi_{b}^{j}(q,\theta , { \overline\theta } ) + 2\left[{\overline d}_{\dot\alpha } \phi^{a\dagger}_{i}(p,\theta,{\overline \theta})\right ] \cdot \right .
\nonumber \\ & & \hspace{-2.5 cm } \left .
\cdot \left[(d^{2}{\overline d}^{\dot\alpha})\phi^{c\dagger } ( p+q - r,\theta,{\overline \theta})\right ] \phi_{d}^{l}(r,\theta , { \overline\theta } ) \phi_{b}^{j}(q,\theta,{\overline\theta } ) + 4\left[(d^{\alpha}{\overline d}_{\dot\alpha})\phi^{a\dagger}_{i } ( p,\theta,{\overline \theta})\right ] \cdot \right .
\nonumber \\ & & \hspace{-2.5 cm } \left .
\cdot \left[d_{\alpha } { \overline d}^{\dot\alpha}\phi^{c\dagger}(p+q - r,\theta,{\overline \theta})\right ] \phi_{d}^{l}(r,\theta , { \overline\theta } ) \phi_{b}^{j}(q,\theta,{\overline\theta } ) \right\ } \equiv \nonumber \\ & & \hspace{-2.5cm}\equiv \int \frac{d^{4}k}{(2\pi)^{4 } } d^{2}\theta d^{2}{\overline\theta } \frac{1}{k^{2}(p - k)^{2}(q+k)^{2}(q+k - r)^{2 } } k_{bdik}^{acjl } ( p , q , r;\theta;{\overline\theta } ) \label{4pointa1}\end{aligned}\ ] ] and the constant @xmath198 is a group theory factor defined by @xmath199 @xmath200 where @xmath201 @xmath202 where @xmath203 \left [ d^{2 } \phi^{j}_{b}(q,\theta,{\overline \theta})\right ] \phi^{c\dagger}_{k}(p+q - r,\theta,{\overline \theta } ) \phi^{l}_{d } ( r,\theta,{\overline \theta } ) + \right . \label{4pointa3}\\ & & \hspace{-3 cm } \left .
+ 8i\sigma^{\mu}_{\alpha{\dot\alpha } } ( p - k - r)_{\mu } \left [ { \overline d}^{\dot\alpha}\phi^{a\dagger}_{i } ( p,\theta,{\overline \theta})\right ] \left [ d^{\alpha } \phi^{j}_{b}(q,\theta,{\overline \theta})\right ] \phi^{c\dagger}_{k}(p+q - r,\theta,{\overline \theta } ) \cdot \right .
\nonumber \\ & & \hspace{-3cm}\cdot\left.\phi^{l}_{d}(r,\theta,{\overline\theta } ) - 16 ( p - k - r)^{2 } \phi^{a\dagger}_{i } ( p,\theta,{\overline \theta})\phi^{j}_{b}(q,\theta,{\overline\theta } ) \phi^{c\dagger}_{k}(p+q - r,\theta,{\overline \theta } ) \phi^{l}_{d } ( r,\theta,{\overline \theta } ) \right\ } \ , ; \nonumber\end{aligned}\ ] ] @xmath204 where @xmath205 \phi^{j}_{b}(q,\theta,{\overline \theta } ) \phi^{c\dagger}_{k}(p+q - r,\theta,{\overline \theta } ) \left [ d^{2 } \phi^{l}_{d}(r,\theta,{\overline \theta } ) \right]+ \right
. \nonumber \\ & & \hspace{-3 cm } \left .
+ 8i\sigma^{\mu}_{\alpha{\dot\alpha } } ( q+k)_{\mu } \left [ { \overline d}^{\dot\alpha}\phi^{a\dagger}_{i } ( p,\theta,{\overline \theta})\right ] \phi^{j}_{b}(q,\theta , { \overline \theta } ) \phi^{c\dagger}_{k } ( p+q - r,\theta,{\overline \theta } ) \left [ d^{\alpha } \phi^{l}_{d}(r,\theta,{\overline\theta } ) \right ] - \right .
\nonumber \\ & & \hspace{-3 cm } \left .
- 16 ( q+k)^{2 } \phi^{a\dagger}_{i } ( p,\theta,{\overline \theta})\phi^{j}_{b}(q,\theta,{\overline\theta } ) \phi^{c\dagger}_{k}(p+q - r,\theta,{\overline \theta } ) \phi^{l}_{d } ( r,\theta,{\overline \theta } ) \right\ } \ , .
\label{4pointa4}\end{aligned}\ ] ] the second kind of contributions correspond to the diagram phi+phiphi+phib ( 150,110 ) in this case the diagrams obtained by crossing are identical to the one depicted here , so that they are accounted for by giving the correct weight to @xmath206 .
one finds @xmath207 where @xmath208 \phi^{j}_{b}(q,\theta,{\overline \theta } ) \phi^{c\dagger}_{k}(p+q - r,\theta,{\overline \theta } ) \left [ d^{2 } \phi^{l}_{d}(r,\theta,{\overline \theta } ) \right]+ \right .
\nonumber \\ & & \hspace{-3 cm } \left .
+ 8i\sigma^{\mu}_{\alpha{\dot\alpha } } ( p - k)_{\mu } \left [ { \overline d}^{\dot\alpha}\phi^{a\dagger}_{i } ( p,\theta,{\overline \theta})\right ] \phi^{j}_{b}(q,\theta , { \overline \theta } ) \phi^{c\dagger}_{k } ( p+q - r,\theta,{\overline \theta } ) \left [ d^{\alpha } \phi^{l}_{d}(r,\theta,{\overline\theta } ) \right ] - \right .
\nonumber \\ & & \hspace{-3 cm } \left . - 16 ( p - k)^{2 } \phi^{a\dagger}_{i } ( p,\theta,{\overline \theta } ) \phi^{j}_{b}(q,\theta,{\overline\theta } ) \phi^{c\dagger}_{k}(p+q - r,\theta,{\overline \theta } ) \phi^{l}_{d } ( r,\theta,{\overline \theta } ) \right\ } \ , .
\label{4pointb}\end{aligned}\ ] ] the next one - loop correction is phi+phiphi+phic ( 160,90 ) this contribution is trivially zero because it contains the product @xmath209 this would not be true in a gauge different from the fermi
feynman gauge , _
i.e. _ with @xmath145 , because in that case there would be projectors acting on the @xmath28 s .
the vanishing of @xmath210 , which is completely trivial in the superfield formulation , corresponds , in the component formulation , to a complicated cancellation among various terms coming from graphs with the same topology .
another subset of diagrams includes phi+phiphi+phid ( 150,110 ) as well as the crossed ones .
there are three inequivalent crossed diagrams .
the result of the calculation consists of the following four terms ( @xmath211 , @xmath212 , @xmath213 , @xmath214 ) @xmath215 where @xmath216 \left [ d^{2 } \phi^{j}_{b}(q,\theta,{\overline \theta})\right ] \phi^{c\dagger}_{k}(p+q - r,\theta,{\overline \theta } ) \phi^{l}_{d}(r,\theta,{\overline \theta } ) + \right . \label{4pointd1 } \\ & & \hspace{-3 cm } \left .
+ 8i\sigma^{\mu}_{\alpha{\dot\alpha } } ( k+r - p)_{\mu } \left [ { \overline d}^{\dot\alpha}\phi^{a\dagger}_{i } ( p,\theta,{\overline \theta})\right ] \left [ d^{\alpha } \phi^{j}_{b}(q,\theta,{\overline \theta } ) \right ] \phi^{c\dagger}_{k}(p+q - r,\theta,{\overline \theta } ) \cdot \right .
\nonumber \\ & & \hspace{-3cm}\cdot\left .
\phi^{l}_{d}(r,\theta,{\overline\theta } ) - 16 ( k+r - p)^{2 } \phi^{a\dagger}_{i } ( p,\theta,{\overline \theta})\phi^{j}_{b}(q,\theta,{\overline\theta } ) \phi^{c\dagger}_{k}(p+q - r,\theta,{\overline \theta } ) \phi^{l}_{d } ( r,\theta,{\overline \theta } ) \right\ } \nonumber \ , ; \end{aligned}\ ] ] @xmath217 where @xmath218 \phi^{j}_{b}(q,\theta,{\overline \theta } ) \phi^{c\dagger}_{k}(p+q - r,\theta,{\overline \theta } ) \left [ d^{2 } \phi^{l}_{d}(r,\theta,{\overline \theta } ) \right ] + \right .
\nonumber \\ & & \hspace{-3 cm } \left .
+ 8i\sigma^{\mu}_{\alpha{\dot\alpha } } ( k - p - q)_{\mu } \left [ { \overline d}^{\dot\alpha}\phi^{a\dagger}_{i } ( p,\theta,{\overline \theta})\right ] \phi^{j}_{b}(q,\theta,{\overline \theta } ) \phi^{c\dagger}_{k}(p+q - r,\theta,{\overline \theta } ) \left [ d^{\alpha}\phi^{l}_{d}(r,\theta,{\overline\theta})\right ] - \right .
\nonumber \\ & & \hspace{-3 cm } \left .
- 16 ( k - p - q)^{2 } \phi^{a\dagger}_{i } ( p,\theta,{\overline \theta})\phi^{j}_{b}(q,\theta,{\overline\theta } ) \phi^{c\dagger}_{k}(p+q - r,\theta,{\overline \theta } ) \phi^{l}_{d } ( r,\theta,{\overline \theta } ) \right\ } \label{4pointd2 } \ , ; \end{aligned}\ ] ] @xmath219 where @xmath220 \left [ d^{2 } \phi^{l}_{d}(r,\theta,{\overline \theta } ) \right ] - \right .
\nonumber \\ & & \hspace{-3 cm } \left .
- 8i\sigma^{\mu}_{\alpha{\dot\alpha } } ( p - k - r)_{\mu } \phi^{a\dagger}_{i}(p,\theta,{\overline \theta } ) \phi^{j}_{b}(q,\theta,{\overline \theta } ) \left [ { \overline d}^{\dot\alpha}\phi^{c\dagger}_{k } ( p+q - r,\theta,{\overline \theta } ) \right ] \left [ d^{\alpha}\phi^{l}_{d}(r,\theta,{\overline\theta})\right ] - \right .
\nonumber \\ & & \hspace{-3 cm } \left . - 16 ( p - k - r)^{2 } \phi^{a\dagger}_{i } ( p,\theta,{\overline \theta})\phi^{j}_{b}(q,\theta,{\overline\theta } ) \phi^{c\dagger}_{k}(p+q - r,\theta,{\overline \theta } ) \phi^{l}_{d } ( r,\theta,{\overline \theta } ) \right\ } \label{4pointd3 } \ , ; \end{aligned}\ ] ] @xmath221 where @xmath222 \left[{\overline d}^{2}\phi^{c\dagger}_{k } ( p+q - r,\theta,{\overline \theta } ) \right ] \phi^{l}_{d}(r,\theta,{\overline \theta } ) - \right .
\label{4pointd4 } \\ & & \hspace{-3 cm } \left .
- 8i\sigma^{\mu}_{\alpha{\dot\alpha } } ( p+q - k)_{\mu } \phi^{a\dagger}_{i}(p,\theta,{\overline \theta } ) \left [ d^{\alpha}\phi^{j}_{b}(q,\theta,{\overline \theta})\right ] \left [ { \overline d}^{\dot\alpha}\phi^{c\dagger}_{k } ( p+q - r,\theta,{\overline \theta } ) \right ] \cdot \right .
\nonumber \\ & & \hspace{-3cm}\cdot\left .
\phi^{l}_{d}(r,\theta,{\overline\theta } ) - 16 ( p+q - k)^{2 } \phi^{a\dagger}_{i } ( p,\theta,{\overline \theta})\phi^{j}_{b}(q,\theta,{\overline\theta } ) \phi^{c\dagger}_{k}(p+q - r,\theta,{\overline \theta } ) \phi^{l}_{d } ( r,\theta,{\overline \theta } ) \right\ } \nonumber \ , .\end{aligned}\ ] ] the last family of one - loop diagrams consists of the one below plus again those obtained by crossing .
phi+phiphi+phie ( 150,110 ) the resulting contributions to the four - point green function are the following four terms ( @xmath223 , @xmath224 , @xmath225 , @xmath226 ) @xmath227 where @xmath228 @xmath229 where @xmath230 @xmath231 where @xmath232 @xmath233 where @xmath234 the sum of all the preceding terms results in a finite and non vanishing total one - loop correction to the four point function .
the final expression contains terms with six different tensorial structures @xmath235 where @xmath236 notice that , since in the fermi
feynman gauge the one - loop corrections to the propagators and vertices are zero , the total cross section is completely determined by the sum of the above contributions , which is non vanishing for on - shell external momenta , _
i.e. _ @xmath143=@xmath237=@xmath238=0 . in the presence of a mass term for the ( anti )
chiral superfields the expressions given above are modified by the presence of the mass in the free propagators and furthermore there are two additional sets of contributions corresponding to the diagrams phi+phiphi+phif ( 150,110 ) phi+phiphi+phig ( 150,110 ) both of these graphs give corrections proportional to @xmath144 , that can be calculated much in the same way as the previous ones .
the infrared divergences found in the calculation of the propagators of the chiral and the vector superfields are due to the fact that the vector superfield is dimensionless , so that it contains in particular , as its lowest component , the scalar @xmath21 that is itself dimensionless and hence has a propagator which behaves , in momentum space , like @xmath239 the contribution of the scalar @xmath21 to the @xmath103 propagator leads to an infrared divergence whenever a diagram contains a loop involving a @xmath180 line . in the fermi
feynman gauge the problem is not present because the choice @xmath98 gives a kinetic matrix of the form of equation ( [ kineticmatrices ] ) in the component expansion .
the corresponding inverse matrix @xmath240 in ( [ inverskinmatr ] ) shows that no @xmath241 free propagator is present . on the contrary any choice @xmath145 produces such a propagator , _
i.e. _ gives a matrix @xmath240 with a non - vanishing @xmath242 entry , leading to the previously discussed infrared problem .
an explicit calculation in components with @xmath145 and without fixing the wess
zumino gauge should display problematic infrared divergences analogous to those encountered here .
these problems with infrared divergences are not peculiar of the @xmath0=4 super yang mills theory , but appear in any supersymmetric gauge theory .
analogous infrared divergences in green functions have been observed in @xcite .
the conclusion proposed in these papers is that there exist no way to remove the infrared divergences , so that the choice @xmath36=1 is somehow necessary , at least for the computation of gauge - dependent quantities .
there are two general theorems concerning infrared divergences in quantum field theory .
the first is the kinoshita
lee nauenberg theorem @xcite , which deals with infrared divergences in cross sections .
it states that no infrared problem is present in physical cross sections of a renormalizable quantum field theory , if the appropriate sums over degenerate initial and final states , associated with soft and collinear particles , are considered .
the second theorem was proved by kinoshita , poggio and quinn @xcite and concerns green functions .
the statement of the theorem is that the proper ( one - particle irreducible ) euclidean green functions with non - exceptional external momenta are free of infrared divergences in a renormalizable quantum field theory .
a set of momenta @xmath243 , @xmath244 is said to be exceptional if any of the partial sums @xmath245 vanishes . the reason for the absence of divergences is that the external momenta , if non - exceptional , play the rle of an infrared regulator in the green functions .
the proof of the theorem is based on dimensional analysis which does not work in the presence of a @xmath246 propagator . in particular
it fails in the case of supersymmetric gauge theories in general gauges , because the adimensionality of the @xmath13 superfield implies that a propagator @xmath246 can appear in loop integrals .
the apparent violation of the kinoshita poggio
quinn theorem can be understood from the viewpoint of the component formulation : supersymmetric gauge theories , being non - polynomial and thus containing an infinite number of interaction terms , are not _ formally _ renormalizable .
the choice of the wess
zumino gauge makes the theory polynomial .
in fact in this case no infrared divergence is found in green functions and the general theorems are satisfied .
however in the case of the @xmath247=4 theory the choice of the wz gauge results in a change of the ultraviolet properties of the model , at least for what concerns gauge dependent quantities , _ e.g. _ the propagators . in theories with less supersymmetry ,
in which ultraviolet divergences are present the problem might be less severe , since the subtraction of the ultraviolet infinities could also cure infrared singularities .
the infrared problems discussed here are however gauge artifacts and can not affect gauge invariant quantities .
the mechanism leading to the cancellation of infrared divergences in gauge independent green functions has not been verified in detail yet , but can be understood starting from the super feynman rules . when the propagator @xmath248 is inserted into a feynman graph it is connected to vertices which carry covariant derivatives .
these derivatives come directly from the form of the action for vertices involving only @xmath13 fields and from the definition of the functional derivatives for vertices involving ( anti ) chiral superfields .
the situation in diagrams for gauge invariant quantities is such that at least one covariant derivative can always be brought to act on the @xmath13 propagator by integrations by parts . in this way
an additional factor of the momentum @xmath34 is generated in the numerator .
more precisely for the infrared problematic term @xmath249 one gets @xmath250 = -4i\frac{k_{\mu}\sigma^{\mu}_{\alpha{\dot\alpha}}}{k^{4 } } { \overline d}^{\dot\alpha}d^{2}\ ] ] and @xmath251 = 4i\frac{k_{\mu}{\overline \sigma}^{\mu{\dot\alpha}\alpha } } { k^{4 } } d_{\alpha}{\overline d}^{2 } \ , .\ ] ] a rigorous and general check of this mechanism in gauge invariant green functions has not been achieved yet . with a gauge choice different from the fermi
feynman gauge the computation of four - point green functions is more complicated .
single diagrams involving vector superfield propagators contain new contributions , some of which are infrared divergent .
the correction @xmath128 is not zero anymore , because there are projection operators acting on the @xmath28-functions .
moreover further diagrams must be included in the calculation of scattering amplitudes at the same order as a consequence of the non - vanishing of the one - loop correction to the propagators and vertices .
for example one must consider diagrams such as as well as diagrams obtained by the insertion of one - loop corrections to the propagators in tree graphs .
notice that the inclusion of these contributions , each one separately infrared divergent , is fundamental in order to get a finite total cross section .
the techniques illustrated in this paper for the calculation of green functions in the @xmath0=1 formalism can be applied to correlation functions of composite operators as well .
green functions of gauge invariant composite operators such as those that form the multiplet of currents ( see @xcite for the explicit form of the multiplet in the abelian case ) play a crucial rle in the correspondence with type iib superstring theory on ads space . the application of @xmath0=1 superspace to this problem has not been considered here , but is under active investigation @xcite .
it can be shown that the extension of the formalism to the case of composite operators is in principle rather straightforward , the fundamental difference being that the complete green functions and not the proper parts must be considered .
i am very grateful to massimo bianchi and giancarlo rossi for many useful discussions and for their comments on the manuscript .
i also wish to thank claudio fusi , yassen stanev and kensuke yoshida for stimulating discussions . | the finiteness properties of the @xmath0=4 supersymmetric yang mills theory are reanalyzed both in the component formulation and using @xmath0=1 superfields , in order to discuss some subtleties that emerge in the computation of gauge dependent quantities .
the one - loop corrections to various green functions of elementary fields are calculated . in the component formulation
it is shown that the choice of the wess zumino gauge , that is standard in supersymmetric gauge theories , introduces ultraviolet divergences in the propagators at the one - loop level .
such divergences are exactly cancelled when the contributions of the fields that are put to zero in the wess
zumino gauge are taken into account . in the description in terms of @xmath0=1
superfields infrared divergences are found for every choice of gauge different from the supersymmetric generalization of the fermi
feynman gauge .
two- , three- and four - point functions of @xmath0=1 superfields are computed and some general features of the infrared problem are discussed .
we also examine the effect of the introduction of mass terms for the ( anti ) chiral superfields in the theory , which break supersymmetry from @xmath0=4 to @xmath0=1 .
it is shown that in the mass deformed model no ultraviolet divergences appear in two - point functions .
it argued that this result can be generalized to @xmath1-point functions , supporting the proposal of a possible of use of this modified model as a supersymmetry - preserving regularization scheme for @xmath0=1 theories .
rom2f/99/01 + + + + + + |
in this work we attempt to calculate the populations of various isotopes of boron , carbon and nitrogen that were measured in a number of experiments at the nscl - msu facility[1 - 2 ] .
the calculation proceeds in two stages . in the first
, primary populations are calculated in a two - component statistical model .
the calculations in the first part are exact although numerical .
these populations are both in particle stable and unstable states . in the second stage
the particle unstable states are allowed to decay .
this is done in a weisskopf formalism .
exact calculations are very long and some approximations had to be introduced . these approximations will be discussed .
after the decays , the populations are compared with experiments .
one motivation for this calculation was that it serves as an application of the two - component statistical model where an exact calculation can be done .
this , therefore , could serve as a benchmark of how far one can trust the predictions of the model .
unfortunately , the particular predictions we are looking for are also affected by the subsequent decays .
this effect is not small .
hence , the predictions are the result of the combination of two models which had to be applied in tandem before experimental data could be compared .
a recent application of the two - component model was the computation of the caloric curve @xcite in nuclei .
the sections are organised as follows .
section ii gives a brief description of the two - component statistical model .
after presenting , in section iii , in words and simple formulae , the overview of the secondary decay calculation , we present in section iv the formalism that we use to model secondary decay . in section
v we present some calculational details , section vi presents the results of the calculation .
summary and conclusion are presented in section vii .
a short appendix of the more complicated formulae are presented in section viii .
for completeness , we present here the essential details of the two - component statistical model .
the one component model was described elsewhere @xcite .
the formalism of the two - component model can also be found in @xcite assume that the system which breaks up after two heavy ions hit each other can be desribed as a hot , equilibrated nuclear system characterised by a temperature @xmath0 and a freeze - out volume @xmath1 within which there are @xmath2 nucleons ( @xmath3 ) .
the partition function of the system is given by @xmath4 here @xmath5 is the number of composites with proton number @xmath6 and neutron number @xmath7 , and @xmath8 is the partition function of a single composite with proton , neutron numbers @xmath9 respectively .
there are two constraints : @xmath10 and @xmath11 .
these constraints would appear to make the computation of @xmath12 prohibitively difficult , but a recursion relation exists @xcite which allows numerical computation of @xmath12 quite easy even for large @xmath13 and @xmath14 .
three equivalent recursion relations exist , any one of which could be used .
for example , one such relation is @xmath15 all nuclear properties are contained in @xmath8 .
it is given by @xmath16 here @xmath17 is the free volume within which the particles move ; @xmath17 is related to @xmath1 through @xmath18 where @xmath19 is the excluded volume due to finite sizes of composites .
we take @xmath17 to be a variable of the calculation , it is set to be equal to @xmath20 where @xmath21 is the normal volume for @xmath22 nucleons , @xmath23 is then varied to obtain the best fit with experimental data .
the quantity @xmath24 is the internal partition function of the composite .
@xmath25 where the summation on the right hand side is the contribution from the discrete spectrum(the cut - off @xmath26 is simply the highest energy level that has been resolved for the given nucleus and is available from data tables ) ; and @xmath27 is the contribution from the continuum . without loss of generality we can write @xmath28 where we have used the abbreviation @xmath29 , to stand for both @xmath6 and @xmath7 ; @xmath30 is usually partly discrete and partly continuous .
we will need both @xmath31 and @xmath30 .
volumes of work are available on @xmath30 .
this is dealt with in detail in appendix 2b of @xcite .
the saddle - point approximation for the density of states assuming a fermi - gas model is ( see eq .
2b-14 in @xcite ) @xmath32 for explanations of how @xmath33 and @xmath34 are to be chosen see @xcite . in the fermi - gas model
the quantity which is exponentiated is simply the total entropy @xmath35 .
thus the density of states is given by a familiar expression @xmath36 where @xmath37 is the pre - factor .
approximate values of @xmath37 are known provided one does not have to concern with very low value of @xmath38 ( which we do need ) . at temperatures
we will be concerned with , @xmath39 in the fermi - gas model is given quite accurately by @xmath40 $ ] .
in the bulk of this paper we adopt this prescription . for upto @xmath41
we write the density of state as @xmath42 , where the low temperature fermi - gas expression for @xmath43 as written above is used .
the energy independent value of the pre - factor is fixed from experimentally known levels : @xmath44 while objections can be raised against this procedure , it achieves three objectives which we wanted to have : ( a ) we did not want to lose all information of the experimentally measured discrete excited states ; ( b ) we did want to take into account the contribution from the continuum and ( c ) with this procedure calculations are fairly simple . although , we will not report on all other formulae for density of states that we also used , our final results for the isotope populations are quite stable within reasonable variations that were tried .
we estimate the continuum contribution as a similar integral from @xmath26 to infinity .
@xmath45 this process is continued upto @xmath41 wherein we can read off energy levels from data tables . for elements above @xmath41 , a parametrised version was used , which is given as @xmath46 } \label{qint}\ ] ] where @xmath47 , @xmath48 , @xmath49 , @xmath50 and @xmath51 .
the first four terms in the right hand side of equation([qint ] ) arise from a parametrised version of the binding energy of the ground state .
the last term arises from an approximation to the fermi - gas formula for level density .
this was also used in @xcite . for protons and
neutrons @xmath52 is 1 .
the average number of particles of a composite is given by @xmath53 however , this population is partly over particle stable states and partly over particle unstable states which will decay into other nuclei before reaching the detectors .
in keeping with the way experimental data are presented , we will compute ratios of yields of different isotopes of boron , carbon , and nitrogen .
to lowest order one can consider the @xmath54 obtained from equation ( [ primary ] ) above , remove the particle unstable fractions , and compare them directly with experiment .
this is shown in the figures as the dotted line with a filled triangle plotting symbol .
these populations contain only particle stable states .
next we consider decay of the particle unstable states . we restrict the secondary decay to be due to emission of six species : neutron , proton , deuteron , @xmath55he , triton , and alpha particles .
any given nucleus @xmath56 from a particle unstable state can in principle go to at most six other nuclei . as the populations are canonically distributed among the various energy levels
, we can calculate the fraction that are in particle stable or unstable states .
if the fraction of nuclei @xmath56 at the first stage in unstable states is @xmath57 , then the number of nuclei @xmath56 left in particle stable states at the stage we call ` upto single decay ' is given by @xmath58 where @xmath59 is the fraction of the once decayed nuclei in unstable states .
we will indicate how to calculate @xmath59 in the next section .
@xmath60 is the width for emission by @xmath61 from @xmath62 and @xmath63 is the total width .
we can then take these revised populations @xmath64 and again compute the ratios .
we label these ` upto single decay ' .
these are reported in the plots as the small dashed line with the diamond plotting symbol .
note : this is just the stable fraction of the population after one stage of decay , the actual population is possibly greater .
after the first decay there may be some fraction in particle unstable states .
these can decay , thereby , changing the population of @xmath56 to @xmath65 .
if we take the ratios now we get what we call ` upto double decay ' , this is denoted by the dot@xmath66dashed line and the square plotting symbol .
again at this stage the @xmath67 represent only the sum of the stable fractions of the populations obtained from the initial distribution , single and double decays .
it is clear the procedure can be continued .
the fraction remaining in particle unstable states will continue to decrease .
we found no significant difference between the ` upto triple decay ' and the ` upto quadruple decay ' calculation .
thus we do not continue beyond .
once again it should be noted that all the plotted populations , @xmath54 , @xmath64 , @xmath67 , @xmath68 etc .
, quote only the stable fractions at freezeout , after single , double , and triple decay respectively .
the formalism for the decay calculation is given in the next section , there the quantities @xmath69 , @xmath60 will be calculated in somewhat greater detail .
the reader who is only interested in the final results could jump to sections vi and vii .
as the heated clusters stream out from the hot source , many of them will be in particle unstable states , these will decay by particle emission , for example , by emitting a neutron , proton , @xmath70 particle etc .
they will then leave a residue nucleus which may be particle stable or unstable ; if it is unstable then it will decay further into another isotope and this process will continue till the residue is produced in a particle stable state .
the primary calculation assumes that thermal equilibrium is achieved at freezeout ; if this is true then the number of composites with @xmath6 protons and @xmath7 neutrons with an energy in the interval @xmath38 and @xmath71 is given by the canonical factor @xmath72 where we have abbreviated @xmath2 to mean @xmath56 , and @xmath73 , from section ii .
, is given as @xmath74 .
the multiplicative constant @xmath75 will , henceforth , be absorbed into the overall normalization constant @xmath76 .
thus from now on the density function is given simply as @xmath77 \label{dens}\ ] ] @xmath76 is a normalization constant such that @xmath78 now of the various levels in a particular nucleus , some will be at a very low energy and as a result will be stable to any form of particle decay .
those that lie above an energy @xmath79 will in general be unstable to decay via emission of a particle @xmath80 ( i.e. , a particle with neutron number @xmath81 and proton number @xmath82 ) , where @xmath83 is the mass of the particle , @xmath84 is the mass of the decaying nucleus @xmath56 , @xmath85 is the mass of the residue left over after decay and @xmath86 is the coulomb barrier for that particle .
note that @xmath80 could represent a variety of particles ; in this note we will consider ` six ' such particles , as mentioned in the introduction . as is evident from equation ( [ exy ] ) ,
different particle decays have different energy thresholds .
consider an isotope @xmath56 , as an example let us take @xmath87 @xmath88 .
as we start from the ground state level and move upwards , we will encounter different thresholds .
the lowest will be the @xmath89 decay threshold at an energy @xmath90 ( in @xmath87 it is at @xmath91 approximately ) , the next higher threshold is for proton decay at @xmath92 ( in @xmath87 it is at @xmath93 approximately ) , and so on ; we will get different thresholds one after the other ( note : the order of different thresholds is different for different isotopes ) .
all nuclei of type @xmath56 which are formed between the ground state and the lowest threshold @xmath94 , will remain as isotopes @xmath56 , this number is given by @xmath95 those that are formed between @xmath94 and the next threshold @xmath96 , will all completely decay by @xmath89 emission , and these nuclei will then appear as nuclei of type @xmath97 and must be added on to the population of isotope @xmath97 .
then , those nuclei of type @xmath56 which are formed between @xmath96 and the next threshold @xmath98 , will decay both by @xmath89 emission and by proton emission . in the next zone
there will be three kinds of decay , and so on .
we now ask , how many of the initial nuclei formed in a particular zone will decay by each of the channels that are available , and how many of the residues formed will be stable or unstable ? to answer the above questions : we start by writing down the number of particles of type @xmath80 with energy between @xmath99 that are emitted , in a time interval between @xmath100 and @xmath101 , by nuclei of type @xmath56 , lying between an energy @xmath102 , leaving behind a residue nucleus @xmath103 ( we may alternatively refer to @xmath80 , @xmath56 , and @xmath103 by simply their mass numbers @xmath104 , @xmath2 and @xmath105 where @xmath106 , @xmath29 and @xmath107 ) @xmath108 where @xmath109 is the number of nuclei of type @xmath56 initially formed at an energy @xmath102 which are still left undecayed after a time @xmath100 , given by @xmath110 and @xmath111 is the weisskopf decay probability per unit time @xcite given by the expression @xmath112 } \frac { \rho_{b}(e - b_{a } - \varepsilon ) \rho_{b}^{0 } } { \rho_{a}(e ) \rho_{a}^{0 } } \label{wisko}\ ] ] in equation ( [ decay1 ] ) , @xmath113 is the the total decay probability per unit time from an energy level @xmath38 of the isotope @xmath2 . in equation ( [ wisko ] ) , @xmath114 is the spin degeneracy factor of the emitted particle , @xmath115 is a constant of a particular decay @xcite , given by @xmath116 where @xmath117 is the mass of a nucleon . in equation ( [ wisko ] ) ,
@xmath118 , is the separation energy of the decay ; @xmath119 } $ ] is the crossection for the reverse reaction to occur , ( i.e. @xmath120 ) .
it is given semiclassically for uncharged particles as @xmath121 } = \pi r_{a}^2\ ] ] and for charged particles as @xmath121 } = \pi r_{a}^2 \frac { \varepsilon - v } { \varepsilon } \theta ( \varepsilon -v ) \label{cros}\ ] ] where @xmath122 is the radius associated with the geometrical crossection of the formation of a from b and @xmath104 .
following the prescription of friedmann and lynch @xcite , @xmath122 is given by @xmath123r_{0 } & \mbox { , for $ a \geq 2 $ } \\ r_{0}(a-1)^{1/3 } & \mbox { , for $ a=1 $ } \end{array } \right.\ ] ] where @xmath124 .
in equation ( [ cros ] ) , @xmath1 is the coulomb barrier for the formation of a from b and @xmath104 . again following @xcite ,
this is written in the touching sphere approximation as @xmath125 where @xmath126 . also in equation ( [ wisko ] ) , @xmath127 , @xmath128 are the respective density of states of the two nuclei .
they have the same form as in equation ( [ dens ] ) .
also @xmath75 and @xmath129 are the respective multiplicative constants for the density of states , as mentioned in section(ii ) .
we note that @xmath130 in equation ( [ num1 ] ) is also equal to the number of nuclei that were initially formed as nuclei of type @xmath2 at an energy between @xmath38 and @xmath71 , and then decayed into nuclei @xmath105 with an excitation energy of @xmath131 . to get the total number of states that decayed from a level @xmath38 by emission of a particle of any allowed energy , we integrate over @xmath132 from its minimum value @xmath133 to its maximum value @xmath134 , and
get @xmath135 where @xmath136 which on integration gives @xmath137 \label{big}\end{aligned}\ ] ] ( the derivation of the above equation is given in the appendix ) , where @xmath138 , @xmath139 , @xmath140 . in the above equation @xmath114 ( equation([wisko ] ) ) , and some of the factors of @xmath141 ( equation([cros ] ) ) have been absorbed into @xmath142 thus @xmath143 we may now integrate out the time to get @xmath144 to get the total number of states that have decayed from nuclei of type @xmath2 by channel @xmath104 we must integrate over @xmath38 from @xmath94 to @xmath145 , @xmath146 this integration is quite involved for as we crossover from one zone of decay @xmath147 to another zone @xmath148 , @xmath149 changes discontinously as a new channel of decay becomes accessible to the nuclei .
thus we break up the integration into @xmath150 zones , corresponding to the @xmath150 real decay zones , and integrate within each zone independently . note that the last zone extends from @xmath151 to @xmath152 , and is thus considerably larger than the other zones
however , at the low temperatures that will be encountered , this zone will be sparsely populated .
thus the following approximation is valid . within each zone , with an energy from @xmath153 to @xmath154 , the integral can be replaced by a mean value expression , @xmath155 where @xmath156 is the mean number of nuclei of type @xmath56 ( or @xmath2 ) that were initially formed at an energy between @xmath153 and @xmath154 , and decayed by the @xmath80 ( or @xmath104 ) channel . in the above equation @xmath157 of course , the left hand side is zero if channel @xmath104 is not open in the region @xmath153 to @xmath154 .
the mean decay rate over all channels is @xmath158 and @xmath159 thus by summing up all the contributions from the six different zones , we get the total number of nuclei that have decayed from isotope @xmath2 by the @xmath104 channel as @xmath160 to find out how many of these have decayed to stable isotopes , we must first calculate from equation ( [ num1 ] ) the stable decay rate @xmath161 .
two cases emerge in this calculation . if @xmath162 , @xmath161 is obtained by integrating over @xmath132 , from @xmath163 to its maximum value @xmath164 , where @xmath165 is the stable level or the lowest threshold @xmath94 of the residue nucleus @xmath105 above which @xmath105 is unstable .
the expression for @xmath161 is obtained from that of @xmath166 in equation ( 18 ) by replacing @xmath167 . if @xmath168 , then @xmath169 .
then , following a similar procedure as above for @xmath170 , we get the total number of nuclei @xmath2 ( or @xmath56 ) lying in an energy range between @xmath171 , that decay by the @xmath104 channel to a stable state as @xmath172 the unstable decay rate from a particular level or zone is the probability of a decay per unit time from @xmath2 to an unstable level or levels of @xmath105 from which further decay can take place .
it is easy to see that they are given simply as the difference of the total decay rate and the stable decay rate i.e. , @xmath173 the derivations and expressions for the full decay rates are given in the appendix .
after a decay has taken place ( @xmath174 ) , we ask what is the population distribution of the residue as a function of its energy ( @xmath175 ) .
this can , in principle , be calculated from equation([num1 ] ) by integrating over @xmath38 and @xmath132 , such that @xmath176 , the energy of the residue , is a constant .
first we make a change of variables from @xmath177 to @xmath178 and then integrate over @xmath38 only .
we get @xmath179 this integration is quite involved .
we assume that the residue population is canonically distributed , but with a new temperature @xmath180 i.e. , @xmath181 there are two unknowns in this formula , the new temperature @xmath180 and the overall normalization constant @xmath182 . to find these two constants
we will impose that the total population of this interim stage ( i.e. @xmath183 ) , and the mean energy of the distribution @xmath184 , be reproduced by this new temperature .
we can obtain formal expressions for the total population of the residue @xmath105 as contributed by the decay of @xmath2 , as well as its mean energy @xmath184 , from equation([canres ] ) as @xmath185 \label{frmln}\end{aligned}\ ] ] @xmath186 \label{frmlx}\end{aligned}\ ] ] where the formal expression for @xmath187 is used in equation([frmlx ] ) .
the numerical value of @xmath188 is taken from equation([atob ] ) .
the numerical value of @xmath184 is found by explicit use of equation ( [ befshady ] ) . from these two equations
we obtain the two constants @xmath189 and @xmath190 .
the numerical value of @xmath184 is derived from equation ( [ befshady ] ) as follows .
@xmath191 in the above equation , the numerical value of @xmath188 is taken from equation([atob ] ) .
we may now change the order of integration to get @xmath192 the @xmath82 integration is now done simply to obtain @xmath193 where @xmath194 is given by @xmath195 \end{aligned}\ ] ] in the ensuing integration over @xmath38 , we , once again , replace the integral with its mean value expression .
@xmath196 where @xmath197 and @xmath198\end{aligned}\ ] ] thus the formal expressions for @xmath187 ( equation([frmln ] ) ) , and @xmath199 ( equation([frmlx ] ) ) , are compared to the actual values obtained for @xmath188 ( equation([atob ] ) ) , and @xmath184 ( equation([actlx ] ) ) , and the two unknowns of equation([canres ] ) are evaluated .
we can now proceed with further decays following the same procedure as before with decay occurring from a canonically distributed population at a temperature @xmath180 .
we can thus model an n@xmath66step decay process by assuming that at each intermediate stage the population is canonically distributed with a new temperature and overall normalization constant .
the decay rates to the next stage are calculated with the new temperature .
following this , the fraction of the population that decays through a particular channel , and the mean energy of the resultant residue nucleus , are calculated .
these are then used to secure the temperature and normalization constant of the next stage of decay .
this process will continue till the fraction of decay to particle unstable states becomes negligible .
from the primary calculation we obtain that @xmath200 nuclei of type @xmath56 ( or @xmath2 ) are formed from the initial multifragmentation
. the population @xmath200 is distributed canonically among the various energy levels as demonstrated by equation([canon ] ) .
if a particular nucleus is at a sufficiently excited state then it will emit a particle @xmath80 ( or @xmath104 ) and leave a residue @xmath201 ( or @xmath105 ) , which may again decay by emitting a particle @xmath202 ( or @xmath203 ) leaving a nucleus @xmath204 ( or @xmath205 ) , and so on untill it finally reaches a nucleus @xmath206 ( or @xmath13 ) in a stable state .
we ask the question that if @xmath207 nuclei of type @xmath2 were intially formed , then how many of these will finally end up as stable nuclei of type @xmath2 , @xmath105 , @xmath205 ... @xmath13 . the contribution of @xmath200 to the final stable population of @xmath2
is given simply by equation([stbpop ] ) as @xmath208 the number of nuclei initially formed as @xmath56 which decay to @xmath209 is given as @xmath210 the mean energy of the newly formed residue nucleus is given by @xmath211 we assume that this population is canonically distributed from an excitation energy of @xmath212 to @xmath145 with a new temperature @xmath180 ( equation [ canres ] ) .
extraction of the new temperature @xmath180 and the overall normalization constant @xmath213 is done as detailed in section(iv ) . in most cases , where this procedure was implemented
, we obtained a new temperature @xmath180 which was lower than the initial temperature @xmath214 ; however , in about @xmath215 of the cases @xmath180 turned out to be higher than @xmath214 ; this occurs when the residue of the decay process is far from the valley of stability .
we can then calculate the number of nuclei that initially started out as @xmath2 s and finally ended up as ` stable ' @xmath105 s as @xmath216 note that in the above equation @xmath217 and @xmath218 are the density of states and lowest decay threshold for the nucleus of type @xmath201 .
this number can also be calculated directly by using the stable decay rates ( equation([sdert ] ) ) @xmath219 the second equation is more correct as it does not depend on the assumption that the residue is canonically distributed .
a comparison of the @xmath220 obtained from the above two equations gives an estimate of the error involved in the assumption of a canonically distributed residue population .
now we ask , what is the number of nuclei of the b s just formed which will decay by emitting a particle @xmath203 to a nucleus of type @xmath205 ; this is calculated simply as @xmath221 the decay rates in the above equation are calculated with the temperature @xmath180 .
we then calculate the mean energy @xmath222 of the new distribution as @xmath223 using these , we continue the process on , by again calculating the temperature and norm of a canonical distribution , which when summed from excitation energy @xmath224 to @xmath145 is equal to @xmath225 , and whose mean energy is equal to @xmath226 .
we can then proceed to find how many of these will be in stable states , how many will decay on further etc .
we continue this process till the contribution from this decay chain , @xmath227 , will give numbers of nuclei negligible compared to the already present number in stable states .
our objective is to calculate the yields of the boron , carbon and nitrogen isotopes measured in the @xmath228 heavy - ion collision at an energy of @xmath229 @xcite . in figures 1 to 5 the data are shown as empty squares .
the method of calculation is simple , first we calculate the primary populations of the isotopes using equation([primary ] ) .
we then remove the unstable fraction of the population , and quote only the stable part .
this is denoted by the dotted line and triangle plotting symbol .
we then incorporate secondary decay by adding on all the populations of nuclei that can reach a stable level of the isotopes by emitting only one of the six particles considered .
we call these the ` upto single decay ' populations and denote them by the small dashed line and diamond plotting symbol .
we then add on all those unstable nuclei which can reach a stable level of the given isotopes by sequentially emitting any two particles of the six considered .
we call these the ` upto double decay ' populations and denote them by the dot@xmath66dashed line and square plotting symbol
. we then add on all those that can reach the isotopes by three particle emissions , called the ` upto triple decay ' population and denoted by the large dashed line and star plotting symbol .
and finally we add on the ` upto quadruple decay ' population denoted by the solid line and circle plotting symbol . as there is negligible difference between
` upto triple decay ' and ` upto quadruple decay ' we stop after ` quadruple decay ' . to fit with experimental data , we have four parameters to tune , the obvious ones being the initial temperature @xmath230 or @xmath0 , the free volume @xmath231 of the primary calculation , the ratio @xmath232 ( as one does not know how much loss due to pre@xmath66equilibrium emission has taken place ) and an overall multiplicative constant @xmath233 ( as we do not know how many nuclei collided in the experiment ) .
the plots are noted to be most sensitive to @xmath230 and @xmath232 .
thus in fitting the data we first set particular values of @xmath230 and @xmath232 , and calculate the multiplicities at all stages of decay ( @xmath231 is varied to get the best possible fit at this temperature and @xmath232 ) .
we then multiply all the multiplicities by an appropriate @xmath233 and take the logarithm .
these are then plotted and compared with @xmath234 obtained from the experiment .
we then vary @xmath230 , @xmath232 and repeat the above procedure till a good fit is obtained .
we present fits for three different temperatures , and different @xmath232 for each temperature .
@xmath231 and @xmath233 are set to obtain the best fit possible for a given @xmath230 and @xmath232 .
we note that the @xmath235 system is one with @xmath236 and @xmath237 thus @xmath238 .
the authors of @xcite state that some pre - equilibrium emission may have taken place . as we do not know what proportion of neutrons and protons
are lost in such a process , we start the calculation with the same @xmath232 as the @xmath235 system .
we start with a @xmath239 and @xmath238 i.e. @xmath240 .
we start the calculation with a low temperature of @xmath241 in figure([fig1 ] ) ( @xmath231 and @xmath233 are varied to get the best fit ) .
we note that overall there is a slight excess of the heavier nirogen isotopes as compared to data and a deficit of the lighter boron isotopes , this implies that the temperature is too low and enough of the light isotopes are not being formed .
we proceed by raising the temperature to @xmath242 , maintaining the same @xmath232 . by now varying @xmath231 and @xmath233
we find an excellent fit with the data ( figure([fig2 ] ) ) .
one may ask at this point , if there is more than one set of parameters which fits the data well . to answer this question we increase , first , the temperature to @xmath243 , maintain the same @xmath232 and redo the calculation . we get a bad fit ( figure([fig3 ] ) ) .
there is an over all deficit in the nitrogen population and an excess in the boron population .
also we note that within a particular @xmath13 there is a deficit in the neutron rich isotopes .
we try to remedy this situation by increasing the @xmath232 ratio .
the best fit at this temperature is obtained at an @xmath244 ( figure([fig4 ] ) ) , but we still obtain an overal deficit in the nitrogen population ; the carbon fit is good , but there still remains an excess in the boron isotopes especially in the neutron rich isotopes . on inspection of the fits ( figures ( [ fig1 ] ) to ( [ fig4 ] ) ) , we note that the best fit is obtained at figure([fig2 ] ) . in this fit @xmath245 , @xmath246 , @xmath247 , and @xmath248 . in this figure
we note that , for the boron populations we get an excellent agreement with the data . in this case
there seems to be little change after single decay . for the carbon isotopes
the agreement is good . for nitrogen
, we have a good fit except for the case of @xmath249 .
another property of the fits noticed is that they do not seem to depend on @xmath2 and @xmath13 independently but rather on the ratio @xmath232 . as a demonstration of this , we plot in figure([fig9 ] )
a fit for @xmath250 and @xmath237 ( i.e. , @xmath251 ) .
we note that we are able to obtain a fit very similar to fig.[fig2 ] , with the same temperature and @xmath252 as in fig.[fig2 ] , but with a slightly lower @xmath233 .
this is very much expected , as in this case each source has a larger number of nucleons than before .
in this note we have presented a secondary decay formalism and performed calculations to fit the populations of various isotopes measured in @xcite .
we obtain very good fits ( fig.[fig2 ] ) with experiment for the boron and carbon isotopes . in the nitrogen isotopes ,
we obtain a good fit except for the case of @xmath249 .
no particular reason could be found for this , but let us go over several approximations ( introduced to keep the calculation at a resonably simple level ) which may have contributed .
actual energy levels from data tables were used only upto @xmath253 ( equation ( [ qreal ] ) ) for the primary populations . for higher masses ,
the emperical mass formula ( equation ( [ qint ] ) ) was used .
the secondary decay is very approximate , instead of calculating decay level to level , we have blurred out such details by using a smoothed level density . for the capture cros - section ( equation([cros ] ) ) , we have used a simple semiclassical formula , assuming that all nuclei are spherically symmetric which is definitely not true . a more precise calculation involving level to level decay
would use a more accurate expression for the cros - sections e.g. , the hauser@xmath66feshbach formalism @xcite @xcite .
still another problem lies in the assumption made in calculating the effects of higher order decay , that the interim populations can be taken to be canonically distributed .
this is true only in first order decay , thus making the higher order contributions subject to some error .
there is also an experimental problem according to the authors of @xcite , the angular distributions were forward peaked , indicating significant emission prior to attainment of thermodynamic equilibrium .
such an emission could affect the populations of the various isotopes .
no doubt , incorporating changes to correct the above mentioned problems will improove the accuracy of the calculation .
however , such changes may make the expressions analytically intractible and one would have to resort to numerical means .
this may slow down the calculation considerably .
the calculations presented in this note take minimal computer time .
inspite of the shortcomings of the calculation presented above , this still remains a good test of the two component statistical model , and shows that such a model can definintely be used to explain certain experimental data quite accurately .
the authors wish to thank c. gale , w. g. lynch , m. b. tsang , s. pratt and n. de takacsy for helpful discussions .
this work was supported in part by the natural sciences and engineering research council of canada and by _ le fonds pour la formation de chercheurs et laide la recherche du qubec . _
* ( i ) derivation of equation [ big ] . * with @xmath263 , @xmath264
. the stable decay rate , i.e. , the decay rate from an energy level @xmath38 of the nucleus @xmath2 to any of the allowed stable levels of @xmath105 is given simply from the above expression by replacing the upper limit to @xmath265 where @xmath266 is the stable threshold of the residue nucleus @xmath105 .
however in the event that @xmath267 then the above mentioned replacement should not be made . in this case
the total decay rate is the same as the stable decay rate . | we compute the populations of isotopes of boron , carbon and nitrogen measured experimentally in intermediate energy heavy ion collisions .
a two component soluble statistical model is used to find the initial populations of different nuclei at a finite temperature .
these initial populations are both in particle stable and particle unstable states .
the particle unstable states then decay . the final populations after these decays are computed and compared with experimental data . |
the non - singlet components of the structure function @xmath4 @xmath5 and of the spin structure function @xmath1 @xmath6 have been investigated in great detail in deep inelastic scattering ( dis ) experiments .
the standard theoretical framework is provided by dglap@xcite . in this approach ,
any of @xmath7 can be represented as a convolution of the coefficient functions and the evolved quark distributions calculated in lo@xcite , nlo@xcite and nnlo@xcite .
combining these results with appropriate fits for the initial quark distributions , provides a good agreement with the available experimental data ( see e.q.@xcite and the recent paper @xcite ) on the other hand , the dglap evolution equations were originally applied in a range of rather large @xmath2 values , where higher - loop contributions to the coefficient functions and the anomalous dimensions are small .
according to the results of refs @xcite , such corrections are becoming more and more essential when @xmath2 is decreasing and dglap should not work so well at @xmath8 .
nevertheless , it turns out that dglap predictions are in a good agreement with available experimental data , leading to the conclusion that the impact of the higher - order corrections is negligibly small for the available values of @xmath2 and those corrections may be relevant at asymptotically small @xmath2 only .
in the present paper we use our previous results @xcite to show that the impact of the high - order corrections on the @xmath3 and @xmath2 -evolutions of the non - singlet structure functions is indeed quite sizable at available values of @xmath2 .
we also show that the reason for the success of phenomenological analysis of the data based on dglap at @xmath9 is related to the sharp @xmath2 -dependence assumed for the initial parton densities , which is able to mimic the role of high - order corrections .
the paper is organized as follows : in sect .
2 we introduce our notations and discuss the difference of our approach with dglap . in sect .
3 the dglap and our predictions for the small-@xmath2 asymptotics are explicitly compared .
it allows us to clarify the role played by the singular terms in @xmath2 in the dglap inputs for the initial parton densities . in sect .
4 we give a detailed numerical comparison of our results with dglap , and discuss also the role played by the regular part of the initial parton densities . as our approach deals with small @xmath2 only , in sect . 5
we suggest a possible method to combine dglap with our approach in order to obtain equally correct expressions for both large and small values of @xmath2 . finally , sect .
5 contains our conclusions .
as the dglap -expressions for the non - singlet structure functions are well - known , we discuss them briefly only . in this approach
, both @xmath10 and @xmath11 @xmath12 can be represented as a convolution @xmath13 of the coefficient functions @xmath14 and the evolved quark distributions @xmath15 .
similarly , @xmath16 can be expressed through the convolution of the splitting functions and the initial quark densities @xmath17 where @xmath18 is the starting point of the @xmath3 -evolution .
it is convenient to represent @xmath19 in the integral form , using the mellin transform : @xmath20\ ] ] where @xmath21 are the non - singlet coefficient functions , @xmath22 the non - singlet anomalous dimensions and @xmath23 the mellin transforms of the initial non - singlet quark densities . in the @xmath2- space , the standard dglap inputs for @xmath24 for the non - singlet parton densities ( see e.g. refs .
@xcite ) include terms which are singular when @xmath25 , as well as regular ones for all @xmath2 .
for example , the fit a of ref.@xcite is chosen as follows : @xmath26 with @xmath27 being normalization factors , @xmath28 , @xmath29 , @xmath30 and @xmath31 .
as the first term @xmath32 in the rhs of eq .
( [ fita ] ) is singular when @xmath25 whereas the second one , @xmath33 is regular , we will address them as the singular and regular parts of the fit respectively . obviously , in the @xmath34 -space eq .
( [ fita ] ) is a sum of pole contributions : @xmath35~,\ ] ] with @xmath36 , so that the first term in eq
. ( [ fitaomega ] ) ( the leading pole ) corresponds to the singular term @xmath37 of eq .
( [ fita ] ) and the second term , i.e. the sum of the poles , corresponds to the interference between the singular and regular terms .
in contrast to the leading pole position @xmath38 , all other poles in eq .
( [ fitaomega ] ) occur at negative values , because @xmath39 . as usual , when @xmath34 is integer : @xmath40 , the integrand of eq .
( [ fdglapmellin ] ) gives the @xmath41-th moment of @xmath42 , @xmath43 .
the standard dglap -technology of calculating @xmath44 corresponds to first calculating @xmath45 and then reconstructing @xmath44 from the moments , choosing appropriate forms for @xmath46 .
presently @xmath21 for the non - singlet structure functions are known with two - loop @xcite and three - loop@xcite accuracy . in order to make the all - order resummation of the double - logarithmic contributions to @xmath10 and @xmath11 , i.e. the contributions @xmath47 to @xmath21 and @xmath22 in eq .
( [ fdglapmellin ] ) , an alternative approach was used in refs .
@xcite , by introducing and solving infrared evolution equations with fixed @xmath48 .
this approach was improved in refs .
@xcite , where single - logarithmic contributions were also accounted for and the qcd coupling was running in the feynman graphs contributing to the non - singlet structure functions .
in contrast to the dglap parametrization @xmath49 , we used in refs .
@xcite another parametrization where the argument of @xmath48 in the quark ladders is given by the time - like virtualities of the intermediate gluons .
arguments in favor of such a parametrization were given in ref .
in particular , it was shown that it coincides with the dglap -parametrization when @xmath50 , but it differs sensibly when @xmath51 . instead of eq .
( [ fdglapmellin ] ) , refs .
@xcite suggest the following formulae for the non - singlet structure functions : in eq .
( [ fnsint ] ) was performed in refs .
@xcite,@xcite , however with simplified assumptions for @xmath48 . ] @xmath52 @xmath53 with @xmath54 so that @xmath18 is the starting point of the @xmath3 -evolution .
the new coefficient functions @xmath55 are expressed in terms of new anomalous dimensions @xmath56 : @xmath57 and the new anomalous dimensions @xmath58 which account for the resummation of the double- and single- logarithmic contributions are @xmath59\ ] ] where @xmath60 finally @xmath61 and @xmath62 in eq .
( [ b ] ) are expressed in terms of @xmath63 , @xmath64 and the color factors @xmath65 , @xmath66 : @xmath67 ~,\ ] ] @xmath68\ ] ] @xmath62 is the mellin transform of @xmath48 with the time - like argument @xmath69 , @xmath70 . the comparison of eq .
( [ fdglapmellin ] ) to eqs .
( [ fnsint ] , [ gnsint ] ) clearly shows the difference between dglap and our approach : first , our approach accounts for the double- and single- logarithmic contributions to all orders in the qcd coupling whereas in the dglap -expressions the coefficient functions and the anomalous dimensions are obtained up to fixed - order accuracy ( two and three loops ) .
then a second difference is related to the treatment of @xmath48 . in dglap ,
the argument of @xmath48 in any feynman graph is @xmath71 , with @xmath72 being the momentum of the ladder quarks , and therefore @xmath48 is not involved in the mellin transform .
this leads to the @xmath3 dependence of the non - singlet structure functions with the exponent @xmath73 . in our approach @xmath48
depends on time - like virtualities of the ladder gluons and therefore , it is involved in the mellin transform .
the inclusion of @xmath48 in the mellin transform explicitly shows that the factorization between the longitudinal and transverse spaces used in dglap is not valid for small @xmath2 .
however , as shown in ref .
@xcite , our approach and dglap converge when @xmath50 where the factorization does hold .
the parametrization of @xmath74 in the form of eq .
( [ a ] ) has recently been used in refs .
@xcite for calculating the double- and single- logarithmic corrections to the bjorken sum rules .
in the first place let us study the small-@xmath2 asymptotics of the non - singlet structure functions .
when @xmath25 , one can use the saddle point method in order to estimate the integrals in eqs .
( [ fnsint],[gnsint ] ) and derive much simpler expressions for the non - singlet structure functions . when the initial parton densities @xmath76 are regular in @xmath2 ( e.g. given by @xmath77 only in eq .
( [ fita ] ) ) , they do not contribute to the saddle point position , which is : @xmath78~,\ ] ] with @xmath79 and @xmath80 , @xmath81 . it immediately leads to the regge asymptotics for the non - singlets : @xmath82 with @xmath83^{1/2 } \big(y/2 + 1/\sqrt{b^{(\pm ) } } \big)}{\pi^{1/2}\ln^{3/2}\xi}~.\ ] ] for @xmath84gev@xmath85 , @xmath86 , so @xmath87 do not depend on @xmath88 .
therefore eq . ( [ as ] ) can be rewritten as : @xmath89 with @xmath90 and the intercepts @xmath91 and @xmath92 ( see @xcite for details ) .
the factors @xmath93 , @xmath94^{1/2}\ ] ] do not depend on @xmath88 .
( [ asympt ] ) predicts the asymptotic scaling for the non - singlet structure functions : in the region @xmath95gev@xmath85 , with @xmath96 depending on one argument @xmath97 instead of @xmath2 and @xmath3 independently . in the opposite case
when @xmath98gev@xmath85 , the scaling is valid for @xmath99 and @xmath100 .
the small-@xmath2 dependence of eq .
( [ asympt ] ) is confirmed by various analyses of the experimental data in refs .
@xcite , while the @xmath3- dependence of these formulae has not been checked yet . on the other hand , under the same hypothesis that @xmath76 is regular ( e.g. given again by @xmath77 in eq .
( [ fita ] ) ) , the small-@xmath2 asymptotics of @xmath101 is well - known ( we drop the inessential pre - factor here ) : @xmath102 obviously , this behavior is different and less steep than the regge asymptotics of eq .
( [ asympt ] ) .
however when the singular part in @xmath75 is also included in the dglap input of eq .
( [ fita ] ) , the dglap asymptotics for the non - singlet structure functions is controlled by the leading singularity @xmath103 of the fit in the integrand in eq .
( [ fitaomega ] ) so that the asymptotics of @xmath104 is also of the regge type : @xmath105 with @xmath64 .
the comparison of eq .
( [ asympt ] ) and eq .
( [ asdglap ] ) shows that both dglap and our approach lead to the regge asymptotic behavior in @xmath2 for the non - singlet structure functions . in particular , eq .
( [ fita ] ) makes the dglap prediction be more singular in @xmath2 than ours , also they differ as far as the predictions for the @xmath3 -behavior are concerned .
however , it is important to stress that our intercepts @xmath106 are naturally obtained by the total resummation of the leading logarithmic contributions , without assuming any singular input for @xmath75 , whereas the dglap intercept @xmath107 in eq .
( [ asdglap ] ) is generated by the phenomenological factor @xmath108 of eq .
( [ fita ] ) , which makes the structure functions grow when @xmath2 decreases , and mimics in fact the total resummation were confirmed by several analyses@xcite of the experimental data ] . in other words ,
the role of the higher - loop radiative corrections on the small-@xmath2 behavior of the non - singlets is , actually , incorporated into dglap phenomenologically , through the initial parton densities fits .
it also means that the singular factors can be dropped from such fits when the coefficient functions account for the total resummation of the leading logarithms and therefore regular functions of @xmath2 only can be chosen for the initial densities @xmath75 .
as discussed in sect .
ii , the regular part in the initial quark distribution in the dglap approach , e.g. @xmath109 in eq . ( [ fita ] ) , corresponds to the sum of pole contributions @xmath110 : @xmath111~.\ ] ] we have introduced here the normalization factor @xmath112 .
obviously , @xmath113 when @xmath25 .
an input @xmath114 means that the shape of the initial quark distribution does not depend on @xmath2 . in order to study the influence of the initial parton densities in more detail ,
we compare numerically our results ( [ gnsint ] ) for two situations : ( a ) a regular fit @xmath115 is used , ( b ) @xmath116 .
we define @xmath117 as follows : @xmath118/g_1^{ns}\big [ \delta q(x ) = n \big]\ ] ] at @xmath119gev@xmath85 . for all numerical studies in the present section
we choose the starting point of the @xmath120 -evolution @xmath121gev and assume @xmath122 gev and @xmath123 .
the results for @xmath124 are plotted in fig . 1 ( curve 1 ) .
it shows that one can approximate @xmath77 by a constant distribution for @xmath125 whereas at larger values of @xmath2 the @xmath2 -dependence of the initial quark distributions is essential . the comparison of our results ( [ gnsint ] ) with dglap when the regular fit @xmath77 is used is also shown in fig . 1 ( curve 2 ) for the ratio @xmath126 defined as @xmath127/g^{ns}_{1~dglap}\big [ \delta q(x ) = \phi(x ) \big]\ ] ] where we choose again @xmath119gev@xmath85 .
this shows that when the input for @xmath75 does not have a singular term , the difference between our results and dglap grows fast for @xmath128 . on the contrary , when the singular term is included into the dglap input , the dglap results grow faster than ours
this is shown in fig . 1 ( curve 3 ) for the ratio @xmath129 : @xmath130/ g^{ns}_{1~dglap}\big [ \delta q(x ) = x^{-\alpha } \phi(x ) \big]~,\ ] ] with @xmath119gev@xmath85
. a fast growth of @xmath131 $ ] for small @xmath2 also contradicts the analyses of experimental data obtained in refs .
@xcite .
our results ( [ fnsint],[gnsint ] ) differ slightly from dglap also for the @xmath3 -evolution , depending on x. this is shown in fig . 2 where the ratio @xmath132 : is plotted for @xmath134 ( curve 1 ) , @xmath135 ( curve 2 ) and for @xmath136 ( curve 3 ) .
2 also suggests that the difference between our and dglap predictions for the @xmath3 -evolution slowly grows with increasing @xmath3 .
we have shown that eqs .
( [ fnsint],[gnsint ] ) account for the resummation of the double- and single logarithmic contributions to the non - singlet anomalous dimensions and the coefficient functions that are leading when @xmath2 is small .
however , the method we have used does not allow us to account for all other contributions which can be neglected for @xmath2 small but become quite important when @xmath2 is not far from 1 .
on the other hand , such contributions are naturally included in dglap , where the non - singlet coefficient function @xmath137 and anomalous dimension @xmath138 are known with the two - loop accuracy : @xmath139 therefore , we can borrow from the dglap formulae the contributions which are missing in eqs .
( [ fnsint],[gnsint ] ) by adding @xmath137 and @xmath138 to the coefficient functions and anomalous dimensions of eqs .
( [ fnsint],[gnsint ] ) . of course , both @xmath137 and @xmath138 contain also terms already included in eqs .
( [ fnsint],[gnsint ] ) , so in order to avoid a double counting these terms should be extracted from @xmath140 . in order to do so ,
let us consider the region of @xmath50 where the effective values of @xmath34 in eqs .
( [ fnsint],[gnsint ] ) are large . in this region
we can expand @xmath56 and @xmath55 into a series in @xmath141 . retaining the first two terms in each series and expressing them through @xmath142 and @xmath143 , we arrive at @xmath144 , @xmath145 with @xmath146~ , \\
\nonumber & & \widetilde{h}_{ns}^{(\pm ) } = \frac{a(\omega)c_f}{4\pi } \big [ 2/\omega + 1 \big ] + \big(\frac{a(\omega)c_f}{4\pi}\big)^2 ( 1/\omega)\big [ 2/\omega + 1 \big]^2 + d^{(\pm)}(\omega)[1/\omega + 1/2]~.\end{aligned}\ ] ] now let us define the new coefficient functions @xmath147 and new anomalous dimensions @xmath147 as follows : @xmath148 + \frac{a(\omega)}{4\pi}\gamma^{(0 ) } + \big(\frac{a(\omega)}{4\pi}\big)^2 \gamma^{(1)}~ , \\
\nonumber & & \hat{c}_{ns}^{(\pm ) } = \big[c_{ns}^{(\pm ) } - \widetilde{c}_{ns}^{(\pm ) } \big ] + 1 + \frac{a(\omega)}{2\pi } c^{(1)}~.\end{aligned}\ ] ] these new , `` implemented '' coefficient functions and anomalous dimensions of eq .
( [ newch ] ) include both the total resummation of the leading contributions and the dglap expressions in which @xmath149 is replaced by @xmath62 . as shown in ref .
@xcite , this parametrization differs from the dglap one , @xmath49 , at small @xmath2 , though both parameterizations coincide when @xmath2 is large .
the main point is that the factorization of the phase space into transverse and longitudinal spaces used in dglap is a good approximation for large @xmath2 only . the implemented coefficient functions @xmath147 and anomalous dimensions @xmath150 of eq .
( [ newch ] ) can be used either in the standard dglap way by letting @xmath34 be integer , i.e. @xmath151 , with @xmath152 and thereby obtaining the moments of the non - singlet structure functions , or alternatively by using @xmath153 in eqs .
( [ fnsint],[gnsint ] ) where @xmath34 is complex . in the latter case ,
@xmath137 and @xmath138 should first be continued analytically from the integer values @xmath151 .
we have verified that the new formulae in eq .
( [ newch ] ) provide a much better agreement with nlo dglap in the region of large @xmath2 .
we notice that our approach is qualitatively close to the one suggested in ref .
@xcite , however they are not quite identical : indeed they first differ in the treatment of @xmath48 which is running in our picture in every feynman graph involved , whereas in ref .
@xcite the dl contributions are obtained from refs .
@xcite where @xmath48 was kept fixed at an unknown scale . to make it running , in ref .
@xcite the parametrization @xmath154 is suggested in the final formulae , which is of course incorrect at small @xmath2 ( see refs .
@xcite for details ) .
in addition , the single- logarithmic contributions were not accounted in ref .
@xcite at all .
as is well known , the conventional dglap approach was originally suggested for the region of rather large values of @xmath2 , where logarithms of @xmath3 gave the most important contributions . at the same time
, dglap neglects the total resummation of logs of @xmath2 . in our approach
we have been able to account for those contributions to the non - singlet structure functions at small @xmath2 .
the results are presented in eqs .
( [ fnsint],[gnsint ] ) . similarly to dglap , the new anomalous dimensions ( [ hns ] ) control the @xmath3 -evolution while new coefficient functions ( [ cns ] ) evolve the initial quark densities from their empirical values @xmath75 at @xmath50 to @xmath51 .
the extrapolation of eqs .
( [ fnsint],[gnsint ] ) into the asymptotic region @xmath25 leads to the regge asymptotic formulae of eqs .
( [ as],[asympt ] ) .
the extrapolation of the dglap expressions with the standard assumption of eq .
( [ fita ] ) for the initial quark distribution @xmath75 also leads to the regge behavior eq .
( [ asdglap ] )
. however , our approach provides the regge behavior because of the total resummation of the leading logarithms of @xmath2 whereas the dglap regge asymptotics is generated by the phenomenological factor @xmath155 in the fit of eq .
( [ fita ] ) for @xmath75 .
this allows us to conclude that the singular part of the dglap input for the initial quark densities mimics the impact of total resummation and can be dropped when the resummation is done , simplifying the structure of the input densities to expressions regular in @xmath2
. it also turns out that the dglap small-@xmath2 asymptotics , using the fit of eq .
( [ fita ] ) for @xmath75 for the non - singlet , is more singular than our predictions .
this contradicts the various analyses of the experimental data of refs .
@xcite , which agree with our predictions .
our asymptotics also differ from dglap as far as the @xmath3 -dependence is concerned : indeed we predict an asymptotic scaling , where the non - singlet structure functions asymptotically depend on one argument @xmath156 instead of two arguments @xmath3 and @xmath2 , separately . giving up the singular factor from the dglap fit eq.([fita ] ) and using its regular part @xmath77 only
, we have compared numerically our formulae eqs .
( [ fnsint],[gnsint ] ) with the nlo dglap results .
such a comparison confirms that in absence of the singular terms the dglap small @xmath2-dependence and @xmath3-dependence of the non - singlets are slower than ours .
the results have been displayed in figs .
finally , in order to obtain an overall approach which is consistent and accurate in both regions of small and large @xmath2 , we have suggested to combine our results and those of dglap by an appropriate definition of the anomalous dimensions and the coefficient functions .
indeed , the expressions of eqs .
( [ newch ] ) for the improved anomalous dimensions and coefficient functions account for both nlo dglap terms and our total resummation of the most important logs of @xmath2 .
therefore , eqs .
( [ fnsint],[gnsint ] ) can be used both at large and small @xmath2 .
then the new inputs for @xmath75 in this approach could be much simpler functions of @xmath2 than in the standard dglap approach .
the work is supported partly by grant rsgss-1124.2003.2
fig . 1 : the @xmath2 -dependence for : @xmath124 of eq . ( [ r1 ] ) ( curve 1 ) , @xmath126 of eq . ( [ r2 ] ) ( curve 2 ) and @xmath129 of eq . ( [ r3 ] ) ( curve 3 ) .
all curves correspond to @xmath157 gev@xmath85 .
g. altarelli and g. parisi , nucl .
phys.b126 ( 1977 ) 297 ; v.n .
gribov and l.n .
lipatov , sov .
j. nucl . phys . 15
( 1972 ) 438 ; l.n.lipatov , sov .
j. nucl .
( 1972 ) 95 ; yu.l .
dokshitzer , sov .
jetp 46 ( 1977 ) 641 .
e.g. floratos , d.a .
ross and c.t .
b129(1977)66 ; a. gonzales - arroyo , c. lopez and f.j .
b153(1979)161 ; a. gonzales - arroyo , c. lopez .
b166(1980)429 ; g. curci , w. furmanski and r. petronzio .
b175(1980)27 ; w. furmanski and r. petronzio .
lett b97(1980)437 ; m. gluck , e. reya and a. vogt .
z.phys.c48(1990)471 ; r. mertig and w.l .
van neerven .
c70(1996)637 ; m. gluck , e. reya , m. stratmann and w.vogelsang .
; j. soffer and o.v . teryaev .
rev.56 ( 1997)1549 ; a.l .
kataev , g. parente , a.v .
phys.part.nucl 34(2003)20 ; nucl.phys.a666(2000)184 ; a.v .
kotikov , a.v .
lipatov , g. parente , n.p
. zotov .
eur.phys.j.c26(2002)51 ; v.g .
krivohijine , a.v .
kotikov , hep - ph/0108224 ; a.v .
kotikov , d.v .
peshekhonov hep - ph/0110229 . | the explicit expressions for the non - singlet dis structure functions @xmath0 and @xmath1 , obtained at small @xmath2 by resumming the leading logarithmic contributions to all orders , are discussed and compared in detail with the dglap evolution for different values of @xmath2 and @xmath3 . the role played by the dglap inputs for the initial parton densities on the small-@xmath2 behavior of the non - singlet structure functions
is discussed .
it is shown that the singular factors included into the fits ensure the regge asymptotics of the non - singlet structure functions and mimic the impact of the total resummation of the leading logarithms found explicitly in our approach .
explicit expressions are presented which implement the nlo dglap contributions with our small -@xmath2 results . |
erich l. lehmann was born in strasbourg on november 20th , 1917 .
he passed away in berkeley , california on the morning of september 12th , 2009 .
his family left germany in 1933 , as the nazis came to power , to settle in switzerland .
he spent five years in zrich and two years in trinity college in cambridge studying mathematics . under the united states french immigration quota
strasbourg was , by then , part of france as a consequence of the versailles treaty he arrived in new york at the end of 1940 .
edmund landau , the famous number theorist , was an acquaintance of the lehmann family and had suggested trinity college as the place erich should go to study mathematics .
landau died in 1938 from a heart attack , but his wife wrote a letter of introduction for erich to take to landau s gttingen colleague richard courant who was now in new york developing what became the courant institute .
courant , having offered the option to `` live in new york or in the united states , '' and erich having opted for the latter , recommended the university of california as an up - and - coming good place .
erich arrived in berkeley , california in january 1 , 1941 .
erich s first order of business was to speak with griffith c. evans , chair of the mathematics department , who immediately accepted him as a probationary graduate student .
the probationary status resulted from erich not having a degree .
evans , who had been recruited from the mathematics department at rice institute now rice university had a broad vision for mathematics and had the intention of hiring ronald a. fisher , whom he knew .
however , a visit by fisher to berkeley did not go well . the news of jerzy neyman s successful visit to the united states , culminating with a set of lectures at the u.s .
department of agriculture , reached evans who in 1937 offered neyman a job in the mathematics department at the university of california without having met him . with the advent of the second world war ,
evans advised erich that it might be a good idea to move from mathematics to some other area perhaps physics or statistics that could be more useful to the war efforts .
erich , not being fond of physics , opted for statistics .
his initial experiences , however , led him to second - guess his decision . in lehmann ( 2008b ) , erich writes that `` statistics did not possess the beauty that i had found in the integers and later in other parts of mathematics . instead , ad hoc methods were used to solve problems that were messy and that were based on questionable assumptions that seemed quite arbitrary . ''
( hereafter , a reference followed by `` b '' indicates book reference in section 9 ; a bracketed reference [ x ] refers to that numbered reference in section 8 ; other references appear at the end of this work . ) after some soul - searching , he decided to go back to mathematics and approached the great logician alfred tarski .
tarski accepted him as a student , but before erich had an opportunity to let evans and neyman know about his decision , neyman offered him a job as a lecturer with some implicit potential for the position to become permanent . feeling that this represented a great opportunity to become part of a community , something that erich very much desired at that point in time , he decided to take the offer and abandoned his plans for returning to mathematics . in 1942
erich received an m.a .
degree in mathematics , and was a teaching assistant in the statistical laboratory from 1942 to 1944 and from 1945 to 1946 .
these early years as a graduate student and a teaching member of the department while sharing office space with charles stein , joseph hodges and evelyn fix helped to forge lifetime friendships and productive collaborations . after he spent the year from august 1944 to august 1945 stationed in guam as an operations analyst in the united states air force , erich returned to berkeley and started working on a thesis problem proposed by pao - lu hsu in consultation with neyman .
the problem was in probability theory some aspect of the moment problem and after obtaining some results and getting ready to write them up , erich discovered that his results were already in markov s work .
the situation became complicated as neyman was invited to supervise the greek elections . before leaving , neyman asked hsu if he could provide another thesis topic for erich .
hsu obliged but was not able to supervise erich s thesis , as he followed hotelling from columbia to north carolina and then decided to go back to china .
neyman turned to george plya at stanford for help .
weekly meetings with plya , commuting between berkeley and stanford , finally yielded a thesis . meanwhile , neyman was back from greece after being relieved of his duties for insubordination .
neyman had felt that the elections were rigged and decided to check by himself .
when asked to stop , he refused .
this turn of events allowed neyman to be back in berkeley for erich s examination .
thus , in june of 1946 , erich obtained his ph.d .
degree with a thesis titled `` optimum tests of a certain class of hypotheses specifying the value of a correlation coefficient . ''
erich was not the first of neyman s berkeley ph.d .
students , but he was the first one to be hired by the mathematics department .
he held the title of assistant professor of mathematics from 1947 to 1950 , and spent the first half of 19501951 as a visiting associate professor at columbia , and as a lecturer at princeton during the second half of that year . partly to allow more time for the tumultuous situation created in berkeley by the anti - communist loyalty oath to settle down , and partly to make a decision on an offer from stanford , erich spent the year of 19511952 as a visiting associate professor at stanford .
erich decided to go back to berkeley , but not before he was able to persuade neyman not to require him to do consulting work for the statistical laboratory .
( stanford s offer explicitly mentioned that erich was not expected to do any applied work . ) on his return to berkeley in 1952 , erich was promoted to associate professor of mathematics , and then in 1954 was promoted to professor of mathematics . in 1955 , after evans stepped down as chair of mathematics , thus providing neyman with his opportunity for a new department of statistics , erich s title changed to professor of statistics . in 1988 , erich became professor emeritus and then from 1995 to 1997 he was distinguished research scientist at the educational testing service ( ets ) . in spite of his retirement in 1988 , erich continued to be professionally active and a regular participant in the social life of the department . despite offers from stanford in 1951 and from the eidgenssische technische hochschule ( eth ) in 1959 , and except for short stints at columbia , princeton , stanford and ets , erich lived in berkeley from his arrival on january 1st , 1941 until his death on september 12th , 2009
erich lehmann s towering contributions to statistics have received many well - deserved accolades .
erich was an elected fellow of the institute of mathematical statistics ( ims ) and of the american statistical association ( asa ) , and he was an elected member of the international statistical institute .
remarkably , he was the recipient of three guggenheim fellowships ( 1955 , 1966 and 1980 ) and two miller institute for basic research professorships ( 1962 and 1972 ) .
the ims honored him as the wald lecturer in 1964the title of his lectures being `` topics in nonparametric statistics . ''
this was followed in 1988 by the committee of presidents of statistical societies ( copss ) r. a. fisher memorial lecture entitled `` model specification : fisher s views and some later strategies . '' in 1975 erich was elected fellow of the american academy of arts and sciences and in 1978 he was elected member to the national academy of sciences .
election as an honorary fellow of the royal statistical society followed in 1986 and the asa recognized him with the wilks memorial award in 1996 .
his life - long work was recognized with two doctorates _ honoris causa _ , the first from the university of leiden in 1985 , and the second from the university of chicago in 1991 .
the honor from leiden carries with it the distinction of being the first dr .
_ h. c. _ granted by the university of leiden to a mathematician in a century , the previous one having been awarded to stieltjes in 1884 . in 1997 , to celebrate erich s 80th birthday , the berkeley statistics department instituted the lehmann fund to provide support for students . in 2000
erich became the first goffried noether award recipient and lecturer for his influential work in nonparametrics .
his noether lecture , entitled `` parametrics versus nonparametrics : two alternative methodologies , '' formed the basis for an invited paper with discussion in the _ journal of nonparametrics _ ( jnps ) in 2009 [ 121 ] .
posthumously , erich received the best jnps paper award for 2009 .
his students and colleagues honored him with a set of reminiscences in 1972 ( j. rojo , ed . ) , a _
festschrift for erich l. lehmann _ organized by bickel , doksum and hodges in 1982 [ see also bickel , doksum and hodges ( 1983 ) ] , and a series of _ lehmann symposia _ , organized by rojo and perez - abreu in 2002 , and rojo in 2004 , 2007 and 2011 .
perhaps surprisingly , although he was honored with the fisher lecture , he never received the honor of being the neyman lecturer .
it may be surmised that erich s lack of affinity for applied work impeded his being so honored .
erich served the profession well .
although initially reluctant to serve as chair of the statistics department at berkeley , he did so from 19731976 .
and he did it very well .
brillinger ( 2010 ) writes : _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ he had always refused previously for a variety of reasons .
he did it so well that i sometimes thought that he must have thought through how a chair should behave and put his conclusions into practice .
for example , to the delight of visitors and others he was in the coffee room each day at 10 a.m. he focused on the whole department staff , students , colleagues and visitors .
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ during 19601961 , erich was ims president and was a leader in the internationalization of the ims [ see , e.g. , lehmann ( 2008b ) and van zwet ( 2011 ) ] .
he was a member of the executive committee of the miller institute ( 19661970 ) , and a member of the committee of visitors to the harvard department of statistics ( 19741980 ) and princeton ( 19751980 ) .
he served as editor of the _ annals of mathematical statistics _ from 19531955 and as associate editor from 19551968 .
he was invited to stay on for a second term as editor but , after accepting , had to decline .
for details see lehmann ( 2008b ) and van zwet ( 2011 ) .
in his youth , erich leo lehmann had a desire to become a writer . in lehmann ( 2008b )
, he wrote , `` _ _ my passion was german literature , my dream to become a writer , perhaps another thomas mann or gottfried keller__. '' surely it was this passion that drove erich to write his successful and influential books .
the list includes : 1 .
* testing statistical hypotheses*. three editions ( 1959 , 1986 , 2005 ) .
the 2005 edition is joint with joseph p. romano .
the 1959 edition was translated into russian ( 1964 ) , polish ( 1968 ) and japanese .
* basic concepts of probability and statistics * , with joseph l. hodges . two editions ( 1964 , 1970 ) . reprinted in 2005 as part of the siam series classics in applied mathematics .
the book was translated into hebrew ( 1972 ) , farsi ( 1994 ) , italian ( 1971 ) and danish ( 1969 ) .
* elements of finite probability * , with joseph l. hodges . two editions ( 1965 , 1970 ) .
nonparametrics : statistical methods based on ranks * , with the assistance of h. j. m. dabrera .
hardcover edition ( 1975 ) by holden - day .
paperback edition ( 1998 ) by prentice - hall , inc . ,
simon & schuster , and then by springer science in 2006 .
the book was translated into japanese ( 1998 ) .
* theory of point estimation*. two editions ( 1983 , 1998with george casella ) .
the 1983 edition was translated into russian ( 1991 ) , and the 1998 edition into chinese ( 2004 ) . 6 .
* elements of large - sample theory * , 1999 . 7 .
* reminiscences of a statistician : the company i kept * , 2008 . additionally , erich collaborated with judith m. tanur on the book _ statistics _ : _ a guide to the unknown_. this book went through several editions and translations [ chinese ( 1980 ) and spanish ( 1992 ) ] .
spin - offs from this book were two other books with similar titles : _ statistics _ : _ a guide to the study of the biological and health sciences _ and _ statistics _ : _ a guide to political and social issues _ , both published in 1977 , and on which erich collaborated .
erich served as co - editor or special editor .
the complete list of books and their translations is given in section 9 .
@cc@ + + [ cols="^,^ " , ] + * erich l. lehmann in 1919 , 1992 and 2004 . * the book _ fisher _ , _ neyman _ , _ and
the creation of classical statistics _ has now been published posthumously by springer , lehmann ( 2011b ) .
erich was finishing the manuscript at the time of his death .
juliet shaffer worked diligently after erich s passing to bring the book to publication form .
fritz scholz continues work on a revision , started before erich s death , of the _ nonparametrics _ : _ statistical methods based on ranks _ book .
the revision incorporates the use of r and the book is expected to be completed in two years .
99 lehmann , e. l. ( 1959 ) .
_ testing statistical hypotheses_. wiley , new york .
hodges , j. l. jr . and lehmann , e. l. ( 1964 ) .
_ basic concepts of probability and statistics_. holden - day , san francisco , ca .
lehmann , e. l. ( 1964 ) .
russian translation of _ testing statistical hypotheses _ , moscow .
hodges , j. l. jr . and lehmann , e. l. ( 1965 ) .
_ elements of finite probability_. holden - day , san francisco , ca .
lehmann , e. l. ( 1968 ) .
_ tesowanie hipotez statvstycznych_. polish translation of _ testing statistical hypotheses_. panstwowe wydawnicto naukowe , warsaw .
hodges , j. l. jr . and
lehmann , e. l. ( 1969 ) .
_ grundbegreger i sandsynhghedsregning og statistik_. danish translation of _ basic concepts of probability and statistics_. nyt nordisk forlag , copenhagen .
hodges , j. l. jr . and lehmann , e. l. ( 1970 ) .
_ basic concepts of probability and statistics _ , 2nd ed
holden - day , san francisco , ca .
hodges , j. l. jr . and lehmann , e. l. ( 1970 ) .
_ elements of finite probability _
holden - day , san francisco , ca .
hodges , j. l. jr . and lehmann , e. l. ( 1971 ) .
italian translation of _ basic concepts of probability and statistics _ , two volumes .
societa editrice il mulino , bologna , italy .
hodges , j. l. jr . and lehmann , e. l. ( 1972 ) .
hebrew translation of _ basic concepts of probability and statistics _
, 2nd ed .
lehmann , e. l. japanese translation of _ testing statistical hypotheses _ 418 pp .
iwanami shoten , tokyo .
lehmann , e. l. ( 1975 ) .
_ nonparametrics : statistical methods based on ranks_. holden - day , san francisco , ca .
tanur , j. m. , mosteller , f. , kruskal , w. h. , lehmann , e. l. , link , r. f. , pieters , r. s. and rising , g. r. , co - eds .
( 1977 ) . _
statistics _ : _ a guide to the study of the biological and health sciences _ 140 pp .
holden - day , san francisco .
tanur , j. m. , ed . , lehmann , e. l. , special ed .
, mosteller , f. , kruskal , w. h. , link , r. f. , pieters , r. s. and rising , g. r. , co - eds .
_ statistics _ : _ a guide to political and social issues _ 141 pp .
holden - day , san francisco .
lehmann , e. l. ( 1978 ) .
japanese translation of _ nonparametrics _ :
_ statistical methods based on ranks_. tokyo , 1978 .
tanur , j. m. , ed . , lehmann , e. l. , special ed . , mosteller , f. , kruskal , w. h. , link , r. f. , pieters , r. s. and rising , g. r. , co - eds .
( 1978 ) . _
statistics _ : _ a guide to the unknown _ , 2nd ed .
holden - day , san francisco .
lehmann , e. l. ( 1983 ) .
_ theory of point estimation_. _ wiley series in probability and mathematical statistics _ : _ probability and mathematical statistics_. wiley , new york .
tanur , j. m. , ed . , lehmann , e. l. , special ed .
, mosteller , f. , kruskal , w. h. , link , r. f. , pieters , r. s. and rising , g. r. , co - eds .
_ statistics _ : _ a guide to the unknown _ , 2nd ed . reprint .
wadsworth & brooks / cole , monterey , ca .
lehmann , e. l. ( 1986 ) .
_ testing statistical hypotheses _ , 2nd ed .
wiley , new york .
tanur , j. m. , mosteller , f. , krusal , w. h. , lehmann , e. l. , link , r. f. , pieters , r. s. and rising , g. r. , co - eds .
_ statistics _ : _ a guide to the unknown _ , 3rd ed .
wadsworth .
tanur , j. m. , ed . , lehmann , e. l. , special ed .
, mosteller , f. , kruskal , w. h. , link , r. f. , pieters , r. s. and rising , g. r. , co - eds .
chinese translation of _ statistics _ : _ a guide to the unknown _ , 2nd ed .
lehmann , e. l. ( 1991 ) .
russian translation of _ theory of point estimation_. tanur , j. m. , ed . ,
lehmann , e. l. , special ed .
, mosteller , f. , kruskal , w. h. , link , r. f. , pieters , r. s. and rising , g. r. , co - eds .
spanish translation of _ statistics _ : _ a guide to the unknown _ , 2nd ed .
alianza editorial , s. a. , spain , hodges , j. l. jr . and lehmann , e. l. ( 1994 ) .
farsi translation of _ basic concepts of probability and statistics_. lehmann , e. l. and casella , g. ( 1998 ) .
_ theory of point estimation _
springer , new york .
lehmann , e. l. ( 1999 ) .
_ elements of large - sample theory_. springer , new york .
lehmann , e. l. and casella , g. ( 2004 ) .
chinese translation of _ theory of point estimation _
china statistics press .
lehmann , e. l. and romano , j. p. ( 2005 ) .
_ testing statistical hypotheses _ , 3rd ed
springer , new york .
lehmann , e. l. ( 2008 ) .
_ reminiscences of a statistician _ : _ the company i kept_. springer , new york
. lehmann , e. l. ( 2011 ) .
_ fisher , neyman , and the creation of classical statistics_. springer , new york .
erich s contributions are multifaceted and too many to do justice to in the allotted space .
a more extensive and careful assessment of his work is provided in rojo ( 2011 ) . here , only a small part of his work will be briefly reviewed .
some of his ground - breaking work in nonparametric statistics is discussed in this issue by van zwet ( 2011 ) . while still a graduate student at berkeley , erich submitted a paper that was published in 1947 [ 2 ] , in which the issue of what to do when a uniformly most powerful ( ump ) test does not exist is discussed .
erich proposed that , due to the many tests available to choose from , one must reduce attention to a class of tests @xmath0 with the property that for any test @xmath1 not in @xmath0 , there is a test @xmath2 in @xmath0 with a power function at least as good as that of @xmath1 . and if @xmath3 and @xmath4 are two tests in @xmath0 , then neither one dominates the other .
in addition , the paper characterizes the class @xmath0 for a special case .
erich recognized that the class @xmath0 may still be too large to offer much relief in finding a good solution and , therefore , other information or principles may be needed to further narrow down the class @xmath0 .
thus , the concept of minimal complete classes , that plays a fundamental role in the theory of statistical decisions of wald ( 1950 ) , was born in this paper . in his book
_ statistical decision functions _
( 1950 ) , wald credits lehmann : _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ the concept of complete class of decision functions was introduced by lehmann , and the first result regarding such classes is due to him [ 30]@xmath5 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ interestingly , neyman was not impressed by this work . in degroot ( 1986 ) erich states : _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ i wrote it up it was just a few pages and said to neyman that i would like to publish it .
he essentially said , `` it s junk
. do not bother . ''
but i sent it in to wilks anyway .
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ some of erich s early work was motivated by the work of hsu ( 1941 ) that dealt with optimal properties of the likelihood ratio test in the context of analysis of variance . in lehmann ( 1959 )
[ 34 ] , erich shows that these optimal properties are consequences of the fact that the test is uniformly most powerful invariant .
in addition , the paper unified optimality results of kiefer ( 1958 ) for symmetrical nonrandomized designs , and optimality results of wald ( 1942 ) for the analysis of variance test for the general univariate linear hypothesis .
hsu ( 1941 ) also proposed a method for finding all similar tests .
lehmann ( 1947 ) [ 3 ] extended hsu s results to the composite null hypothesis problem , and ideas in hsu ( 1941 ) motivated the concept of completeness in lehmann and scheff ( 1950 ) [ 12 ] .
lehmann and scheff ( 1950 ) [ 12 ] and lehmann and scheff ( 1955 ) [ 26 ] provided a comprehensive study of the concepts of similar regions and sufficient statistics .
together with lehmann and stein ( 1950 ) [ 11 ] , where uniformly minimum variance unbiased estimators are discussed in the sequential sampling context , these papers provide the final word on certain problems in hypotheses testing and estimation .
hodges and lehmann ( 1950 , 1951 , 1952 ) [ 10 , 13 , 16 ] , provided minimax estimators for several examples and the admissibility of minimax estimators and connections with bayes estimators were discussed . in hodges and lehmann ( 1950 ) [ 10 ] , a minimax estimator for the probability of success @xmath6 in a binomial experiment is obtained by considering the bayes estimator with respect to a beta conjugate prior that yields a bayes estimator with constant risk .
the minimax estimator thus found is admissible due to the uniqueness of the bayes estimator .
the results are extended to the case of two independent binomial distributions and a minimax estimator is obtained for the difference of the probability of successes when the sample sizes are equal .
the question of whether a minimax estimator exists for the difference of the success probabilities for unequal sample sizes remains an open problem .
the papers also consider the nonparametric case , and methods for deriving nonparametric minimax estimators are provided under certain conditions .
the concept of complete classes having been formalized by wald ( 1950 ) , the paper also shows that , for convex loss functions , the class of nonrandomized estimators is essentially complete .
hodges and lehmann ( 1951 ) [ 13 ] used a different approach to obtain minimax and admissible estimators when the loss function is a weighted squared error loss .
the method requires the solution of a differential inequality involving the lower bound for the mean squared error .
various sequential problems were discussed and minimax estimators were derived .
hodges and lehmann ( 1952 ) [ 16 ] proposed finding estimators whose maximum risk does not exceed the minimax risk by more than a given amount @xmath7 .
under this restriction it was proposed to find the _ restricted bayes solution _ with respect to some prior distribution @xmath8 .
that is , find @xmath9 that minimizes @xmath10 subject to @xmath11 .
conditions were discussed for the existence of restricted bayes estimators and several examples were provided that illustrate the method . it was argued that wald s theory can be extended to obtain results for these restricted bayes procedures .
wald ( 1950 ) obtained the existence of least favorable distributions under the assumption of a compact parameter space .
lehmann ( 1952 ) [ 17 ] addressed this issue and , in the case of hypothesis testing and , more generally , in the case where only a finite number of decisions are available , lehmann weakened the conditions for the existence of least favorable distributions .
lehmann and stein ( 1953 ) [ 20 ] proved the admissibility of the most powerful invariant test when testing certain hypotheses in the location parameter family context .
erich s work on hypothesis testing is well known . here
some aspects of that work are briefly reviewed .
lehmann ( 1947 ) [ 3 ] and lehmann and stein ( 1948 ) [ 5 ] studied the problem of testing a composite ( null ) hypothesis .
the 1947 paper extends the work of scheff ( 1942 ) .
suppose that @xmath12 is a @xmath13-dimensional parameter space .
let @xmath14 be the subset of @xmath12 given by @xmath15 , for one @xmath16 .
then the null hypothesis @xmath17 is an example of a composite ( null ) hypothesis with one constraint , and the parameters @xmath18 , are nuisance parameters .
neyman ( 1935 ) provided type b regions for the case of a single nuisance parameter .
these results were extended by scheff to the case of several nuisance parameters ( under @xmath19 ) , and scheff provided sufficient conditions for these type b regions to also be type b@xmath20 ( uniformly most powerful unbiased ) regions .
lehmann ( 1947 ) [ 3 ] utilized neyman and pearson s ( 1933 ) and hsu s ( 1945 ) methods to determine the totality of similar regions and extended scheff s results to obtain uniformly most powerful tests against one - sided alternatives .
hsu s method was also employed to obtain ump regions in cases , for example , location and scale exponential and uniform distributions , where neyman and pearson s method does not apply .
the above approach is not as fruitful in the case of more than one constraint , but results of hsu ( 1945 ) are useful in this regard . in lehmann and stein ( 1948 )
[ 5 ] the problem of testing a composite hypothesis against a single alternative is addressed by relaxing the condition of similarity to one requiring only that @xmath21 or all @xmath22 , where @xmath23 denotes the critical region of the test .
adapting the neyman
pearson lemma to hold in this case , sufficient conditions for the existence of most powerful tests were derived .
the results for student s problem , with composite null hypothesis given by the normal family with mean @xmath24 and unknown variance , and the simple alternative hypothesis given by the normal distribution with known parameters were somewhat surprising ; see lehmann ( 2008b ) , page 48 . lehmann ( 1950 , 1959 , 2006 )
[ 9 , 34 , 118 ] deal with the likelihood ratio principle for testing .
although this principle is `` intuitive '' and provides `` reasonable '' tests , it is well known that it may fail .
the papers examine different aspects of the problem focusing on the optimality of the likelihood ratio test in some cases , and in its total failure in other cases .
lehmann ( 1959 ) [ 34 ] considered a class of invariant tests endowed with an order that satisfies certain properties .
it was then shown that , in this case , the likelihood ratio test s optimality properties follow directly from the fact that the test is uniformly most powerful invariant .
see also section 4.1 . in lehmann ( 2006 )
[ 118 ] and lehmann ( 1950 ) [ 9 ] , properties of tests produced by other approaches are examined and compared to the likelihood ratio tests .
for example , when the testing problem remains invariant with respect to a transitive group of transformations , the _ likelihood averaged or integrated with respect to an invariant measure approach _ in lehmann ( 2006 ) [ 118 ] produces tests that turn out to be uniformly at least as powerful as the corresponding likelihood ratio test , with the former being strictly better except when the two coincide ; and in the absence of invariance , the proposed approach continues to improve on the likelihood ratio test for many cases .
lehmann ( 1950 ) [ 9 ] was discussed in section 4.1 .
lehmann s work on orderings of probability distributions was motivated in part from the need to study properties of power functions .
thus , lehmann ( 1955 ) [ 27 ] discussed the stochastic and monotone likelihood ratio orderings .
the latter plays a fundamental role in the theory of uniformly most powerful tests and both can be characterized in terms of the function @xmath25 ; see , for example , lehmann and rojo ( 1992 ) [ 98 ] .
it is this function @xmath26 that also plays a fundamental role in the lehmann alternatives and , hence , is also connected with the cox proportional hazards model and has now spilled over to the literature on receiving operating characteristic ( roc ) curves .
a different collection of partial orderings between distributions @xmath27 and @xmath28 can be defined in terms of the function @xmath29 .
bickel and lehmann ( 1979 ) [ 64 ] considered the dispersive ordering defined by requiring that @xmath30 for all @xmath31 , and considered several of its characterizations .
this concept is equivalent , under some conditions , to a tail - ordering introduced by doksum ( 1969 ) .
this function , @xmath32 , is also useful in comparing location experiments ( lehmann ( 1988 ) [ 85 ] ) .
lehmann ( 1966 ) [ 47 ] introduced concepts of dependence for random variables @xmath33 .
this work has attracted a lot of attention in the literature from applied probabilists and statisticians alike .
erich believed in the frequentist interpretation of probability and in the neyman
pearson wald school of optimality , but recognized that both perspectives have their limitations .
see , for example , page 188 of lehmann ( 2008b ) .
bickel and lehmann ( 2001 ) [ 110 , 111 ] discussed some of the philosophical shortcomings of a frequentist interpretation of probability .
erich felt that optimality considerations achieve solutions that may lack robustness and other desirable properties .
his work on foundational issues focused on the following : ( i ) model selection ; ( ii ) frequentist statistical inference ; ( iii ) bayesian statistical inference ; and ( iv ) exploratory data analysis . restricting attention to ( ii ) , ( iii ) and ( iv ) , erich viewed the trichotomy as being ordered by the level of model assumptions made .
thus , ( iv ) is free of any model assumptions and allows the data to speak for itself , while the frequentist approach relies on a probability model to evaluate the procedures under consideration .
the bayesian approach , in addition , brings in the prior distribution .
erich felt that none of these approaches is perfect .
motivated by this state of affairs , lehmann ( 1985 , 1995 ) [ 82 , 104 ] developed ideas that bridge the divide created by the heated philosophical debates .
lehmann ( 1985 ) [ 82 ] discussed how the neyman pearson
wald approach contributes to the exploration of underlying data structure and its relation with bayesian inference .
lehmann ( 1995 ) [ 104 ] continued with this line of thought : _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ in practice , the three approaches can often fruitfully interact , with each benefiting from considerations of the other points of view .
it seems clear that model - free data analysis , frequentist and bayesian model - based inference and decision making each has its place .
the question appears not to be as it is often phrased which is the correct approach but in what circumstances
each is most appropriate .
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ erich s balanced view of foundational issues is appealing .
his work reflects the belief that no single paradigm is totally satisfactory . rather than exacerbating their differences through heated debates
, he proposed that a fruitful approach is possible by consolidating the good ideas from ( ii ) , ( iii ) and ( iv)with ( iii ) serving as a bridge that connects all three .
although his original position was solidly in the frequentist camp , he shifted , somewhat influenced by classical bayesian ideas .
however , he felt that a connection with the radical bayesian position was more challenging .
he states in lehmann ( 1995 ) [ 104 ] that `` _ _ bridge building to the radicalbayesian position is more difficult__. '' a definition of the radical bayesian position is not provided , but it can be surmised that this refers to a paradigm that insists on the elicitation of a prior distribution at all costs . in lehmann ( 2008b ) , he writes : _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ however , it seems to me that the strength of these beliefs tends to be rather fuzzy , and not sufficiently well defined and stable to assign a definite numerical value to it . if , with considerable effort , such a value is elicited , it is about as trustworthy as a confession extracted through torture .
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
i first attended u.c .
berkeley during the fall of 1978 .
my first course was statistics 210 a
the first quarter of theoretical statistics .
the recollections of my days as a student during that first quarter , followed by two more quarters of theoretical statistics statistics 210 b and c all taught by erich , are very vivid . during that first academic year
, i was very impressed with erich s lecturing style .
he would present the material without unnecessarily dwelling too long on technical details , and in such a way that connections with previous material seemed virtually seamless .
it was quite enjoyable to follow `` the story '' behind the theory .
his lectures were so perfectly organized even when only using a few notes on his characteristic folded - in - the - coats - pocket - yellow sheets ! regarding teaching , erich wrote in lehmann ( 2008b ) : _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ while i eschewed very large courses , i loved the teaching that occurred at the other end of the spectrum .
working on a one - on - one basis with ph.d .
students was , for me , the most enjoyable and rewarding aspect of teaching .
at the same time , it was an extension of my research , since these students would help me explore areas in which i was working at the time .
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ this love for one - on - one teaching produced a total of 43 ph.d .
students . curiously , two of erich s ph.d .
students obtained their degrees from columbia rather than from berkeley . that these students graduated from columbia , rather than from berkeley , resulted from a confluence of circumstances .
although erich had received an invitation from wald to visit columbia during the 19491950 academic year , erich had to postpone his visit to columbia for the following year since neyman took a sabbatical during the 19491950 academic year .
after wald s tragic and untimely death , two of wald s students approached erich with a request to become his students .
these students are marked with an asterisk in the following table that presents the names and dissertation titles , by year of degree , for all 43 of erich s ph.d .
students . 1 .
* colin ross blyth * + _ i. contribution to the statistical theory of the geiger muller counter _ ; + _ ii .
on minimax statistical decision procedures and their admissibility .
* fred charles andrews * + _ asymptotic behavior of some rank tests for analysis of variance . _ + * jack laderman * * + _ on statistical decision functions for selecting one of k populations .
* hendrik salomom konijn * + _ on the power of some tests for independence .
* allan birnbaum * * + _ characterizations of complete classes of tests of some multiparametric hypotheses , with applications to likelihood ratio tests . _ + * balkrishna v. sukhatme * + _ testing the hypothesis that two populations differ only in location . _ 5 . *
v. j. chacko * + _ testing homogeneity against ordered alternatives .
_ 6 . * piotr witold mikulski * + _ some problems in the asymptotic theory of testing statistical hypotheses .
_ 7 . * madan lal puri * + _ asymptotic efficiency of a class of c - sample tests . _ + * krishen lal mehra * + _ rank tests for incomplete block designs .
paired - comparison case . _ + * subha bhuchongkul sutchritpongsa * + _ class of non - parametric tests for independence in bivariate populations . _ + * shishirkumar shreedhar jogdeo * + _ nonparametric tests for regression models . _ 8 . *
peter j. bickel * + _ asymptotically nonparametric statistical inference in the multivariate case . _ + * arnljot hyland * + _ some problems in robust point estimation .
* milan kumar gupta * + _ an asymptotically nonparametric test of symmetry . _ + * madabhushi raghavachari * + _ the two - sample scale problem when locations are unknown . _ + * ponnapalli venkata ramachandramurty * + _ on some nonparametric estimates and tests in the behrens fisher situation . _ + * vida greenberg * + _ robust inference in some experimental designs .
kjell andreas doksum * + _ asymptotically minimax distribution - free procedures . _ + * william harvey lawton * + _ concentration of random quotients . _ 11 . * shulamith gross * + _ nonparametric tests when nuisance parameters are present . _ + * bruce hoadley * + _ the theory of large deviations with statistical applications . _ + * gouri kanta bhattacharyya * + _ multivariate two - sample normal scores test for shift . _ + * james nwoye adichie * + _ nonparametric inference in linear regression .
_ + * dattaprabhakar v. gokhale * + _ some problems in independence and dependence . _ 12 . * frank rudolph hampel * + _ contributions to the theory of robust estimation .
* wilhelmine von turk stefansky * + _ on the rejection of outliers by maximum normed residual . _ + * neil h. timm*co - advisors erich leo lehman and leonard marascuilo + _ estimating variance covariance and correlation matrices from incomplete data . _ + * louis jaeckel * + _ robust estimates of location .
friedrich wilhelm scholz * + _ comparison of optimal location estimators .
_ + * dan anbar * + _ on optimal estimation methods using stochastic approximation procedures .
* michael denis stuart * + _ components of 2 for testing normality against certain restricted alternatives . _ + * claude l. guillier
* + _ asymptotic relative efficiencies of rank tests for trend alternatives . _ + * sherali mavjibhai makani * + _ admissibility of linear functions for estimating sums and differences of exponential parameters .
howard j. m. dabrera * + _ rank tests for ordered alternatives .
* hyun - ju yoo jin * + _ robust measures of shift . _ 18 . * amy poon davis * + _ robust measures of association .
* jan f. bjornstad * + _ on optimal subset selection procedures .
* william paul carmichael * + _ the rate of weak convergence of a vector of u - statistics generated by a single sample . _ + * david draper * + _ rank - based robust analysis of linear models . _ 21 . * wei - yin loh * + _ tail - orderings on symmetric distributions with statistical applications .
* marc j. sobel * + _ admissibility in exponential families .
javier rojo * + _ on lehmann s general concept of unbiasedness and the existence of l - unbiased estimators . _
999 lehmann , e. l. ( 1946 ) .
une proprit optimale de certains ensembles critiques du type a1 .
_ c. r. acad .
paris _ * 223 * 567569 .
lehmann , e. l. ( 1947 ) . on families of admissible tests .
_ ann . math .
statist . _
* 18 * 97104 .
lehmann , e. l. ( 1947 ) . on optimum tests of composite hypotheses with one constraint .
statist . _
* 18 * 473494 .
lehmann , e. l. and scheff , h. ( 1947 ) . on the problem of similar regions .
usa _ * 33 * 382386 .
lehmann , e. l. and stein , c. ( 1948 ) .
most powerful tests of composite hypotheses
. i. normal distributions . _
_ * 19 * 495516 .
lehmann , e. l. ( 1949 ) .
some comments on large sample tests . in
_ proceedings of the berkeley symposium on mathematical statistics and probability _ 451457 .
california press , berkeley .
lehmann , e. l. ( 1949 ) .
recent publications : elementary statistical analysis . _
amer . math . monthly _ * 56 * 429430 .
lehmann , e. l. and stein , c. ( 1949 ) .
on the theory of some nonparametric hypotheses .
statist . _
* 20 * 2845 .
lehmann , e. l. ( 1950 ) .
some principles of the theory of testing hypotheses .
_ ann . math .
statist . _
* 21 * 126 .
hodges , j. l. jr . and lehmann , e. l. ( 1950 ) . some problems in minimax point estimation .
statist . _
* 21 * 182197 .
lehmann , e. l. and stein , c. ( 1950 ) .
completeness in the sequential case .
statist . _
* 21 * 376385 .
lehmann , e. l. and scheff , h. ( 1950 ) .
completeness , similar regions , and unbiased estimation .
i. _ sankhy _ * 10 * 305340 .
hodges , j. l. jr . and lehmann , e. l. ( 1951 ) .
some applications of the cramr
rao inequality . in _ proceedings of the second berkeley symposium on mathematical statistics and probability _ 1322 .
california press , berkeley .
lehmann , e. l. ( 1951 ) .
consistency and unbiasedness of certain nonparametric tests .
_ ann . math .
statist . _
* 22 * 165179 .
lehmann , e. l. ( 1951 ) .
a general concept of unbiasedness .
statist . _
* 22 * 587592 .
hodges , j. l. jr . and lehmann , e. l. ( 1952 ) .
the use of previous experience in reaching statistical decisions . _ ann .
statist . _
* 23 * 396407 .
lehmann , e. l. ( 1952 ) . on the existence of least favorable distributions .
_ * 23 * 408416 .
lehmann , e. l. ( 1952 ) .
testing multiparameter hypotheses .
statist . _
* 23 * 541552 .
lehmann , e. l. ( 1953 ) .
the power of rank tests .
statist . _
* 24 * 2343 .
lehmann , e. l. and stein , c. m. ( 1953 ) .
the admissibility of certain invariant statistical tests involving a translation parameter .
_ * 24 * 473479 .
chernoff , h. and lehmann , e. l. ( 1954 ) .
the use of maximum likelihood estimates in @xmath34 tests for goodness of fit .
statist . _
* 25 * 579586 .
hodges , j. l. jr . and lehmann , e. l. ( 1954 ) . matching in paired comparisons .
statist . _
* 25 * 787791 .
hodges , j. l. jr . and lehmann , e. l. ( 1954 ) .
testing the approximate validity of statistical hypotheses .
b _ * 16 * 261268 .
anderson , t. w. , cramer , h. , hodges , j. l. , freeman , jr .
h. a. , lehmann , e. l. , mood , a. m. and stein , c. ( 1955 ) . the life of abraham wald . in _ selected papers in statistics and probability . _ mcgraw - hill , new york .
bahadur , r. r. and lehmann , e. l. ( 1955 ) .
two comments on `` sufficiency and statistical decision functions . '' _ ann .
_ * 26 * 139142 .
lehmann , e. l. and scheff , h. ( 1955 ) .
completeness , similar regions , and unbiased estimation .
_ sankhy _ * 15 * 219236 .
lehmann , e. l. ( 1955 ) .
ordered families of distributions . _
_ * 26 * 399419 .
hodges , j. l. jr . and lehmann , e. l. ( 1956 ) .
two approximations to the robbins monro process . in _ proceedings of the third berkeley symposium on mathematical statistics and probability , 19541955 , vol . i _
california press , berkeley .
hodges , j. l. jr . and lehmann , e. l. ( 1956 ) .
the efficiency of some nonparametric competitors of the @xmath35-test .
_ * 27 * 324335 .
lehmann , e. l. ( 1957 ) .
a theory of some multiple decision problems .
_ * 28 * 125 .
lehmann , e. l. ( 1957 ) .
a theory of some multiple decision problems .
_ * 28 * 547572 .
lehmann , e. l. ( 1958 ) .
significance level and power .
statist . _
* 29 * 11671176 .
fix , e. , hodges , j. l. jr . and lehmann , e. l. ( 1959 ) . the restricted chi - square test . in _ probability and statistics : the harald cramr volume _
( u. grenander , ed . ) 92107 .
almqvist & wiksell , stockholm .
lehmann , e. l. ( 1959 ) .
optimum invariant tests .
_ * 30 * 881884 .
hodges , j. l. jr . and lehmann , e. l. ( 1961 ) .
comparison of the normal scores and wilcoxon tests . in _ proc .
4th berkeley sympos .
math . statist . and prob .
california press , berkeley , ca .
lehmann , e. l. ( 1961 ) .
some model i problems of selection .
statist . _
* 32 * 9901012 .
hodges , j. l. jr . and lehmann , e. l. ( 1962 ) .
rank methods for combination of independent experiments in analysis of variance .
statist . _
* 33 * 482497 .
hodges , j. l. jr . and
lehmann , e. l. ( 1962 ) .
probabilities of rankings for two widely separated normal distributions . in _ studies in mathematical analysis and related topics
_ 146151 .
stanford univ . press , stanford , ca .
hodges , j. l. jr . and lehmann , e. l. ( 1963 ) .
estimates of location based on rank tests .
statist . _
* 34 * 598611 .
lehmann , e. l. ( 1963 ) .
robust estimation in analysis of variance
statist . _
* 34 * 957966 .
lehmann , e. l. ( 1963 ) . a class of selection procedures based on ranks .
_ * 150 * 268275 .
lehmann , e. l. ( 1963 ) .
asymptotically nonparametric inference : an alternative approach to linear models .
statist . _
* 34 * 14941506 .
lehmann , e. l. ( 1963 ) .
nonparametric confidence intervals for a shift parameter .
* 34 * 15071512 .
lehmann , e. l. ( 1964 ) .
asymptotically nonparametric inference in some linear models with one observation per cell . _
statist . _
* 35 * 726734 .
lehmann , e. l. ( 1965 ) .
on the non - verifiability of certain parametric functions .
primen . _ * 10 * 758760
. lehmann , e. l. ( 1966 ) . on a theorem of bahadur and goodman .
_ ann . math .
statist . _
* 37 * 16 .
lehmann , e. l. ( 1966 ) .
some concepts of dependence .
statist . _
* 37 * 11371153 .
hodges , j. l. jr . and lehmann , e. l. ( 1967 ) .
moments of chi and power of @xmath35 . in _ proc .
fifth berkeley sympos .
math . statist . and probability ( berkeley , calif .
, 1965/66 ) , vol .
i : statistics _ 187201 .
california press , berkeley , ca .
hodges , j. l. jr . and
lehmann , e. l. ( 1967 ) . on medians and quasi medians .
_ j. amer .
assoc . _ * 62 * 926931 .
hodges , j. l. jr . and lehmann , e. l. ( 1968 ) . a compact table for power of the @xmath35-test .
statist . _
* 39 * 16291637 .
lehmann , e. l. ( 1968 ) .
hypothesis testing . in _ international encyclopedia of the social sciences _
( d. l. sills , ed . )
macmillan , new york .
bickel , p. j. and lehmann , e. l. ( 1969 ) .
unbiased estimation in convex families
statist . _
* 40 * 15231535 .
hodges , j. l. jr . and lehmann , e. l. ( 1970 ) . deficiency .
_ ann . math .
statist . _
* 41 * 783801 .
hodges , j. l. jr . and lehmann , e. l. ( 1973 ) .
wilcoxon and @xmath35 test for matched pairs of typed subjects .
_ j. amer .
* 68 * 151158 .
le cam , l. and lehmann , e. l. ( 1974 ) .
j. neyman : on the occasion of his 80th birthday .
statist . _
* 2 * vii xiii .
bickel , p. j. and lehmann , e. l. ( 1974 ) .
measures of location and scale . in _ proceedings of the prague symposium on asymptotic statistics _ ( _ charles univ . _ ,
_ prague _ , 1973 ) , _ vol .
i _ ( j. hajek , ed . ) 2536 .
charles univ . ,
prague .
bickel , p. j. and lehmann , e. l. ( 1975 ) .
descriptive statistics for nonparametric models .
i. introduction .
statist . _
* 3 * 10381044 .
lo kam , l. and lehmann , l. ( 1975 ) .
professor jerzy neyman . on the occasion of his 80th birthday .
_ fiz .- mat .
* 18 * 152156 .
bickel , p. j. and lehmann , e. l. ( 1975 ) .
descriptive statistics for nonparametric models .
ii . location .
_ * 3 * 10451069 .
bickel , p. j. and lehmann , e. l. ( 1976 ) .
descriptive statistics for nonparametric models .
statist . _
* 4 * 11391158 .
lehmann , e. l. and shaffer , j. p. ( 1977 ) . on a fundamental theorem in multiple comparisons .
_ j. amer .
assoc . _ * 72 * 576578 .
lehmann , e. l. ( 1978 ) .
henry scheff , 19071977 .
_ internat .
_ * 46 * 126 .
lehmann , e. l. and shaffer , j. p. ( 1979 ) .
optimum significance levels for multistage comparison procedures .
statist . _ * 7 * 2745 .
bickel , p. j. and lehmann , e. l. ( 1979 ) .
descriptive statistics for nonparametric models .
iv . spread . in _ contributions to statistics _ 3340 .
reidel , dordrecht .
anderson , t. w. , chung , k. l. and lehmann , e. l. ( 1979 ) .
pao lu hsu : 19091970 .
statist . _ * 7 * 467470 .
lehmann , e. l. ( 1979 ) .
hsu s work on inference .
_ * 7 * 471473 .
daniel , c. and lehmann , e. l. ( 1979 ) .
henry scheff , 19071977 .
_ * 7 * 11491161 .
lehmann , e. l. ( 1980 ) .
efficient likelihood estimators .
statist . _ * 34 * 233235 .
lehmann , e. l. ( 1980 ) .
the work of pao lu hsu on statistical inference .
_ knowledge practice math .
_ * 3 * 68 .
anderson , t. w. , chung , k. l. and lehmann , e. l. ( 1980 ) .
pao lu hsu : 19101970 .
_ knowledge practice math .
_ * 3 * 35 .
lehmann , e. l. ( 1981 ) .
an interpretation of completeness and basu s theorem .
_ j. amer .
assoc . _ * 76 * 335340 .
bickel , p. j. and lehmann , e. l. ( 1981 ) .
a minimax property of the sample mean in finite populations .
statist . _
* 9 * 11191122 .
hodges , j. l. jr . and lehmann , e. l. ( 1982 ) .
minimax estimation in simple random sampling . in _ statistics and probability : essays in honor of c. r. rao
_ ( kallianpur et al . , eds . ) .
north - holland , amsterdam .
lehmann , e. l. and reid , c. ( 1982 ) .
in memoriam : jerzy neyman ( 18941981 ) . _ amer .
statist . _ * 36 * 161162 .
lehmann , e. l. ( 1982 ) .
. _ encycl .
* 2 * 20792087 .
hodges , j. l. and lehmann , e. l. ( 1983 ) .
hodges lehmann estimators .
sci . _ * 3 * 31803183 .
lehmann , e. l. ( 1983 ) .
estimation with inadequate information .
_ j. amer .
assoc . _ * 78 * 624627 .
lehmann , e. l. ( 1983 ) .
least informative distributions . in _ recent advances in statistics
_ 593599 . academic press , new york .
lehmann , e. l. ( 1983 ) .
comparison of experiments for some multivariate normal situations . in _ studies in econometrics ,
time series , and multivariate statistics _
( karlin et al . , eds . ) 491503 . academic press , new york .
lehmann , e. l. ( 1984 ) .
specification problems in the neyman pearson wald theory . in _ statistics : an appraisal .
proceedings 50th anniversary conference _ ( h. a. david and h. t. david , eds . ) 425436 .
iowa state univ .
press , ames , ia .
lehmann , e. l. ( 1985 ) .
the neyman
pearson lemma .
_ * 6 * 224230 .
lehmann , e. l. ( 1985 ) .
the neyman
pearson theory after fifty years . in
_ proceedings of the berkeley conference in honor of jerzy neyman and jack kiefer , vol .
i _ ( l. le cam and r. a. olshen , eds . ) 114 .
wadsworth , belmont , ca .
lehmann , e. l. ( 1988 ) .
sci . _ * 9 * 386391 .
lehmann , e. l. ( 1988 ) .
statistics an overview . _
_ * 8 * 683702 .
lehmann , e. l. ( 1988 ) .
comparing location experiments .
statist . _
* 16 * 521533 .
lehmann , e. l. and shaffer , j. p. ( 1988 ) .
inverted distributions . _
statist . _ * 42 * 191194 .
lehmann , e. l. ( 1989 ) .
group families . in _ encycl . statist .
_ 7071 .
diaconis , p. and lehmann , e. l. ( 1990 ) .
contributions to mathematical statistics . in _ a statistical model _ : _ frederick mosteller s contributions to statistics , science , and public policy _ ( s. e. fienberg , ed . ) 5980 .
springer , new york .
lehmann , e. l. ( 1990 ) .
verifiability and strong verifiability . in _
zapiski nauchnykh seminarov leningradskogo otdeleniya matematicheskogo instituta imeni v. a. steklova akademii nauk sssr _
( _ lomi _ ) ( ibragimov et al . , eds . ) * 184 * 182188 .
lehmann , e. l. ( 1990 ) .
model specification : the views of fisher and neyman , and later developments .
_ statist .
sci . _ * 5 * 160168 .
lehmann , e. l. and loh , w .- y .
pointwise versus uniform robustness of some large - sample tests and confidence intervals .
j. statist . _
* 17 * 177187 .
lehmann , e. l. ( 1990 ) .
comment on lindley : the present position in bayesian statistics . _
sci . _ * 5 * 8283 .
lehmann , e. l. and scheff , h. ( 1990 ) . in _ dictionary of scientific biography _
* 18 * supplement ii .
lehmann , e. l. and neyman , j. ( 1990 ) . in _ dictionary of scientific biography _
* 18 * supplement ii .
( reproduced as : neyman , j. ( 2008 ) . _
complete dictionary of scientific biography _ * 18 * 669675 .
charles scribner s sons , detroit .
gale virtual reference library . web .
14 june 2011 . )
lehmann , e. l. ( 1991 ) .
introduction to student ( 1908 ) : the probable error of a mean . in _ breakthroughs in statistics ,
( s. kotz and n. l. johnson , eds . ) 2932 .
springer , new york .
lehmann , e. l. ( 1991 ) .
introduction to neyman , j. and pearson , e. s. ( 1933 ) : on the problem of the most efficient tests of statistical hypotheses . in _ breakthroughs in statistics ,
i _ ( s. kotz and n. l. johnson , eds . ) 6772 .
springer , new york .
lehmann , e. l. and scholz , f. w. ( 1992 ) .
ancillarity . in _
current issues in statistical inference : essays in honor of d. basu_. _ institute of mathematical statistics lecture notes
monograph series _ * 17 * 3251 .
ims , hayward , ca .
lehmann , e. l. and rojo , j. ( 1992 ) .
invariant directional orderings .
statist . _
* 20 * 21002110 .
lehmann , e. l. ( 1993 ) .
the fisher , neyman
pearson theories of testing hypotheses : one theory or two ?
_ j. amer .
assoc . _ * 88 * 12421249 .
lehmann , e. l. ( 1993 ) .
the bertrand
borel debate and the origins of the neyman
pearson theory . in _ statistics and probability .
a raghu raj bahadur festschrift _ ( ghosh et al . , eds . ) 371380 .
wiley eastern ltd .
, new delhi .
lehmann , e. l. ( 1993 ) .
mentors and early collaborators : reminiscences from the years 19401956 with an epilogue . _ statist .
_ * 8 * 331341 .
lehmann , e. l. ( 1994 ) .
jerzy neyman ( 18941981 ) . in _ biographical memoirs of the nat .
acad . of sci . _ * 63 * 395420 . the national academies press , washington , dc .
lehmann , e. l. ( 1990 ) . in _
zapiski nauchnykh seminarov leningradskogo otdeleniya matematicheskogo instituta imeni v. a. steklova akademii nauk sssr _
( _ lomi _ ) ( ibragimov et al . , eds . ) * 184 * 182188 .
lehmann , e. l. ( 1995 ) .
foundational issues in statistics : theory and practice .
_ foundations of science _ * 1 * 4549
. lehmann , e. l. ( 1995 ) .
neyman s statistical philosophy .
statist . _
* 15 * 2936 .
dedicated to the memory of jerzy neyman .
lehmann , e. l. ( 1996 ) . the creation and early history of the berkeley statistics department . in _
statistics , probability and game theory _ ( t. ferguson , ed . ) .
_ institute of mathematical statistics lecture notes
monograph series _
* 30 * 139146 .
ims , hayward , ca .
lehmann , e. l. ( 1997 ) .
le cam at berkeley . in _
festschrift for lucien le cam _ 297304 .
springer , new york .
lehmann , e. l. ( 1997 ) .
testing statistical hypotheses : the story of a book . _
* 12 * 4852 .
lehmann , e. l. ( 1999 ) .
`` student '' and small - sample theory . _
* 14 * 418426 .
bickel , p. j. and lehmann , e. l. ( 2001 ) .
frequentist inference . in _
international encyclopedia of the social and behavioral sciences _
pergamon , oxford .
bickel , p. j. and lehmann , e. l. ( 2001 ) .
frequentist interpretation of probability . in _ international encyclopedia of the social and behavioral sciences _
( n. j. smelser and paul b. baltes , eds . ) 57965798 .
pergamon , oxford .
lehmann , e. l. ( 2004 ) .
optimality and symposia : some history . in _ the first erich l. lehmann symposium
optimality_. _ institute of mathematical statistics lecture notes
monograph series _ * 44 * 110 .
ims , beachwood , oh . lehmann , e. l. , romano , j. p. and shaffer , j. p. ( 2005 ) . on optimality of stepdown and stepup multiple test procedures .
_ * 33 * 10841108 .
lehmann , e. l. and romano , j. p. ( 2005 ) .
generalizations of the familywise error rate .
statist . _
* 33 * 11381154 .
arrow , k. j. and lehmann , e. l. ( 2005 ) .
harold hotelling . in _ biographical memoirs_. _ national academy of sciences _ * 87 * 221233 . the national academies press , washington , dc .
lehmann , e. l. ( 2006 ) .
hodges , joseph lawson , jr . in _ encyclopedia of statistical science _
, 2nd ed . 31793180 .
lehmann , e. l. ( 2006 ) .
scheff , henry . in _ encyclopedia of statistical science _ , 2nd ed .
74727474 .
lehmann , e. l. ( 2006 ) . on likelihood
ratio tests . in _ the second erich l. lehmann symposium optimality _ ( j. rojo , ed . ) .
_ institute of mathematical statistics lecture notes
monograph series _ * 49 * 19 .
ims , beachwood , oh . lehmann , e. l. ( 2008 ) . on the history and use of some standard statistical models . in _ probability and statistics : essays in honor of david a. freedman _ ( d. nolan and t. speed , eds . ) . _
_ * 2 * 114126 .
ims , beachwood , oh . diaconis , p. and lehmann , e. l. ( 2008 ) .
comment on `` on student s 1908 article : the probable error of a mean , '' by s. l. zabell .
_ j. amer .
assoc . _ * 103 * 1619 .
lehmann , e. l. ( 2009 ) .
parametric versus nonparametrics : two alternative methodologies .
_ j. nonparametr .
* 21 * 397405 .
lehmann , e. l. ( 2009 ) . rejoinder to `` parametric versus nonparametrics : two alternative methodologies ( with discussion ) . '' _
j. nonparametr .
* 21 * 425426 .
lehmann , e. l. ( 2009 ) .
some history of optimality . in _
optimality_. _ institute of mathematical statistics lecture notes
monograph series _ * 57 * 1117 .
ims , beachwood , oh .
erich s sensitivity toward others , contagious zest for life , gentle spirit , fundamental contributions to statistics and remarkable contributions to human resources development , have been recorded , chronicled and honored through various mechanisms .
after his death , erich s life was celebrated with a memorial service that took place at the berkeley women s faculty club on november 9th , 2009 .
the service was well attended .
his family , friends , students , collaborators and colleagues paid homage .
peter bickel organized a memorial session during the 2010 joint statistical meetings in vancouver ( persi diaconis , juliet shaffer and peter bickel speakers ) .
the session was very well attended with standing room only .
the respect and appreciation for erich was international .
willem van zwet organized a memorial session during the 73rd ims annual meeting in gothenburg , sweden in 2010 ( david cox , kjell doksum , willem van zwet , speakers ) , and peter bickel gave a lecture during the latin american congress of probability and mathematical statistics ( clapem ) in venezuela , november 2009 , in remembrance of erich lehmann .
recordings of various erich talks are freely accessible to the public for viewing .
these include lectures he gave during the second and third lehmann symposia at rice university .
obituaries by peter bickel ( 2009 ) and david brillinger ( 2010 ) provide additional information about the life and work of erich l. lehmann .
other sources that present fascinating accounts of erich s work and life include lehmann ( 2008b ) , degroot ( 1986 ) and reid ( 1982 ) . a collection of selected works edited by the author will soon be published by springer .
the _ selected works of e. l. lehmann _ provides an extended bibliography and , through invited vignettes , examines more closely the various facets of his work . | through the use of a system - building approach , an approach that includes finding common ground for the various philosophical paradigms within statistics , erich l. lehmann is responsible for much of the synthesis of classical statistical knowledge that developed from the neyman
pearson wald school . a biographical sketch and a brief summary of some of his many contributions are presented here .
his complete bibliography is also included and the references present many other sources of information on his life and his work . . |
class ii methanol masers are unique tracers of the early stages of high - mass star formation ( see e.g. * ? ? ?
the other maser transitions commonly observed in high - mass star formation regions ( main - line oh and 22 ghz water ) are also observed in different environments , such as late - type stars and low - mass star formation regions .
in contrast , despite a number of sensitive and extensive searches no class ii methanol masers have been found associated with anything other than high - mass star formation regions @xcite .
the majority of class ii methanol maser sources are not associated with ultra - compact hii ( uchii ) regions @xcite .
this indicates that either the masers are associated with objects that are not massive enough to form detectable hii regions , or that they are associated with high - mass stars prior to the formation of an uchii region .
numerous observations at millimeter through mid - infrared wavelengths ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) suggest that many class ii methanol maser sources are associated with a very early evolutionary phase in the high - mass star formation process .
hindered rotation in the methanol molecule leads to a very large number number of rotational and vibrational transitions in the millimetre and submillimetre regime .
a large number of these transitions have been observed in interstellar space and a select few show maser action .
the transitions in which masers are observed have been empirically classified into two groups @xcite .
the class ii methanol masers are closely associated with sites of high - mass star formation and related objects , including infrared sources , hii regions and often oh or water masers . the 6.7 ghz @xmath2 transition is the strongest and most common class ii maser transition , having been observed towards more than 900 regions within our galaxy @xcite .
class i methanol masers are found near high - mass star formation regions , but offset from the hii regions by a few tenths of a parsec @xcite .
the 44.0 ghz @xmath3 transition is the strongest and most common class i maser transition .
this paper focuses solely on class ii methanol masers and hereafter it is implicit that all references to `` methanol masers '' refer to class ii masers . to date eighteen different transitions ( or transition groups ) , associated with eight different methanol transition series
have been observed as class ii methanol masers and their frequencies , along with the discovery reference are listed in table [ tab : classii ] .
counting the total number of known class ii methanol maser transitions is somewhat subjective .
the distinction which we have made is that where there are two or more transitions at similar frequencies which are always ( or nearly always ) either present , or absent together from sources we have considered these to be a transition group . for example , all observations to date of the 38.3 ( @xmath4 ) and 38.5 ghz ( @xmath5 ) transitions show that where emission from one of these is detected , the other is also and with similar intensity @xcite .
hence we consider the 38.3 and 38.5 ghz masers to represent a single transition group .
in contrast the 6.7 , 107.0 and 156.6 ghz transitions are all members of the @xmath6a@xmath7 series , however , many sources which show emission in the 6.7 ghz transition are not observed in either of the other two , so we count these as three separated maser transitions .
the @xmath8e series at frequencies near 157 ghz , is another example of a single transition group @xcite , although the 148.1 ghz @xmath9e is counted as a separate transition as it stands apart both in frequency , and in the physical conditions it requires .
the two lowest frequency methanol maser transitions ( @xmath2 at 6.7 ghz and @xmath10 at 12.2 ghz ) have been detected towards a large number of sources , but the majority of the other transitions are only observed towards a handful of sources ( see * ? ? ? * and references therein ) .
current models of class ii methanol masers @xcite find that the 6.7 and 12.2 ghz transitions are strongly inverted over a wide range of physical conditions , explaining both their high - brightness temperatures and common occurrence .
the models successfully predict masing in all of the other class ii transitions for which it has been observed , but generally the physical conditions implied are much more restricted @xcite .
so the presence of multiple class ii maser transitions within one source can , through comparison with models , be used to constrain the physical conditions in the masing region ( under the assumption that the various transitions all arise from the same location ) .
the observations were made using the swedish eso submillimetre telescope ( sest ) between 2000 june 22 - 27 .
we observed a total of 13 different methanol transitions , 7 of which are known class ii methanol maser transitions and the remaining 6 of which are predicted class ii methanol maser transitions @xcite .
the frequencies of the observed transitions range from 84 to 265 ghz .
these transitions fall into 11 different transition groups , as the @xmath11 and @xmath12 represent a single transition group in each case .
all of the observations were targeted towards the class ii methanol maser source g345.01 + 1.79 ( more precisely g345.010 + 1.792 ) at @xmath13 = 16@[email protected]@xmath16 ; @xmath17 = -40@[email protected]@xmath20
@xcite . the onsource integration time for each of these transitions ranged from 30 to 60 minutes and the resulting rms per spectral channel range from 0.3 to 2.3 jy .
the observations were made by frequency switching with a throw of 6 mhz and a switching time of 2 minutes .
the spectra were obtained using the high - resolution acousto - optic spectrometer , which has a total bandwidth of 86 mhz and a spectral resolution of 80 khz , resulting in velocities resolutions of between 0.28 and 0.09 @xmath21 ( for the lowest and highest frequency transitions observed , respectively ) .
the intensity scale was converted from kelvin to jy by assuming at sensitivity of 25 jy k@xmath22 for the transitions at frequencies less than 150 ghz , 30 jy k@xmath22 for transitions at frequencies between 150 and 200 ghz and 41 jy k@xmath22 for transitions at frequencies higher than 200 ghz .
the data were processed using per bergman s xs package , following the standard procedure for radio - millimetre single dish spectral line observations .
the details of the transitions observed and the achieved sensitivities are summarized in table [ tab : obssum ] . because we were searching for narrow
maser lines the spectra were not smoothed prior to measuring the rms .
emission was detected at 9 of the 13 methanol transitions observed .
these observations represent the first astronomical detection of maser emission from the 104.1 ghz ( @xmath0 ) transition and the 216.9 ghz ( @xmath1 ) transition .
the spectra of all of the methanol maser transitions detected during the current sest observations are shown in fig .
[ fig : stack ] .
it should be noted that the class ii methanol maser reported here is a different transition from the 104.3 ghz ( @xmath23 ) class i methanol maser previously discovered by @xcite .
spectra of the emission in these two new class ii methanol maser transitions are shown in fig [ fig : new ] . in both cases
the maser emission consists of a narrow ( fwhm @xmath24 0.5 @xmath21 ) peak at a velocity of around -22 @xmath21 .
this is the approximate velocity at which most of the rare class ii methanol masers observed towards g345.01 + 1.79 peak .
observations of the 104.1 ghz transition of methanol have previously been made towards the archetypal class ii methanol maser region w3(oh ) by @xcite .
no emission was detected in the w3(oh)/(h@xmath25o ) region at the sensitivity level achieved for their 104.1 ghz observations .
the 216.9 ghz spectrum shows an additional broader peak at approximately -13 @xmath21 , coinciding with the velocity at which thermal methanol emission has previously been observed in this source @xcite .
although the signal to noise in each of these spectra is moderately low , subsequent sest observations focusing on studying the thermal emission in this region have confirmed the presence of the maser emission in each of these two transitions @xcite .
the observations of the 104.1 and 216.9 ghz transitions of @xcite , were made in april 2003 , slightly less than three years after the observations reported here . in 2003
the 104.1 ghz maser had a peak flux density of @xmath26 3.8 jy , consistent ( within the relative uncertainties ) with the flux density of the current sest observations . in 2003
the 216.9 ghz transition had a peak flux density of @xmath26 10.3 jy , significantly greater than the @xmath26 6 jy reported in the current observations . comparing the 216.9 ghz spectrum in figure 1 in @xcite with our figure [ fig : new ]
shows that relative intensity of the maser and thermal components has not change dramatically .
given that we do nt expect to see any significant variations in the thermal emission , the observed difference in maser intensity is likely primarily due to calibration and/or pointing errors in the current observations , rather than significant variations in the maser .
@xcite detected maser emission from the 231.3 ghz ( @xmath27 ) transition in g345.01 + 1.79 with a peak flux density of @xmath262 jy .
no maser emission was detected in our sest observations , however , as they are significantly less sensitive ( rms 1.5 jy ) than the observations of @xcite , this is expected .
we have fitted gaussian profiles to both the maser emission , and ( where seen ) the thermal emission for each transition .
the results of these fits are summarised in table [ tab : gauss ] .
for all of the transitions it was necessary to fit more than one gaussian component to adequately describe the maser emission ( i.e. residuals comparable with the noise ) .
many of the transitions have the strongest components at velocities close to -22.2 and -21.4 @xmath21 ( e.g. 86.6 , 86.9 and 104.1 ghz ) , while others show their strongest emission at a velocity of around -21.7 @xmath21 ( e.g. 107.0 , 108.9 , 156.6 and 156.8 ghz ) . even accounting for the relatively low signal to noise in some of the spectra
it is clear from this figure that although the maser emission in all transitions peaks at a velocity of around -22 @xmath21 , it can not be described by a single gaussian component .
uncertainty in the absolute frequency calibration of the spectrometer and in the adopted rest frequencies of some of the transitions is approximately 0.2 @xmath21 , nevertheless , the difference in the shape and width of the emission at different frequencies requires the presence of two or more different maser components .
g345.01 + 1.79 has been observed at 14 of the 15 class ii methanol maser transition ( or transition groups ) frequencies listed in table [ tab : classii ] .
the exception is the @xmath28 ( 29.0 ghz ) transition , for which there is only one published study @xcite . including the two new class ii transitions reported here
, maser emission has been detected in g345.01 + 1.79 for 15 of the 16 known class ii transitions ( or transition groups ) observed in this source , the exception being the @xmath29 23.1 ghz , for which the sensitive observations of @xcite placed a 3@xmath30 upper limit of 0.5 jy . some other well studied hii regions ( such as w3(oh ) and ngc6334f ) also show a large number of class ii methanol masers , but no others exhibit as many as g345.01 + 1.79 .
it is not clear why g345.01 + 1.79 is such an exceptional source .
@xcite identified two distinct class ii methanol maser sites ( separated by 19@xmath31 , with non - overlapping velocity ranges ) in this general region .
g345.010 + 1.792 has methanol emission covering the velocity range -24 to -16 @xmath21 , while g345.012 + 1.797 masers covers -16 to -10 @xmath21 @xcite . g345.010 + 1.792 is associated with an uchii region , and all of the rare / weak methanol maser transitions are associated with this source .
@xcite detected three water masers in this region , one associated with each maser site and a third which is not associated with any other known maser transitions .
interestingly , the weakest water maser in the region is associated with the strongest methanol maser site ( g345.010 + 1.792 ) , which also has an associated oh maser @xcite , while the secondary methanol maser site is associated with the strongest water maser . in this paper
, we are focusing purely on the strongest methanol maser site ( g345.010 + 1.792 ) .
in general the strongest emission from class ii methanol maser transitions in this source is at velocities near -22 @xmath21 .
in contrast , when it was first observed in the early 1990 s , the peak velocity of the 6.7 ghz methanol maser was at -17 @xmath21 @xcite .
the intensity of that spectral feature has declined over the intervening 20 years ( see e.g. * ? ? ?
* ; * ? ? ?
* ) , and the peak of the 6.7 ghz maser emission is now at @xmath26 -21 @xmath21 .
the 19.9 ghz methanol maser is now the only transition which peaks at -17 @xmath21 @xcite , and in contrast to the other rare / weak methanol maser transitions no emission is observed at velocities less than -19 @xmath21 .
the complex , non - linear relationship between the physical conditions of the masing gas and the intensity of the observed maser means that it is generally not possible to uniquely infer the former from the latter . where more than one maser transition is observed from the same volume of gas
, it must be the case that a complete theoretical maser model will be able to find conditions that simultaneously invert all of the detected transitions in the observed intensity ratios .
@xcite used information on 9 class ii methanol maser transitions , collected with a variety of instruments , over a period of several years , to constrain the physical conditions in g345.01 + 1.79 , using the model of @xcite .
they found the observed ratio of maser intensities to be consistent with cool gas ( 30k ) at a density of @xmath32 @xmath33 with a methanol column density of @xmath34 @xmath35 , being pumping by the radiation from warm ( 175k ) dust .
modelling such as that undertaken by @xcite requires a number of assumptions .
in particular , it is assumed that the emission from each transition arises from the same region of gas and that the emitting region is the same size for each transition . where information is collated from observations made with different instruments over a period of years , it is also necessary to ignore the effects of temporal variability . for g345.01
+ 1.79 @xcite showed that the 6.7 , 12.2 , 85.5 and 86.6 ghz transitions are all coincident to within 1@xmath31 ( the relative positional accuracy of the observations ) , and where it has been possible to test it the assumption that the different maser transitions are co - spatial with greater precision this appears to be generally true @xcite . the assumption that the emitting regions are the same size in each transition is not likely to be correct , but it requires very long baseline interferometry observations to make the necessary measurements , and at present that is only practical for the 6.7 and 12.2 ghz transitions . although the variability of class ii methanol masers is not as extreme as that observed for water masers , variations on timescales of months or years are common @xcite . for the weaker millimetre methanol masers variability
is also expected , possibly even to a greater extent due to the lower degree of saturation in these transitions , and evidence of variations in the millimetre masers in g345.01 + 1.79 have been observed @xcite . of the many thousands of methanol transitions which fall at centimeter or millimeter wavelengths , a small subset ( around 20 )
exhibit pronounced maser characteristics .
many of the maser lines can be grouped into series with the same patterns of @xmath36 and @xmath37 quantum numbers .
for example , the @xmath2 maser at 6.7 ghz is part of a series of @xmath6a@xmath7 masers including the @xmath38 maser at 107.0 ghz and the @xmath39 maser at 156.6 ghz .
this is consistent with the models , which also predict maser action in the @xmath40 transition at 57.0 ghz ( which can not be observed by ground - based telescopes due to atmospheric absorption by molecular oxygen at this frequency ) .
one can think of the @xmath41 ladder of energy levels as being slightly overpopulated with respect to the nearby @xmath42 energy levels , with the resulting population inversions being responsible for the maser action .
the new maser transitions detected in the current work are the first members of two new series of methanol transitions to exhibit maser action , and as such represent a more stringent test of the models .
our detection of maser action in the @xmath1 transition at 216.9 ghz is consistent with the modelling of @xcite and @xcite , which also include the @xmath43 120.2 ghz , @xmath44 168.6 , and @xmath45 265.3 ghz transitions from the same series in the list of maser candidates .
the newly observed @xmath0 maser at 104.1 ghz is part of the same series as the @xmath46 7.3 ghz and @xmath47 55.7 ghz transitions , both of which were identified as class ii methanol maser candidates in these models . in the models presented in these papers the 104.1 ghz maser candidate was below the threshold intensity for listing , indicating that its excitation to observable levels requires more extreme conditions .
detailed modelling of the methanol masers in g345.01 + 1.79 is beyond the scope of this letter , and will be the subject of a future publication .
in addition to the data presented here , previous and subsequent observations with sest and other instruments have obtained detections and limits on the emission from more than three dozen known or potential class ii methanol transitions .
here we restrict ourselves to a preliminary exploration which utilises the data from these observations , in combination with published 6.7 and 12.2 ghz spectra , following the approach outlined in @xcite .
the significant differences in the detailed spectral structure of the methanol maser emission in different transitions near velocities of -22 @xmath21 , demonstrates the presence of multiple , partially blended maser components .
this implies that within g345.01 + 1.79 there are multiple sites with differing physical conditions which are giving rise to emission in rare methanol transitions .
hence it is unlikely that any single model fit will be able to explain the observed intensities of all of the transitions , and this is indeed the case . to adequately explain the observed ( and the one non - detected ) class ii methanol transitions in g345.01 + 1.79 using the @xcite model we require emission from at least two regions with similar / overlapping line - of - sight velocities but different physical parameters
this is inferred from the differences in the line profiles of the observed transitions ( e.g. figure [ fig : stack ] shows the 86.6 , 86.9 and 104.1 ghz transitions are double - peaked , while other transitions are single - peaked ) and in their peak velocities ( see table [ tab : gauss ] ) .
applying models which assume that emission in all the transitions is co - spatial can not produce a single , self - consistent solution in cases such as this .
for example in g345.01 + 1.79 we observe both maser transitions which require low densities ( e.g. the 86.6 and 86.9 ghz ) and transitions which prefer higher densities ( e.g. the 85.6 and 104.1 ghz ) for their efficient excitation .
we suggest that the reason for g345.01 + 1.79 exhibiting maser emission in an unusually large number of different methanol transitions , may be that the coexistence of masing regions covering such a broad range of physical conditions is rare .
we have discovered two , new class ii methanol maser transitions towards the g345.01 + 1.79 star formation region .
the emission from these transitions is relatively weak in g345.01 + 1.79 and it is likely that a only small number of high - mass star formation regions will show maser emission in these transitions .
g345.01 + 1.79 clearly has great potential for helping us to study the physical conditions in high - mass star formation regions at high resolution , through observations of the large number of class ii methanol maser transitions .
the primary limitations on the current observations are due to relatively low signal to noise and spatial blending .
high resolution observations of the weaker millimetre methanol masers to obtain detailed information on their distribution , to constrain the spot size and determine whether the different transitions are coincident have recently become much more feasible with the commencement of alma operations .
such observations offer the prospect of unlocking the potential of multi - transition maser observations and providing important insights into the processes which occur in high - mass star formation regions .
this research has made use of nasa s astrophysics data system abstract service .
travel to sest for observations was provided under the access to major research facilities program which is supported by the commonwealth of australia under the international science linkages program .
ams was partly supported by russian state contract no .
14.518.11.7064 and the russian foundation for basic research ( grants 10 - 02 - 00589-a and 11 - 02 - 01332-a ) .
dmc and pdg acknowledge financial support for this work from the australian research council .
salii , s.v . , sobolev , a.m. , kalinina , n.d . ,
ellingsen , s.p . ,
cragg , d.m . ,
godfrey , p.d . , harjunp , p. , zinchenko , i.i .
, 2003 , in communications of the konkoly observatory , the interaction of stars with their environment ii , ed .
kiss , m. kun & v. knyves , 75 tsunekawa , s. , ukai , t. , toyama , a. , takagi , k. , 1995 , microwave frequencies of the ch@xmath48oh molecule in the frequency range from 7 to 200 ghz , report for the grant - in - aid for scientific research on priority areas ( interstellar matter , 1991 - 1994 ) of the ministry of education , science , and culture , japan .
toyama university , japan .
.class ii methanol maser transitions .
the rest frequencies have been rounded at the 100 khz significance level .
the g345.01 + 1.79 reference is the most sensitive high resolution observation available for the specific transition .
[ cols="^,<,^ , < , < " , ] | we have used the swedish eso submillimetre telescope ( sest ) to search for new class ii methanol maser transitions towards the southern source g345.01 + 1.79 . over a period of 5 days we observed 11 known or predicted class ii methanol maser transitions .
emission with the narrow line width and characteristic velocity of class ii methanol masers ( in this source ) was detected in 8 of these transitions , two of which have not previously been reported as masers . the new class ii methanol maser transitions are the @xmath0 transition at 104.1 ghz and the @xmath1 transition at 216.9 ghz .
both of these are from transition series for which there are no previous known class ii methanol maser transitions .
this takes the total number of known class ii methanol maser series to 10 , and the total number of transitions ( or transition groups ) to 18 .
the observed 104.1 ghz maser suggests the presence of two or more regions of masing gas with similar line of sight velocities , but quite different physical conditions .
although these newly discovered transitions are likely to be relatively rare , where they are observed combined studies using the australia telescope compact array and the atacama large millimeter array offer the prospect to be able to undertake multi - transition methanol maser studies with unprecedented detail . |
analytical expressions of many feynman diagrams contain polylogarithmic functions of various kinds ( nielsen polylogarithms , harmonic polylogarithms@xcite , harmonic sums@xcite , multiple zeta values , euler sums@xcite , etc ) .
but there exist feynman integrals which can not be described only in terms of polylogarithms . at two - loop level ,
the discontinuity of the off - shell massive `` sunrise '' diagram with different masses is expressed by elliptic functions . at three ( or more ) loop level
the situation worsens .
these diagrams contain nested multiple elliptic integrals ; the current mathematical knowledge of such integrals is scarce or missing . at this preliminary stage ,
_ `` experimental mathematics '' _ is the best tool . in other words , high - precision numerical values of the integrals
are fitted@xcite with various candidate analytical expressions until an agreement is found .
the equalities are then checked up to hundreds or thousands of digits . in this way the right analytical expression is identified beyond any reasonable doubt .
this may be the starting point of the subsequent search for a rigorous proof of the result , task which may take months of hard work@xcite . in ref .
we work out a very high - precision value of the 3-loop scalar master integral of the `` sunrise '' diagram @xmath2 of fig.1 ; then we fit that value with products of elliptic integrals , checking the equality with a precision of thousand of digits . in this paper
we continue the work on the 3-loop integral , and we simplify the 3-loop analytic result by using some identities between elliptic integrals .
next we review and develop the approach used for fitting and identifying the 3-loop integral and we apply it to the 4-loop scalar master integral of the `` sunrise '' diagram @xmath3 of fig.1 .
the analytical calculation of this master integral is also of physical interest , because it is the simplest non - trivial master integral of the 4-loop electron @xmath0-@xmath1 ( of which a high - precision numerical calculation is under way@xcite ) .
very likely , the analytical constants which appear in the expression of @xmath3 should also appear in the analytical expression of 4-loop @xmath0-@xmath1 .
we calculate an high - precision value of @xmath3 and we are able to fit this value with an expression containing two new elliptic constants , checking the equality with a precision of thousand of digits .
we apply this procedure also to the values of some 4-dimensional lattice integrals , and we identify their analytical expressions ; surprisingly , they contain the same elliptic constants of the 4-loop integral .
the plan of the paper is the following : in section [ treloop ] we simplify the results of ref . by using identities between elliptic integrals . in section [ quattroloop ]
we study the 4-loop `` sunrise '' integral .
we collect a number of results on a `` simplified '' version of the integrals involved .
then we use these results as a guide for identifying the candidate analytical expressions suitable for fitting the 4-loop results . in section [ risultati ]
we show the analytical results found for the 4-loop integrals . in section [ quattrolattice ] we fit the values of some 4-dimensional lattice integrals with the same analytical constants discovered in the 4-loop integrals . in section [ conclusions ]
we give our conclusions .
in ref . we considered the feynman diagram @xmath4 with equal masses @xmath5 , and on - shell external momentum ( see fig.1 ) @xmath6 where @xmath7 by using an hyperspherical representation for the integral , we found that the value of @xmath2 could be expressed as a sum of various double elliptic integrals , the simplest being @xmath8 @xmath9 we were not able to calculate @xmath10 in analytical form .
therefore we evaluated it at very high precision and we tried to fit the numerical value with various kinds of analytical expressions . in ref .
we found that @xmath11 where @xmath12 is the first of the two elliptic integrals @xmath13 we were able to fit the values of @xmath2 in 2 and 4 dimensions : @xmath14 @xmath15 , and were checked up to 30000 , 40000 and 1200 digits , respectively .
now we note that the arguments @xmath16 of the elliptic integrals are _
singular values_. in the context of elliptic integrals @xmath17 is a called singular value if @xmath18 where @xmath19 is an integer or a rational number .
the arguments @xmath16 of elliptic integrals are singular values for @xmath20 and @xmath21 respectively , @xmath22 that is @xmath23 @xmath24 the values of elliptic integrals of second kind of are obtained following ref .
@xmath25 by using the values @xcite @xmath26 @xmath27 where @xmath28 and expressing @xmath29 by using , we are able to rewrite and in the very compact form @xmath30 @xmath31 were also shown by david broadhurst in his beautiful talk given in bielefeld@xcite . after the talk
, our unpublished results were shown to david broadhurst .
for sake of completeness we recall here some unpublished observations that suggested us the form of .
first we calculated analytically the _ simplest _ double elliptic integral : @xmath32 ^ 2= 3.437\;592\;909{\ldots } \ , \ ] ] which factorizes in the square of the elliptic integral @xmath12 .
then we observed that in the diagram @xmath33 the value of the square of the external momentum @xmath34 is @xmath35 , which is different from the threshold ( @xmath36 ) or the pseudothresholds ( @xmath37 ) .
we expected that the analytical expression is much simpler for the on - threshold diagram than for the off - threshold diagram .
so we considered the above graph with one mass changed : @xmath38 .
now the value of @xmath39 is on a pseudothreshold ( which are at @xmath40 ) .
the integral analogous to @xmath10 is @xmath41 we were able to fit the numerical value of @xmath42 with @xmath43 ^ 2 = \dfrac{\gamma^6\left(\frac{1}{3}\right)}{2^\frac{14}{3}\pi^2}\ .\ ] ] subsequently , as we expected the form of @xmath10 to be more complicated than @xmath42 , we tried also products of @xmath12 with different arguments , and we found .
now we consider the single - scale 4-loop self - mass diagram @xmath44 of fig.1 , in the case of equal masses @xmath5 and on shell external momentum @xmath39 .
this diagram has 3 master integrals , the simplest being @xmath45 as already said in the introduction , this is also one of the several master integrals which appear in the calculation of 4-loop @xmath0-@xmath1 .
the first observation is that , fortunately , the value of @xmath39 in @xmath3 is already on a pseudothreshold ( which are @xmath39,@xmath46,@xmath47 ) ; therefore to simplify the analytical structure of the integral we have not to use the mass change of the previous section [ tricka3 ] .
we need an high - precision numerical value of this integral in order to obtain a meaningful fit .
this is worked out by using the difference equation method presented in refs . .
summarizing , one raises to @xmath48 one denominator of @xmath49 the function @xmath50 satisfies the fourth - order difference equation @xmath51 where @xmath52 and @xmath53 is the one - loop integral @xmath54 contains in the r.h.s . the integral
obtained from @xmath3 by contracting one line , which factorizes into the product of 4 one - loop tadpoles .
the solution of compatible with the large-@xmath48 boundary condition @xmath55 can be written as @xmath56 .
the functions @xmath57 , @xmath58 and @xmath59 are respectively the two solutions of the homogeneous equation compatible with the above large-@xmath48 behaviour and one particular solution of the nonhomogeneous equation . the constants @xmath60 and @xmath61 are obtained from the 3-loop self - mass integrals belonging to the diagram obtained from @xmath3 by deleting one line , that is @xmath2 .
the amount of calculations needed to work out and solve the systems of difference equations is high , so that the calculations have been performed by means of an automatic tool , the program described in ref . .
in two dimension one finds @xmath62 and in the limit @xmath63 @xmath64 all the above numerical constants were calculated with a precision of over 2400 digits ; for the sake of space we show here only the first 25 . by using an hyperspherical representation for the integral , @xmath65 and
the finite part of @xmath66 contain _ triple _ elliptic integrals , the simplest being @xmath67 we prefer to redistribute the arguments of the @xmath68 functions and consider the similar integral @xmath69 and the companion integral @xmath70 where @xmath71 are the two zeroes of @xmath72 .
@xmath73 and @xmath74 are the 4-loop integrals analogues of the 3-loop integral @xmath10 , and we have to find their analytical expressions .
first of all we consider the simplest triple elliptic integral with structure similar to : @xmath75 by changing one limit of integration over @xmath76 to the zero of @xmath77 we obtain a companion integral analogous to @xmath78 we expect that the study of the simpler constants @xmath79 and @xmath80 can help us understand the analytical expressions of @xmath73 and @xmath74 . integrating over @xmath76 @xmath81 @xmath82 integrating over @xmath83 and @xmath84 @xmath85 @xmath86 no analytical expression
is known for @xmath87 . trying to understand the reason of that , we study the following family of integrals : @xmath88 we consider here only the integrals which have results containing elliptic constants . at the value
@xmath89 there is the integral @xmath90 equivalent to ; note that it factorizes in a square of @xmath91 . at the value @xmath92
we find numerically that the six integrals @xmath93 @xmath94 @xmath95 @xmath96 @xmath97 can be expressed in terms of @xmath79 and @xmath80 . at the value @xmath98
we find a surprise : six integrals factorizes into the fourth power of @xmath99 .
@xmath102 @xmath103 @xmath104 @xmath105 the analytical results have been fitted numerically and checked up to 200 digits of precision .
we find results factorized at level 2 and 4 , but not at level 3 ( odd ) .
this behaviour reminds us of the non - factorization of values of riemann @xmath106-function at odd integers , and suggest to consider the constants @xmath79 and @xmath80 as irreducible objects
. we can relate @xmath79 and @xmath80 to multidimensional @xmath106-like quadruple series .
let us consider the integrals @xmath107 these integrals correspond to the integrals , , , for @xmath108 , @xmath109 , @xmath1 , respectively , and their values are @xmath110 , @xmath111 , @xmath112 .
we apply the change of variable @xmath113 where @xmath114 are the jacobi theta functions , then @xmath115 expanding in series the @xmath116 functions , and integrating over @xmath117 term - by - term , @xmath118 becomes a quadruple series @xmath119 for @xmath120 the series converges , so that @xmath121 where the prime means that the origin @xmath122 must be excluded in the summation .
coming back to the 4-loop integral , if we close the 4-loop self - mass diagram @xmath3 by connecting together the two external lines , we obtain a 5-loop vacuum diagram .
the 5-loop vacuum diagram can be decomposed into two 2-loop self - mass diagrams connected together .
therefore in @xmath123 dimensions its value is given by the integral of the square of the 2-loop self - mass diagram @xmath124 .
the vacuum diagram is expected to have the same analytical structure of @xmath3 , but with higher transcendentality .
@xmath125 satisfies a second order nonhomogeneous differential equation ( see ref . for more details ) .
the corresponding homogeneous differential equation has two solutions @xmath126 and @xmath127 .
in order to reduce the transcendentality we may substitute @xmath125 with @xmath126 or @xmath127 .
the analytical expressions of @xmath128 and @xmath129 are given in the appendix of ref . ; for example , if @xmath130 they read @xmath131 in ref .
we found also that @xmath132 we consider here the integrals of products of @xmath133 : we have checked up to 2400-digits of precision the following equalities : @xmath134 and @xmath135 therefore we have identified some one - dimensional integral representations of the numerical constants @xmath73 and @xmath74 .
are not satisfactory elementary definitions of @xmath73 and @xmath74 , because of the complexity of the arguments of the @xmath12 function in .
our aim is to find out integral representations of @xmath73 and @xmath74 as simple as and .
we make a comparison between the integrals of products of @xmath12 of the family and the integrals of products of @xmath133 .
corresponds to @xmath136 we try to modify the integrand of so that the result contains the constant @xmath137 .
our many year experience with the analytical calculation of 3-loop @xmath0-@xmath1 suggests that such a constant is usually associated with integrals containing the polynomial @xmath138 in the denominator .
this is a factor of @xmath139 .
therefore we try to consider a `` cubic '' modification of the usual elliptic integral @xmath140 .
one fruitful choice is @xmath141{(1-t^3)(1-mt^3)^2}}\ , \qquad e_c(m)=\int_0 ^ 1 \dfrac{dt\;\sqrt[3]{1-mt^3}}{\sqrt[3]{1-t^3}}\ , \ ] ] or , equivalently , expressing them in terms of the hypergeometric function @xmath142 now we calculate the numerical values of the integrals obtained by replacing @xmath12 with @xmath143 in , and we look for relations with the constants @xmath73 and @xmath74 .
luckily , we find @xmath144 we stress the tremendous simplification obtained by going from the usual description with elliptic integrals to the `` cubic '' version .
for the sake of brevity we define the following constants @xmath145 now we look for relations between the numerical values of the above constants @xmath146 , @xmath147 and @xmath148 and the numerical values of @xmath149 and @xmath150 .
we find that @xmath151 or , alternatively , @xmath152 we note the appearance of the factor @xmath153 , similar to the appearance of @xmath154 in . by using the integer - relation search program
pslq@xcite we have been able to fit the numerical result of with the analytical expression @xmath155 the equalities and are the main result of this paper ; they have been checked up to 2400 digits of precision .
note that constant @xmath147 does not appear in and .
considering lattice perturbation theory , at one loop level one finds these integrals@xcite @xmath156 and @xmath157 some years ago , while visiting the department of physics of parma , york schrder pointed out to the author that , whether 3-loop @xmath0-@xmath1 was known in analytical form , no analytical result was known for the `` simple '' lattice 1-loop tadpole @xmath158 .
puzzled by this fact , and noting that @xmath158 can be reduced to a triple elliptic integral , we have tried to relate the numerical values of @xmath158 and @xmath159 to the new constants @xmath146 , @xmath147 and @xmath148 . working with only 10-digits precision numbers
we have discovered numerically that @xmath160 that is @xmath161 by using again pslq , we have also found that @xmath162 values of @xmath158 and @xmath159 with 400 digits of precision are quoted in ref . .
very kindly , york schrder provided us 16000-digits values . by using these numbers ,
we have checked and up to 2400 digits of precision , the maximum precision of our values of @xmath146 and @xmath148 .
we note also that the integral can be rewritten into the so - called watson integral in 4-dimensions ( see refs . )
@xmath163 from one obtains @xmath164
in this paper we have at last identified beyond any reasonable doubt the analytical constants which appear in the simplest non - trivial 4-loop @xmath0-@xmath1 master integral .
we have also discovered that the _ same _ constants appear in some 4-dimensional lattice integrals .
clearly we still do not know a rigorous proof of these relations , but , once the form of the results is known , we hope that proofs will be easier to find ( see the very recent papers@xcite ) .
while we were completing this paper , kindly david broadhurst sent us a copy of his new paper@xcite . in that paper ,
several elliptic integral evaluations of bessel moments are performed . in particular , a proof of our and is given , as well as of and .
constants analogous to our @xmath79 and @xmath80 are found , @xmath165 and @xmath166 , as well as @xmath167 and @xmath168 .
in addition , several relations between bessel moments are found , and some evaluations of double elliptic integrals are done .
we thanks enrico onofri and giuseppe amoretti for helping us to retrieve the unpublished 3-loop results from a remote off - line computer , on occasion of the talk @xcite .
9 e. remiddi and j. a. m. vermaseren , _ int .
j. mod .
* a15 * ( 2000 ) 725 . j. a. m. vermaseren , _ int . j. mod .
phys . _ * a14 * ( 1999 ) 2037
. j. blumlein and s. kurth , _ phys .
* d60 * ( 1999 ) 014018 .
d. j. broadhurst and d. kreimer , ( 1997 ) 403 .
s. laporta , , 115 ( 2002 ) .
d. h. bailey , j. m. borwein , d. broadhurst , m. l. glasser , _ j. phys . a : math . theor . _
* 41 * ( 2008 ) 205203 ; preprint arxiv:0801.0891v2 [ hep - th ] .
d. broadhurst , preprint arxiv:0801.4813v3 [ hep - th ] .
`` singular values of elliptic integrals in quantum field theory '' talk given by d. broadhurst at the international workshop `` loops and legs in quantum field theory '' , sondershausen , 25 april 2008 .
`` numerical calculation of electron g-2 at 4 loops in qed '' , talk presented by s. laporta at 45th course of the international school of subnuclear physics , erice , sicily , 30 aug 2007 , http://www.ccsem.infn.it/issp2007/docs_nt/laporta_erice.pdf j. m. borwein and p. b. borwein , _ pi and the agm : a study in analytic number theory and computational complexity_(wiley , new york , 1987 ) .
`` reciprocal pslq and the tiny nome of bologna '' , talk given by d. broadhurst at the international workshop `` frontiers in perturbative quantum field theory '' , bielefeld , 14 june 2007 .
s. laporta and e. remiddi , _ nucl .
phys . _ * b704 * , 349 ( 2005 ) .
s. laporta , , 5087 ( 2000 ) .
s. laporta , , 188 ( 2001 ) .
s. capitani , _ phys .
rept . _ * 382 * , 113 ( 2003 ) .
g. s. joyce , _ j. phys . a : math.gen._ * 36 * , 911 ( 2003 ) . m. l. glasser and a. j. guttmann , _ j. phys .
a : math . gen . _ * 27 * , 7011 ( 1994 ) .
h. r. p. ferguson and d. h. bailey , rnr technical report rnr-91 - 032 . | in this paper we continue the work begun in 2002 on the identification of the analytical expressions of feynman integrals which require the evaluation of multiple elliptic integrals .
we rewrite and simplify the analytical expression of the 3-loop self - mass integral with three equal masses and on - shell external momentum .
we collect and analyze a number of results on double and triple elliptic integrals . by using very high - precision numerical fits ,
for the first time we are able to identify a very compact analytical expression for the 4-loop on - shell self - mass integral with 4 equal masses , that is one of the master integrals of the 4-loop electron @xmath0-@xmath1 .
moreover , we fit the analytical expressions of some integrals which appear in lattice perturbation theory , and in particular the 4-dimensional generalized watson integral . |
the collection , @xmath9 , of closed subsets of a topological space @xmath10 , ordered by inclusion , forms a distributive complete lattice often referred to as the _ closed set lattice _ of @xmath10 .
if a lattice @xmath11 is isomorphic to the closed set lattice of some topological space @xmath10 , we say that @xmath10 is a _ topological representation _ of @xmath11 .
it is natural to ask : [ quest : first question ] which lattices have topological representations ?
seymour papert was the first to characterize such lattices as those which are complete , distributive , and have a base consisting of irreducible elements , @xcite .
we can also ask how much of the topological structure of a space is encoded in its closed set lattice . following wolfgang thron , @xcite
, we say that two topological spaces @xmath10 and @xmath12 are _ lattice - equivalent _ if their closed - set lattices are order - isomorphic . clearly , homeomorphic spaces are lattice - equivalent , but the converse fails .
this then leads us to : [ quest : second question ] which classes @xmath2 of topological spaces are such that any two lattice - equivalent spaces @xmath10 and @xmath13 are homeomorphic , i.e. : @xmath14 sober topological spaces are exactly those that can be fully reconstructed from their closed set lattices , @xcite ; therefore the class @xmath15 of sober spaces is a natural choice in answer to question [ quest : second question ] .
furthermore , any topological space @xmath10 is lattice - equivalent to its sobrification @xmath16 and so it follows that @xmath2 can not contain a non - sober space @xmath10 and its sobrification @xmath16 at the same time , in other words , @xmath15 is a _ maximal _
choice for @xmath2 .
the two questions above can also be asked in the context of a particular class of topological spaces .
the ones we have in mind were introduced by dana scott , @xcite , and are known collectively as _
domains_. the characteristic feature of domains is that they carry a partial order and that their topology is completely determined by the order .
more precisely , let @xmath0 be a poset and @xmath17 a subset of @xmath0 .
one says that @xmath17 _ scott open _ , if ( i ) @xmath17 is an upper set , and ( ii ) @xmath17 is inaccessible by directed joins .
the set @xmath18 of all scott opens of @xmath0 forms the _ scott topology _ on @xmath0 , and @xmath19 is called the _ scott space _ of @xmath0 . in what follows , for a poset @xmath0 , we write @xmath1 to always mean the lattice of scott - closed subsets of @xmath0 .
we may now relativize our definitions to the context of scott spaces .
we say that a lattice @xmath11 has a _ scott - topological representation _ if @xmath11 is isomorphic to @xmath1 for some poset @xmath0 , and ask : [ prob : first problem ] which lattices have scott - topological representations ?
although some work has been done on this problem , @xcite , as of now it remains open .
in the special case of _ continuous domains _ a very pleasing answer was given independently by jimmie lawson , @xcite , and rudolf - eberhardt hoffmann , @xcite .
they showed that a lattice @xmath11 has a scott - topological representation @xmath1 for some continuous domain @xmath0 if and only if @xmath11 is completely distributive . in the order - theoretic context
the second question reads as follows : [ prob : second problem ] which classes of posets @xmath2 satisfy the condition @xmath20 a class @xmath2 of posets is said to be _
@xmath4-faithful _ if the above condition holds ( @xcite[p .
2170 , remark 2 ] ) .
the following classes of posets are known to be @xmath4-faithful : 1 .
@xmath21 of complete semilattices , @xcite , containing in particular all complete lattices ; 2 .
@xmath22 of dcpos whose scott topologies are sober , containing in particular @xmath23 , the class of all continuous dcpos .
a _ dcpo - completion _ of a poset @xmath0 is a dcpo @xmath24 together with a scott - continuous mapping @xmath25 such that for any scott - continuous mapping @xmath26 into a dcpo @xmath27 there exists a unique scott - continuous mapping @xmath28 satisfying @xmath29 .
it was shown in @xcite that the dcpo - completion , @xmath30 , of a poset @xmath0 always exists ; furthermore @xmath31 .
hence the class , @xmath32 , of all posets is _ not _ @xmath4-faithful .
indeed any class of posets that is strictly larger than @xmath8 , the class of dcpos , is not @xmath4-faithful .
this means that we can restrict attention entirely to dcpos , in particular , it leads one to ask : ( @xcite , ( * ? ? ? * remark 2 ) ) [ prob : ho - zhao problem ] is @xmath8 @xmath4-faithful ?
this question was dubbed the _ ho - zhao problem _ in @xcite .
the authors of this paper claimed that the two dcpos @xmath33,\leq ) \text { and } \psi = ( \{[0,a ] \mid 0 < a \leq 1\},\subseteq)\ ] ] show that @xmath8 is not @xmath4-faithful .
however , it is easy to see that @xmath34,\leq)$ ] so that @xmath35,\leq)$ ] . on the other hand , @xmath36,\leq)_\bot$ ] which is not isomorphic to @xmath37 .
this failure is only to be expected as both @xmath38 and @xmath39 are continuous dcpos and we already noted that @xmath23 is @xmath4-faithful .
this paper comprises two parts . in section [ sec : positive ]
we present a positive result by introducing the class @xmath40 of _ dominated dcpos _ and showing it to be @xmath4-faithful .
importantly , @xmath40 subsumes all currently known @xmath4-faithful classes listed above . in the second part ( sections [ sec : johnstones - countereg][sec : hoxi - structure ] )
we show that the answer to the ho - zhao problem is negative .
we construct a dcpo @xmath41 which is not dominated , and derive from it a dominated dcpo @xmath42 so that @xmath43 but @xmath44 .
the construction makes use of johnstone s famous example of a dcpo @xmath45 whose scott topology is not sober ( @xcite ) . to familiarize the reader with johnstone s dcpo @xmath45 , we revisit it in section [ sec : johnstones - countereg ] , highlighting its peculiarities .
this prepares us for the counterexample @xmath41 , presented in section [ sec : hoxi - informal ] . for quick accessibility
we keep the description of @xmath41 in section [ sec : hoxi - informal ] informal .
intrepid readers who are keen to pursue the detailed construction of @xmath41 and how it works in answering the ho - zhao problem may then continue their odyssey into section [ sec : hoxi - structure ] . for notions from domain theory
we refer the reader to @xcite .
a nonempty subset @xmath24 of a topological space @xmath46 is called _ irreducible _ if whenever @xmath47 for closed sets @xmath27 and @xmath48 , @xmath49 or @xmath50 follows .
we say that @xmath24 is _ closed irreducible _ if it is closed and irreducible .
the set of all closed irreducible subsets of @xmath10 is denoted by @xmath51 .
two facts about irreducible sets will be important below : * if @xmath24 is irreducible then so is its topological closure . *
the direct image of an irreducible set under a continuous function is again irreducible . in this paper
we are exclusively interested in dcpos and the scott topology on them .
when we say `` ( closed ) irreducible '' in this context we always mean ( closed ) irreducible with respect to the scott topology .
it is a fact that every directed set in a dcpo is irreducible in this sense , while a set that is directed and closed is of course a principal ideal , that is , of the form @xmath52 .
peter johnstone discovered in 1981 , @xcite , that a dcpo may have irreducible sets which are not directed , and closed irreducible sets which are not principal ideals ; we will discuss his famous example in section [ sec : johnstones - countereg ] and someone not familiar with it may want to have a look at it before reading on . given that we may view irreducible sets as a generalisation of directed sets , the following definition suggests itself : a dcpo @xmath53 is called _ strongly complete _ if every irreducible subset of @xmath54 has a supremum . in this case
we also say that @xmath54 is an _
scpo_. a subset of an scpo is called _ strongly closed _ if it closed under the formation of suprema of irreducible subsets .
note that despite this terminology , strongly complete partial orders are a long way from being complete in the sense of lattice theory . the basis of questions [ quest : first question ] and [ quest : second question ] in the introduction is the well - known ( dual ) adjunction between topological spaces and frames , which for our purposes is expressed more appropriately as a ( dual ) adjunction between topological spaces and coframes : @xmath55[r]{$\gamma:\mathbf{top}$}\ar@/^/[r]_\bot & \makebox[2em][l]{$\mathbf{cofrm}^{\mathrm{op}}:\mathrm{spec}$}\ar@/^/[l ] } \ ] ] here @xmath56 is the functor that assigns to a coframe its set of _ irreducible _ elements , that is , those elements @xmath57 for which @xmath58 implies @xmath59 or @xmath60 , topologized by closed sets @xmath61 .
starting with a topological space @xmath46 we obtain the _ sobrification _
@xmath16 of @xmath10 by composing @xmath56 with @xmath4 .
concretely , the points of @xmath16 are given by the closed irreducible subsets of @xmath10 and the topology is given by closed sets @xmath62 where @xmath63 . in order to present problems [ prob : first problem ] and [ prob : second problem ] in a similar fashion ,
it seems natural to replace @xmath64 with the category of _ dcpo spaces _
, i.e. , directed - complete partially ordered sets equipped with the scott topology .
however , while we know that @xmath56 yields topological spaces which are dcpos in their specialisation order , it is _ not the case _ that the topology on @xmath65 equals the scott topology with respect to that order ; all we know is that every @xmath66 is scott - closed . rather than follow a topological route ,
therefore , we reduce the picture entirely to one concerning _ ordered sets_. to this end we restrict the adjunction above to the category @xmath67 of _ monotone convergence spaces _ (
* definition ii-3.12 ) ) on the topological side and compose it with the adjunction between monotone convergence spaces and dcpos @xmath68_\bot & \mathbf{mcs}:\mathrm{so}\ar@/^/[l ] } \ ] ] which assigns to a dcpo its scott space and to a space its set of points with the _
specialisation order_. we obtain a functor from @xmath8 to @xmath69 which assigns to a dcpo its coframe of scott - closed subsets and we re - use the symbol @xmath4 for it rather than writing @xmath70 . in the other direction ,
we assign to a coframe the _ ordered set _ of irreducible elements , where the order is inherited from the coframe . to emphasize the shift in perspective , we denote it with @xmath71 rather than @xmath56 or @xmath72
. it will also prove worthwhile to recall the action of @xmath71 on morphisms : if @xmath73 is a coframe homomorphism , then @xmath74 maps an irreducible element @xmath57 of @xmath75 to @xmath76 .
altogether we obtain the following ( dual ) adjunction : @xmath55[r]{$\gamma:\mathbf{dcpo}$}\ar@/^/[r]_\bot & \makebox[2em][l]{$\mathbf{cofrm}^{\mathrm{op}}:{\mathrm{irr}}$}\ar@/^/[l ] } \ ] ] its unit @xmath77 maps an element @xmath78 of a dcpo @xmath54 to the principal downset @xmath52 , which is always closed and irreducible and hence an element of @xmath79 .
the counit @xmath80 ( as a concrete map between coframes ) sends an element @xmath81 of a coframe @xmath11 to the set @xmath82 which is clearly closed under directed suprema , hence an element of @xmath83 .
combining @xmath71 with @xmath4 yields a monad on @xmath8 which we denote with @xmath84 .
concretely , it assigns to a dcpo @xmath54 the set @xmath85 of closed irreducible subsets ordered by inclusion .
we call this structure the _ order sobrification of @xmath54_. if @xmath86 is a scott - continuous function between dcpos , and @xmath87 is a closed irreducible set , then @xmath88 is the scott closure of the direct image @xmath89 ( which is again irreducible as we noted at the beginning of this section ) .
the monad unit is given by the unit of the adjunction and the multiplication @xmath90 maps a closed irreducible collection .
] of closed irreducible subsets to their union .
we will see in section [ sec : hoxi - informal ] that this order - theoretic monad is _ not _ idempotent ; this is an important difference to its topological counterpart , the sobrification monad .
the concrete description of @xmath91 as union allows us to conclude immediately : [ prop : scpo ] for any dcpo @xmath54 , @xmath85 is strongly complete .
there is more to discover about our adjunction and the union operation : [ prop : union ] let @xmath54 be a dcpo . 1 .
if @xmath27 a closed set of @xmath54 then @xmath92 .
2 . if @xmath93 is irreducible as a subset of @xmath85 then @xmath94 is irreducible as a subset of @xmath54 .
if @xmath95 is scott - closed as a subset of @xmath85 then @xmath96 is scott - closed as a subset of @xmath54 .
if @xmath27 is a closed set of @xmath54 then @xmath97 is strongly closed .
\(1 ) this follows from the general description of the counit .
\(2 ) assume we have @xmath98 with @xmath99 scott - closed subsets of @xmath54 .
then @xmath100 by ( 1 ) and because every member of @xmath93 is irreducible .
furthermore , @xmath97 and @xmath101 are scott - closed subsets of @xmath85 , so by assumption , one of them must cover the irreducible collection @xmath93 , say @xmath97 .
it then follows that @xmath94 is contained in @xmath27 .
\(3 ) this follows for categorical reasons because the union operation is the same as @xmath102 , but a direct proof is instructive for what follows : if @xmath103 is a directed subset of @xmath94 then each @xmath104 must belong to some @xmath105 .
because each @xmath106 is a lower set of @xmath54 we have @xmath107 and because @xmath93 is a lower set of @xmath85 it follows that @xmath108 .
the collection @xmath109 is directed and by assumption its supremum @xmath110{$\scriptstyle\uparrow$}}}}s$ ] belongs to @xmath93 as @xmath93 is scott - closed .
if follows that @xmath111{$\scriptstyle\uparrow$}}}}s$ ] belongs to @xmath94 .
\(4 ) let @xmath93 be an irreducible subset of @xmath97 .
then by ( 2 ) we have that @xmath94 is irreducible . since @xmath94 is contained in the closed set
@xmath27 , so is its closure @xmath24 . we know that @xmath24 is again irreducible and
therefore this is the supremum of @xmath93 and it belongs to @xmath97 .
strong closedness is crucial for the following result : [ lem : strong ] let @xmath54 be a dcpo and @xmath112 .
1 . @xmath113 .
2 . if @xmath95 is strongly closed then equality holds .
the first statement is trivial by the definition of the counit @xmath114 .
for the second , let @xmath24 be a closed irreducible subset of @xmath96 .
we need to show that @xmath24 is an element of @xmath95 .
every element @xmath78 of @xmath24 belongs to some @xmath115 . as we argued in part ( 3 ) of the preceding proof
, it follows that for every @xmath116 , @xmath117 .
we claim that the collection @xmath118 is irreducible as a subset of @xmath85 .
this will finish our proof as we clearly have that @xmath24 is the supremum of @xmath93 and by assumption , @xmath95 is closed under forming suprema of irreducible subsets .
so let @xmath93 be covered by two closed collections @xmath119 , in other words , every @xmath52 belongs to either @xmath120 or @xmath121 .
it follows that each @xmath116 belongs to either @xmath122 or @xmath123 and these two sets are scott - closed by part ( 3 ) of the preceding proposition . because @xmath24 is irreducible , it is already covered by one of the two , and this implies that @xmath93 is covered by either @xmath120 or @xmath121 .
we approach question [ prob : second problem ] via the monad @xmath84 .
starting from the assumption @xmath124 we immediately infer @xmath125 and the question then becomes whether this isomorphism implies @xmath126 .
our counterexample will demonstrate that in general the answer is `` no '' but in this section we will exhibit a new class @xmath40 of _ dominated dcpos _ for which the answer is positive , that is , we will show : @xmath127 before we do so , let us check that invoking the monad does not change the original question , in other words , the assumption @xmath128 is neither stronger nor weaker than @xmath124 : for arbitrary dcpos @xmath54 and @xmath129 , @xmath130 .
the implication from right to left is trivial , so assume we are given an order isomorphism @xmath131 .
the idea for an isomorphism @xmath132 from @xmath133 to @xmath134 is very simple : given @xmath135 we compute @xmath97 , the collection of all closed irreducible sets contained in @xmath27 .
each of these can be replaced with its counterpart in @xmath129 via the given isomorphism @xmath136 . in @xmath129 , then , we simply take the union of the collection @xmath137 . using the maps that are provided to us by the adjunction
, we can express @xmath132 as follows : @xmath138^{\varepsilon_{\gamma(d ) } } & \gamma({\widehat{d}})\ar[r]^{\gamma(i^{-1 } ) } & \gamma({\widehat{e}})\ar[r]^{\bigcup } & \gamma(e ) } \ ] ] for an inverse , we follow the same steps , starting at @xmath134 : @xmath139^{\varepsilon_{\gamma(e ) } } & \gamma({\widehat{e}})\ar[r]^{\gamma(i ) } & \gamma({\widehat{d}})\ar[r]^{\bigcup } & \gamma(d ) } \ ] ] in order to show that these are inverses of each other we use the fact that @xmath97 is _ strongly _ closed which we established in proposition [ prop : union](4 ) . since the concept of strong closure is purely order - theoretic we get that the direct image under @xmath136 is again strongly closed .
this is crucial as it allows us to invoke lemma [ lem : strong](2 ) .
the computation thus reads : @xmath140 the other composition , @xmath141 , simplifies in exactly the same way to the identity on @xmath134 , and since all maps involved are order - preserving , we have shown that the pair @xmath142 constitutes an order isomorphism between @xmath133 and @xmath134 .
our new version of question [ prob : second problem ] requires us to infer @xmath126 from @xmath128 , and the most direct approach is to find a way to recognize _ purely order - theoretically _ inside @xmath85 those elements which correspond closed irreducible subsets of the form @xmath52 with @xmath143 .
as our counterexample @xmath41 to the ho - zhao problem will show , this is not possible for general dcpos .
the purpose of the present section is to exhibit a class of dcpos for which the direct approach works . given @xmath144 we write @xmath145 if there is @xmath116 such that @xmath146 .
we write @xmath147 for the set @xmath148 . clearly , an element @xmath24 of @xmath85 is of the form @xmath52 if and only if @xmath149 holds , but this is not yet useful since the definition of @xmath150 makes explicit reference to the underlying dcpo @xmath54 .
we can , however , record the following useful facts : let @xmath54 be a dcpo and @xmath87 closed and irreducible . 1 .
@xmath151 2 .
@xmath152 is irreducible as a subset of @xmath85 .
the first statement is trivial because @xmath152 contains all principal ideals @xmath52 , @xmath116 .
the proof of the second statement is essentially the same as that of lemma [ lem : strong ] .
let @xmath54 be a strongly complete partial order and @xmath153 .
we write @xmath154 if for all closed irreducible subsets @xmath24 of @xmath54 , @xmath155 implies @xmath156 .
we say that @xmath143 is _ @xmath157-compact _
if @xmath158 , and denote the set of @xmath157-compact elements by @xmath159 .
note that this is an _ intrinsic _ definition of a relation on @xmath54 without reference to any other structure .
it is reminiscent of the way - below relation of domain theory but note that it is defined via _ closed _ irreducible sets .
this choice has the following consequence , which is definitely not true for way - below : [ prop : prec - downset ] let @xmath54 be an scpo and @xmath143 .
the set @xmath160 is scott - closed .
let @xmath161 be a directed set of elements , each of which is @xmath157-below @xmath78 .
we need to show that @xmath111{$\scriptstyle\uparrow$}}}}_{i\in i}a_i$ ] is also @xmath157-below @xmath78 . to this end
let @xmath24 be a closed irreducible subset of @xmath54 with @xmath162 .
since @xmath163 for all @xmath164 we have that every @xmath165 belongs to @xmath24 and because @xmath24 is closed , @xmath111{$\scriptstyle\uparrow$}}}}_{i\in i}a_i\in a$ ] follows .
recall from proposition [ prop : scpo ] that @xmath85 is strongly complete for any dcpo @xmath54 , so on @xmath85 we can consider both @xmath150 and @xmath157 .
we observe : @xmath145 implies @xmath166 for all @xmath167 .
this holds because the supremum of a closed directed collection of closed irreducible subsets is given by union .
our aim now is to give a condition for dcpos @xmath54 which guarantees the reverse implication .
a dcpo @xmath54 is called _ dominated _ if for every closed irreducible subset @xmath24 of @xmath54 , the collection @xmath152 is scott - closed in @xmath85 .
we are ready for the final technical step in our argument : a dcpo @xmath54 is dominated , if and only if @xmath166 implies @xmath145 for all @xmath167 .
if : let @xmath168 .
by assumption we have @xmath169 and in proposition [ prop : prec - downset ] we showed that the latter is always scott - closed .
only if : let @xmath166 .
we know that @xmath152 is irreducible and by assumption closed .
we also know that @xmath151 so it must be the case that @xmath170 .
we are ready to reap the benefits of our hard work .
all of the following are now easy corollaries : for a dominated dcpo @xmath54 , the only @xmath157-compact elements of @xmath85 are the principal ideals @xmath52 , @xmath143 .
also , the unit @xmath77 is an order isomorphism from @xmath54 to @xmath171 .
let @xmath54 and @xmath129 be dominated dcpos . the following are equivalent : 1 .
2 . @xmath173 .
3 . @xmath174 .
the class @xmath40 of dominated dcpos is @xmath4-faithful .
let us now explore the reach of our result and exhibit some better known classes of dcpos which are subsumed by @xmath40 .
[ prop : complete semilattice is dominated ] every scpo is dominated . if @xmath145 for closed irreducible subsets of an scpo , then by definition @xmath146 for some @xmath116 and hence @xmath175 .
if @xmath176 is a directed family in @xmath152 , then @xmath111{$\scriptstyle\uparrow$}}}}_{i\in i}a_i\subseteq{\mathord{\downarrow}}{\mathop{\bigvee{}^{\,\makebox[0pt]{$\scriptstyle\uparrow$}}}}_{i\in i}(\bigvee a_i)$ ] , and the element @xmath111{$\scriptstyle\uparrow$}}}}_{i\in i}(\bigvee a_i)$ ] belongs to @xmath24 because @xmath24 is a scott - closed set .
[ cor : all - dominated ] the following are all dominated : 1 .
complete semilattices and complete lattices ; 2 .
dcpos which are sober in their scott topologies ; 3 .
@xmath85 for any dcpo @xmath54 .
\(1 ) holds because complete semilattices have suprema for all bounded sets , which is all that is needed in the proof of the preceding proposition .
\(2 ) holds because the only closed irreducible subsets of sober dcpos are principal ideals .
\(3 ) holds because we showed in proposition [ prop : scpo ] that @xmath85 is always strongly complete .
recall that a topological space is called _ coherent _ if the intersection of two compact saturated sets is again compact .
it is called _ well - filtered _ if whenever @xmath177 is a filtered collection of compact saturated sets contained in an open @xmath178 , then some @xmath179 is contained in @xmath178 already .
[ lem : ub(a ) is compact saturated ] let @xmath54 be a dcpo which is well - filtered and coherent in its scott topology . then for any @xmath180
, the set @xmath181 of upper bounds of @xmath24 is compact saturated . since @xmath182
, it is saturated . for any finite nonempty @xmath183 ,
@xmath184 is compact since @xmath54 is coherent .
thus @xmath185 forms a filtered family of compact saturated sets whose intersection equals @xmath181 .
if it is covered by a collection of open sets then by well - filteredness of @xmath54 some @xmath184 is covered already , but the latter is compact by coherence so a finite subcollection suffices to cover it .
[ prop : filtered and coherent implies dominated ] every dcpo @xmath54 which is well - filtered and coherent in its scott topology is dominated .
let @xmath176 be a directed family in @xmath152 . by the preceding proposition , the sets @xmath186 are compact saturated and
form a filtered collection .
suppose for the sake of contradiction that @xmath187 .
since @xmath188 is scott open , by the well - filteredness of @xmath54 it follows that there exists @xmath189 such that @xmath190 .
this contradicts the fact that every @xmath191 is bounded by an element of @xmath24 . at the juncture
, the curious reader must be wondering why we have not yet given an example of a dcpo which is not dominated .
now , as we showed , any dcpo whose scott topology is sober is dominated and so a non - dominated dcpo , if it exists , must be non - sober .
peter johnstone , @xcite was the first to give an example of such a dcpo and it is thus natural to wonder whether it is dominated or not . in order to answer this question we explore his example @xmath45 in some detail in the next section .
this will also help us to construct and study our counterexample to the ho - zhao conjecture .
we denote with @xmath192 the _ ordered set _ of natural numbers augmented with a largest element @xmath193 .
when we write @xmath194 we mean the natural numbers as a set , that is , _ discretely ordered_. johnstone s counterexample @xmath45 , depicted in figure [ fig : johnstones - non - sober - dcpo ] , is based on the ordered set @xmath195 , that is , a countable collection of infinite chains .
we call the chain @xmath196 the _ @xmath197-th component _ of @xmath45 . similarly , for a fixed element @xmath198 we call the set @xmath199 the _
@xmath200-th level_. so @xmath201 is the unique element in the @xmath197-th component on level @xmath200 .
we call the elements on the @xmath193-th level _ limit points _ , and refer to all the others as _ finite elements _ ( although they are not `` finite '' in the sense of domain theory ) .
the order on @xmath45 is given by the product order plus the stipulation that the limit point @xmath202 of the @xmath197-th component be above every element of levels @xmath203 to @xmath197 .
it will be convenient later to write this formally as follows : * @xmath204 if @xmath205 ( @xmath206 , @xmath207 ) * @xmath208 if @xmath209 ( @xmath210 ) johnstone s non - sober dcpo @xmath45 . ] as the hasse diagram makes clear , the only non - trivial directed sets of @xmath45 are the chains contained in a component , with supremum the component s limit point .
since every finite element @xmath201 is also below the limit points @xmath211 , where @xmath212 , but not below any other finite element outside its own component , we see that none of them is _ compact _ in the sense of domain theory .
it follows that @xmath45 is highly non - algebraic , and indeed it could nt be algebraic as algebraic dcpos are always sober spaces in their scott topology .
let us take a closer look at the scott topology on @xmath45 .
the two defining conditions of scott - closed sets manifest themselves in the following properties : * closed sets are lower sets : if the set contains a limit point @xmath202 , then it must contain the component @xmath213 and all levels @xmath214 . *
closed sets are closed under the formation of limits : if the set contains infinitely many elements of any one component then it must contain the limit point of that component .
taken together , we obtain the following principle ( which plays a crucial role in our construction as well ) : * if a scott - closed subset of @xmath45 contains infinitely many limit points , then it equals @xmath45 .
this is because such a set contains infinitely many levels by the first property of scott - closed sets , which means that it contains infinitely many elements of every component , and therefore contains the limit points of all components by the second property .
applying the first property again we see that it contains everything .
it is now easy to see that @xmath45 itself is an irreducible closed set : if we cover @xmath45 with two closed subsets then at least one of them must contain infinitely many limit points . by @xmath215
that set then is already all of @xmath45 .
of course , @xmath45 is not the closure of a singleton since there is no largest element so we may conclude , as peter johnstone did in @xcite , that @xmath45 is not sober in its scott topology . in the terminology of the last section
, it also follows that @xmath45 is not strongly complete .
the order sobrification @xmath216 of @xmath45 is depicted in figure [ fig : sobrification - of - johnstone ] .
we will prove in subsection [ subsec : irred in johnstone dcpo ] that the picture is correct , that is , @xmath45 itself is indeed the only closed irreducible subset of @xmath45 which is not a principal downset .
the order sobrification @xmath216 of johnstone s non - sober dcpo . ] for the moment , however , let us check whether @xmath45 is already a counterexample to the ho - zhao conjecture .
unfortunately , this is not the case , the scott topologies of @xmath45 and @xmath216 are not isomorphic .
we see this as follows : the extra point at the top of @xmath217 can not be reached by a directed set ; it is _ compact _ in the sense of domain theory .
also , the image of @xmath45 under @xmath218 forms a scott - closed subset @xmath27 of @xmath216 .
together this means that the largest element of @xmath219 is compact , while the largest element of @xmath220 is not : it is the directed limit of the closed sets @xmath221 , @xmath222 . as an aside
, we may also observe that the order sobrification of @xmath216 would add yet another point , placed between the elements of @xmath27 and the top element @xmath216 which shows that order sobrification is not an idempotent process .
in fact , our construction of @xmath41 addresses exactly this point , by making sure that the new elements that appear in the order sobrification are not compact and do not lead to new scott - closed subsets . the construction of @xmath41 may be viewed as an infinite process .
we begin with @xmath45 and for every finite level @xmath223 of @xmath45 , we add another copy @xmath224 , whose infinite elements are identified with the elements of @xmath223 . no order relation between the _ finite _ elements of two different @xmath224
, @xmath225 is introduced .
now the process is repeated with _
each _ finite level of _ each _
@xmath224 , adding a further @xmath226 many copies of @xmath45 .
we keep going like this _ ad infinitum _ and in this way ensure all elements are limit elements . to make this a bit more precise , let @xmath227 be the set of lists of natural numbers .
( instead of `` list '' we also use the word `` string '' . )
we denote lists by @xmath228 and the empty list by @xmath229 .
the result of adding the element @xmath200 to the front of list @xmath230 we write as @xmath231 , and for decomposing a list @xmath230 into its first element and the rest we use the notation @xmath232 and @xmath233 ( `` head '' and `` tail '' ) .
note that @xmath234 and @xmath235 .
the top of one component of @xmath41 . ]
we use @xmath227 to index the many copies of johnstone s example that make up our dcpo @xmath41 . in a first step ,
we let @xmath236 be the disjoint union of all @xmath237 , @xmath238 .
we label individual elements of @xmath236 with triples @xmath239 in the obvious way .
on @xmath236 we consider the equivalence relation @xmath240 which identifies the finite element @xmath241 of @xmath237 with the infinite element @xmath242 of @xmath243 , for all @xmath244 and @xmath238
. we may now define @xmath41 as the quotient of @xmath236 by @xmath240 .
we let the order on @xmath41 be the quotient order , that is , the smallest preorder such that the quotient map from @xmath236 to @xmath41 is monotone .
we will show in the next section that this is in fact an order and that @xmath41 is a dcpo .
figure [ fig : tree ] gives an impression of the resulting ordered set by showing the top of one component of @xmath41 .
the constituent copies of @xmath45 are still very much present in @xmath41 : for a fixed @xmath238 we may collect all elements of the form @xmath241 , where @xmath206 , @xmath198 , into a substructure of @xmath41 that is order - isomorphic to @xmath45 .
we call it the _ sheet _ @xmath237 .
on @xmath245 we have again the _ finite levels _ @xmath246 consisting of the points @xmath247 for each @xmath222 .
glueing together the copies of @xmath45 in this way does not create any new directed sets other than those ( essentially ) contained in a single sheet @xmath237 . from this
it follows that the characterisation of scott - closed subsets given in section [ sec : johnstones - countereg ] above is still valid .
the principle @xmath215 that we derived from this , however , now has greater reach : since every finite level in any sheet is simultaneously the set of limit points of a subsequent sheet , it holds that a scott - closed subset of @xmath41 can not contain infinitely many elements of _ any _ level without containing all of them .
a close analysis of the situation ( which we carry out in the next section ) then shows that @xmath41 has one closed irreducible subset @xmath248 , which is not a point closure , for _ every _
@xmath238 , to wit , the sheet @xmath237 and everything below it .
the top sheet of the order sobrification of @xmath41 . ] the ( inclusion ) order among the @xmath248 is exactly as indicated in figure [ fig : tree ] which implies in particular that none of them is compact in the sense of domain theory when viewed as elements of the order sobrification @xmath42 .
more precisely , we have that @xmath248 is the limit ( i.e. , the closure of the union ) of the chain @xmath249 .
if we look just at the top sheet @xmath250 in the order sobrification then we obtain the structure displayed in figure [ fig : sober - hoxi ] . the set @xmath27 of point closures ( indicated as open circles in figure [ fig : sober - hoxi ] ) no longer forms a scott - closed subset of @xmath42 .
because a closed set must be downward closed , @xmath251 contains all @xmath252 , @xmath222 , and because it must be closed under taking limits , it contains @xmath253 as well , which means it equals all of @xmath42 .
the same considerations hold on any sheet @xmath254 , @xmath238 , and we see that the scott topology of @xmath42 is the same as the sobrification topology , which in turn is always isomorphic to the topology of the original space , in our case the scott topology of @xmath41 . thus we have shown : [ thm : h = sobh ] the scott topologies of @xmath41 and @xmath42 are isomorphic .
on the other hand , @xmath41 and @xmath42 are clearly not isomorphic as ordered sets : the latter has a largest element whereas the former does not .
thus we have : [ cor : dcpo - not - faithful ] the category @xmath8 is not @xmath4-faithful .
the structure of @xmath41 is sufficiently intricate to warrant a more formal examination of the claims made in the previous section . as explained informally above , we begin with the structure @xmath236 , the disjoint union of infinitely many copies @xmath237 of johnstone s dcpo , indexed by the set @xmath227 of strings of natural numbers . on @xmath236
we define the equivalence relation @xmath240 generated by @xmath255 for all @xmath206 , @xmath256 .
explicit , then we can write the definition also as @xmath257 . ]
we note that each equivalence class with respect to @xmath240 contains precisely the two elements @xmath258 and @xmath259 which appear in the definition , except for the elements of the form @xmath260 ( the maximal elements on sheet @xmath250 ) which are each in a class by themselves . for the arguments that follow below it is important to remember that @xmath240 connects elements from
_ different _ sheets , to wit , @xmath237 and @xmath261 .
the quotient preorder @xmath262 is defined as the transitive closure of @xmath263 . any preorder gives rise to an equivalence relation but in our case this is just @xmath240 again : the equivalence relation @xmath264 derived from the quotient preorder is equal to @xmath240 . by definition ,
two elements @xmath241 and @xmath265 are related by @xmath262 if and only if there is a sequence of elements @xmath266 , @xmath267 such that @xmath268 .
we consider the first two steps of such a sequence ; since an element of the form @xmath269 is maximal in the sheet @xmath270 , the beginning of the sequence has to look as follows : @xmath271 with either @xmath272 and @xmath273 , or @xmath212 and @xmath274 . in the first case
the sequence can continue with @xmath275 but then we are in sheet @xmath276 where the head element of the index @xmath277 is strictly larger than in @xmath231 . in the second case , we are already in a sheet where the index @xmath230 is strictly shorter than the starting index @xmath231 .
this means that two elements @xmath241 and @xmath265 are related by @xmath264 only if they are related by @xmath240 already .
we can now define @xmath41 as the quotient @xmath278 together with the quotient order @xmath279 .
we denote the quotient map with @xmath280 . restricted to any sheet @xmath237 ,
the map @xmath280 is an order - isomorphism .
this follows from the fact that the equivalence relation connects elements from different sheets only .
this observation justifies speaking of the `` sheet @xmath237 '' irrespective of whether we are in @xmath41 or in @xmath236 .
[ p : h - is - a - dcpo ] @xmath41 is a dcpo .
elements of the form @xmath258 are maximal in the sheet @xmath237 , so in order to move from a finite element @xmath241 to a strictly larger one there is no other way than finding a larger element within @xmath237 , or to pass through an infinite element @xmath281 within @xmath237 and then to switch over to the element @xmath282 on sheet @xmath261 ( and to carry on from there ) .
however , the strings in @xmath227 are of finite length , so the process of switching sheet can only be performed finitely often .
it follows that any directed set of @xmath41 contains a cofinal subset entirely contained in a single sheet .
because sheets are isomorphic to johnstone s dcpo @xmath45 by the preceding proposition , the supremum of the directed set can be found there as well .
this proof also shows the following : the quotient map @xmath280 is scott - continuous .
we are ready to show the main technical tool for establishing that @xmath41 is indeed a counterexample to the ho - zhao conjecture : the scott topology on @xmath41 equals the quotient topology derived from the scott topology on @xmath236 and the equivalence relation @xmath240 .
since @xmath280 is scott - continuous , we have that @xmath283 is scott - open in @xmath236 for any scott - open set @xmath178 of @xmath41 .
it follows that @xmath178 belongs to the quotient topology .
conversely , if @xmath178 is a member of the quotient topology , then @xmath283 is scott - open in @xmath236 , which means that the set @xmath284 is scott - open for every @xmath238 .
since we showed in the proof of proposition @xmath285 that directed sets of @xmath41 are essentially contained in a single sheet , it follows that @xmath178 is inaccessible by directed suprema ; in other words , scott - open since @xmath178 is upper .
below we will have to make several ( and sometimes quite subtle ) constructions of open ( and closed ) sets on @xmath41 . by the proposition above we are allowed to do this by exhibiting a scott - open set @xmath286 for each sheet @xmath237 such that @xmath287 if and only if @xmath288 .
the compatibility with @xmath240 together with the principle @xmath215 identified for johnstone s dcpo leads to the following principle for @xmath41 : * if a scott - closed subset of @xmath41 contains infinitely many elements from any level @xmath223 on any sheet @xmath237 , then it contains all of @xmath223 .
this means that the closed ( resp . , open )
sets @xmath289 ( where @xmath290 is a closed set of @xmath41 ) are _ more specialized _ than just being scott - closed ( respectively , scott - open ) .
closed sets of the form @xmath289 are called _ locally closed _ sets ( respectively , _ locally open _ ) .
we will need to know exactly what such sets look like and the next proposition spells this out .
the characterisation follows from the definition of scott - closedness , the order on @xmath45 , and @xmath291 .
see figure [ fig : generic - closed ] for an illustration .
the three parts of a generic locally closed set . ]
[ p : closed ] a locally closed set @xmath292 is either all of @xmath45 or can be uniquely written as the union of the following three parts : @xmath293 \mbox{\emph{ii:}}&\bigcup_{1\leq p\leq h}l_p&\mbox{where } h\geq m_k&\mbox{(the lower rectangle)}\\[1ex ] \mbox{\emph{iii:}}&\bigcup_{j=1}^l{\mathord{\downarrow}}(n_j , q_j)&\mbox{where } n_1<n_2<\ldots < n_l\\ & & \mbox{and } n_j >
h\mbox { for } j=1,\ldots , l&\mbox{(the finite columns ) } \end{array}\ ] ] if any of the numbers @xmath294 , @xmath295 and @xmath296 is equal to zero , then the corresponding part is empty , but because of the way the order is defined on @xmath45 , every component @xmath297 from part i forces the presence of a lower rectangle of at least height @xmath298 .
the purpose of part ii is to account for the possibility that @xmath248 contains more complete rows than is required by the presence of limit points .
we already know that @xmath45 is not sober in its scott topology since @xmath45 is closed and irreducible but not a point closure but we would like to establish that this is the only such subset .
we do this in a fashion that prepares us for the more complicated case of @xmath41 .
[ p : irrs ] the closed irreducible subsets of @xmath45 are @xmath103 itself and the closures of singleton sets .
assume that @xmath24 is a closed proper subset of @xmath45 .
we distinguish two cases : case 1 : @xmath24 contains no limit points .
let @xmath299 be a maximal element of @xmath24 .
we claim that @xmath300 is scott - closed .
indeed , since @xmath24 contains no infinite chains at all we do nt need to worry about closure under limits .
downward closure also holds because there is no order relationship between finite elements from different components of @xmath45 .
we therefore have the decomposition @xmath301 which shows that @xmath24 can only be irreducible if @xmath302 and @xmath303 .
case 2 : @xmath24 contains limit points . by the principle @xmath215 , there are only finitely many limit points in @xmath24 , so let @xmath202 be the one with the highest component number @xmath197 .
for a decomposition we set @xmath304 and @xmath305 , and again we have @xmath306 .
if we can show that @xmath27 is closed then the rest of the argument is as in the previous case .
closure under directed sups is clear because the only non - trivial chains of @xmath45 are those infinite subsets that are contained in a single component , and we have left all components of @xmath24 unchanged except for @xmath213 but there we have removed a cofinal part .
downward closure of @xmath27 follows because @xmath307 consists only of the limit points @xmath308 and none of these were members of @xmath24 to start with .
[ prop : johnstone is dominated ] johnstone s dcpo @xmath45 is dominated .
the only nontrivial directed sets in @xmath216 are the chains @xmath309 and their supremum is @xmath310 .
if a closed irreducible subset @xmath24 of @xmath216 contains such a chain it must also contain its supremum , and because the supremum is a principal downset , it belongs to @xmath152 . by corollary [ cor :
all - dominated ] , @xmath216 is dominated . from the preceding result ,
@xmath45 is dominated .
hence @xmath311 ; otherwise , because both @xmath45 and @xmath216 are dominated , @xmath312 would imply @xmath313 , which we know is false .
this argument then provides an alternate explanation as to why the pair of dcpos @xmath45 and @xmath216 is not a counterexample to the ho - zhao conjecture .
let us now apply a similar technique in order to identify precisely all irreducible subsets of @xmath41 .
our aim is to show the following : [ p : characterisation ] the closed irreducible subsets of @xmath41 are of the following two forms : 1 .
downsets of individual elements of @xmath41 ; 2 .
downsets of levels @xmath314 , where @xmath238 , @xmath222 , or @xmath315 .
note that we could have expressed the second form as `` _ _ downsets of levels @xmath316 _ _ '' without the need to treat @xmath317 as a special case but for the arguments that follow we have found it advantageous to think of the level as consisting of finite elements .
the proof of the proposition is in three parts .
the sets mentioned in the previous proposition are closed and irreducible .
sets of the first kind can also be seen as closures of singleton subsets and such sets are always irreducible .
sets of the second kind are clearly irreducible by principle @xmath291 .
the only concern is whether they are closed . for this
we have to show that a set @xmath318 is locally closed on each sheet @xmath319 .
we use the characterisation provided in proposition [ p : closed ] : if @xmath320 then @xmath321 consists of part ii only , namely , the levels @xmath322 . in case @xmath323 , @xmath324 is all of @xmath325 .
if @xmath326 for some @xmath327 , then the limit points of @xmath328 are identified with the level @xmath329 of @xmath237 by the equivalence relation @xmath240 .
therefore @xmath324 is either empty ( if @xmath330 ) or all of @xmath328 ( if @xmath331 ) . in either case
it is locally closed .
this argument repeats itself if @xmath332 is extended with further numbers . in all other cases ( i.e.
, if @xmath230 is not a suffix of @xmath333 ) , @xmath324 is empty .
[ p : clipping ] if a closed irreducible subset of @xmath41 contains a maximal element @xmath241 but not all of the level @xmath334 , then it equals the downset of @xmath241 .
assume that @xmath335 contains @xmath241 as a maximal element but not every element on the corresponding level @xmath334 .
we may assume that @xmath230 is of minimal length and that on @xmath237 , @xmath200 is maximal with these properties ( i.e. , if @xmath336 then the level @xmath337 contains no element of @xmath24 ) .
as in the proof of proposition [ p : irrs ] , we need to distinguish two cases : case 1 : @xmath338 : first observe that since @xmath200 is maximal in @xmath248 , part i of @xmath248 is empty .
we claim that there is a closed subset @xmath27 which does not contain @xmath241 such that @xmath339 . as in the proof of [ p : irrs ] we can then invoke irreducibility to conclude that @xmath340 and hence @xmath341 .
we define @xmath27 via its locally closed sets @xmath342 , @xmath343 . here
we go : if @xmath320 then we first observe that because @xmath241 is maximal in @xmath24 and @xmath338 , @xmath241 is one of the generators of part iii of @xmath248 .
this means we can let @xmath344 and still have a subset that conforms with the structure exhibited in proposition [ p : closed ] .
if @xmath326 for some @xmath327 , then the limit points of @xmath328 are identified with the level @xmath329 of @xmath237 by the equivalence relation .
since we altered @xmath248 only on level n , we can restrict attention to the case where @xmath345 .
the element @xmath241 appears in @xmath346 as @xmath347 and since we removed the former from @xmath248 we must remove the latter from @xmath348 , where it was one of the generators of part i of @xmath348 .
( note that we have assumed that @xmath24 does not contain the whole level @xmath334 , so @xmath348 does not contain all limit points of @xmath346 .
) however , in order to define @xmath349 it is not sufficient to remove the limit point @xmath347 as the result would not be closed under directed suprema .
however , we can remove cofinally many elements of component @xmath350 , making sure we are not interfering with part ii of @xmath348 .
the result is that the infinite column generated by @xmath347 has been replaced with a finite column generated by @xmath351 where @xmath352 with @xmath295 the height of the `` lower rectangle '' in the language of proposition [ p : closed ] . here . ]
the construction continues in the same vein as we add more numbers to the front of the string @xmath353 .
we always have to convert an infinite column into a finite one in case the generating limit point was removed in the parent sheet .
if @xmath230 is not a suffix of @xmath333 , then we do nt need to manipulate @xmath324 at all in order to obtain @xmath354 . in summary
, we have constructed the individual sets @xmath354 in such a way that they conform to the characterisation of locally closed set in proposition [ p : closed ] , that they are compatible with the equivalence relation @xmath240 , and that all elements in @xmath355 are below @xmath241 .
crucially , @xmath356 does not contain the element @xmath241 we started out with .
case 2 : @xmath357 : in this case we can also assume that @xmath358 because otherwise the maximal element @xmath258 is equivalent to @xmath259 on sheet @xmath261 , contradicting our assumption that the length of @xmath230 is minimal . since we know that @xmath24 does not contain the whole level @xmath359 , it contains only a finite subset of it by principle @xmath291 and @xmath260 is a generator of part i of @xmath360 . as before ,
we convert the infinite column generated by @xmath260 to a finite column generated by some @xmath361 and continue with this process on subsequent sheets . for the third and final step in our proof of [ p : characterisation ] we turn our attention to closed sets that do not contain any `` isolated '' maximal elements , that is , sets that contain the full level @xmath334 whenever they contain the element @xmath241 for some @xmath206 .
levels are indexed by a string ( indicating the sheet ) and a number , but since the equivalence relation identifies @xmath316 with @xmath362 we only need strings of numbers to index them .
we say that level @xmath363 is below @xmath364 if @xmath365 .
this induces an order @xmath366 between strings of numbers generated by the following two rules : * @xmath367 for all @xmath222 , @xmath238 * @xmath368 for all @xmath369 , @xmath238 with respect to @xmath366 , @xmath227 is a tree with root @xmath229 , in other words , @xmath370 is linearly ordered for all @xmath238 .
it is easy to write out the chain above @xmath230 and to check that it is upwards closed , so must be all of @xmath370 : @xmath371 a hasse diagram of @xmath372 would look exactly the same as the one shown in figure [ fig : tree ] ( though there we used a different labelling scheme ) .
[ p : tree ] a closed irreducible subset @xmath24 of @xmath41 can only contain one complete level among its maximal elements .
let @xmath364 be a maximal level contained in @xmath24 .
we claim that @xmath373 is scott - closed .
closure under suprema of directed sets is immediate since we are subtracting a downset .
the issue at hand is downward closure , so assume there is @xmath374 , i.e. , there is @xmath375 with @xmath376 .
the element @xmath78 belongs to some level @xmath377 and since @xmath378 is linearly ordered with respect to @xmath366 , we must have @xmath379 or @xmath380 . in the former case ,
@xmath381 because we assumed @xmath364 to be maximal in @xmath24 . in the latter case
we have @xmath365 by the definition of @xmath366 . in either case
we get a contradiction to the assumption that @xmath382 .
we decompose @xmath24 as @xmath383 and irreducibility now implies that the second component must be empty .
having identified exactly which subsets of @xmath41 are closed irreducible with respect to the scott topology , we can now deliver on our promise , made at the end of section [ sec : positive ] , of giving an example of a non - dominated dcpo .
[ prop : hoxi is not dominated ] the dcpo @xmath41 is not dominated . by the preceding proposition , @xmath384 and for every @xmath385 ,
furthermore , for each @xmath385 , @xmath387 and hence @xmath388 is a directed family in @xmath389 . but @xmath390 equals @xmath41 and it is not the case that @xmath391 .
thus , @xmath41 fails to be dominated . reassured by the above result , we now proceed to the next subsection to establish that the dcpos @xmath41 and @xmath42 form the desired counterexample to the ho - zhao conjecture .
recall that the unit @xmath392 of the order sobrification monad maps @xmath78 to @xmath52 and is scott - continuous .
we will employ this in our proof of theorem [ thm : h = sobh ] : we have the map @xmath393 described in the previous section ; it maps a closed set @xmath48 to @xmath394 . because @xmath395 is scott - continuous we have the map @xmath396 in the opposite direction .
we show that they are inverses of each other . for the first calculation let @xmath48 be a scott - closed subset of @xmath41 .
@xmath397 for the other composition , let @xmath398 be a scott - closed subset of @xmath42 . @xmath399 our proof will be complete if we can show the reverse of the first implication in the calculation above .
for this we use our knowledge about the elements of @xmath42 , that is , the irreducible closed subsets of @xmath41 established in the previous subsection ( proposition [ p : characterisation ] ) . for irreducible subsets of the form @xmath52
the reverse of @xmath400 is trivially true , so assume that @xmath24 is the downset of some level @xmath334 for @xmath238 , @xmath222 , and @xmath401 .
in particular we have @xmath402 for the elements
@xmath241 of @xmath334 .
because all elements of level @xmath197 on sheet @xmath346 are below @xmath403 , we have @xmath404 for all @xmath206 , and hence @xmath405 as the latter is downward closed .
finally , @xmath406 so @xmath407 follows as desired .
the case @xmath315 is similar .
we have shown that the class of dominated dcpos is @xmath4-faithful and also seen that it is quite encompassing ( corollary [ cor : all - dominated ] ) .
one may wonder whether there are other natural classes of dcpos on which @xmath4 is faithful or whether @xmath40 is in some sense `` maximal . '' strictly speaking , the answer to this question is no , since the singleton class @xmath408 is ( trivially ) @xmath4-faithful , yet as we showed not contained in @xmath40 .
what is needed , then , is a proper definition of `` maximal '' before any attempt can be made to establish its veracity .
this research was started when the first author visited the second in the autumn of 2015 , supported by grant sc7 - 1415 - 08 from the _ london mathematical society_. the second author would like to thank the participants of dagstuhl seminar 15441 on _ duality in computer science _ for their comments on an early presentation of the counterexample @xmath41 .
he is also grateful to paul taylor and xiaodong jia for discussing and clarifying the categorical setting that gives rise to the order sobrification monad .
g. gierz , k.h .
hofmann , k. keimel , j.d .
lawson , m.w .
mislove , and d.s .
_ continuous lattices and domains_. number 93 in encyclopedia of mathematics and its applications .
cambridge university press , cambridge , 2003 . | given a poset @xmath0 , the set , @xmath1 , of all scott closed sets ordered by inclusion forms a complete lattice .
a subcategory @xmath2 of @xmath3 ( the category of posets and scott - continuous maps ) is said to be @xmath4-faithful if for any posets @xmath0 and @xmath5 in @xmath2 , @xmath6 implies @xmath7 .
it is known that the category of all continuous dcpos and the category of bounded complete dcpos are @xmath4-faithful , while @xmath3 is not . ho & zhao ( 2009 ) asked whether the category @xmath8 of dcpos is @xmath4-faithful . in this paper , we answer this question in the negative by exhibiting a counterexample . to achieve this
, we introduce a new subcategory of dcpos which is @xmath4-faithful .
this subcategory subsumes all currently known @xmath4-faithful subcategories . with this new concept in mind ,
we construct the desired counterexample which relies heavily on johnstone s famous dcpo which is not sober in its scott topology . |
atomic nuclei are bound by the strong , short - range nucleon - nucleon interactions which impose the average nucleon - nucleon center - to - center distances comparable to the nucleon sizes . as a consequence , nuclear spatial density distributions , say @xmath8 , can be considered well localised in space , described / delimited by abstract auxiliary surfaces , here and below denoted with the symbol @xmath9 .
these surfaces can be defined , for the sake of the following discussion , e.g. by the relation @xmath10 , where @xmath11 can be thought of as the density in the nuclear centre .
it then follows , as one of the consequences of the short - range of the nuclear interactions , that the nucleonic density falls rapidly to zero , and , by the same token , that the nuclear mean - field potentials acting on any single nucleon , rapidly vanish when the nucleons leave the nuclear interior ( i.e. outside of @xmath9 ) .
these arguments allow one to introduce auxiliary geometrical surfaces to facilitate discussing certain _ global _ nuclear properties here : the geometrical forms of the nuclear matter distributions which , at this level of the discussion , can be thought of as a simple classical visualisation of the quantum probability density distributions . however , the problem acquires a much better defined quantum mechanical sense when the symmetry aspects can be discussed by combining the symmetry properties contained within a certain quantum formalism ( usually represented by a quantum hamiltonian approximating the properties of the physical system in question ) and the theory of the group representations . in this article
we will present the calculations which can be seen as an example of the applications of the nuclear microscopic many - body theories .
our hamiltonian of choice will be a microscopic nuclear two - body hamiltonian with the interactions in the gogny form and the problem will be solved within the mean - field theory formalism . generally , the better the mean - field theory approximation , the more precise become the theoretical predictions of the experimental data and the more relevant the examining of the geometrical symmetries associated with nuclear surfaces , @xmath9 , the latter describing not only the symmetries of the spatial distribution of the nuclear _ densities _ but , more importantly , the symmetries of the _ spatial behaviour of the mean - field potential _ , say @xmath12 , the latter having a direct impact on the single particle levels . in particular , the single particle energy spectra respect the degeneracies which are _ in principle _ measurable straightforward mean - field calculations show that originally @xmath13-degenerate orbitals loose their degeneracy very quickly . as an example , in the case of the cited minute deformation , the energy difference between the extreme @xmath14-members of the @xmath15 orbital , @xmath16 approaches 2 mev or more , an estimate which clarifies the next - to - impossibility of discovering the presence of nuclear symmetries through the single - nucleon degeneracy criteria .
] and equal to the dimensions of the irreducible representations of @xmath17 . in the case of the tetrahedral symmetry
these degeneracies are equal either 4 or 2 independently of the size of the related mass multiple moment @xmath18 . to be able to briefly discuss the symmetry - related physics - problems when aiming at the discovery of the point - group symmetries of the compact nuclear systems , let us consider a point - group @xmath17 with the elements @xmath19 under action of which the considered surface @xmath9 happens to be invariant .
the invariance mentioned implies the commutation - relations in the form @xmath20=0,\;\forall\,k , $ ] where the operators @xmath21 form a representation of the group @xmath17 .
when this happens , the group @xmath17 turns into a symmetry group of the hamiltonian and , as a consequence within the mean - field approximation , also into the symmetry group of the considered nuclear system .
this offers the possibility of predicting some measurable consequences since the implied single particle wave functions transform as irreducible representations of the group in question and thus the electromagnetic selection rules can be _ in principle _ derived .
such an approach is , however , not straightforward at all because of the mathematical complications which involve constructions of the determinant - type many - body wave functions and the projection on the irreducible representations of the studied symmetry groups in the appropriately constructed many - body spaces . in the next section we summarise briefly an alternative quantum approach which does not involve the just mentioned mathematical complications .
our strategy followed in the present article will be to profit from the simplicity of these alternative considerations and approach the physical manifestations of the symmetry issue on the semi - quantitative level .
whereas the above arguments suggest a research way straightforward to follow , in practice the corresponding analysis is complicated by the necessity of analysing experimentally the branching ratios of very specific transitions whose big number is usually not known at the present time rotational patterns ... except that in contrast to all other nuclei , the corresponding levels are _ not _ connected by the usually strong @xmath22-transitions and this because the underlying quadruple moments vanish as well as the implied collective reduced transition probabilities . ] . therefore demonstrating the presence of non - trivial geometrical symmetries in nuclei
requires proposing certain specifically designed experiments most likely on the step - by - step basis and thus the study of the discussed symmetry problem becomes a lengthy , multi - step process .
it is then of importance to calculate beforehand the leading features of the nuclear spectra in order to focus the possible experimental investigation on the verification of the predicted structures such as the characteristic sequences of the low - lying energy levels which can be considered characteristic for the examined symmetry ( see below ) and verify the reduced transition probabilities and branching ratios resulting from the symmetry considerations .
analogues of the collectively rotating tetrahedral - symmetric nuclei are known to exist in nature , in the domain of molecular physics , where the tetrahedral symmetry molecules have been studied for a long time @xcite .
the corresponding formalism related to the group - theory aspects uses the fact that the wave - functions , solutions of the tetrahedral - symmetric quantum rotor hamiltonian of the system composed of an even number of fermions , must belong to one of the five irreducible representations of the @xmath23-group .
since these aspects are relatively seldom discussed in the nuclear physics literature let us introduce a few underlying notions which will allow appreciating the simplicity of the symmetry discussions when addressing the rotational properties ; actually we will limit ourselves to using the conclusions of such a formulation known from the other authors ( see below ) .
the so - called ` usual ' case of the nuclear quantum rotor is obtained through quantising the classical tri - axial top energy expression in which case the classical angular momenta , @xmath24 , @xmath25 and @xmath26 of a rotating uniform tri - axial ellipsoid are replaced by the corresponding operators @xmath27 , @xmath28 and @xmath29 , respectively : @xmath30 where the adjustable constants @xmath31 , @xmath32 and @xmath33 are sometimes called ` effective moments of inertia ' . to be able to construct a quantum rotor hamiltonian generalised to higher orders in terms of the angular momentum operators it is convenient to introduce spherical - tensor operator - basis in the form @xmath34 with the help of which any operator constructed out of powers of the angular momentum can be obtained .
the notation above employing the symbol `` @xmath35 '' refers to the usual clebsch - gordan coupling and @xmath36 according to standard notation in terms of spherical tensors ; each of the corresponding tensors appears as an @xmath37-order polynomial in terms of @xmath38 .
for instance for the so - called generalised quantum rotor hamiltonian ( cf . ref .
@xcite ) we find , using somewhat schematic notation , the following general form @xmath39 \big\ } , \label{eqn.03}\ ] ] where the adjustable constants @xmath40 can be constructed in such a way that the operator on the left hand side above is hermitian .
it is then straightforward to verify with the help of the usual clebsch - gordan coupling techniques that the lowest - order tetrahedral - symmetry quantum - rotor hamiltonian takes the form @xmath41}_{\displaystyle\hat{h}_{{\rm t}_d}(3 ) } = \hat{h}_{\rm sph.}(2 ) + \hat{h}_{{\rm t}_d}(3 ) , \label{eqn.04}\ ] ] where @xmath42 and @xmath43 are quadratic and third order operators constructed out of components of @xmath38 , of the second and third ranks , respectively .
the above object is invariant under the symmetry operations of the @xmath44 group since the first term above is a scalar under the rotation group ( and thus invariant under any point - group operations ) whereas the second term is invariant specifically under the symmetry operations of the @xmath44-group .
the parameters of such an hamiltonian can be adjusted to simulate the nuclear rotational properties , cf .
@xcite for certain mathematical specificities of the problem .
the matrix representations of the hamiltonian in ( [ eqn.04 ] ) can be straightforwardly obtained within the basis of the wigner functions @xmath45 and the corresponding solutions constructed numerically together with their classification in terms of the five irreducible representations , the scalar ones , @xmath46 and @xmath47 , as well as @xmath48 and @xmath49 and @xmath50 .
however even without performing such calculations explicitly one may expect that for the doubly - magic tetrahedral - symmetry even - even nuclei the states belonging to the totally - symmetric representation @xmath46 lie low in the energy scale and it becomes of a primary interest to identify the spins and parities of such states on the basis of the group - theoretical approach on the one - hand side and to calculate the energies of such states microscopically using our mean - field approach , on the other .
for the latter - the rest of this article will be devoted to a description of the corresponding results whereas the solution of the former problem is well known ( cf .
@xcite , in particular the discussion summarising the relevant considerations in the appendix a thereof ) .
the @xmath46-representation contains in its low - spin part the sequence of states with the following spin - parity combinations : @xmath51 as it is discussed in more detail below , in all the three doubly - magic tetrahedral - symmetry nuclei , those and only those states are predicted to form the yrast line ( alternatively , to lie close to the yrast states ) whereas all other states appear higher in the energy scale .
this property underlies once again somewhat paradoxically , since at the level of the two - body formulation of the nuclear hamiltonian there is formally no place for the point - group symmetry concepts the importance of the concepts of the geometrical symmetries .
the latter are the fundamental tool in the studies of molecular systems and yet their importance can be seen to extend to the realm of the microscopic _ nuclear _ many - body approaches based on the strong interactions .
our results in the form of the energy - vs .- spin diagrams below , in comparison to the sequence in ( [ eqn.05 ] ) , provide an illustration of the above observation . at the same time
these results suggest that investing in the experimental studies of the nuclei in question is very timely , especially from the point of view of the examining of their octupole properties : the reduced @xmath52 transitions and , if the absolute values turn out to be too difficult to obtain , in terms of the corresponding relative transition probabilities and branching ratios .
the present article focuses on this type of the calculations which may facilitate discovering of one of those exotic symmetries the tetrahedral one , see e.g. ref.@xcite and references therein . with the help of group theory
, one can deduce what kind of spin - parity combinations of the nuclear energy levels should be expected e.g. at and near the yrast line , the mathematical arguments used in the nuclear context fully analogous to the ones known from molecular physics . however , in order to obtain the possibly reliable knowledge of the characteristic features of the spectra and of the transition probabilities , the method used should preferably be based on a realistic microscopic approach , possibly already tested in the past in a different context .
we have recently analysed theoretically the quantum spectra of the tetrahedrally deformed nuclei with doubly - closed tetrahedral - shell configurations @xcite , where both the spherical and inversion symmetries are broken and the efficient quantum number projection method from the most general mean - field states , ref.@xcite , becomes important .
it has been confirmed that the calculated low - lying states compose a collective band built out of states with some specific spin - parity combinations , which is expected by the group theory considerations .
moreover , when the tetrahedral deformation increases , the character of the yrast sequence changes from the approximately linear energy - vs .- spin dependence to the parabolic one , just like in the case of the increasing quadrupole deformation . in our previous study @xcite ,
a schematic separable - type interaction has been used , employing among others a realistic woods - saxon mean - field potential . in this article
we present the results obtained using hartree - fock - bogoliubov ( hfb ) method with the gogny interactions @xcite .
we focus the discussion on the realistic calculations with the tetrahedral - symmetry solutions .
some preliminary results were published in ref.@xcite .
we believe that our microscopic results discussed below can be seen as realistic many - body theory predictions of the structure of the low - lying spectra in the three nuclei illustrated here which coincide structurally with the spectra predicted for the tetrahedral - symmetry rotors .
we also believe that it is very much worthwhile to attempt the experimental identification of the presence of such states in the nuclei which can be reached under the present day experimental conditions - taking into account , as indicated earlier , that the collective @xmath53 and @xmath22 transitions are most likely non - existing ( negligibly small ) under the presence of the exact tetrahedral symmetry and thus the only collective transition will be of the octupole nature , as also discussed in some more detail below .
we have developed a specifically designed computer program to perform the hfb and projection calculations with the gogny effective interaction . in the present article
we follow the techniques of calculations described already in the previous publications , cf .
e.g. refs.@xcite , see also @xcite and references therein . in particular , the hfb equations are solved using the expansion in terms of the harmonic oscillator basis .
one of the characteristic features of the gogny interaction is that the central two - body potential is represented by the sum of terms composed of gaussians , for which the matrix elements can be calculated analytically using the hypergeometric functions @xcite .
the coulomb interaction is accurately approximated by the eight gaussian terms and treated in the same way as the central interaction .
the matrix elements of the zero - range term are calculated with the help of the 61-points gauss - hermite quadrature , after having verified that the desired accuracy has been attained . in order to examine the deformation dependence of the nuclear energy
, we have used quadratic constraints @xcite in terms of the hermitian - symmetrised multipole moment operators , @xmath54 \right\}_i , \quad \mu \ge 0 , \label{qmom}\ ] ] where @xmath55 denotes the spherical harmonics .
if some specific values of constraints are necessary we use the so - called augmented lagrangian method @xcite .
the @xmath56 center - of - mass constraints @xmath57 are always imposed .
we use the deformation parameters defined through @xmath58 so defined deformation parameters correspond , to the lowest order , to the usual parameterization of the nuclear surfaces represented by the spherical harmonic expansion @xmath59 , \label{eqn.08}\ ] ] where @xmath60 is the nuclear radius and the numerical factor @xmath61 , a function of the actual set of the deformation parameters @xmath62 , is introduced to guarantee that the nuclear volume does not depend on the deformation . in this article
we focus on the results with the axial - quadrupole constraint operators @xmath63 ( alternatively @xmath64 ) and the tetrahedral one , the latter given by @xmath65 ( alternatively @xmath66 ) , but we also consider the constraints related to other octupole moments @xmath67 .
traditionally the d1s parameterization @xcite of the gogny interactions has been frequently used in the literature , cf .
however , a few newer forms of the parameterization , such as d1n @xcite and d1 m @xcite , have been proposed recently .
we are going to illustrate the differences between them for the nuclear potential energies within the hfb calculations .
after obtaining the constrained hfb state @xmath68 , we perform the full quantum number projection from it to obtain the the projected wave function , @xmath69 according to the standard definitions and notation .
the amplitude @xmath70 and the energy eigenvalue @xmath71 are obtained by the so - called hill - wheeler _ ansatz _
@xcite , @xmath72 where the kernels are defined by @xmath73 { \cal n}_{k , k'}^{inz(\pm ) } \end{array } \right\ } = \langle \phi | \left\ { \begin{array}{c } \hat{h } \\[2 mm ] 1 \end{array } \right\ } \hat{p}_{kk'}^i \hat{p}^n \hat{p}^z \hat{p}_{\pm } | \phi \rangle .
\label{eqn.11}\ ] ] since the tetrahedral deformation breaks both the axial and inversion symmetries , we will perform the final calculations using the three dimensional angular momentum and parity projections , @xmath74 and @xmath75 , simultaneously - unless - for the comparison purposes , the angular - momentum projection will be switched off .
the particle number projection ( neutrons , @xmath76 , and/or protons , @xmath77 ) is optionally employed if the system is in the superfluid phase .
there are , however , ambiguities for the treatment of the density - dependent term to evaluate the hamiltonian kernel , see e.g. ref.@xcite . we follow ref
. @xcite and adopt the projected density prescription for the number projection @xcite and the conventional mixed ( or transition ) density prescription for the angular momentum and parity projections . concerning some specificities of our projection calculations , for more details c.f .
@xcite , the value of the basis cut - off parameter is chosen at the level of @xmath78 , i.e. , only the canonical basis states whose occupation probabilities are larger than @xmath78 are considered . for solving the hill - wheeler equation ,
the norm cut - off parameter @xmath79 is used , i.e. , the eigenstates of the norm - kernel whose norm eigenvalues are smaller than @xmath79 are excluded . as it has been found out in ref.@xcite ,
it is important to include the time - odd components in the hfb state in order to obtain a reliable estimate of the moment of inertia . for this purpose
we employ a small perturbation term , ref.@xcite , in the form which resembles the cranking hamiltonian , @xmath80 where we use @xmath81 kev and the direction vector @xmath82 along the @xmath83-axis in the present work .
it should be noticed that the results do not depend on the actual value of the frequency parameter @xmath84 , as long as its value is small enough @xcite .
note , however , that one can not use too small a frequency , because then the time - odd components are too small and excluded by the norm cut - off : the value 20 kev is a result of a compromise between these two factors .
it has also been confirmed in ref .
@xcite that for the non - zero tetrahedral deformation the calculated result does not depend on the choice of direction vector @xmath82 .
a spherically - symmetric cartesian harmonic oscillator basis is employed with the oscillator frequencies @xmath85 , the condition required to perform the accurate angular momentum projection .
all the basis states with the oscillator quantum numbers @xmath86 , with @xmath87 are retained .
an example of the results of the gogny - hfb calculations with the axial - quadrupole constraint ( @xmath88 ) is shown in fig.[fig.01 ] in the form of functions of the deformation parameter @xmath64 for @xmath0zr nucleus using the d1s parametrisation of the gogny interaction hamiltonian .
the results using the model space specified by @xmath89 , 12 , and 16 are compared .
it can be seen that the energy differences between the spherical and prolate minima are larger by about 1 mev in the calculation with @xmath90 than in that with @xmath91 .
it is therefore desirable to use as large @xmath92 as possible , but the large model space requires large computational effort , especially for the quantum number projection calculations . in this article
we present the results primarily with @xmath93 ( i.e. 13 shells ) or @xmath94 ( i.e. 9 shells ) as compromise choices , verifying each time that the acceptable stability of the final result with respect to the basis cut off has been achieved .
zr as a function of the deformation parameter @xmath64 for an increasing model space defined by the maximum oscillator shells @xmath958 , 12 , and 16 .
, width=302 ] one of the main purposes of the present article is to study the spectroscopic properties of the tetrahedral ` doubly - magic ' nuclei which happen to carry the ` ideal ' tetrahedral deformation .
for this purpose we choose the tetrahedral doubly magic nuclei of ref.@xcite , @xmath96zr@xmath97 , @xmath98zr@xmath99 , and @xmath100zr@xmath101 .
they contain relatively small numbers of nucleons and are therefore relatively less demanding for the calculations with the full quantum number projections .
potential energy curves for @xmath0zr , calculated with the constrained gogny - hfb approach are shown in fig.[fig.02 ] .
we employ three variants of the parameterization of the gogny interaction hamiltonian : d1s , d1n , and d1 m . here and in the following ,
we choose the origin of the energy scale ( zero energy ) as the spherical hfb energy ( without projection ) for each case .
as can be seen from the figure , all the three variants of the parameterisation give similar potential energy curves .
our prediction is ( cf .
the bottom panel of the figure ) that the ground - state has a pure tetrahedral deformation with @xmath102 and @xmath103 ; its energy lies approximately @xmath104 mev below the energy of spherical - shape configuration .
( top ) and of the deformation parameter @xmath105 ( bottom ) for @xmath0zr ( @xmath106 ) .
the results with three variants of the gogny interactions labelled , d1s , d1n , and d1 m are compared .
the model space is defined by @xmath9512 .
, width=302 ] we found that the single - particle energy gaps with @xmath106 are very large so that the pairing correlations vanish for both the neutrons and the protons .
there are a few hfb or hf - bcs calculations with zero - range skyrme interactions for the tetrahedral deformation in this nucleus @xcite . compared with the result of hf - bcs with the siii interaction in ref .
@xcite , the energy gain of the tetrahedral minimum in our calculations is slightly smaller , while it is larger than that of refs.@xcite . it should be noted that the definition of our deformation parameters is different from those in the cited references ; their deformation parameter corresponding to our @xmath107 is a factor @xmath108 larger for @xmath109 .
moreover the other authors use @xmath110=1.2@xmath111 fm instead of the calculated radius parameter as in eq.([eqn.07 ] ) .
we believe that our definition is more appropriate to calculate the magnitude of the deformation , especially when the specific nuclear structure effects such as skin and/or halo become important for certain nuclei .
we have tested that the obtained tetrahedral - deformed configuration is really the minimum by restarting the hfb iterations from initial states which are perturbed in such a way that they have small extra moments @xmath112 with @xmath113 and @xmath114 and break the tetrahedral symmetry . after performing a sufficiently high number of iterations ,
the obtained minimum deformations coincide with the original tetrahedral - symmetric solution .
the higher - rank @xmath115 non - zero multipole moments of the present hfb minimum with the d1s parametrisation are listed in table [ tab.01 ] .
the lowest order tetrahedral - symmetry multipole - moment is of the rank @xmath116 and is specified by the only non - zero component @xmath18 .
the next order tetrahedral - symmetry moments allowed by the group considerations must have the rank @xmath117 .
the tetrahedral - symmetry at this rank is specified by a simultaneous combination of two moments _ viz .
_ @xmath118 and @xmath119 related by a fixed coefficient , i.e. : @xmath120 ( cf . ref .
as it can be seen from the table , this geometrical symmetry relation is verified up to a six digit accuracy illustrating at the same time certain aspects of precision achieved in our numerical calculations .
recall that the tetrahedral group is a subgroup of the octahedral group , and therefore the tetrahedral - symmetry states are characterised by a simultaneous presence of both the tetrahedral and octahedral deformations .
in other words , the non - zero tetrahedral - symmetry moments at the tetrahedral - symmetry minima are usually accompanied by finite octahedral - symmetry moments @xcite .
let us remind the reader at this point that , after the cited reference , the lowest order octahedral deformation has the rank @xmath121 and is determined by the combined @xmath122 and @xmath123 multipole moments according to the relation @xmath124 the next order octahedral - symmetry moments are of the rank @xmath125 and are determined by a combination of @xmath126 and @xmath127 which are the only allowed non - zero multipole moments of this rank . as it can be seen from table [ tab.01 ] , the tetrahedral - symmetry properties of the minimum configuration are perfectly respected by our hfb calculations represented in terms of the mass multiple moments with the ( at least ) six decimal digit accuracy . [ similar analysis has been performed for the selfconsistent skyrme hfb calculations in ref . @xcite . ]
cccccc @xmath18 & @xmath118 & @xmath119 & @xmath128 & & @xmath129 & @xmath130 & @xmath131 & @xmath131 & & @xmath122 & @xmath132 & @xmath133 & @xmath126 & @xmath134 & @xmath135 @xmath136 & @xmath137 & @xmath137 & @xmath138 & @xmath139 & @xmath139 cccccc @xmath105 & @xmath140 & @xmath141 & @xmath142 & & @xmath143 & @xmath144 & @xmath145 & @xmath145 & & @xmath146 & @xmath147 & @xmath148 & @xmath149 & @xmath150 & @xmath151 @xmath152 & @xmath153 & @xmath154 & @xmath155 & @xmath156 & @xmath157 since the multipole moments of the increasing rank @xmath115 are usually given by quickly increasing numbers , as seen in table [ tab.01 ] , it is rather difficult to imagine the information about shapes of the nuclear surfaces which such moments carry .
because of this it will be convenient to represent the same information using the nuclear deformations introduced by eqs.([eqn.07 ] ) , see also the surrounding text .
this is done with the help of table [ tab.02 ] showing that within at least nine decimal places the numerical results do represent the mathematical correlations expected on the basis of the group theory considerations .
recently , the state - of - the - art quantum number projection and configuration mixing calculations for quadrupole degrees of freedom have been performed in ref.@xcite with the gogny d1s interaction for the same nucleus .
the reflection symmetry has been assumed and no tetrahedral deformation has been taken into account in ref . @xcite . in the latter reference ,
the mean - field states determined by variation after number - projection calculation ( pnp - vap ) are employed to calculate the potential energy curves rather than the hfb states , but the results of the energy curves are similar to ours : the state with no quadrupole deformation @xmath102 is the lowest one in this case since no octupole deformations were considered and the prolate and oblate minima appear as excited states .
let us mention that the form of the curves and in particular the prolate and oblate deformation minima in ref.@xcite may seem different from ours .
however , we have verified that this is because of the different definition of the deformation parameter , i.e. , @xmath110=1.2@xmath111 fm used in ref.@xcite in their definition instead of ours in eq.([eqn.07 ] ) . in particular our calculated r.m.s .
radii for @xmath0zr are slightly larger , the obtained values being @xmath110=1.243@xmath111 fm at the spherical hfb state and @xmath110=1.321@xmath111 fm at the near super - deformation of @xmath64=0.563 .
although the spherical minimum is the lowest in energy with the pnp - vap calculation , the effect of the angular momentum projection is so large that the quadrupole deformed state becomes the ground state according to calculations including the projection and configuration mixing , which seems to be consistent with the experimental data@xcite .
the correlation energy gained by the angular momentum projection is shown to be down by @xmath158 mev for various quadrupole deformed states in ref.@xcite , which is confirmed also in our calculations .
state as functions of tetrahedral deformation for @xmath0zr - compared with the hfb energy curve .
the d1s parametrisation has been employed in this case ; the results of calculations with two model spaces defined by @xmath958 and 12 are presented . , width=302
] it is very instructive to examine the effects of the projection also for the tetrahedral deformation .
the energies of the @xmath159 state have been calculated using the projection technique from the hfb states subject to the @xmath65-constraint as a function of @xmath105 .
the results are shown in fig.[fig.03 ] . in these calculations ,
the numbers of the mesh points in the gaussian quadratures of the projectors are @xmath160 for the euler angles @xmath161 .
there are large single - particle energy gaps at the non - zero tetrahedral deformation for both the neutrons and protons at the particle number @xmath162 .
the tetrahedral - minimum gogny - hfb states in @xmath0zr are thus not superfluid , both for the neutrons and protons , so that the particle - number projection is not necessary in this particular case . in the discussed case
only the results with the d1s interaction are presented because those with the two other variants of the parameterisation are very similar . as it is seen from fig.[fig.03 ] , where the @xmath159 projected energy curves are plotted as functions of @xmath105 , the correlation energy gain relative to the spherical configuration due to the angular momentum projection is very large and amounts to about 10 mev for the tetrahedral deformation @xmath163 , the effects which are much larger than those for the quadrupole deformations . in fig.[fig.03 ]
we have included the two results with @xmath958 and 12 ; it is seen that the correlation energy of the quantum number projection can be accurately estimated with the smaller model space @xmath89 .
zr as in fig.[fig.03 ] but applying the single - constraint condition for each of octupole components @xmath164 ( @xmath165 ) as discussed in the text .
the model space is @xmath958 in this case .
, width=529 ] the experimentally known octupole reduced transition probabilities in the zirconium mass - region are about the largest ever measured in nuclear physics with the value for @xmath2zr reaching as much as @xmath52=57@xmath1664 w.u .
( cf . ref .
@xcite for a comparative table and the original references ) .
the predicted tetrahedral configurations ( non - zero octupole @xmath105 deformations with vanishing quadrupole deformation components ) are very low - lying in the energy scale and in some of these nuclei are predicted to be very likely tetrahedral - deformed in their ground - states , @xcite .
this remark has a couple of consequences whose consideration may be instructive .
firstly , if the considered tetrahedral configurations are indeed ground - state ones the generally difficult measurements of the reduced transition probabilities in question needed for further study of the discussed phenomena may turn out to be slightly easier .
secondly , the large @xmath52 values come in this case principally from ` pure ' octupole contributions as opposed to the situation in which octupole _ and _ quadrupole deformations coexist and it becomes of particular interest to examine the possible effects of all the @xmath67 moments to the final reduced transition probabilities . with this goal in mind it becomes instructive to test the total energy behaviour in function of the other octupole - deformation parameters .
we have calculated the hfb as well as the projected energies in function of these components treated as constraints for @xmath0zr .
the results are shown in fig.[fig.04 ] .
they are obtained by using the smaller model space with @xmath958 ( for the sake of completeness , the result for the tetrahedral deformation @xmath105 , which is the same as fig.[fig.03 ] , is also included ) . in order to be able to extract the sought information in a way which resembles the one - axis projection cuts like those presented so far we have imposed a single active - constraint condition e.g. , for calculating the hfb energy as a function of @xmath167 , the components @xmath168 with @xmath169 are constraint to be zero and then the angular - momentum projection is applied .
for all the four constraints studied in the case of @xmath0zr , the hfb energy gain relative to the spherical configuration is similar , but the largest one corresponds to the tetrahedral deformation .
the latter defines a real local hfb minimum at @xmath170 which turns out to be the absolute one , according to our calculations .
compared with this minimum energy , the lowest energies for other @xmath171 , @xmath167 and @xmath172 constraints are 272 , 318 and 354 kev higher , respectively .
we may associate this feature with the presence of the four - fold degeneracies of the single - nucleon levels at the tetrahedral - symmetric shapes as implied by the presence of the four - dimensional irreducible representations of the tetrahedral group what results in stronger ( tetrahedral ) shell - effects and stronger chances to generate the bigger level spacings .
it may be instructive to discuss a possible impact of the quadrupole deformation developing along the paths corresponding to the @xmath173 constraints discussed above .
as for the tetrahedral deformation , @xmath105 , the quadrupole deformation is always vanishing thus the tetrahedral symmetry is preserved along the corresponding static minimum path . for the @xmath171 constraint ,
only @xmath64 is non - vanishing and gradually increases up to about 0.045 at the largest value of constraint in fig.[fig.04 ] .
similarly , for the @xmath167 constraint , both @xmath64 and @xmath174 gradually increase up to about 0.05 and 0.035 , respectively ; for the @xmath172 constraint , only @xmath64 gradually decreases down to about @xmath175 .
thus , one can safely say that the effect of the static quadrupole deformation is minor . the analogous results for the angular - momentum projected energies are similar among themselves .
the obtained lowering of the energy due to projection is very significant : indeed , the projected nuclear energies are lowered by about 10 mev _ for each _ of the four deformations as illustrated in fig.[fig.04 ] . among the lowest projected energies along the paths corresponding to the four constraints , those for the @xmath171 and @xmath167 constraints coincide within 1 kev and the others are 673 and 901 kev higher , respectively , for the @xmath105 and @xmath172 constraints .
we conclude that in the nucleus studied _ all the octupole deformation components are comparably important _ according to our projected gogny - hfb calculations . the above presented results for the hfb local minima can be compared with those in refs .
@xcite , the latter suggesting as well that the nucleus @xmath0zr is soft against not only the tetrahedral deformation ( @xmath105 ) but also other components of the octupole deformation . to prepare the way for the possible experimental tests of the symmetry properties introduced so far , one of the first issues which comes to one s mind is that of the spin - parity sequence of the energy levels in their consecutive appearance when spin increases , especially at- and close to the yrast line where the excited levels are usually populated most easily . in this context it will be of particular importance to examine such features expected for the pure tetrahedral symmetry as an ideal limiting case when trying to obtain the experimental confirmation of the presence of the discussed symmetry in nuclei . in our previous study @xcite , we have investigated the @xmath48-vs.-@xmath176 pattern of the lowest collective excitations as a function of the tetrahedral deformation . it has been shown that at small deformations the energy vs. spin sequence is approximately linear , whereas at increasing deformation the energy vs. spin relation approaches the usual parabolic form .
qualitatively , according to our calculations , the deformations at which a transition between the two patterns take place get relatively smaller with increasing nuclear mass .
zr in function of spin calculated by the projection method from the hfb minimum state with @xmath170 , left , and from a constrained hfb state with @xmath177 @xmath178 corresponding to a larger deformation , @xmath1790.194 , right .
the gogny d1s interaction is used with the model space defined by @xmath180 . here and in the following ( except fig.[fig.06 ] )
the effect of the perturbative time - odd term discussed in the text , c.f .
eq.([eqn.10 ] ) , with @xmath181 kev has been included .
[ for the comments about certain characteristic features of these diagrams see the text .
] , title="fig:",width=264 ] zr in function of spin calculated by the projection method from the hfb minimum state with @xmath170 , left , and from a constrained hfb state with @xmath177 @xmath178 corresponding to a larger deformation , @xmath1790.194 , right .
the gogny d1s interaction is used with the model space defined by @xmath180 . here and in the following ( except fig.[fig.06 ] ) the effect of the perturbative time - odd term discussed in the text , c.f .
eq.([eqn.10 ] ) , with @xmath181 kev has been included .
[ for the comments about certain characteristic features of these diagrams see the text .
] , title="fig:",width=264 ] the excitation - energy spectrum obtained by the projection technique applied at the hfb minimum is shown in fig.[fig.05 ] , left .
the d1s parameterization has been used here .
the tetrahedral deformation at the minimum is relatively small , @xmath1790.114 , and thus the spectrum has approximately linear energy vs. spin dependence .
for comparison we have calculated the analogous excitation spectrum for a larger tetrahedral deformation assuming @xmath182 @xmath183 which corresponds to @xmath1790.194 , cf .
fig.[fig.05 ] , right . at the latter deformation
, the resulting spectrum has still a ` transitional character ' i.e. a nearly linear energy vs. spin dependence but the fluctuations of the level positions around an imaginary straight line representing the average level positions are weaker .
a couple of observations need to be emphasised .
firstly , the levels shown in fig.[fig.05 ] represent _ all _ the low - energy states which either form the yrast line or lie close to it .
all levels not shown in the figure , in particular the states at spins @xmath184 and 5@xmath185 , lie much higher in the energy scale forming a sequence roughly parallel to the approximately linear sequences shown ; the latter is positioned approximately 6-to-8 mev higher in the energy scale !
this form of the behaviour has been presented and discussed in some detail earlier and the interested reader is referred to @xcite .
let us stress that the levels ` coming down ' in our gogny - hfb calculations ( the ones which are found in the figure ) are exactly those characterised by the spin - parity combinations characteristic of the @xmath46 irreducible representation of the tetrahedral group as listed in eq.([eqn.05 ] ) .
this observation can be seen as the sign of high relevance and usefulness of the notions of point - group representation - theory in that the characteristic features of the spectra can be traced back to the information about the irreducible representations of the groups in question .
secondly , let us stress that our theoretical predictions are limited by the fact that they are obtained within the mean - field theory and are focussed on the tetrahedral symmetry minima .
thus they do not include any information about neither collective rotation effects associated with the higher - lying super - deformed or oblate - shape configurations nor e.g. the low - lying 1-particle 1-hole excitations which are expected to form the characteristic sequences with the spins @xmath186 ; in fact the latter sequences are known experimentally in several nuclei in the zirconium region .
similarly to those latter ones , other particle - hole excitations , such as @xmath187 are expected to lie relatively low in the energy scale and possibly compete with the states to which our calculations are limited . in order to illustrate the influence of the perturbative time - odd term in the hamiltonian which resembles the cranking term , the result of projection from the hfb state without the ` cranking '
are shown in fig.[fig.06 ] .
the slope of the excitation energy with respect to spin is increased by roughly about 15% through this inclusion of the time - odd term .
this effect is not so large compared to the case of the rotational spectra in @xmath188er studied in ref .
@xcite , where the moment of inertia increases by about 50% when the effect of ` cranking ' is taken into account .
, right , but without the perturbative time - odd term ( for the definition of this term see text ) .
comparison shows that although the overall structure of the two spectra is very similar nevertheless the energies of the calculations with including the perturbative time - odd term lie noticeably lower in energy . , width=264 ] in our previous work @xcite , we have employed a rather simple parameterisation of the hamiltonian in which the single - particle potential has been chosen in the form of the universal woods - saxon potential combined with the residual interactions of separable multipole - multipole type . this latter interaction is composed of terms with multipolarity @xmath1892 , 3 , and 4 for the particle - hole channel and with @xmath1890 and 2 for the particle - particle ( pairing ) channel .
the radial form - factors have been chosen of the surface type , i.e. proportional to the derivative of the woods - saxon potential , for the particle - hole channel , and the volume - type , i.e. proportional to the usual @xmath190 factors , for the pairing channel .
the strengths of the interactions for the particle - hole channel have been chosen selecting the so - called self - consistent value as in ref.@xcite , while the strength of the monopole pairing interaction has been determined in such a way as to reproduce the observed even - odd mass difference . , right , here calculated by employing a simplified model with the woods - saxon potential combined with the separable multipole - multipole interactions as in ref.@xcite instead of the gogny interactions .
, width=264 ] the ratio of the monopole and quadrupole interactions has been chosen as 13.5 ( see ref .
@xcite for details ) , what reproduces satisfactorily the moments of inertia for nuclei in the rare earth region .
figure [ fig.07 ] illustrates the spectrum calculated with this briefly recapitulated scheme of ref.@xcite , using the tetrahedral deformation equal that of the projected minimum with the gogny d1s interaction .
as it is seen the calculated spectrum is quantitatively similar .
comparison shows that the rotational excitations of the tetrahedral - symmetry nuclei are not much sensitive to the detailed form of the effective interaction and that the schematic interaction can describe them rather well ; this has been also known for the collective spectra of quadrupole deformed systems @xcite . corresponding to the constraint value @xmath191 @xmath178 .
the dotted line represents a reference @xmath192 parabola passing through the origin and the first excited @xmath193 state .
, width=264 ] it is instructive to examine the evolution of the rotational spectra at increasing tetrahedral deformation with the gogny interactions as compared with the simplified interactions of ref . @xcite . in fig.[fig.08 ] we show an example of the calculated spectrum using the constraint value @xmath194 @xmath183 corresponding to @xmath195 , the deformation which is considerably larger than the deformation associated with the calculated hfb minimum . as it can be seen from the figure , the first @xmath193 energy is considerably lowered at this large deformation , and the states with the same spin with opposite parities are almost degenerate composing a sequence very close to the ` usual ' @xmath196-parabola .
we have verified that the results are very similar to the ones obtained with the simple model of ref .
@xcite as long as the same deformations are used . in this section
we discuss the results for the doubly - magic tetrahedral - symmetry nucleus @xmath2zr whose single - particle energy - gap properties and related diagrams have been discussed e.g. in refs .
@xcite and @xcite .
the potential energy curves as functions of @xmath64 for @xmath2zr calculated using the hfb and employing the three gogny - type interactions are shown in fig.[fig.09 ] , analogous to fig.[fig.02 ] for @xmath0zr ( in obtaining these results the angular - momentum projection has _ not _ been employed ) .
it can be seen that the tetrahedral configuration has the lowest energy except for the d1n parametrisation , in which case the oblate - deformed minimum is slightly lower in energy .
it is worth noticing that the potential energy curves with the three forms of parameterisation are considerably different in this nucleus .
the single - particle energy gap at @xmath197 is not as large as the one at @xmath198 and the energy gain of the tetrahedral deformation is rather small as it is shown in the lower panel in fig.[fig.09 ] .
( top ) and of @xmath105 ( bottom ) for @xmath2zr ( tetrahedral magic gap @xmath197 ) .
the results with the three gogny interactions , d1s , d1n , and d1 m are compared .
the model space is defined by @xmath9512 . , width=302 ] however , the situation changes dramatically when the angular - momentum projection is included .
the energy gain with the tetrahedral deformation becomes as large as 10 mev ; the projected energy curve of the @xmath199 state as a function of @xmath105 is very similar to that in the case of @xmath0zr ( not shown ) .
the slightly smaller single - particle gap at @xmath197 implies the fact that the neutron pairing correlations survive at the tetrahedral minimum which leads to the extra correlation energy by the number projection for neutrons , which is about @xmath200 mev .
the numbers of mesh points employed for the gaussian quadratures when constructing the projectors are @xmath201 for the three euler angles , and @xmath202 for the gauge angle .
the projection calculation from the superconducting mean - field states is much more time - consuming numerically , and this is also why a smaller model space with @xmath90 has been used for this calculation .
the calculated lowest energy levels for @xmath2zr are shown in fig.[fig.10 ] , which are obtained by employing full - projection techniques from the hfb minimum .
the hfb energy curve in fig.[fig.09 ] is very shallow and the minimum deformation at @xmath203 might be considered small . however , the fact that the energy landscape is flat implies the existence of the low energy large amplitude motion with a possibly strong , in the present context the _ dynamical _ effect of the nuclear shape with the tetrahedral symmetry . in @xmath2zr .
the gogny d1s parameterisation is used with the model space defined by @xmath89 . for comments about the characteristic features of these predictions
see the text .
, width=264 ] it is of interest to examine the structure of the energy spectrum with such a relatively small tetrahedral deformation .
let us notice that the lowest energy sequence is still given by the spin - parity combinations of eq.([eqn.05 ] ) , however this time the approximate degeneracies as the ones present in the case of @xmath0zr nucleus discussed earlier do not appear anymore . instead ,
as it can be seen from the right - hand side of the figure , the @xmath204 , @xmath193 , ( @xmath205 , @xmath206 ) , ( @xmath207 , @xmath208 , @xmath209 ) , @xmath210 , states or groups of states are grouped resembling the structure of the zero , one , two , three , @xmath210 , -phonon multiplets , based on the elementary mode of the @xmath193 phonon . although the anharmonicity is non - negligible , this kind of the phonon - like grouping pattern is characteristic of the near spherical tetrahedral spectra , see ref .
@xcite .
the nucleus @xmath2zr is stable and many of its transitions have been known for many years , see e.g. ref.@xcite .
it is believed to be spherical , and its decay pattern contains a few sequences resembling the @xmath211-configurations as e.g. @xmath212 .
indeed sequences of this type are rather abundantly seen in many spherical nuclei .
however , there is no way of saying something about the shape of nuclei which involve the coupling of the @xmath213-nucleon core be it either spherical or , alternatively , ( e.g. tetrahedral ) deformed , with the core - spin @xmath214 and the above mentioned particle - hole type @xmath215-configurations .
any wave function @xmath216 will appear as spherical in the corresponding clebsch - gordan coupling independently of whether its underlying intrinsic moments are compatible with the spherical- or non - spherical ( e.g. tetrahedral ) shape .
in other words : there is no way of concluding about the underlying deformation or sphericity of the @xmath215-structures other then through a dedicated measurement of the related multipole moments in the sequence ; this information being usually very scarce in the literature , some authors simply _ assume _ the sphericity rather then demonstrating it via experiment .
trying to impose any non - trivial constraining conditions which would aim at pinning down the presence of the non - zero charge multipole moments would inevitably need to address the octupole @xmath52-transitions since the ideal tetrahedral - symmetric nucleus carries neither collective @xmath22- nor @xmath53-moments and for this very reason it will appear as spherical in any analysis which takes into considerations only the @xmath56 and 2 multipolarity .
but the @xmath217-properties of the lowest @xmath193 state in @xmath2zr have been measured and its @xmath52 reduced transition probabilities of @xmath218 w.u .
are among the largest ever measured and stronger than those in e.g. @xmath219pb .
moreover , the @xmath197 isotones have comparably strong reduced @xmath52 transition probabilities , see table 1 and the discussion in ref.@xcite , where the references to the experimental information can also be found .
the observed lowest @xmath193 state has an excitation energy 1.897 mev , which is much lower than the value calculated within the present gogny formalism , 3.85 mev , using the projection from the state with @xmath220 .
this suggests that the _ calculated here _
@xmath193 state is not the collective vibrational excitation in @xmath2zr but to draw conclusions about the vibrational nature of the discussed state we would need to extend the formalism e.g. to include the particle - vibration coupling , what goes beyond the scope of this article .
in this section we present the mechanism of competing symmetries , focussing on the tetrahedral symmetry for the case of the neutron - rich , unstable , tetrahedral doubly - magic nucleus @xmath4zr . indeed , according to the previously published results , with the neutron number @xmath223 , one obtains a big single particle gap at a significant @xmath105-deformation , what allows for its qualification as the next tetrahedral - magic number , cf .
ref.@xcite .
the low energy part of the excitation scheme of this nucleus has been studied in ref.@xcite employing the woods - saxon potential together with the simplified , separable two - body schematic interaction like the one used for the case of @xmath0zr ( fig.[fig.07 ] ) .
similarly as for the other isotopes discussed in this article , we compare the potential energy curves as functions of @xmath64 calculated using d1s , d1n and d1 m variants of the parametrisation of the gogny interaction , cf .
fig.[fig.11 ] .
( top ) and @xmath105 ( bottom ) for @xmath4zr ( @xmath223 ) .
the results correspond as before to the three choices of parameterisation of the gogny interaction , d1s , d1n , and d1 m .
the model space is defined by @xmath9512 .
, width=302 ] all the curves are rather similar showing three minima along the @xmath64-projection axis with the oblate ( @xmath224 ) , spherical ( the lowest ) and the prolate , nearly super - deformed quadrupole minimum at @xmath225 . however , the spherical minimum in this representation turns out to be unstable with respect to the tetrahedral deformation , with the energy gain of about 0.7 mev at the static tetrahedral deformation @xmath226 .
in contrast to the nucleus @xmath0zr , where the tetrahedral minimum is significantly lower in energy as compared to the other minima , in the present case all the three minima have rather similar energies , though with the significant separating potential barriers . the effect of full projection for the @xmath199 state is also large for this nucleus and the energy gain relative to the spherical shape is about 10 mev again ( not shown ) , which is very similar to analogous results for @xmath0zr and @xmath2zr .
the numbers of the mesh points employed for the gaussian quadratures of the projectors are same as in the case of @xmath2zr .
the pairing is non - vanishing for neutrons and , as before , we have used the smaller model space ( @xmath958 ) . in @xmath4zr .
the gogny d1s parametrisation is used with the model space defined by @xmath89 .
let us notice that the pattern of the spin - parity combinations of this lowest - lying sequence resembles again the sequence of eq.([eqn.05 ] ) , see discussion following fig.[fig.05 ] .
, width=264 ] finally , the low - lying excitations energies for increasing spin calculated using the full - projection based on the hfb minimum are displayed in fig.[fig.12 ] for @xmath4zr . in this nucleus
the value of the tetrahedral static - equilibrium deformation for the hfb minimum , @xmath227 , is similar to that of @xmath0zr .
consequently , the two energy schemes look rather similar and because of the smallness of the tetrahedral deformation in both cases the energy vs. spin dependence is approximately linear .
however , the @xmath4zr nucleus is considerably heavier than the other zirconium isotopes discussed in this article and the excitation energy of the first @xmath193 state is less than 3 mev ; it is the smallest in sequence of the calculated isotopes .
compared to the calculations with the schematic woods - saxon plus multipole - multipole interactions of ref.@xcite the results with the gogny d1s interaction used here are quantitatively similar , indicating that the collective excitations can be well described by the simple schematic hamiltonians as it is shown in the case of @xmath0zr , see figs.[fig.05 ] and [ fig.07 ] ( except for the vibrational states whose inclusion in the analysis would require a dedicated extension of the present formalism ) .
in the present article we have investigated the low - energy excitation patterns for doubly - magic tetrahedral - symmetry nuclei by employing the simultaneous particle - number , parity and angular - momentum projection method with the gogny interaction .
the pure tetrahedral - symmetry shape is expected to appear as the ground state for all the tetrahedral double magic zr isotopes studied here : @xmath0zr@xmath1 , @xmath2zr@xmath3 , and @xmath4zr@xmath5 .
more precisely , calculations of the nuclear potential energy with the gogny - hfb method show that only for @xmath2zr and with the gogny d1n interaction the lowest energy state is not of tetrahedral symmetry , whereas all the others are .
the energy gain in terms of increasing tetrahedral deformation is rather weak with the gogny - hfb method , depending slightly on the actual parameterisation used , and amounts to only a few hundred kev for @xmath2zr and less than one mev for @xmath0zr and @xmath4zr .
however , the effects of possibly small static tetrahedral deformations must not be ignored .
they are usually accompanied with the flat energy landscapes in the @xmath105-direction implying large amplitude oscillations and generating possibly large dynamical presence of the tetrahedral symmetry in the considered nuclei . among the three zr isotopes
, @xmath0zr has the most stable tetrahedral minimum , i.e. , the energy difference between the tetrahedral and other minima is the largest and exceeds 3 mev .
when the effect of the angular momentum projection is included the situation dramatically changes .
the correlation energy of the projection is very large for the octupole deformation including the tetrahedral one ( @xmath105 ) , typically about 10 mev for zr isotopes , which is much larger than that for the quadrupole deformation .
therefore the energy landscapes of the projected @xmath199 states change very importantly .
in particular , it should be expected that the calculated energy minima associated with the tetrahedral shapes will correspond to larger tetrahedral deformation when the angular momentum projection is applied .
the spectrum for the collective excitations based on the tetrahedral shape is also calculated by the projection method .
the calculated hfb minimum corresponds to rather small tetrahedral deformation , @xmath228 , and the energy vs. spin dependence is nearly linear - what constitutes a possible element of the experimental testing of the discussed properties . in the present work we have focussed on the microscopic calculations of the low - energy spectra in nuclei with the pure tetrahedral deformation ( pure tetrahedral symmetry )
however , more generally , some other deformations leading to lowering of the total nuclear energy would be superimposed , e.g. , triaxial quadrupole deformation and/or the pear shape ( @xmath171 ) deformation . in order to study such a more general case
, one may like to combine the full - projection and configuration mixing .
such an extension is under progress .
last but not least , we could fully confirm the very specific result that the lowest energy sequence of states either forming the yrast line or lying very close to the yrast states has the spin - parity combinations characteristic of the @xmath46 irreducible representation of the tetrahedral point - group , @xmath44 .
all other , ` next - lowest lying states ' , are predicted to form the sequences approximately parallel but appearing significantly ( several mev ) higher in the energy scale .
this grouping of states seems to indicate again the usefulness of the information coming from the theory of representations of the symmetry groups of the mean - field hamiltonians .
however , these predictions apply only to the states which can be interpreted as rotational ones , associated with the tetrahedral - symmetry minima ; such states are expected to compete with e.g. particle - hole non - collective excitations as well as vibrational collective states .
this work is supported in part by grant - in - aid for scientific research no .
25@xmath229949 and no .
22540285 from japan society for the promotion of science , and by the polish - french copin collaboration under project number 04 - 113 .
99 herzberg g , _ molecular spectra and molecular structure _ , + vol .
ii , _ infrared and raman spectra of polyatomic molecules _ , ( 1945 ) , + vol .
iii , _ electronic spectra and electronic structure of polyatomic molecules _ , ( 1966 ) ( d. van nostrand company inc . , new york ) | we report on the results of the calculations of the low energy excitation patterns for three zirconium isotopes , _ viz .
_ @xmath0zr@xmath1 , @xmath2zr@xmath3 and @xmath4zr@xmath5 , reported by other authors to be doubly - magic tetrahedral nuclei ( with tetrahedral magic numbers @xmath6=40 and @xmath7=40 , 56 and 70 ) .
we employ the realistic gogny effective interactions using three variants of their parametrisation and the particle - number , parity and the angular - momentum projection techniques . we confirm quantitatively that the resulting spectra directly follow the pattern expected from the group theory considerations for the tetrahedral symmetric quantum objects .
we also find out that , for all the nuclei studied , the correlation energy obtained after the angular momentum projection is very large for the tetrahedral deformation as well as other octupole deformations .
the lowering of the energies of the resulting configurations is considerable , i.e. by about 10 mev or even more , once again confirming the significance of the angular - momentum projections techniques in the mean - field nuclear structure calculations . |
let @xmath4 be a field and @xmath5 a quiver of dynkin type @xmath6 .
let @xmath7 denote the bounded derived category of finite dimensional @xmath8-modules .
let @xmath9 denote the auslander - reiten translate of @xmath7 and let @xmath10 denote the shift . for @xmath11 the _ @xmath0-cluster category _ associated to @xmath8 is the orbit category @xmath12 this category was introduced in @xcite and has been studied by the authors @xcite , thomas @xcite , wralsen @xcite and zhu @xcite .
it is known that @xmath13 is triangulated @xcite , krull - schmidt and has almost split triangles @xcite .
the @xmath0-cluster category is a generalisation of the cluster category .
the cluster category was introduced in @xcite ( for type @xmath14 ) and @xcite ( general hereditary case ) , and can be regarded as the case @xmath3 of the @xmath0-cluster category .
keller has shown that the @xmath0-cluster category is calabi - yau of dimension @xmath15 @xcite .
we remark that such calabi - yau categories have also been studied in @xcite .
one of the aims of the definition of the cluster category was to model the fomin - zelevinsky cluster algebra @xcite representation - theoretically .
we show that @xmath16 can be realised geometrically in terms of a category of arcs in a punctured polygon with @xmath2 vertices .
this generalises a result of schiffler @xcite , who considered the case @xmath3 .
we remark that the punctured polygon model for the cluster algebra of type @xmath1 appears in work of fomin , schapiro and thurston @xcite as part of a more general set - up , building on @xcite which consider links between cluster algebras and teichmller theory .
also , such a geometric realisation of a cluster category first appeared ( with a construction for type @xmath17 in the case @xmath3 ) in @xcite .
our approach is based on the idea of the _ @xmath0th power _ of a translation quiver introduced in @xcite .
we show that , with a slight modification of the definition for @xmath18 , the auslander - reiten quiver of @xmath16 can be realised as a connected component of the @xmath0th power of the auslander - reiten quiver of @xmath19 . in section [ se : toralexample ] we show that , if this modification is not made , the square of the auslander - reiten quiver of @xmath20 has a connected component whose underlying topological space is a torus .
let @xmath5 be a quiver of underlying dynkin type @xmath1 .
the vertices of @xmath5 are labelled @xmath21 and the arrows are @xmath22 ( @xmath23 ) together with @xmath24 ; see figure [ dnquiver ] . ]
[ dnquiver ] we now recall the auslander - reiten quiver of the cluster category @xmath25 ( see @xcite , @xcite ) . it is a stable translation quiver built from @xmath26 copies of @xmath5 .
we denote it by @xmath27 .
the vertices of @xmath27 are @xmath28 .
the arrows are @xmath29 finally , the translation @xmath9 is given by @xmath30 we use the convention that @xmath31 .
note that the switch described here only occurs for odd @xmath26 .
as an example , we draw the quivers @xmath27 for @xmath32 and @xmath33 ; see figures [ fi : quiverd3 ] and [ fi : quiverd4 ] .
the translation @xmath9 is indicated by dotted lines ( it is directed to the left ) . ] ] we recall the notion of the @xmath0-th power of a translation quiver ( cf .
@xcite ) . if @xmath34 is a translation quiver with translation @xmath9 , then the quiver @xmath35 , the _ @xmath0-th power of @xmath34 _ , is the quiver whose objects are the same as the objects of @xmath34 and whose arrows are the sectional paths ( in @xmath34 ) of length @xmath0 .
a path @xmath36 is said to be _ sectional _ if @xmath37 for @xmath38 ( in the cases where @xmath39 is defined ) , cf .
@xcite .
one of our goals is to realise the auslander - reiten quiver for the @xmath0-cluster category of type @xmath1 in terms of the @xmath0th power of the auslander - reiten quiver of a cluster category of type @xmath40 . to be able to do this ,
we introduce a new class of sectional paths .
let @xmath41 be the translation quiver defined above , with vertices @xmath42 .
we say that a sectional path @xmath36 ( where @xmath43 ) is _ restricted _ if there is no @xmath44 such that @xmath45 for some @xmath46 while @xmath47 or such that @xmath48 for some @xmath46 while @xmath49 .
[ re : restricted ] note that unless @xmath18 , the restricted sectional paths of length @xmath0 in @xmath34 are exactly the sectional paths of length @xmath0 .
we can see this as follows .
firstly , it is clear that any sectional path of length @xmath50 is necessarily restricted .
suppose that @xmath51 .
let @xmath36 be sectional , and suppose that there is an @xmath52 such that @xmath45 and @xmath47
. then @xmath53 . in case
@xmath54 we have @xmath55 , and it follows that the original path is not sectional , a contradiction . similarly , if @xmath56 , we have @xmath57 , and again the original path is not sectional . hence the only sectional paths that are not restricted are the paths of the form @xmath58 and @xmath59 ( @xmath60 ) .
with this new notion we are now ready to introduce a restricted version of the translation quiver @xmath61 .
we define a translation quiver @xmath62 as follows .
the vertices of @xmath62 are the same as the vertices of @xmath63 , i.e. @xmath64 , the arrows are the restricted sectional paths of length @xmath0 in @xmath63 and the translation is @xmath65 . for any @xmath0 ,
the pair @xmath62 is a stable translation quiver .
we firstly note that the unrestricted version , @xmath66 , is a stable translation quiver by @xcite . by remark
[ re : restricted ] , the quiver @xmath62 is the same as @xmath66 if @xmath67 , so the result follows in this case .
now assume that @xmath18 and fix a vertex @xmath68 in @xmath27 . to show that @xmath62 is a translation quiver , we need to show that there is a restricted sectional path of length @xmath69 from @xmath70 to @xmath68 if and only if there is a restricted sectional path of length @xmath69 from @xmath71 to @xmath70 . since the restricted sectional paths in @xmath27 of length @xmath69 starting or ending at @xmath68 are the same as the sectional paths provided @xmath68 is not of the form @xmath72 or @xmath73 , we are reduced to this case . if @xmath74 , the sectional paths of length @xmath69 ending in @xmath68 are @xmath75 and @xmath76 .
the sectional paths of length @xmath69 starting at @xmath77 are @xmath78 and @xmath79 .
the second path only in each case is not restricted , so we see that there is a restricted sectional path of length @xmath69 from @xmath70 to @xmath68 if and only if there is a restricted sectional path of length @xmath69 from @xmath71 to @xmath70 .
the argument in case @xmath80 is similar .
hence @xmath62 is a translation quiver . by construction ,
no vertex is projective and @xmath65 is defined on all vertices ( since @xmath9 is ) .
therefore , @xmath62 is stable .
let @xmath81 , with @xmath82 .
we recall that @xcite the derived category of a quiver of dynkin type @xmath1 has vertices @xmath83 and arrows given by @xmath84 and @xmath85 for @xmath86 , and @xmath87 and @xmath88 , where @xmath89 is arbitrary .
we also have that @xmath90 while @xmath91 , otherwise .
let @xmath92 be the quiver with vertices @xmath93 the arrows are given by @xmath84 and @xmath85 for @xmath86 , and @xmath87 and @xmath88 , where @xmath94 is arbitrary and the addition is modulo @xmath2 .
we also define @xmath95 it follows from the construction of @xmath16 and the above description of the derived category that @xmath96 is the auslander - reiten quiver of @xmath16 ( and , in particular , is a stable translation quiver ) .
the vertices of the auslander - reiten quiver @xmath97 of @xmath19 are @xmath98 the arrows are given by @xmath84 and @xmath85 for @xmath99 , and @xmath87 and @xmath88 , where @xmath44 is arbitrary and the addition is modulo @xmath2 .
we also have @xmath100 we define a map @xmath101 from @xmath102 to @xmath103 as follows .
we set @xmath104 whenever @xmath105 or @xmath106 , @xmath0 is odd and @xmath26 is even . otherwise , we have @xmath107 or @xmath108 and we set @xmath109 here we restrict @xmath44 to lie in the set @xmath110 .
we use the usual convention that , for a real number @xmath68 , @xmath111 denotes the largest integer @xmath4 such that @xmath112 .
let @xmath113 .
here we adopt the convention that @xmath114 . with @xmath101 defined as above , we have that @xmath115 .
first we note that it is clear from the definition of @xmath101 that @xmath116 .
let @xmath117 and suppose that @xmath118 .
suppose first that @xmath119 .
write @xmath120 for @xmath121 .
we have @xmath122 in @xmath123 , and it follows that @xmath124 so @xmath125 . if @xmath126 or @xmath108 then @xmath127 and we are done .
let @xmath34 denote the full subquiver of @xmath128 induced by @xmath129 , and let @xmath130 be the ( surjective ) map obtained by restricting the codomain of @xmath101 to @xmath129 .
we will show that @xmath130 is an isomorphism from @xmath92 to @xmath34 and that @xmath34 is a connected component of @xmath128 .
[ closed ] let @xmath131 and suppose that @xmath132 is a restricted sectional path of length @xmath0 in @xmath97
. then @xmath133 .
if @xmath119 we can argue as in @xcite to see that @xmath70 is either @xmath134 , @xmath135 or @xmath136 , where the last possibility only arises if @xmath137 . if @xmath126 or @xmath108 a similar argument shows that @xmath70 must be @xmath138 . if @xmath18 we have the sectional paths @xmath139 and @xmath140 , but , by definition , these are not restricted sectional paths .
the map @xmath141 is an isomorphism of quivers .
since @xmath142 and @xmath130 is surjective , it follows that @xmath130 is bijective .
the arrows in @xmath92 are of the form @xmath84 , @xmath143 , @xmath87 or @xmath88 .
the arrows in @xmath144 are of the form @xmath145 , @xmath146 , @xmath147 or @xmath148 ( see the proof of lemma [ closed ] ) .
it follows that @xmath130 is an isomorphism of quivers .
the map @xmath141 is an isomorphism of translation quivers .
its image , @xmath34 , is a connected component of @xmath128 . by lemma
[ closed ] , both statements of the theorem will follow if we can show that , for all @xmath149 , @xmath150 , since this will also imply that the image of @xmath130 is closed under @xmath65 .
we firstly note that if @xmath151 then @xmath152 while @xmath153 . since @xmath154 and @xmath155 , the result holds .
so we are left with the case where @xmath107 or @xmath108 .
we break this down into cases , considering first the case where @xmath107 .
* case ( a ) * : @xmath0 odd and @xmath26 even .
+ in this case we have that @xmath2 is even , and @xmath156 is even , so for any @xmath157 , @xmath158 while @xmath159 . since @xmath160 and @xmath161 , we are done . * case ( b ) * : @xmath0 is even .
+ in this case we have that @xmath2 is odd , so for @xmath162 or @xmath108 , we have : @xmath163 since @xmath0 is even , @xmath156 is even , so @xmath159 for all @xmath44 .
\(i ) suppose first that @xmath164 .
then @xmath165 .
it follows that either @xmath160 and @xmath161 or @xmath166 and @xmath167 . in either case
we see that @xmath168 .
\(ii ) suppose next that @xmath169 .
then @xmath170 .
it follows that either @xmath166 and @xmath161 or @xmath160 and @xmath167 . in either case
we see that @xmath168 .
\(iii ) finally , suppose that @xmath171 .
then @xmath172 , and @xmath173 is even while @xmath174 is odd ( using here the fact that @xmath175 ) .
it follows that @xmath176 and @xmath161 and thus that @xmath168 .
* case ( c ) * : @xmath177 both odd .
+ in this case we have that @xmath2 is odd , so for @xmath162 or @xmath108 , we have : @xmath178 since @xmath26 and @xmath0 are both odd , @xmath156 is odd , so @xmath179 ( i ) suppose first that @xmath164 . then @xmath165 .
it follows that either @xmath160 and @xmath161 or @xmath166 and @xmath167 . in either case
we see that @xmath168 .
\(ii ) suppose next that @xmath169 .
then @xmath170 .
it follows that either @xmath166 and @xmath161 or @xmath160 and @xmath167 . in either case
we see that @xmath168 .
\(iii ) finally , suppose that @xmath171 .
then @xmath172 , and , as in case ( b)(i ) , @xmath173 is even and @xmath180 , which means in this case that it is also even .
it follows that @xmath181 and @xmath161 and thus that @xmath168 .
we therefore have : [ thm : d - component ] the translation quiver @xmath92 can be realised as a connected component of the restricted @xmath0th power of the translation quiver @xmath97 . since @xmath16 is equivalent to the additive hull of the mesh category of @xmath92 , we obtain the following corollary .
the @xmath0-cluster category of type @xmath1 can be realised as the additive hull of the mesh category of a connected component of the restricted @xmath0th power of the auslander - reiten quiver of the cluster category of type @xmath40 .
for @xmath51 it is enough to take the usual @xmath0th power .
we give an example of the theorem in the case where @xmath33 and @xmath18 .
the theorem tells us that @xmath182 is isomorphic to a connected component of @xmath183 . in figure
[ fi : d7usual ] we show the translation quiver @xmath184 with the vertices of @xmath185 shown in circles . in figure
[ fi : d4ind7 ] we isolate the connected component @xmath34 of @xmath183 induced by @xmath129 , and in figure [ fi : d4usual ] we indicate the translation quiver @xmath182 with the usual labelling of its vertices . ] of the translation quiver @xmath184 ] ]
in this section , we give a geometric realisation of the @xmath0-cluster category of type @xmath1 .
to do so , we use certain @xmath0-arcs in a punctured @xmath2-gon .
thus we are generalising the notion of tagged edges of schiffler @xcite for the cluster category of type @xmath1 and the notion of @xmath0-diagonals of our work on @xmath0-cluster categories of type @xmath17 @xcite .
let @xmath187 be a punctured @xmath188-gon in the plane ( later we shall specialise to the case where @xmath189 ) .
we label the vertices of @xmath187 clockwise . for @xmath190 , we denote by @xmath191 the boundary path @xmath192 , going clockwise around the boundary ( taking the vertices @xmath193 ) . if @xmath194 , we let @xmath195 be the whole boundary path @xmath196 and @xmath197 denote the trivial path at @xmath44 consisting of the vertex @xmath44 .
the _ length _ @xmath198 of the boundary path @xmath191 is the number of vertices it runs through . here , we count both the starting and end point unless @xmath199 .
in particular , @xmath200 and @xmath201 .
as an example of a boundary path of length @xmath202 , we have indicated @xmath203 inside a punctured @xmath204-gon in figure [ fig:7gon ] . and arcs @xmath205 and @xmath206 for @xmath207 , and @xmath208 not the clockwise neighbour of @xmath44 , an _ arc _
@xmath209 is a line from @xmath44 to @xmath208 that is homotopic to the boundary path @xmath191 .
if @xmath208 is the clockwise neighbour of @xmath44 , there is no arc clockwise from @xmath44 to @xmath208 other than the boundary path @xmath191 .
if @xmath194 , we always tag the arc by @xmath210 or @xmath211 , as in schiffler s work @xcite .
such arcs are denoted by @xmath212 and @xmath213 .
we will occasionally write @xmath214 to denote an arbitrary arc and call it a _ tagged arc_. in that case , if @xmath207 , then @xmath214 will only stand for the arc @xmath209 . as an example , the arcs @xmath205 and @xmath215 of a punctured @xmath204-gon are pictured in figure [ fig:7gon ] .
addition of subscripts for the @xmath209 will always be modulo @xmath188 . in what follows , we will use a slightly generalised version of a polygon .
we will allow arcs @xmath214 and sides @xmath216 of the polygon @xmath187 as sides of a polygon .
we will say that such a ( generalised ) polygon is _ degenerate _ if it has more sides than vertices .
note that such polygons may or may not contain the puncture . in the remainder
, we will in particular be interested in the following types of generalised polygons and generalised degenerate polygons obtained from the regular @xmath188-gon @xmath187 .
* type ( i ) : * a combination of an arc @xmath209 with the boundary path @xmath191 , or of an arc @xmath209 with the boundary path @xmath217 , @xmath207 , where in the former case , @xmath218 and in the latter case , @xmath219 .
such a polygon has @xmath198 vertices ( respectively , @xmath220 vertices ) .
* type ( ii ) : * a combination of two arcs @xmath209 , @xmath221 with the boundary path @xmath222 , or of @xmath221 , @xmath223 with @xmath191 , where @xmath44 , @xmath208 , @xmath4 are all distinct and lie in clockwise order on @xmath187 .
furthermore , in the former case , @xmath224 and in the latter case , @xmath225 . such a polygon has @xmath226 vertices ( respectively , @xmath227 vertices ) . * type ( iii ) : * a combination of an arc @xmath228 with the boundary path @xmath195 , or with @xmath197 . in the first case ,
the polygon is has @xmath229 sides and @xmath188 vertices . in the latter case
, the polygon has one side and one vertex . *
type ( iv ) : * a combination of an arc @xmath228 with an arc @xmath209 and the boundary path @xmath217 , or a combination of @xmath228 with a boundary path @xmath191 and the arc @xmath230 ( where we always have @xmath207 ) . in the former case , @xmath219 and in the latter case , @xmath231 .
such a polygon has @xmath232 sides and @xmath220 vertices ( respectively , @xmath227 sides and @xmath198 vertices ) .
note that we can view type ( iii ) as the limit @xmath233 of type ( i ) and type ( iv ) as the limit @xmath234 or @xmath235 of type ( ii ) .
we show each of these four types in figure [ fig : polygons ] .
[ def : m - arc ] let @xmath214 be an arc of @xmath187 .
if @xmath207 , we say that @xmath209 is an _
@xmath0-arc _ if the following hold : \(i ) @xmath209 and @xmath191 form a @xmath236-gon for some @xmath4 , \(ii ) @xmath209 and @xmath217 form a @xmath237-gon for some @xmath238 .
if @xmath194 , @xmath228 is a _ tagged @xmath0-arc _ if @xmath228 and @xmath195 form a degenerate @xmath236-gon for some @xmath4 . the parts ( i ) and ( ii ) in the definition of an @xmath0-arc also apply to the case @xmath194 if we use the boundary paths @xmath195 and @xmath197 .
namely , @xmath228 is a tagged @xmath0-arc if @xmath228 and @xmath195 form a degenerate @xmath236-gon for some @xmath4 and if @xmath228 and @xmath197 form a degenerate @xmath50-gon .
let @xmath187 be a punctured @xmath204-gon and @xmath18 .
the arc @xmath205 is a @xmath69-arc ( cf .
figure [ fig:7gon ] ) , since the arc @xmath205 together with the boundary path @xmath203 forms a @xmath202-gon ( i.e. @xmath239 ) whereas @xmath205 and @xmath240 form a @xmath241-gon ( i.e. @xmath242 ) .
each of the arcs @xmath243 forms an @xmath244-gon together with @xmath245 , and thus also is a @xmath69-arc .
we now define @xmath0-moves generalising the @xmath0-rotation for type @xmath17 of @xcite and the elementary moves for type @xmath1 of @xcite .
let @xmath187 be a punctured @xmath188-gon .
@xmath0-move _ arises when there are two arcs in @xmath187 with a common end - point such that the two arcs and a part of the boundary bound an unpunctured @xmath246-gon , possibly degenerate .
if the angle from the first arc to the second at the common end - point is negative ( i.e. clockwise ) , then we say that there is an @xmath0-move taking the first arc to the second .
more precisely , it is a move of one of the following forms : \(i ) @xmath247 if @xmath209 , @xmath222 and @xmath221 form an @xmath246-gon , @xmath248 .
\(ii ) @xmath249 if @xmath209 , @xmath250 and @xmath251 form an @xmath246-gon , @xmath252 .
\(iii ) @xmath253 if @xmath209 , @xmath228 and @xmath217 form a degenerate @xmath246-gon , @xmath254 .
\(iv ) @xmath255 if @xmath228 , @xmath230 and @xmath191 form a degenerate @xmath246-gon ; @xmath256 . in figure
[ fig : m - moves ] , we illustrate the four types of @xmath0-moves inside a heptagon , i.e. @xmath33 , @xmath18 .
-moves inside a heptagon ] our goal is to model the @xmath0-cluster category @xmath257 geometrically . to do so , we will from now on assume that @xmath189 , so the polygon @xmath187 has @xmath2 vertices .
let @xmath187 be a punctured polygon with @xmath2 vertices and let @xmath207 .
then the two conditions of definition [ def : m - arc ] are equivalent , i.e. @xmath209 and @xmath191 form a @xmath236-gon for some @xmath4 if and only if @xmath209 and @xmath217 form an @xmath237-gon for some @xmath238 .
we are now ready to define a translation quiver using the punctured polygon , @xmath187 .
let @xmath258 be the quiver whose vertices are the tagged @xmath0-arcs of @xmath187 and whose arrows are given by @xmath0-moves .
let @xmath259 be the map sending an arc @xmath214 to @xmath260 if @xmath207 or @xmath0 is even .
if @xmath194 and @xmath0 is odd , we set @xmath261 . in other words
, if @xmath207 or @xmath0 is even , @xmath259 rotates a tagged arc anti - clockwise around the center . in case @xmath194 and
@xmath0 is odd , @xmath259 rotates the tagged arc anti - clockwise around the center and changes its tag .
figure [ fig : geom-4 - 2 ] shows the example @xmath262 ( we will see shortly , cf .
theorem [ thm : odot - transl ] , that @xmath263 is a translation quiver ) . ]
[ lm : sect - paths ] the sectional paths of length @xmath0 in @xmath97 are of the form \(i ) @xmath264
if @xmath265 , \(ii ) @xmath266 and @xmath267 if @xmath268 , \(iii ) @xmath269 if @xmath270 and @xmath271 exists , \(iv ) @xmath272 and @xmath273 if @xmath274 exists . for ( i ) and ( iii ) we can argue as in @xcite , using the vertices @xmath275 of @xmath97 instead .
the other cases follow with the same argument , using the assumption that the paths are restricted , i.e. excluding the sectional paths @xmath276 and @xmath277 .
[ thm : odot - transl ] the quiver @xmath278 is a translation quiver isomorphic to to the auslander - reiten quiver of @xmath279 . it is enough to show that @xmath278 is isomorphic to the image @xmath34 of the map @xmath130 from section [ se : mthpower ] and that , under the isomorphism , the map @xmath259 on @xmath278 corresponds to @xmath65 on @xmath34 .
recall that the vertices of @xmath34 are @xmath280 ( using the convention that @xmath0 divides @xmath108 ) , recalling that @xmath189 .
in other words , @xmath129 is the subset in @xmath281 of the vertices whose second coordinate is divisible by @xmath0 .
we will now define a map @xmath282 , where @xmath283 denotes the set of vertices of @xmath278 .
note that @xmath0-arcs in @xmath283 going through two distinct vertices are always of form @xmath284 .
on such @xmath0-arcs , @xmath285 is defined as follows : @xmath286 where @xmath287 modulo @xmath188 and @xmath288 . on arcs
@xmath228 , @xmath285 is defined as follows .
@xmath289 ( where @xmath287 modulo @xmath188 ) . to see that @xmath285 is a bijection , we divide @xmath283 up into @xmath26 types of arcs .
let @xmath290 be the set of arcs of the form @xmath291 ( @xmath292 ) , i.e. the arcs homotopic to a boundary path of length @xmath246 .
then @xmath285 sends each element of @xmath290 to a vertex of the top row of @xmath129 , @xmath293 ( where @xmath294 ) .
it is straightforward to check that @xmath285 induces a bijection from the set @xmath290 to the top row of @xmath129 .
more generally , for @xmath288 , let @xmath295 be the set of arcs of the form @xmath296 , i.e. the set of arcs homotopic to a boundary path of length @xmath236 .
since @xmath297 , @xmath285 sends the arcs in @xmath295 to the @xmath4th row ( from the top ) of @xmath129 ( @xmath298 ) .
clearly , this is also a bijection . furthermore , the arcs @xmath228 are sent to the two last rows of @xmath129 , also bijectively .
thus , we have that @xmath285 is a bijection from @xmath283 to @xmath129 .
next , we observe that the arrows given by the @xmath0-moves are the same as the arrows in @xmath34 : for arcs in @xmath295 with @xmath299 , an @xmath0-move sends @xmath284 to @xmath300 or @xmath284 to @xmath301 whereas a restricted sectional path of length @xmath0 sends @xmath302 to @xmath303 ( type ( i ) in lemma [ lm : sect - paths ] ) or to @xmath304 ( type ( iii ) in lemma [ lm : sect - paths ] ) . for arcs in @xmath305
, an @xmath0-move sends @xmath306 to @xmath212 , to @xmath213 or to @xmath307 whereas a restricted sectional path of length @xmath0 sends @xmath308 to @xmath309 , to @xmath310 or to @xmath311 ( types ( ii ) and ( iii ) in lemma [ lm : sect - paths ] ) .
finally , arcs @xmath228 are sent to @xmath312 by @xmath0-moves , and restricted sectional paths of length @xmath0
send @xmath309 to @xmath313 and @xmath310 to @xmath313 ( type ( iv ) in lemma [ lm : sect - paths ] ) .
furthermore , the translation maps correspond : on @xmath295 ( with @xmath314 ) @xmath315 ( subscripts taken @xmath193 ) and on the @xmath316 first rows from the top , @xmath317 ( first entries taken @xmath193 ) . if @xmath56 then @xmath318 while @xmath319 and @xmath320 .
if @xmath54 then @xmath321 while @xmath322 it follows that @xmath323 for all @xmath44 .
a similar argument applies to the tagged arcs @xmath213 . since @xmath324
, we see that @xmath325 for all @xmath44 .
we have seen that @xmath285 induces an isomorphism of quivers and @xmath326 for all arcs @xmath214 .
it follows that @xmath278 is a translation quiver and that @xmath285 is an isomorphism of translation quivers .
in this section we give an example of a toral translation quiver arising from the cluster category @xmath327 of type @xmath328 .
the auslander - reiten quiver of @xmath327 , @xmath329 , is shown in figure [ fi : quiverd4 ]
. a connected component of its ( unrestricted ) square , @xmath330 is shown in figure [ fi : toralquiver ] .
the underlying topological space @xmath331 ( in the sense of gabriel and riedtmann ; see ( * ? ? ?
* p51 ) ) is a torus . ]
we have seen in theorem [ thm : d - component ] that the auslander - reiten quiver of the @xmath0-cluster category of type @xmath1 is a connected component of the restricted @xmath0-th power @xmath128 . in this section ,
we describe the other components arising in the restricted @xmath0-th power of the translation quiver @xmath333 .
we consider the following subset of the vertices of the quiver @xmath128 : @xmath336 such a set @xmath337 is a union of rows in the quiver @xmath128 .
we show that for each @xmath338 , the translation quiver generated by @xmath337 ( i.e. the full subquiver induced by @xmath337 , together with @xmath65 ) is a connected component of @xmath128 .
this is done by first showing that @xmath337 is closed under @xmath65 and under taking restricted sectional paths of length @xmath0 .
this tells us that @xmath337 is a union of connected components of @xmath128 .
then we show that @xmath337 is connected , hence is a single component .
\2 ) the set @xmath337 is closed under restricted sectional paths of length @xmath0 : we have seen that these paths are of the form @xmath339 or @xmath340 , cf . lemma [ lm : sect - paths ]
. in particular , the new second entry is still congruent to @xmath4 modulo @xmath0 .
\3 ) the subset @xmath337 is connected : note that @xmath0 is coprime to @xmath2 .
hence , the @xmath65-orbit of any vertex @xmath341 ( @xmath342 ) is the same as the @xmath9-orbit of @xmath341 . in other words
, we can use @xmath65 to get everywhere in any given row of @xmath337 , in particular in the row through @xmath343 . using the arrows and starting at @xmath343
, we can get to any other row of @xmath337 .
now by definition , @xmath337 is the union of @xmath344 rows , namely the rows @xmath345 , @xmath346 up to @xmath347 .
each row is of length @xmath2 .
it is clear from the arrows in @xmath337 that @xmath337 is isomorphic to the auslander - reiten quiver of @xmath335 .
the statement follows from theorem [ thm : d - component ] , proposition [ prop : a - components ] and the observation that the vertices of @xmath352 are exhausted by the subsets @xmath337 ( @xmath353 ) together with the vertices of @xmath263 . | we show that the @xmath0-cluster category of type @xmath1 is equivalent to a certain geometrically - defined category of arcs in a punctured regular @xmath2-gon .
this generalises a result of schiffler for @xmath3 .
we use the notion of the @xmath0th power of a translation quiver to realise the @xmath0-cluster category in terms of the cluster category .
[ section ] [ lemma]proposition [ lemma]corollary [ lemma]theorem [ lemma]remark [ lemma]definition [ lemma]example [ lemma]example |
the solar interior , photosphere and atmosphere are coupled by magnetic fields .
it is therefore important to gain insights about the magnetic field structure in all layers of the sun and solar atmosphere .
direct and accurate measurements of the magnetic field vector are typically carried out only on the photosphere .
although measurements in higher layers are available for a few individual cases , _
e.g. _ in the chromosphere by solanki _
( 2003 ) and in the corona by , the line - of - sight integrated character of such chromospheric and coronal magnetic field measurements complicates their interpretation @xcite .
knowledge of the magnetic field in the corona is essential , however , to understand basic physical processes such as the onset of flares , coronal mass ejections and eruptive prominences .
inferences of the coronal magnetic field can be obtained by extrapolating measurements of the photospheric magnetic field vector ( _ e.g. _ observed by _
hinode_/sot , solis or the upcoming sdo / hmi instruments ) into the corona .
because the magnetic pressure dominates the plasma pressure in active - region coronae , making the plasma @xmath1 low , [ see work by and , which discuss the plasma beta over active regions and over the quiet sun , respectively ] , these extrapolations neglect non - magnetic forces and assume the coronal magnetic field @xmath2 to be force - free , such that it obeys : @xmath3 equation ( [ jxb ] ) implies that the electric current density @xmath4 is parallel to the magnetic field @xmath5 .
starting more than a quarter century ago @xcite , different mathematical methods and numerical implementations have been developed to solve the nonlinear force - free equations ( [ divb ] ) and ( [ jxb ] ) for the solar case .
see , for example , @xcite for review papers and @xcite for evaluations of the performance of corresponding computer programs with model data .
the codes use the magnetic field vector ( or quantities derived from the magnetic field vector ) on the bottom boundary of a computational domain as input .
one would like to prescribe the measured photospheric data as the bottom boundary of nonlinear force - free fields ( nlfff ) codes , but there is a problem : the observed photospheric magnetic field is usually not force - free .
the relatively high plasma @xmath1 in the photosphere means that non - magnetic forces can not be neglected there and that such photospheric magnetic field data are not consistent with well known force - free compatibility conditions defined in @xcite .
recently , developed a scheme that mitigates this problem , in which the inconsistent and noisy photospheric vector magnetograms used as bottom boundary conditions are preprocessed in order to remove net magnetic forces and torques and to smooth out small - scale noise - like magnetic structures .
the resulting magnetic field data are sufficiently force - free and smooth for use with extrapolation codes , but also are found to bear a high resemblance to chromospheric vector magnetic field data .
this leads us to the question whether we can constrain the preprocessing tool further by taking direct chromospheric observations , such as h@xmath0 images , into consideration .
we will investigate this topic in the present work .
in this section , we briefly discuss the criteria on the photospheric boundary data that are required for consistency with a force - free extrapolation of the overlying coronal magnetic field .
, , , and show how moments of the lorentz force , integrated over a volume of interest , define constraints on the closed surface bounding this volume . as explained in detail in the sense of these relations
is that on average a force - free field can not exert pressure on the boundary or shear stresses along axes lying in the boundary . for the coronal magnetic field extrapolation calculations discussed here , a localized region of interest , such as an active region , is typically selected for analysis .
the extrapolation algorithms applied to the coronal volume overlying such localized regions of interest require boundary conditions , and , except at the lower ( photospheric ) boundary , these boundary conditions are usually chosen to be consistent with potential fields and thus do not possess magnetic forces or torques . in these cases ,
the consistency criteria reduce to conditions on the lower boundary only : 1 . on average force - free fields can not exert pressure on the boundary + @xmath6 2 . on average force - free fields can not create shear stresses along axes lying in the boundary + @xmath7
these relations must be fulfilled in order to be suitable boundary conditions for a nonlinear force - free coronal magnetic field extrapolation .
we define dimensionless numbers , @xmath8 @xmath9 in order to evaluate how well these criteria are met .
ideally , it is necessary for @xmath10 for a force - free coronal magnetic field to exist . pointed out that the magnetic field is probably not force - free in the photosphere , where @xmath2 is measured because the plasma @xmath1 in the photosphere is of the order of unity and pressure gradient and gravity forces are not negligible . the integral relations ( [ prepro1])-([prepro5 ] ) are not satisfied in this case in the photosphere and the measured photospheric field is not a suitable boundary condition for a force - free extrapolation .
investigations by metcalf _
( 1995 ) revealed that the solar magnetic field is not force - free in the photosphere , but becomes force - free about @xmath11 above the photosphere .
the problem has been addressed also by who pointed out that care has to be taken when extrapolating the coronal magnetic field as a force - free field from photospheric measurements , because the force - free low corona is sandwiched between two regions ( photosphere and higher corona ) with a plasma @xmath12 , where the force - free assumption might break down .
an additional problem is that measurements of the photospheric magnetic vector field contain inconsistencies and noise .
in particular the components of @xmath2 transverse to the line of sight , as measured by current vector magnetographs , are more uncertain than the line - of - sight component . as measurements in higher layers of the solar atmosphere ( where the magnetic field is force - free ) are not routinely available , we have to deal with the problem of inconsistent ( with the force - free assumption as defined by equations ( [ prepro1])([prepro5 ] ) ) photospheric measurements . a routine which uses measured photospheric vector magnetograms to find
suitable boundary conditions for a nonlinear force - free coronal magnetic field extrapolation , dubbed `` preprocessing '' , has been developed by .
the preprocessing scheme of involves minimizing a two - dimensional functional of quadratic form similar to the following : @xmath13 where @xmath14 , \label{defl_1 } \\
l_2 & = & \left [ \left(\sum_p x \left(b_z^2-b_x^2-b_y^2 \right ) \right)^2 + \left(\sum_p y \left(b_z^2-b_x^2-b_y^2 \right ) \right)^2 \right .
\nonumber \\ & & \left .
\hspace*{0.8em } + \left(\sum_p y b_x b_z -x b_y b_z \right)^2 \right ] , \\
l_3 & = & \left [ \sum_p \left(b_x - b_{xobs } \right)^2
+ \sum_p \left(b_y - b_{yobs } \right)^2 \right .
\nonumber \\ & & \left .
+ \sum_p \left(b_z - b_{zobs } \right)^2 \right ] , \label{defl_3 } \\
l_4 & = & \left [ \sum_p \left(\delta b_x \right)^2 + \left(\delta b_y \right)^2 + \left(\delta b_z \right)^2 \right ] .
\label{defl_4}\end{aligned}\ ] ] the surface integrals as defined in equations ( [ prepro1])([prepro5 ] ) are here replaced by a summation @xmath15 over all grid nodes @xmath16 of the bottom surface grid .
we normalize the magnetic field strength with the average magnetic field on the photosphere and the length scale with the size of the magnetogram .
each constraint @xmath17 is weighted by a yet undetermined factor @xmath18 . the first term ( @xmath19=1 )
corresponds to the force - balance conditions ( [ prepro1])-([prepro2 ] ) , the next ( @xmath19=2 ) to the torque - free condition ( [ prepro3])-([prepro5 ] ) .
the following term ( @xmath19=3 ) contains the difference of the optimized boundary condition with the measured photospheric data and the next term ( @xmath19=4 ) controls the smoothing .
the 2d - laplace operator is designated by @xmath20 and the differentiation in the smoothing term is achieved by the usual 5-point stencil . the last term ( @xmath21 )
has not been used in preprocessing so far and will be introduced in the next section .
the aim of the preprocessing procedure is to minimize @xmath22 so that all terms @xmath17 if possible are made small simultaneously .
this minimization procedure provides us iterative equations for @xmath23 ( see for details ) .
as result of the preprocessing we get a data set which is consistent with the assumption of a force - free magnetic field in the corona but also as close as possible to the measured data within the noise level .
nonlinear force - free extrapolation codes can be applied only to low plasma @xmath1 regions , where the force - free assumption is justified .
this is known not to be the case in the photosphere , but is mostly true for the upper chromosphere and for the corona in quiescent conditions .
the preprocessing scheme as used until now modifies observed photospheric vector magnetograms with the aim of approximating the magnetic field vector at the bottom of the force - free domain , _
i.e. _ , at a height that we assume to be located in the middle to upper chromosphere . in this study
, we investigate whether the use of chromospheric fibril observations as an additional constraint in the preprocessing can bring the resulting field into even better agreement with the expected chromospheric vector field .
we discuss this idea in the next section .
the idea is to specify another term ( @xmath24 ) in equation ( [ deflprep ] ) which measures how well the preprocessed magnetic field is aligned with fibrils seen in h@xmath0 . as a first step we have to extract the directions of the fibrils , say @xmath25 and @xmath26 out of the h@xmath0 images , where @xmath27 is a unit vector tangent @xmath28 to the chromospheric fibrils projected onto the solar photosphere ( representing the field direction with a 180-degree ambiguity ) . for simplicity one might rebin @xmath25 and @xmath26 to the same resolution as the vector magnetogram . in regions where we can not identify clear filamentary structures in the images we set @xmath29 .
these regions are only affected by the other , classical terms of the preprocessing functional ( [ deflprep ] ) .
the angle of the projected magnetic field vector on the xy - plane with the h@xmath0 image is @xmath30 where @xmath31 is the projection of the magnetic field vector in the xy - plane and @xmath32 are the directions of the chromospheric h@xmath0 fibrils .
the preprocessing aims for deriving the magnetic field vector on the bottom boundary of the force - free domain , which is located in the chromosphere .
the chromospheric magnetic field is certainly a priori unknown and as initial condition for the preprocessing routine we take @xmath33 from the photospheric vector magnetogram .
we define the functional : @xmath34 please note that the term @xmath35 in equation ( [ defl5 ] ) weights the angle with the magnetic field strength , because it is in particular important to minimize the angle in strong field regions .
the space dependent function @xmath36 is not a priori related to the magnetic field strength .
@xmath37 can be specified in order to indicate the confidence level of the fibril direction - finding algorithm ( see _ e.g. _ , for the description of a corresponding feature recognition tool ) . for the application to observational data
@xmath37 will be ( with appropriate normalization ) provided by this tool .
it is likely , however , that the direction of the h@xmath0 fibrils can be identified more accurately in strong magnetic field regions , but this is not an a priori assumption . in section [ opti_halpha ]
we investigate the influence of different assumptions for @xmath37 .
we take the functional derivative of @xmath38 @xmath39 for a sufficiently small time step @xmath40 we get a decreasing @xmath38 with the iteration equations @xmath41 the aim of our procedure is to make all terms in functional ( [ deflprep ] ) small simultaneously .
there are obvious contradictions between some of the @xmath17 terms , such as between the @xmath42 ( photospheric data ) and @xmath43 ( smoothing ) terms .
an important task is to find suitable values for the five parameters @xmath18 which control the relative weighting of the terms in equation ( [ deflprep ] ) .
the absolute values do not matter ; only the relative weightings are important .
we typically give all integral relations of the force and torque conditions ( [ prepro1])-([prepro5 ] ) the same weighting ( unity ) .
to fulfill these consistency integrals is essential in order to find suitable boundary conditions for a nonlinear force - free extrapolation . in principle
it would be possible to examine different values for the force - free term @xmath44 and torque - free term @xmath45 -or even to give six different weightings for the six integral relations- but giving all integrals the same weighting seems to be a reasonable choice .
the torque integrals depend on the choice of the length scale @xmath46 and giving the same weighting to all integrals requires @xmath47 . for the length scale normalization used here @xmath48
this leads to @xmath49 .
we will test our newly developed method with the help of a model active region in the next section .
layer , center : model - photospheric magnetic field , bottom : model - photospheric magnetic field after classical preprocessing with @xmath50.,width=529 ] we test our extended preprocessing routine with the help of an active region model recently developed by van ballegooijen _
( 2007 ) in this model line - of - sight photospheric measurements from soho / mdi have been used to compute a potential field .
a twisted flux rope was then inserted into the volume , after which the whole system was relaxed towards a nonlinear force - free state with the magnetofrictional method described in . the van ballegooijen _ et al . _ ( 2007 ) model is force - free throughout the entire computational domain , except within two gridpoints of the bottom boundary .
hereafter , we refer to the bottom of the force - free layer as the `` model chromosphere '' ( see the top panel of figure [ magnetograms1 ] ) . on the bottom boundary ( see the central panel of figure [ magnetograms1 ] ) , hereafter referred to as the `` model photosphere '' , the model contains significant non - magnetic forces and the force - free consistency criteria ( [ prepro1])-([prepro5 ] ) are not satisfied .
these forces take the form of vertical buoyancy forces directed upward , and have been introduced by van ballegooijen _ et al . _
( 2007 ) to mimic the effect of a reduced gas pressure in photospheric flux tubes . the nature of these forces is therefore expected to be similar to those observed on the real sun . for a more detailed discussion we refer to metcalf _
( 2007 ) . both the chromospheric ( @xmath51 ) as well as the photospheric magnetic field vector ( @xmath52 ) from the van ballegooijen _ et al . _ ( 2007 ) model have been used to test four sophisticated nonlinear force - free extrapolation codes in a blind algorithm test by metcalf _
the codes computed nonlinear force - free codes in a @xmath53 box , which is about at the upper limit current codes can handle on workstations .
we briefly summarize the results of metcalf _ et al . _
( 2007 ) as : * nlfff - extrapolations from model - chromospheric data recover the original reference field with high accuracy . *
when the extrapolations are applied to the model - photospheric data , the reference field is not well recovered .
* preprocessing of the model - photospheric data to remove net forces and torques improves the result , but the resulting accuracy was lower than for extrapolations from the model - chromospheric data .
the poor performance of extrapolations using the unprocessed model - photospheric data is related to their inconsistency with respect to the force - free conditions ( [ prepro1])-([prepro5 ] ) .
the central panel of figure [ magnetograms1 ] shows the photospheric magnetic field and the central panel of figure [ magnetograms1a ] illustrates the difference between the model - chromospheric and model - photospheric fields .
it is evident that there are remarkable differences in all components of the magnetic field vector . for real data we usually can not measure the chromospheric magnetic field vector directly ( which was possible for @xcite model data ) and we have to apply preprocessing before using the data as input for force - free extrapolation codes .
force - free extrapolations using preprocessed data from the model photosphere ( as lower panels of figures [ magnetograms1 ] and [ magnetograms1a ] ) , while encouraging , were not completely satisfactory , in light of the results being worse than when the model - chromospheric data were used as boundary conditions . in
what follows , we will use an artificial h@xmath0 image created from the model chromosphere to test a modified preprocessing scheme , and compare the results to the classical ( original ) preprocessing scheme .
we use the model - chromospheric magnetic field @xmath54 to derive the direction vectors of the artificial h@xmath0 images . for the model case we can simply use the chromospheric model field to specify the direction vectors @xmath25 and @xmath26 , which contain only information regarding the direction of the horizontal components of the magnetic field ( including a @xmath55 ambiguity , but no information about the magnetic field strength . for real data
this information can be derived from high - resolution h@xmath0 images using feature recognition techniques , _
e.g. _ the ridge detector of .
, right panel : @xmath56 ) with the model chromosphere in dependence of the preprocessing parameters @xmath57 and @xmath58 .
we found a maximum correlation at @xmath59 and @xmath60.,width=529 ] we tested more than @xmath61 possible combinations of @xmath62 and @xmath63 using the model - photospheric field as input , and computed the pearson correlation coefficient between the preprocessed results and the model - chromospheric field . only @xmath64 and @xmath56
were used in computing the correlation coefficient , because the correlation of the longitudinal ( _ i.e. _ , the line - of - sight ) component is in general higher than that of the transverse components , due to @xmath65 not being affected by the ambiguity - problem and the noise being much lower than in the other directions .
we computed @xmath66 combinations of @xmath67 between @xmath68 with a step size of @xmath69 . hereafter a local maximum around @xmath70 appeared .
this region was analyzed in more detail by using these two values as new initial guess .
to do this , we tried another @xmath66 combinations around this pair with a reduced step size of @xmath71 in the positive as well as the negative direction .
then the absolute maximum of the correlation coefficients for both , @xmath72 and @xmath73 appeared at @xmath74 ( see figure [ corr34 ] ) .
the bottom panel of figure [ magnetograms1 ] shows the corresponding preprocessed photospheric magnetic field .
fibrils identified from the model chromosphere . the fibrils give us information about the transverse components ( @xmath64 and @xmath56 ) of the chromospheric magnetic field .
the fibrils contain a @xmath75 ambiguity and do not provide any information about the chromospheric magnetic field strength .
the bottom panels show from left to right the different weighting functions @xmath76 , respectively .
regions where @xmath37 is higher are more important is the @xmath38-preprocessing - term ( [ defl5 ] ) which controls the influence of the h@xmath0-fibrils.,width=529 ] in the following we aim to find suitable parameters for including information from h@xmath0 images into the preprocessing .
our main aim is to investigate the effects of additional chromospheric information . to exclude side effects we therefore keep the combination of @xmath44-@xmath58 found in the previous section to be able to clearly investigate the effect of the additional term @xmath38 . in principle
one could vary all @xmath18 simultaneously .
we can not exclude that there might exist a better combination of @xmath44 to @xmath77 with better agreement of our preprocessed field and the model chromospheric field .
this is , however , not the aim of this work , because this is not a suitable way to deal with real data , because there is no model chromosphere to test the result .
it is not possible to provide an optimal parameter set suitable for all vector magnetographs .
the optimal combination has to be carried out for different instruments separately .
we expect that an optimal parameter set for a certain instrument and particular region will be also useful for the preprocessing of other regions of the same kind ( say active regions ) observed with the same instrument .
we test our methods with `` model fibrils '' extracted from the model chromosphere shown in the top panel of figure [ fig_halpha ] .
we define @xmath78 used in equation ( [ defl5 ] ) as one of the following : 1 .
we assume that at every point of our h@xmath0 image gives us the exact orientation of the magnetic field ( which is indeed the case , as we calculated it from the chromospheric model data ) and fix our weighting with @xmath79 .
we assume that the photospheric magnetic field magnitude gives us the importance of the h@xmath0 information at each point and use + + @xmath80 .
+ we scale @xmath81 to a maximum value of @xmath82 .
( see figure [ fig_halpha ] bottom left panel . )
we do as in the previous case , but assume now , that only points in the magnetogram where the field magnitude is greater than 50 % of the maximum contribute to the h@xmath0 preprocessing .
so , we define + @xmath83 + ( see figure [ fig_halpha ] bottom center panel . )
4 . in our last case
we assume in the same way as in the previous one , but now only points in the magnetogram where the field magnitude is greater than 10 % of the maximum contribute to the preprocessing .
all these grid points are weighted with @xmath82 and the rest with zero .
in other words , one defines + @xmath84 + ( see figure [ fig_halpha ] bottom right panel . )
we now figure out the optimal value of @xmath77 in equation ( [ deflprep ] ) for the four different weighting functions @xmath85 .
initially , we use a step size of @xmath86 and then , around the first appearing maximum , we reduced it to @xmath87 .
this is to find a more precise optimal value of @xmath77 .
we calculate the pearson correlation coefficient between the chromospheric reference field ( @xmath51 ) and the minimum solution of the preprocessing routine ( @xmath88 ) .
this provides us the optimal values of @xmath77 for the different weighting functions , see second row in table [ tab : testparms2 ] .
preprocessing with different weighting functions .
top : @xmath89 , center : @xmath81 , bottom : @xmath90 , see text.,width=529 ] and the h@xmath0-preprocessed fields as shown in figure [ magnetograms2].,width=529 ] table [ tab : testparms2 ] lists some metrics related to the various preprocessing schemes , including the dimensionless numbers @xmath91 and @xmath92 from equations ( [ eps_force ] ) and ( [ eps_torque ] ) , the values of the various @xmath17 from section [ sec : preprocessing ] , and the averaged angles between the preprocessing results and the model - chromospheric field .
the first three rows of the table list the model chromosphere ( @xmath93 ) and photosphere ( @xmath94 ) data and the classical preprocessing scheme ( @xmath95 ) .
when using the unprocessed model - photospheric data ( @xmath94 ) , it is clear that the force - free consistency criteria ( as represented by @xmath96 , @xmath97 , and @xmath98 ) are not fulfilled and are orders of magnitude higher than for the chromospheric data ( @xmath93 ) .
consequently , we can not expect the extrapolation codes to result in a meaningful nonlinear force - free field in the corona , as discussed in metcalf _
( 2007 ) .
the remaining rows in tables [ tab : testparms2 ] and [ tab : extraparms ] list the results for the cases where the h@xmath0 preprocessing was used .
a qualitative comparison of the h@xmath0-preprocessed magnetograms ( shown in figure [ magnetograms2 ] ) with the model chromosphere ( shown in the top panel of figure [ magnetograms1 ] ) indicates a strong resemblance for all three magnetic field components , but certainly not a perfect match .
difference images between the h@xmath0-preprocessed magnetograms and the model chromosphere ( shown in the top panel of figure [ magnetograms1 ] ) are present in figure [ magnetograms2a ] .
the resemblance using the h@xmath0 preprocessing scheme is much improved when compared to the magnetograms resulting from the classical preprocessing scheme .
table [ tab : extraparms ] displays metrics of the resulting nonlinear force - free extrapolations using each preprocessing scheme.[multiblock footnote omitted ] as expected , the extrapolation codes perform poorly when the unprocessed boundary ( @xmath94 ) is used .
in particular , the resulting magnetic energy @xmath99 of this case ( normalized to the energy of the reference solution ) is only 65% of the correct answer , making it almost impossible to estimate the free magnetic energy in the solution available for release during eruptive processes such as flares and coronal mass ejections .
taking preprocessing into account ( rows 3 - 7 in both tables ) significantly improves the result .
the force - free consistency criteria ( @xmath100 ) are adequately fulfilled for all preprocessed cases and are even better ( lower values ) than the model chromospheric field .
this is naturally , however , because the preprocessing routine has been developed in particular to derive force - free - consistent boundary conditions from inconsistent ( forced , noisy ) photospheric measurements .
the classical preprocessing @xmath101 has already reduced the angle to the model h@xmath0 fibrils ( last two columns of table [ tab : testparms2 ] ) by almost a factor of two , even though no information about the chromosphere has been used .
if we include chromospheric information , ( see figure [ fig_halpha ] ) in our preprocessing routine ( @xmath102 , rows 4 - 7 ) the angle of the preprocessed field with the h@xmath0 images reduces significantly . the second to last row in table [ tab : testparms2 ] contains the average angle and in the last column the angle has been weighted by the magnetic field , which means that @xmath103 measures mainly how well the magnetic field and the chromospheric fibrils are aligned in regions of a high magnetic field strength . for the purpose of coronal magnetic field extrapolations the strong field regions are essential .
if we include all information from the h@xmath0 image , as done in row 4 for @xmath89 we find that the magnetic field and the fibrils are almost parallel in the entire region .
this is the ideal case , however , as fibrils have been identified all over the region with the same excellent accuracy .
for observed data it is more likely that the direction of the fibrils will be identifiable with high accuracy only in bright and magnetically strong regions .
this effect is taken into account in rows 5 - 7 of both tables . in the last two rows we take the chromospheric data only into account where the magnetic field strength is larger than @xmath104 and @xmath105 of the maximum field strength , respectively .
naturally , the average angle @xmath106 of the chromospheric fibrils with the preprocessed magnetic field becomes larger than for the ideal case .
we find , however , that the angle @xmath107 remains relatively low in strong field regions , except for the case @xmath90 .
we can easily understand that @xmath90 ( chromospheric information ignored where the magnetic field is less than @xmath108 of its maximum ) provides less accurate results , because the area where chromospheric data have been taken into account , is only a very small fraction of the entire region ( see figure [ fig_halpha ] lower central panel ) .
case @xmath90 has few nonzero points .
these points are , however , in the regions with the strongest magnetic field strength .
the @xmath38 terms minimizes the angle between magnetic field and chromospheric fibrils only in these nonzero points .
this local correction does , however , influence the magnetogram globally , because the @xmath109 and @xmath110 term contain global measures and the @xmath111 terms couples neighbouring points . as a consequence
the preprocessing result is different from classical preprocessing , even if the @xmath38 term is nonzero only for a limited number of pixel . for observational data
the weighting @xmath112 ( last row in the tables , areas with less than @xmath105 ignored ; see also [ fig_halpha ] lower right panel ) seems to be more realistic . in this case
the overall average angle is not better than for classical preprocessing , but is different by only about @xmath113 when preferential weighting is given to the more important strong field regions .
the ultimate test regarding the success of our extended preprocessing scheme is to use the preprocessed field as boundary conditions for a nonlinear force - free coronal magnetic field extrapolation .
the results are presented in table [ tab : extraparms ] , row 3 for classical preprocessing and rows 4 - 7 for h@xmath0 preprocessing .
we find that all preprocessed fields provide much better results than using the unprocessed data . for classical preprocessing
we get the magnetic energy @xmath99 correct with an error of @xmath114 ( for unprocessed data we got an error of @xmath115 ) . taking the h@xmath0 information into account
improves the result and the magnetic energy is computed with an accuracy of @xmath116 or better , even for the cases where we used chromospheric information only in parts of the entire regions .
.results of the various preprocessing schemes : the model chromosphere and photosphere ( first two rows ) , classical preprocessing ( third row ) , and the h@xmath0 preprocessing cases ( last four rows ) .
column 1 identifies the data set , columns 2 and 3 the value of @xmath77 and the weighting scheme used for the h@xmath0 preprocessing cases .
columns 4 - 7 provide the value of the functionals @xmath117 , @xmath118 as defined in equations ( [ defl_1])-([defl_4 ] ) and ( [ defl5 ] ) , respectively . in columns 8 and 9
we show how well the force - free and torque - free consistency criteria ( @xmath119 ) as defined in equations ( [ eps_force ] ) and ( [ eps_torque ] ) are fulfilled .
the last two columns contain the averaged angle ( @xmath120 ) of the field with the model chromospheric data and a magnetic field weighted average angle ( @xmath121 ) with @xmath122 as defined in equation ( [ angle_phi ] ) .
[ cols="^,^,^,^,^,^,^,^,^ , > , > " , ]
within this work we developed an improved algorithm for the preprocessing of photospheric vector magnetograms for the purpose of getting suitable boundary conditions for nonlinear force - free extrapolations .
we extended the preprocessing routine developed by , which is referred to here as `` classical preprocessing '' .
the main motivation for this work is related to the fact that active - region coronal magnetic fields are force - free due to the low @xmath1 coronal plasma , but the magnetic field vector can be measured with high accuracy only on the photosphere , where the plasma @xmath1 is about unity and non - magnetic forces can not be ignored .
our original ( `` classical '' ) preprocessing removes these non - magnetic forces and makes the field compatible with the force - free assumption leading to more chromospheric - like configurations . in this study
, we have found that by taking direct chromospheric observations into account ( such as by using fibrils seen in h@xmath0 images ) , the preprocessing is improved beyond the classical scheme .
this improved scheme includes a term which minimizes the angle between the preprocessed magnetic field and the fibrils .
we tested our method with the help of a model active region developed by van ballegooijen _
( 2007 ) , which includes the forced photospheric and force - free chromospheric and coronal layers .
this model has been used by metcalf _
( 2007 ) for an inter - comparison of nonlinear force - free extrapolation codes .
the comparison revealed that the model coronal magnetic field was reconstructed very well if chromospheric magnetic fields have been used as input , but in contrast the reconstructed fields compared poorly when unprocessed model - photospheric data were used .
classical preprocessing significantly improves the result , but the h@xmath0 preprocessing developed in this paper is even better as the main features of the model corona are reconstructed with high accuracy .
our extended preprocessing tool provides a fair estimate of the chromospheric magnetic field , which is used as boundary condition for computing the nonlinear force - free coronal magnetic field .
in particular , the magnetic energy in the force - free domain above the chromosphere agrees with the model corona within @xmath116 , even if only strong - field regions of the model chromosphere , where the fibrils can be identified with highest accuracy , influence the final solution . from these tests
we conclude that our improved preprocessing routine is a useful tool for providing suitable boundary conditions for the computation of coronal magnetic fields from measured photospheric vector magnetograms as provided for example from _
the combination of preprocessing and nonlinear force - free field extrapolations seem likely to provide accurate computation of the magnetic field in the corona .
we will still not get the magnetic field structure in the relative thin layer between / in the photosphere and the chromosphere correct , because here non - magnetic forces can not be neglected due to the finite @xmath1 plasma .
although this layer is vertically thin ( _ e.g. _ , 2 vertical grid points in the van ballegooijen _ et al . _ ( 2007 ) model compared to 256 vertical grid points in the corona ) it contains a significant part of the total magnetic energy of the entire domain , see metcalf _ et al . _
( 2007 ) . unfortunately
, this part of the energy can not be recovered by force - free extrapolations , because the region is non - force - free .
our improved preprocessing routine includes chromospheric information and therefore provides us with a closer approximation of the chromospheric magnetic field .
this leads to more accurate estimates of the total magnetic energy in the corona
. a further improvement of the preprocessing routine could be done with the help of additional observations , _
e.g. _ the line - of - sight chromospheric field , as planned for solis .
one could include these measurement directly in the @xmath123-term ( [ defl_3 ] ) either as the only information or in some weighted combination with the photospheric field measurement .
an investigation of the true 3d - structure of the thin non - force - free layer between photosphere and chromosphere requires further research .
first steps towards non - force - free magnetohydrostatic extrapolation codes @xcite might help to reveal the secrets of this layer .
non force - free magnetic field extrapolations will require additional observational constraints , because the magnetic field , the plasma density and pressure must be computed self - consistently in one model . the work of t. wiegelmann was supported by dlr - grant 50 oc 0501 and j.k .
thalmann got financial support by dfg - grant wi 3211/1 - 1 .
m. derosa , t. metcalf , and c. schrijver were supported by lockheed martin independent research funds .
we acknowledge stimulating discussions during the fourth nlfff - consortium meeting in june , 2007 in paris .
we briefly summarize our nonlinear force - free extrapolation code here , which has been used to compute the 3d magnetic fields .
we solve the force - free equations ( [ divb ] ) and ( [ jxb ] ) by optimizing ( minimizing ) the following functional : @xmath124 \ ; d^3x \label{defl1},\ ] ] where @xmath125 and @xmath126 are weighting functions .
it is obvious that ( for @xmath127 ) the force - free equations ( [ divb ] ) and ( [ jxb ] ) are fulfilled when @xmath128 is zero .
the optimization method was proposed by wheatland , sturrock , and roumeliotis ( 2000 ) and further developed in wiegelmann and neukirch ( 2003 ) .
here we use the implementation of wiegelmann ( 2004 ) which has been applied to data in wiegelmann et al .
( 2005 ) . in this article
, we used a recent update including of our code that included a multi - scale approach ( see metcalf et al .
( 2007 ) for details ) .
this version of the optimization code was also used with the ( same as in this paper ) model - chromospheric , photospheric and classical preprocessed photospheric magnetic field vector as part of an inter - code - comparison in @xcite . for alternative methods to solve the force - free equations ( [ divb ] ) and ( [ jxb ] )
see the review papers by @xcite and references therein .
in order to quantify the degree of agreement between the extrapolated vector fields of the input model field ( * b * , _ i.e. _ , the extrapolated chromospheric ( reference ) field ) and the nonlinear force - free solutions ( * b * , _ i.e. _ , the extrapolated preprocessed photospheric field ) that are specified on identical sets of grid points , we use five metrics in table [ tab : extraparms ] that compare either local characteristics or the global energy content in addition to the force and divergence integrals .
these measures have been developed in schrijver et al .
( 2006 ) and subsequently been used to evaluate the quality of force - free and non - force - free extrapolation codes @xcite . where @xmath140 is the total number of vectors in the volume , and @xmath141 the angle between * b * and * b * at point @xmath132 .
it is entirely a measure of the angular differences of the vector fields , i. e. @xmath142 if * b * @xmath143 * b * , @xmath144 if they are anti - parallel , and @xmath134 if @xmath135 at each point . unlike the first two metrics , perfect agreement of the two vector fields results in @xmath147 . for an easier comparison with the others
, we list @xmath148 , so that all measures reach unity for a perfect match . | the solar magnetic field is key to understanding the physical processes in the solar atmosphere .
nonlinear force - free codes have been shown to be useful in extrapolating the coronal field upward from underlying vector boundary data .
however , we can only measure the magnetic field vector routinely with high accuracy in the photosphere , and unfortunately these data do not fulfill the force - free condition .
we must therefore apply some transformations to these data before nonlinear force - free extrapolation codes can be self - consistently applied . to this end , we have developed a minimization procedure that yields a more chromosphere - like field , using the measured photospheric field vectors as input .
the procedure includes force - free consistency integrals , spatial smoothing , and newly included in the version presented here an improved match to the field direction as inferred from fibrils as can be observed in , _
e.g. _ , chromospheric h@xmath0 images .
we test the procedure using a model active - region field that included buoyancy forces at the photospheric level .
the proposed preprocessing method allows us to approximate the chromospheric vector field to within a few degrees and the free energy in the coronal field to within one percent . |
shell - type supernova remnants ( snrs ) are prime candidates for acceleration sites of galactic cosmic rays @xcite . the energy density of cosmic rays ( @xmath2 ) can be explained if @xmath3% of the kinetic energy released by supernova explosions is transferred to accelerated particles .
also , diffusive shock acceleration theory , which is expected to be working at snr shocks , can naturally explain the power - law type spectrum of cosmic rays .
observationally , nonthermal x - rays and tev gamma rays have been detected from snrs , particularly from relatively young ones such as cassiopeia a ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , rx j1713.7@xmath03946 ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , and rx j0852.0@xmath04622 ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , providing us with evidence that particles are indeed accelerated at least up to @xmath4 tev .
the snr rx j0852.0@xmath04622 ( also known as [email protected] or vela jr . ) is one of the snrs from which both nonthermal x - rays and tev gamma rays are detected .
the snr was discovered in x - rays by _ rosat _ in the southeastern corner of the vela snr @xcite .
its x - ray spectrum was later confirmed to be of nonthermal origin by _ asca _
observations @xcite .
the x - ray emission has an almost circular morphology with a diameter of @xmath5 with rim brightening especially in the north , northwestern , and western parts . at tev energies , the cangaroo @xcite and h.e.s.s .
@xcite collaborations reported the detection .
similar to the well - studied snr rx j1713.7@xmath03946 , the h.e.s.s .
telescope spatially resolved the emission from rx j0852.0@xmath04622 @xcite .
the most recent and reliable estimates of age and distance of the snr come from the measurement of expansion of the shell by @xcite using _ xmm - newton _ data taken over a span of 6.5 yr .
their estimates of the age and distance are 17004300 yr and 750 pc , respectively . at this distance
, the angular size of the snr ( @xmath6 in radius ) translates to a radius of @xmath7 pc .
it is widely believed that rx j0852.0@xmath04622 is a remnant of a core - collapse ( type ib / ic / ii ) supernova explosion since studies of an x - ray point source , cxou j085201.4@xmath0461753 , located close to the snr center suggest that the source is a neutron star ( e.g. * ? ? ?
* ; * ? ? ?
however , @xcite pointed out the x - ray source is more likely related to a planetary nebula they found at the location in radio data .
the broadband spectrum of the nonthermal emission can be explained either by a leptonic scenario where the gamma rays are produced by inverse compton scattering of high - energy electrons , or by a hadronic scenario where @xmath8 decays are responsible for the gamma rays @xcite . the large area telescope ( lat ) aboard the _ fermi gamma - ray space telescope _
@xcite can provide important information on gev gamma - ray emission which is crucial to constrain the possible emission models . in this letter ,
we report on results of gamma - ray observations of the snr rx j0852.0@xmath04622 by the _ fermi _ lat .
the _ fermi gamma - ray space telescope _ , launched on 2008 june 11 , started continuous survey - mode observations on 2008 august 4 .
the lat , with its wide field - of - view ( fov ) of 2.4 sr , observes the whole sky in every @xmath9 hr in this mode .
besides the wide fov , the lat has better detector performance than its predecessors , large effective area of @xmath10 ( on axis at 1 gev ) and point spread function better than @xmath11 ( for 68% containment ) at 1 gev .
a detailed description of the instrument can be found in @xcite .
here we analyzed _ fermi _ lat data taken for 30 months , from 2008 august 4 to 2011 february 8 .
the data analysis was performed using the standard analysis software package , the lat science tools v9r21p0 .
we selected `` diffuse class '' events , which are those reconstructed with high probability of being photons and are recommended for standard point - source analyses . in the analysis ,
the instrumental response functions ( irfs ) `` pass6 v11 '' were used . to reduce the background from the earth limb @xcite , events with earth zenith angle greater than @xmath12 were eliminated .
we performed a morphological and spectral analysis of the region that contains snr rx j0852.0@xmath04622 . all the observational results except for the counts map shown in figure [ fig : map ] (
a ) and ( b ) are based on binned maximum likelihood analysis .
the emission model for the maximum likelihood analysis includes snr rx j0852.0@xmath04622 , vela x , point sources listed in the first _ fermi _ lat ( 1fgl ) catalog @xcite , the galactic diffuse background , and the isotropic background component which accounts for the extragalactic diffuse emission and residual detector background .
one of the 1fgl sources , 1fgl j0854.0@xmath04632 , is located inside the snr rx j0852.0@xmath04622 , and was excluded from the model . the spatial template for vela x
is assumed to be a uniform disk as reported by @xcite .
the galactic diffuse and isotropic background models we adopted are _
gll_iem_v02_p6_v11_diffuse.fit _ and _ isotropic_iem_v02_p6_v11_diffuse.txt _ , respectively .
the analysis was performed on events within a region - of - interest ( roi ) of a @xmath13 square centered at ( @xmath14 , @xmath15 = ( @xmath16 , @xmath17 ) . in counts maps of _ fermi _ lat data ,
a gamma - ray source was found at the location of snr rx j0852.0@xmath04622 .
figure [ fig : map ] ( a ) shows a _ fermi _ lat counts map of @xmath18 gev events . to take advantage of better angular resolution and weaker galactic diffuse emission
, we also present a counts map of @xmath19 gev events in figure [ fig : map ] ( b ) .
no background subtraction is applied to the counts maps .
emission can be seen at the location of snr rx j0852.0@xmath04622 .
figure [ fig : map ] ( c ) presents the test statistic ( ts ) map of the source calculated using events above 10 gev .
ts is defined as @xmath20 , where @xmath21 and @xmath22 are the likelihood of the null hypothesis ( without a source ) and that of a point source hypothesis , respectively @xcite . to generate the ts map
, we put a point source at each position in the sky and obtained ts for each case .
the ts map indicates significant emission from the location coincident with rx j0852.0@xmath04622 above the backgrounds .
the spatial distribution of the emission shown in figure [ fig : map ] appears to be extended and compatible with the snr seen in other wavelengths . to test the spatial extension and quantify it
, we assumed a uniform disk as a spatial template and performed a maximum likelihood fit to @xmath18 gev events with _
the emission was found to be extended significantly with @xmath23 , where @xmath24 is the likelihood for the case in which a disk is adopted as a spatial template for the source .
the best fit radius is @xmath25 centered at ( @xmath14 , @xmath26 ) = ( @xmath27 , @xmath28 ) .
the ts for the source detection is 228 with the best - fit disk template .
the location and size are consistent with those of snr rx j0852.0@xmath04622 seen in radio , x - rays , and tev gamma rays , and , therefore , support the identification of the emission with the snr .
we also tried fitting the data using the h.e.s.s .
image @xcite as a spatial template . in this case
, we obtained @xmath29 , which is only slightly different from that obtained for the disk shape .
the current statistics do not allow us to study further the morphology of the gev emission such as spatial structures inside the snr . in what follows , we use the h.e.s.s .
image as the default spatial template to study the spectral properties of the source .
the spectral information of the source was extracted by performing maximum likelihood fits with _ gtlike _ for the roi . in the energy region below 1 gev , we found that the systematic errors mainly from the galactic diffuse model become @xmath30% of the flux obtained for rx j0852.0@xmath04622 , and thus we only report results obtained above 1 gev .
firstly , we analyzed all the events above 1 gev together to obtain the spectral parameters in the entire band .
we assumed a power - law function as a spectral model for rx j0852.0@xmath04622 .
the 1fgl sources except for the vela pulsar were also modeled with power laws and their parameters are allowed to vary .
the spectrum of the vela pulsar is represented with a power law with an exponential cutoff , whose parameters ( normalization , index , and cutoff energy ) are also treated as free parameters . to the spectrum of the galactic diffuse model , a power - law function was multiplied in order to allow for any energy - dependent discrepancy between the model and data .
the photon index of rx j0852.0@xmath04622 is @xmath1 , and the flux integrated between 1 gev and 300 gev is @xmath31 . in order to obtain spectral points at each energy
, we divided the data between 1 gev and 300 gev into four log - equal energy bands , and performed maximum likelihood fits for each of them .
the parameters for the point sources in the roi , except for the vela pulsar , are fixed at the best - fit values obtained in the entire - band fit above .
the fitting parameters here are the normalization of rx j0852.0@xmath04622 , that of the vela pulsar , and that of the galactic diffuse emission .
the spectral points obtained in each band are summarized in table [ tab : spec_points ] and are plotted in figure [ fig : spec ] together with the power - law function from the entire band fit .
the major contributions to the systematic errors quoted above come from uncertainties in the effective area calibration of the _ fermi _ lat and imperfect modeling of the galactic diffuse emission .
uncertainties in the effective area are estimated based on observations of the vela pulsar @xcite and earth limb gamma rays @xcite .
here we assumed 5% at @xmath32 , 20% above @xmath33 , and interpolation of the two values in - between .
the accuracy of the galactic diffuse model was evaluated based on the discrepancies between the best - fit model and data at each location of the roi as was done by @xcite .
we found that they are @xmath34 , and we changed the normalization of the galactic diffuse model by @xmath35 from the best - fit values and added the differences of the parameters before and after the renormalization to the systematic errors .
ccc 2.04 & @xmath36 & 84.2 + 8.49 & @xmath37 & 86.1 + 35.3 & @xmath38 & 102.4 + 147 & @xmath39 & 40.9 ccccccccc hadronic & 1.8 & 50 & 1.8 & 10 & 100 & 0.1 & @xmath40 & @xmath41 + leptonic & & & 2.15 & 25 & 12 & 0.1 & & @xmath42
gamma - ray emission mechanisms for rx j0852.0@xmath04622 have been actively discussed in the literature since its detection in the tev band .
two types of scenarios have mainly been considered to account for the nonthermal emission from rx j0852.0@xmath04622 ( e.g. * ? ? ?
one is the so - called hadronic scenario in which gamma rays are dominantly radiated through decays of @xmath8 mesons produced in collisions between accelerated protons / nuclei and ambient gas .
the other is the leptonic scenario in which inverse compton scattering of relativistic electrons with photon fields such as the cosmic microwave background ( cmb ) is the main emission mechanism for gamma rays .
we calculated gamma - ray emission models for both the hadronic and leptonic scenarios . in the calculations , protons and electrons
are injected with a constant rate and fixed spectral form of @xmath43 .
\end{aligned}\ ] ] injected particles suffer cooling via such processes as ionization , synchrotron , or inverse compton scattering , and their spectra are deformed .
the spectra of radiating particles ( @xmath44 ) are obtained by solving the kinetic equations for electrons and protons : @xmath45 + q_{e , p}(p ) , \end{aligned}\ ] ] where the term @xmath46 denotes the momentum loss rate .
the equations are solved for @xmath47 yr , and then photon spectra from electrons and protons are computed for each radiation process .
the distance to the snr is assumed to be @xmath48 pc @xcite . as the seed photons for inverse compton scattering , we considered the cmb as well as the interstellar radiation field in infrared and optical bands .
the spectrum of the interstellar radiation is taken from galprop @xcite at the location of the snr .
figure [ fig : modeling ] ( a ) shows the hadronic model together with radio , x - ray , and gamma - ray data .
the parameters adopted for the calculation are summarized in table [ tab : model_param ] .
the indices of proton and electron spectra are assumed to be the same .
for the model curves in figure [ fig : modeling ] ( a ) , the proton spectrum has an index of @xmath49 and a high - energy cutoff at @xmath50 . in this case , the total kinetic energy of protons above 10 mev is @xmath51 erg , where @xmath52 denotes the hydrogen gas density in the post - shock region .
the synchrotron spectrum from electrons with the same index @xmath53 is consistent with the radio spectral index @xmath54 deduced from the two flux points in figure [ fig : modeling ] @xcite . note that the synchrotron spectrum index and electron index are related as @xmath55 .
the flux of the synchrotron component is controlled by the amount of electrons and the magnetic field strength .
the stronger the magnetic field is , the larger the number of electrons should be , and the inverse compton flux becomes higher . in the hadronic scenario ,
the magnetic field should be at least @xmath56 so that inverse compton scattering becomes negligible in the gev - to - tev domain . in the case of @xmath57
adopted here , the total kinetic energy of electrons becomes @xmath58 erg , and the ratio of injected particle numbers at @xmath59 is @xmath60 . in the case of the leptonic scenario ,
the electron index is constrained by the _ fermi _
lat spectrum .
considering inverse compton scattering in the thomson regime , the photon index and electron index are related as @xmath55 .
therefore , the observed photon index @xmath61 leads to a soft electron spectrum with @xmath62 .
with this index , however , the synchrotron spectrum can not fit the radio and x - ray data well .
we then adopted a harder electron spectrum within the range allowed by the statistical and systematic errors in the _ fermi _ lat spectral points .
the model curves in figure [ fig : modeling ] ( b ) are the leptonic models when the electron index is @xmath63 with a high - energy cutoff at @xmath64 .
the magnetic field strength can be determined as @xmath65 so that the synchrotron to inverse compton flux ratio matches the data .
electron bremsstrahlung emission is calculated to be @xmath66 at 10 gev .
both models face difficulty .
the leptonic model requires a weak magnetic field of order @xmath67 , which may be contradicted by observations .
the width of the filaments in the shell observed in _
chandra _ gives a magnetic field estimate of @xmath68 @xcite .
these results are interpreted as evidence for magnetic field amplification by streaming of accelerated cosmic rays .
however , it is possible to reconcile a high magnetic field with the leptonic model if gev gamma rays are radiated not only from the filamentary structures seen by _
chandra _ , but also from other regions in the snr where the magnetic field may be weaker .
the hadronic model requires an unrealistically large energy of protons if the gas density is as small as @xmath69 .
thermal x - ray emission works as a probe to estimate the gas density .
however , no clear detection is reported so far mainly because it is difficult to separate emission from rx j0852.0@xmath04622 and the vela snr , which overlaps and emits strong thermal emission in the soft x - ray band . using _
asca _ data ,
@xcite placed an upper limit of @xmath70 , where @xmath71 and @xmath72 are the distance to rx j0852.0@xmath04622 and the filling factor of a sphere taken as the emitting volume in the region used for their spectral analysis , respectively . in the case of @xmath73 , which corresponds to rather thick shell width
, the upper limit becomes problematic since @xmath74 reaches @xmath75 erg , which means that almost all the kinetic energy released by the supernova explosion must go to accelerated protons .
the proton index @xmath49 for the hadronic scenario is somewhat harder than @xmath76 expected in the test - particle regime of diffusive shock acceleration with a strong shock .
therefore , it is of interest to compare the observed spectrum with models based on diffusive shock acceleration in the nonlinear regime .
@xcite has calculated the broadband sed of rx j0852.0@xmath04622 with their time - dependent , nonlinear kinetic theory for cosmic - ray acceleration .
their model curve is plotted over the gev - to - tev data in figure [ fig : spec ] .
the _ fermi _ lat spectrum appears harder than the model by @xcite , but the discrepancy is not large compared to the systematic errors .
it is of interest to compare the gamma - ray spectrum of rx j0852.0@xmath04622 with that of the similar object , rx j1713.7@xmath03946 @xcite .
the two snrs have roughly the same age , size , and similar radio , x - ray and tev gamma - ray spectra . the _ fermi _ lat spectrum of rx j1713.7@xmath03946 has a photon index of @xmath77 , which is more preferable for leptonic models than hadronic models @xcite . in the case of rx j0852.0@xmath04622 , a leptonic origin is not favored but can not be ruled out .
the difference between the two similar objects could be an important point to be further explored to probe gamma - ray production in snrs .
further studies of _ fermi _ lat data in the future , particularly better modeling of the galactic diffuse emission , will reduce the uncertainties and will allow us to probe particle acceleration in the snr in greater detail .
the _ fermi _ lat collaboration acknowledges support from a number of agencies and institutes for both development and the operation of the lat as well as scientific data analysis .
these include nasa and doe in the united states , cea / irfu and in2p3/cnrs in france , asi and infn in italy , mext , kek , and jaxa in japan , and the k. a. wallenberg foundation , the swedish research council and the national space board in sweden . additional support from inaf in italy and cnes in france for science analysis during the operations phase is also gratefully acknowledged . | we report on gamma - ray observations of the supernova remnant ( snr ) rx j0852.0@xmath04622 with the large area telescope ( lat ) aboard the _ fermi gamma - ray space telescope_. in the _ fermi _ lat data , we find a spatially extended source at the location of the snr .
the extension is consistent with the snr size seen in other wavelengths such as x - rays and tev gamma rays , leading to the identification of the gamma - ray source with the snr .
the spectrum is well described as a power law with a photon index of @xmath1 , which smoothly connects to the h.e.s.s .
spectrum in the tev energy band .
we discuss the gamma - ray emission mechanism based on multiwavelength data .
the broadband data can be fit well by a model in which the gamma rays are of hadronic origin .
we also consider a scenario with inverse compton scattering of electrons as the emission mechanism of the gamma rays .
although the leptonic model predicts a harder spectrum in the _ fermi _ lat energy range , the model can fit the data considering the statistical and systematic errors . |
the well - known regular bartnik - mckinnon ( bm ) solutions @xcite and the corresponding non - abelian black hole solutions @xcite , are asymptotically flat , static spherically symmetric solutions of su(2 ) einstein - yang - mills ( eym ) theory .
they are unstable solutions , sphalerons @xcite , and are characterized by the number of nodes of the gauge field .
besides these spherically symmetric solutions there are also asymptotically flat , static regular and black hole solutions , which possess only axial symmetry @xcite .
these are characterized by two integers , the node number of their gauge field function(s ) , and the winding number with respect to the azimuthal angle , denoted @xmath2 .
the spherically symmetric solutions have winding number @xmath5 , while the only axially symmetric solutions have winding number @xmath6 .
the @xmath6 black hole solutions possess an event horizon with a slight elongation along the symmetry axis @xcite .
all these eym black hole solutions exist for arbitrarily large horizon size . in su(2 )
einstein - yang - mills - higgs ( eymh ) theory with a triplet higgs field , gravitating monopole solutions and black holes with monopole hair arise @xcite . as in eym
theory , the regular eymh solutions are characterized by two integers , the node number of the gauge field and the azimuthal winding number @xmath2 , which corresponds to the topological charge of the monopoles . the black hole solutions are characterized in addition by their horizon size .
these eymh black hole solutions exist only up to a maximal value of the horizon size @xcite , unlike the known eym black hole solutions .
eymh theory allows for further static axially symmetric solutions , representing gravitating monopole - antimonopole pair , chain and vortex solutions @xcite , which can be characterized by the azimuthal winding number @xmath2 , and by a second integer @xmath7 , related to the polar angle . for the monopole - antimonopole chains , which arise in flat space for @xmath5 and 2
, the integer @xmath7 corresponds to the number of nodes of the higgs field ( and thus the number of poles on the symmetry axis ) , while in vortex solutions , which arise in flat space for winding number @xmath8 , the higgs field vanishes ( for even @xmath7 ) on @xmath9 rings centered around the symmetry axis @xcite .
to all these regular solutions associated black hole solutions should exist , obtained so far only for a monopole - antimonopole pair @xcite . in the limit of vanishing higgs expectation value , eymh solutions approach ( after rescaling ) eym solutions @xcite .
interestingly , when @xmath10 and @xmath11 , new regular eym solutions appear as limiting solutions @xcite .
these static axially symmetric solutions have been characterized by the integers @xmath0 , where @xmath12 .
they have been constructed numerically for @xmath13 , @xmath10 , and @xmath14 , @xmath15 @xcite . unlike the @xmath3 eym solutions
, the @xmath4 solutions always appear in pairs . in this letter
we construct the corresponding eym black hole solutions .
as expected , a branch of black hole solutions is associated with each regular solution .
intriguingly , the two branches of black hole solutions from a pair of regular solutions merge and end at a maximal value of the horizon size . in the isolated horizon framework
@xcite , non - abelian black hole solutions can be interpreted as bound states of regular solutions and schwarzschild black holes @xcite . furthermore , the isolated horizon framework yields a relation for the mass of non - abelian black hole solutions , representing it as the sum of the mass of the regular solution and the horizon mass of the black hole solutions @xcite .
we here show that this relation is also valid for the new black holes .
in particular , the two regular solutions of a given @xmath1 and @xmath2 are connected via this mass formula . in section
ii we present the eym action , the axially symmetric ansatz and the boundary conditions . in section iii we address their asymptotic and horizon properties .
we present our numericl results in section iv , and we give our conclusions in section v.
we consider the su(2 ) eym action @xmath16 with ricci scalar @xmath17 , field strength tensor @xmath18 \ , \label{fmn}\ ] ] gauge potential @xmath19 , and gravitational and yang - mills coupling constants @xmath20 and @xmath21 , respectively .
variation of the action ( [ action ] ) with respect to the metric @xmath22 leads to the einstein equations , variation with respect to the gauge potential @xmath23 to the gauge field equations . in isotropic coordinates the static axially symmetric metric reads @xcite @xmath24 where the metric functions @xmath25 , @xmath7 and @xmath26 are functions of the coordinates @xmath27 and @xmath28 , only .
the @xmath29-axis ( @xmath30 , @xmath31 ) represents the symmetry axis .
regularity on the @xmath29-axis requires @xmath32 there . for the gauge field
we employ the ansatz @xcite @xmath33 \ . \label{gf1}\ ] ] here the symbols @xmath34 , @xmath35 and @xmath36 denote the dot products of the cartesian vector of pauli matrices , @xmath37 , with the spatial unit vectors @xmath38 respectively .
the gauge field functions @xmath39 , @xmath40 , depend on the coordinates @xmath27 and @xmath28 , only . for @xmath41 and @xmath42 , @xmath43 the spherically symmetric black hole solutions @xcite are recovered , while for @xmath3 , @xmath6 , one obtains the axially symmetric solutions of @xcite .
the new black hole solutions reported here are obtained for @xmath4 . in the limit of vanishing horizon radius
they converge pointwise to the globally regular solutions with the same integers @xmath1 and @xmath2 @xcite .
the globally regular solutions are related to eymh solutions with @xmath44 in the limit of vanishing higgs field @xcite .
the ansatz is form - invariant under the abelian gauge transformation @xcite @xmath45 we fix the gauge by choosing the gauge condition @xcite @xmath46 to obtain asymptotically flat solutions which are regular at the horizon and possess the proper symmetries , we need to impose appropriate boundary conditions @xcite .
the horizon of the non - abelian black hole solutions resides at a surface of constant radial coordinate @xmath47 . at the horizon
we impose the boundary conditions @xmath48 at infinity we impose @xmath49 and on the @xmath29-axis we impose @xmath50
we introduce the dimensionless coordinate @xmath51 and horizon radius @xmath52 . defining the mass @xmath53 of the black hole solutions via the komar integral , the dimensionless mass @xmath54 ,
is determined by the derivative of the metric function @xmath25 at infinity @xmath55 from the equations of motion it follows @xcite , that the kretschmann scalar is finite at the horizon , and that the surface gravity @xmath56 @xcite , @xmath57 is constant , as required by the zeroth law of black hole physics .
expansion of the metric functions near the horizon in the form @xmath58 where @xmath59 , yields the dimensionless surface gravity @xmath60 related to @xmath56 by @xmath61 .
we introduce the area parameter @xmath62 @xcite , defined via the dimensionless area of the black hole horizon @xmath63 , @xmath64 the deformation of the horizon is revealed , when the circumference of the horizon along the equator , @xmath65 , is compared to the circumference of the horizon along the poles , @xmath66 , @xmath67 since the black hole solutions have @xmath68 ( in general ) . the isolated horizon framework @xcite yields an intriguing relation between the adm mass @xmath69 of a black hole with area parameter @xmath62 and the mass @xmath70 of the corresponding globally regular solution @xcite , @xmath71 where the ( dimensionless ) horizon mass @xmath72 is defined via @xmath73 the isolated horizon formalism further suggests to interpret a non - abelian black hole as a bound state of a regular solution and a schwarzschild black hole @xcite , @xmath74 where @xmath75 is the adm mass of the schwarzschild black hole with area parameter @xmath62 , and @xmath76 represents the binding energy of the system , @xmath77
subject to the above boundary conditions , we solve the system of seven coupled non - linear partial differential equations numerically . to map spatial infinity to the finite value @xmath78 , we employ the radial coordinate @xmath79 the numerical calculations are based on the newton - raphson method , and are performed with help of the program fidisol @xcite .
the equations are discretized on a non - equidistant grid in @xmath80 and @xmath28 .
typical grids used have sizes @xmath81 , covering the integration region @xmath82 and @xmath83 .
we construct numerically the black hole solutions for ( @xmath3 , @xmath84 ) , ( @xmath13 , @xmath85 ) and ( @xmath14 , @xmath86 ) . for @xmath3 and fixed @xmath2 a branch of black hole solutions emerges from the single globally regular solution , when the horizon radius @xmath87 is increased from zero @xcite .
this branch of black hole solutions exists for arbitrarily large horizon size . when @xmath88 and @xmath89 , however , a pair of globally regular solutions exists for a given set @xmath0 .
thus we find two branches of black hole solutions emerging from the corresponding globally regular solutions , when the horizon radius is increased from zero .
surprisingly , the two branches do not exist for arbitrarily large horizon size .
instead they merge and end at a maximal value of the horizon size , i.e. , at the maximal value of the area parameter @xmath90 , which depends on @xmath1 and @xmath2 .
we show the adm mass @xmath69 as a function of the area parameter @xmath62 in fig . 1
. the mass increases monotonically with increasing @xmath62 . for the new solutions with @xmath88 , the maximal horizon size increases with increasing @xmath2 . at the same time
, the mass difference of the two regular solutions increases , indicating a possible correlation between the soliton mass difference and the maximal horizon size . considering the shape of the horizon , we observe that the deviation from spherical symmetry is small , though .
we observe a ratio of circumferences for @xmath3 solutions of up to @xmath91 , and for @xmath13 and @xmath14 solutions of up to @xmath92 and @xmath93 , respectively , as seen in fig . 2 .
thus the new black holes have a slightly more deformed horizon .
the inverse surface gravity @xmath94 of the black hole solutions is exhibited in fig .
3 . for @xmath3 black hole solutions the inverse surface gravity increases monotonically ( except for @xmath5 ) as a function of the area parameter @xmath62 . for @xmath13 and @xmath14 black holes it increases along the lower branch , and beyond the transition to the upper branch it decreases back to zero .
our numerical results indicate , that the mass relation ( [ ash ] ) , obtained in the isolated horizon formalism , also holds for the new non - abelian black hole solutions .
( in fig . 1 for each branch the corresponding regular solution is the reference point for the integration . )
consequently , the mass of the regular solution on the upper branch is related to the mass of the regular solution on the lower branch via the horizon mass integral , performed along both branches .
a similar result was obtained previously for eymh black holes with dipole hair @xcite . in @xcite the mass formula @xmath95
was derived and shown to hold for spherically symmetric eym solutions in @xcite .
we checked that the same formula also holds for the @xmath3 axially symmetric solutions . in fig .
4 we illustrate the binding energy of the non - abelian black hole solutions . the binding energy is always calculated w.r.t . the corresponding regular solution . consequently , for @xmath13 and @xmath14 black holes the binding energy along the upper branches is smaller than the binding energy along the lower branches .
the difference in binding energy of the black hole solutions for a given set of @xmath0 at the maximal value of the area parameter represents the mass difference of the corresponding pair of regular solutions .
we have constructed numerically new static axially symmetric black hole solutions of eym theory and investigated their properties .
these solutions are characterized by the set of integers @xmath0 , related to the polar and azimuthal angles , respectively . in particular , we have obtained black hole solutions for @xmath13 , @xmath85 , and @xmath14 , @xmath86 .
the @xmath96 solutions ( with one node ) exist most likely for any integer @xmath97 and for arbitrary horizon size .
the new @xmath98 and @xmath99 solutions have lower bounds on @xmath2 , @xmath100 , and upper bounds on the horizon size . for each
allowed set @xmath0 two branches of black hole solutions emerge from the two globally regular solutions and merge and end at the maximal value of the horizon size .
the existence of an upper bound of the horizon radius is a surprising and interesting new feature for eym solutions .
previously a maximal value of the horizon size was known only in theories with an in - built length scale already in minkowski space , hence it was conjectured , that `` there is no bound on the horizon radius of hairy , static eym black holes '' @xcite .
the underlying reason for the occurrence of this maximal horizon size for the new eym black holes is yet to be understood .
we expect , that the @xmath98 and @xmath99 regular and black hole solutions represent only the first sequences of new solutions , and conjecture the existence of @xmath0 regular and black hole solutions also for higher values of @xmath1 .
the ( sperically symmetric ) @xmath3 black holes are unstable @xcite , and there is all reason to believe , that the new @xmath0 solutions are unstable as well . considering the new static axially symmetric solutions from the isolated horizon formalism point of view
, we have verified the mass relation between the regular and the black hole solutions , showing that the black hole mass is given by the sum of the mass of the regular solution and the horizon mass .
in particular , the mass of the regular solution on the upper branch is related to the mass of the regular solution on the lower branch via the horizon mass integral , performed along both branches . interpreting the non - abelian black holes as bound states of regular solutions and schwarzschild black holes @xcite
, we have obtained the binding energy of these bound systems .
we note that the globally regular solutions were first obtained in eymh theory as limiting solutions for vanishing higgs field @xcite .
hence we expect that eymh black holes can be found from the new eym black holes by gradually switching on the higgs field . rotating eym black hole solutions based on the static @xmath3 black hole solutions are known @xcite .
it appears straightforward to construct rotating eym black hole solutions based on the static @xmath13 and @xmath14 black hole solutions .
in contrast , rotating regular eym solutions do not appear to exist @xcite , though recently the first rotating regular eymh solutions have been constructed @xcite .
n. straumann and z. h. zhou , phys . lett .
* b237 * ( 1990 ) 353 ; * b243 * ( 1990 ) 33 ; + m. s. volkov and d. v. galtsov , phys . lett . *
b341 * ( 1995 ) 279 ; + g. lavrelashvili , and d. maison , phys . lett .
* b343 * ( 1995 ) 214 ; + m. s. volkov , o. brodbeck , g. lavrelashvili and n. straumann , phys .
b349 * ( 1995 ) 438 .
b. kleihaus and j. kunz , phys .
* 86 * ( 2001 ) 3704 ; + b. kleihaus , j. kunz , and f. navarro - lrida , phys
* d66 * ( 2002 ) 104001 ; phys .
* 90 * ( 2003 ) 171101 ; phys . rev .
* d69 * ( 2004 ) 064028 ; phys .
lett . * b599 * ( 2004 ) 294 . | we construct new black hole solutions in einstein - yang - mills theory .
they are static , axially symmetric and asymptotically flat . they are characterized by their horizon radius and a pair of integers @xmath0 , where @xmath1 is related to the polar angle and @xmath2 to the azimuthal angle .
the known spherically and axially symmetric eym black holes have @xmath3 . for @xmath4 ,
pairs of new black hole solutions appear above a minimal value of @xmath2 , that increases with @xmath1 .
emerging from globally regular solutions , they form two branches , which merge and end at a maximal value of the horizon radius . the difference of their mass and their horizon mass equals the mass of the corresponding regular solution , as expected from the isolated horizon framework . |
the origin of fermion masses and mixings is an outstanding open problem in particle physics . in the standard model ( sm ) ,
the higgs mechanism is responsible for the electroweak symmetry breaking and mass generation via _ ad - hoc _
yukawa couplings .
there are reasons to believe that the sm is not the final model and a complete study of the coupling of the lightest boson , which we will call the higgs boson , to fermions can provide hints on new physics beyond the sm . in a recent letter @xcite
, we performed a realistic simulation of the process @xmath6 , where @xmath7 can be an electron , muon or tau neutrino , in the environment of a future linear collider with a center - of - mass energy of @xmath8 gev with an accumulated luminosity of 1 ab@xmath9 , based on the tesla design @xcite .
in particular , we noticed that weak gauge boson fusion is the dominant contribution to the subset of diagrams containing the higgs boson for @xmath10 gev at @xmath11 gev and hence this process is sensitive to the higgs boson couplings to @xmath5-quarks .
we have used a similar technique to investigated the possibility of detecting deviations from the sm in the higgs couplings to @xmath12-leptons at future @xmath13 colliders @xcite , which can be improved by using @xmath12 spin correlations @xcite . in this brief report
, we update the result @xcite by combining it with the results of a detailed analysis of the process @xmath3 .
this process has a very clean final state , with no missing energy and easy reconstruction , which can compensate for its smaller rate .
it certainly must be included in a global analysis of the @xmath0 vertex . in order to perform our analysis
, we will assume that the higgs boson has already been discovered at the large hadron collider and concentrate on the determination of its coupling to @xmath5-quarks . in extensions of the sm with extra scalars and pseudoscalars
, the lightest spin-0 particle can be an admixture of states without a definite parity .
hence , we parametrize the general @xmath0 coupling as : @xmath14 where @xmath15 gev , @xmath16 is the @xmath5-quark mass and @xmath17 , @xmath18 in the sm . in cp - violating extensions of the sm ,
deviations of these parameters may be generated at tree - level @xcite .
we will present results considering @xmath4 and @xmath5 as independent parameters and also for the cases of fixed @xmath19 , free @xmath5 and fixed @xmath20 , free @xmath4 .
there is a region of insensivity around circles in the @xmath21 plane since we ca nt at this level of analysis disentangle the effects of @xmath4 and @xmath5 . in the sm
the @xmath22 process is determined by 50 feynman diagrams .
only 2 of these ( higgs radiative production , @xmath23 , and vector boson fusion , @xmath24 ) can be considered as signal , where deviations from sm higgs couplings to @xmath5-quarks show up .
the remaining diagrams are not changed by new physics in the higgs sector .
the total sm cross section for the process @xmath22 is approximately @xmath25 fb for @xmath26 gev at @xmath8 gev , with cuts in the scattering angle between initial and final electrons@xmath27 , and in the final electron - positron invariant mass , @xmath28 gev .
the angular cut avoids the region of @xmath29 around the pipeline , used by the beam pipe itself and by the small angle monitor ( sam ) for luminosity measurements .
the invariant mass cut insures that the electrons and positrons will be detected in the lead glass calorimeter @xcite .
backgrounds from @xmath30 , @xmath31 , etc , can be reduced to the @xmath32 fb level by appropriate cuts@xcite . in our analysis
we will assume sm couplings of the higgs to the electroweak gauge bosons . a full analysis taking into account non
- sm hww and hzz couplings would introduce many more parameters and is beyond the scope of this paper .
we performed our monte carlo simulation by generating observables represented as series in the @xmath4 and @xmath5 couplings multiplied by kinematical factors : @xmath33 where @xmath34 is any observable and the @xmath35 terms are purely kinematical structures which do not contain any @xmath4 and @xmath5 dependence and results from the amplitude squaring and phase space integration . in this way
we only simulate the process once for each observable . in our case ,
terms linear in @xmath5 parameter vanish and hence @xmath36 .
we have studies the following observables : transverse @xmath5-quark momentum @xmath37 , @xmath38 invariant mass @xmath39 , @xmath40 , where @xmath41 is the scattering angle between the @xmath5-jet and initial beam directions , and the @xmath42 correlation , defined by @xmath43 .
the event sample reproducing the expected statistics at tesla was generated using our monte carlo package while the detector response was simulated with the code simdet version 3.01@xcite .
we assume an efficiency for @xmath44jet pair reconstruction of @xmath45 % , which is based on the @xmath5-tag algorithm , as assumed in ref . @xcite . in our simulations
we used @xmath46 gev . in figure [ distributions ]
we show , for comparison purposes , the differential distributions in the invariant @xmath38 mass and @xmath40 , for the total sm contribution and for the higgs contribution only ( including interference with sm ) . notice the @xmath47-boson peak and the smaller higgs peak in the @xmath39 distribution , and how the higgs contribution dominates around its peak .
cm = 1 -4.5 cm = 1 = 1 -5.5 cm in order to demonstrate the effect of different values of the parameters @xmath4 and @xmath5 , we show in figure [ ab ] the @xmath40 distribution arising from the higgs contribution for the sm ( @xmath19 , @xmath20 ) compared to the case with @xmath17 , @xmath48 .
we see that the shapes are very similar , as expected , but the levels can be noticeable different .
another important aspect is the assumption about the detector performance and possible sources of the systematic uncertaintes .
we include the anticipated systematic errors of 0.5% in the luminosity measurement , 1% in the acceptance determination , 1% in the branching ratios , and 1% in the background substraction , and assume the gaussian nature of the sytematics @xcite . to place bounds on the @xmath0 couplings , we use a standard @xmath49-criterion to analyse the events .
after various kinematical distributions were examined , we found that the most strict bounds are achieved from @xmath40 distribution by dividing the distribution event samples into 10 bins .
the experimental error @xmath50 for the @xmath51 bin is given by : @xmath52 where @xmath53 and @xmath54 is the sum in quadrature of the systematic uncertainties mentioned above .
below we present our final results for a tesla - like environment @xcite with a center - of - mass energy of 500 gev and for @xmath55 gev .
we investigated three possible scenarios for the luminosities : 100 fb@xmath9 , 1 ab@xmath9 and 10 ab@xmath9 .
the bounds that can be obtained at 95% confidence level from @xmath22 process on the @xmath56 and @xmath5 parameters are : @xmath57 for the case of @xmath20 and free @xmath58 and @xmath59 for the case of @xmath60 and free @xmath5 .
we will combine these limits with our previous bounds from the @xmath2 process according to the following procedure .
given two bounds on the same quantitity @xmath61 , say @xmath62 and @xmath63 , the combined bound will be @xmath64 and the statistical weights @xmath65 and @xmath66 are given by @xmath67 where @xmath68 and @xmath69 are the number of events for each process . with these prescription
we obtain the improved bounds : @xmath70 for the case of @xmath20 and free @xmath58 and @xmath71 for the case of @xmath72 and free @xmath5 .
these results can be roughly scaled for moderate variations in the higgs boson mass around @xmath73 gev by multiplying the bounds by a factor @xmath74 .
we have performed an update in our previous constraints on deviations of the @xmath0 coupling from its sm value by including a complete analysis of the sensitivity due to the process @xmath75 at the next generation of linear colliders .
these deviations are predicted by many extensions of the standard model .
we improved our previous bounds by roughly @xmath76% .
we showed that future @xmath77 linear collider experiments will be able to probe deviations of @xmath0 coupling .
the weak gauge boson fusion process is instrumental for achieving such a precision . for a tesla - like environment , we are able to constrain the couplings at the level of a few percent for the @xmath4 parameter ( for fixed @xmath5 ) and tens of percent for the @xmath5 parameter ( for fixed @xmath4 ) .
these results are comparable to the study performed in @xcite , where a global fit analysis for @xmath78 fb@xmath9 and @xmath8 gev has resulted in a relative accuracy of @xmath79% in the @xmath80 yukawa coupling . for comparison , the top quark yukawa coupling can be determined with a statistical accuracy of 16% at the lhc for @xmath81 gev @xcite .
in our analysis we can not disentangle the contributions from deviations in the @xmath4 and @xmath5 parameters .
however , notice that the partial width @xmath82 is proportional to @xmath83 .
our distributions have a different dependence @xmath84 therefore , if an independent measurement of @xmath82 is obtained ( for instance , from on - mass - shell higgs production in higgs - strahlung or in a muonic collider ) , one would be able to separate the @xmath4 and @xmath5 contributions and obtain an explicit indication of cp violation in the higgs sector .
the work of a. chalov and a. likhoded is supported by the russian foundation for basic research , grants ss-1303.2003.2 and 04 - 02 - 17530 , and the russian education ministry grant e02 - 3.1 - 96 .
r. rosenfeld would like to thank cnpq for partial financial support .
the authors would like to thank a. belyaev for valuable remarks .
# 1#2#3ann . rev .
nucl . part .
sci . * # 1 * ( # 3 ) # 2 # 1#2#3comments on nucl . part .
# 1 * ( # 3 ) # 2 89_cp violation , _ edited by c. jarlskog ( world scientific , singapore , 1989 ) # 1#2#3electronic conference proceedings * # 1 * , # 2 ( # 3 ) # 1#2#3eur .
j. c * # 1 * ( # 3 ) # 2 79_proceedings of the 1979 international symposium on lepton and photon interactions at high energies , _ fermilab , august 23 - 29 , 1979 , ed . by t. b. w. kirk and h. d. i. abarbanel
( fermi national accelerator laboratory , batavia , il , 1979 87_proceeding of the 1987 international symposium on lepton and photon interactions at high energies , _ hamburg , 1987 , ed . by w. bartel and r. rckl ( nucl .
b , proc .
3 ) ( north - holland , amsterdam , 1988 ) # 1#2#3 * # 1 * ( # 3 ) # 2 72_proceedings of the xvi international conference on high energy physics _ ,
chicago and batavia , illinois , sept . 6 13 , 1972 , edited by j. d. jackson , a. roberts , and r. donaldson ( fermilab , batavia , il , 1972 ) # 1#2#3int
. j. mod .
phys . a * # 1 * ( # 3 ) # 2 # 1#2#3jhep * # 1 * ( # 3 ) # 2 # 1#2#3j .
b * # 1 * ( # 3 ) # 2 # 1#2#3j .
g * # 1 * ( # 3 ) # 2 # 1#2#3mod .
lett . a * # 1 * ( # 3 ) # 2 # 1#2#3nature * # 1 * ( # 3 ) # 2 # 1#2#3nuovo cim .
* # 1 * ( # 3 ) # 2 # 1#2#3nucl .
a * # 1 * ( # 3 ) # 2 # 1#2#3nucl .
b * # 1 * ( # 3 ) # 2 # 1#2#3nucl .
suppl . *
# 1 * ( # 3 ) # 2 # 1#2#3nucl .
suppl . *
# 1 * ( # 3 ) # 2 # 1#2#3#4pisma zh .
* # 1 * ( # 3 ) # 2 [ jetp lett .
* # 1 * ( # 3 ) # 4 ] # 1#2#3phys .
# 1 * ( # 3 ) # 2 # 1#2#3phys .
lett . a * # 1 * ( # 3 ) # 2 # 1#2#3phys .
b * # 1 * ( # 3 ) # 2 # 1#2#3phys .
* # 1 * ( # 3 ) # 2 # 1#2#3phys .
c * # 1 * ( # 3 ) # 2 # 1#2#3phys .
d * # 1 * ( # 3 ) # 2 # 1#2#3phys .
* # 1 * ( # 3 ) # 2 # 1#2#3phys . rep . *
# 1 * ( # 3 ) # 2 # 1#2#3prog .
# 1 * ( # 3 ) # 2 # 1#2#3prog . part .
* # 1 * ( # 3 ) # 2 # 1#2#3rev .
* # 1 * ( # 3 ) # 2 # 1 rp # 1 9025th international conference on high energy physics , singapore , aug . 2 - 8 , 1990 # 1#2#3zeit .
c * # 1 * ( # 3 ) # 2 # 1#2#3zeit .
d * # 1 * ( # 3 ) # 2 | we perform an update of our previous analysis of the constraints on possible deviations of @xmath0 coupling parametrized as @xmath1 , arising from a scalar - pseudoscalar mixing , where the process @xmath2 was used . in this paper
we include a complete simulation of the process @xmath3 and combine these results to obtain tighter bounds on the deviations of the parameters @xmath4 and @xmath5 from their standard model values that could be measured at the next linear collider
. 25.0 cm pacs categories : 98.80.cq 2 |
in figure [ gapvsangle ] we plot the gap size as a function of fermi surface angle for the @xmath8 sample .
the value of gap , extracted from the symmetrized edc s shown in the article , is half of the peak - to - peak distance or the change in slope of the edc . because of the importance of proper orientation for the determination of the nodal velocity
, we developed an alignment protocol .
first , we map the complete fs of each sample in the first bz .
we define three angles , which can be manipulated , as shown in the inset of fig .
[ nnprocedure](a ) .
@xmath85 , @xmath86 , and @xmath87 @xmath88 define a nodal cut .
the red strait line represents the analyzer opening .
second , the angle @xmath89 is adjusted to give a symmetric spectrum .
third , we performed measurements with an intentional shifts @xmath90 and @xmath91 angle to ensure truly perfect alignment . from each measurement we extracted the @xmath92space distance @xmath93 between the two fermi points .
figure [ nnprocedure](a ) presents @xmath94 for @xmath95 variations in steps of 0.5 a degrees .
figure [ nnprocedure](b ) depicts @xmath96 for @xmath97 rotations again in steps of 0.5 degrees . due to the geometry of the fs
the nodal distance should be shortest when the alignment is perfect .
indeed , in both cases , the shortest distance was measured when @xmath98 .
this procedure was repeated for each and every measured sample .
.tight - binding coefficients and basis functions used to fit the experimental data .
the second column lists the coefficient of each term in ev for the @xmath6 and @xmath8 samples , following the convention : @xmath99 [ cols="<,<,<,>",options="header " , ] bernhard edegger , v. n. muthukumar , claudius gros , and p.w .
anderson , phys .
lett * 96 * , 207002 ( 2006 ) , 10.1103/physrevlett.96.207002 ; e. pavarini , i. dasgupta , t. saha - dasgupta , o. jepsen , and o. k. andersen , phys .
87 , 047003 ( 2001 ) , 10.1103/physrevlett.87.047003 .
x. j. zhou , t. yoshida , a. lanzara , p. v. bogdanov , s. a. kellar , k. m. shen , w. l. yang , f. ronning , t. sasagawa , t. kakeshita , t. noda , h. eisaki , s. uchida , c. t. lin , f. zhou , j. w. xiong , w. x. ti , z. x. zhao , a. fujimori , z. hussain and z .- x .
shen , nature * 423 * , 398 ( 2003 ) , doi:10.1038/423398a | we report angle - resolved photoemission spectroscopy ( arpes ) on two families of high temperature superconductors ( ca@xmath0la@xmath1)(ba@xmath2 la @xmath3)cu@xmath4o@xmath5 with @xmath6 ( @xmath7 k ) and @xmath8 ( @xmath9 k ) .
the fermi surface ( fs ) is found to be independent of @xmath10 or @xmath11 , and its size indicates extreme sample - surface overdoping .
this universal fs allowes the comparison of dynamical properties between superconductors of similar structure and identical doping , but different @xmath12 .
we find that the high - energy ( @xmath13 mev ) nodal velocity in the @xmath8 family is higher than in the @xmath6 family .
the implied correlation between @xmath14 and the hopping rate @xmath15 supports the notion of kinetic energy driven superconductivity in the cuprates .
we also find that the antinodal gap is higher for the @xmath8 family .
the recent synthesis of charge compensated ( ca@xmath0la@xmath1)(ba@xmath2la@xmath16)cu@xmath4o@xmath5 ( clblco ) single crystals facilitates an investigation of the relationship between their dynamical properties , such as the electronic dispersion relation @xmath17 and their thermodynamic property @xmath18 , while applying subtle crystal structure changes @xcite .
since the valence of ca and ba is equal , @xmath10 has a minute effect on crystal structure but a large effect on @xmath18 .
therefore , clblco allows experiments where the correlations between @xmath18 and a single parameter are explored .
experiments of such nature can reveal the mechanism for cuprates superconductivity . in the present work
, we measure the electron dispersion @xmath17 of two extreme samples of clblco crystals , using angle - resolved photoemission spectroscopy ( arpes ) , and look for correlations between properties of @xmath17 and @xmath18 . in particular , we focus on the nodal velocity . previously , similar studies could only be done by comparing cuprates with very different structures and levels of disorder @xcite .
clblco is similar to ybco in crystal structure , but has no oxygen chain ordering and is tetragonal for all @xmath10 and @xmath11 @xcite .
this simplifies the arpes interpretation .
while @xmath10 alters the calcium - to - barium ratio , the lanthanum content in the chemical formula remains constant .
we define four clblco `` families '' as samples with different @xmath10 , namely , @xmath19 .
the parameter @xmath11 signifies the oxygen level , which drives the system between different phases . by varying @xmath10 and @xmath11 in the chemical formula
, one can generate phase diagrams that are similar in shape yet differ in the maximum of @xmath18 , @xmath20 , and @xmath21 , and in the critical oxygen level at which the nature of the phase diagram changes . the phase diagram is presented in fig .
[ fig1](a ) @xcite .
it is worth noting that the only structural properties that vary with @xmath10 or @xmath11 are the cu - o - cu buckling angle , bond length , and cuo@xmath22 plane doping efficiency @xmath23 .
the crystallographic parameters were measured with powder neutron diffraction @xcite .
the buckling angle decreases by @xmath24 degrees as @xmath10 increases between families .
the bond length varies from @xmath25 for @xmath8 to @xmath26 at @xmath6 .
the doping efficiency is determined by in - plane @xmath27o nuclear quadrupole resonance ( nqr ) amitprb10 .
the variation in the number of holes on an oxygen @xmath28 is given by @xmath29 , where @xmath30 is defined as the doping at which @xmath21 starts to drop [ see fig .
[ fig1](a ) ] @xcite .
the super - exchange parameter @xmath31 for each clblco family was previously determined with muon spin rotation ( @xmath32sr ) ( magnetization ) versus temperature measurements @xcite and with two magnon raman scattering @xcite .
figure [ fig1](b ) depicts the super - exchange @xmath31 and glass temperature @xmath20 ( both from @xmath32sr ) , and @xmath18 , all normalized by @xmath14 , as a function of @xmath33 .
a universal phase diagram appears , demonstrating that @xmath14 scales like @xmath31 @xcite , which implies that @xmath14 is determined by the overlap of the orbital occupied by electrons on neighboring sites .
orbital overlaps also determine the hopping parameter @xmath34 , and the scaling of @xmath14 with @xmath31 meaning that kinetic energy controls the superconducting phase transition .
however , @xmath31 is determined in the afm phase , which is far , in terms of doping , from the superconducting phase .
a question arises : are the orbital overlaps important in the superconducting phase as well ? in this phase @xmath34 can be measured . here
, we extract @xmath34 from @xmath35 as the velocity in the nodal direction .
we find correlations between @xmath12 and @xmath34 , and confirm the famous relation @xmath36 eskes - tj .
this suggests that the band structure is rigid as a function of doping , as suggested by recent resonance inelastic x - ray scattering experiments @xcite . by the same token
, we also measure the antinodal gaps and compare them with hamiltonian parameters .
the arpes experiments were performed on the sis beam - line at the swiss light source on clblco single crystals .
these unique crystals were grown using the traveling floating zone method . a detailed discussion about growth and characterization of these crystals is given in @xcite .
for this experiment , samples with @xmath6 and @xmath8 were used .
the samples were mounted on a copper holder with silver glue to improve electrical conductivity .
the fermi level and resolution were determined from the polycrystalline copper sample holder .
the samples were cleaved _ in situ _ using a glued - on pin at @xmath37 k. circularly polarized light with @xmath38 ev was used .
the spectra were acquired with a vg scienta r4000 electron analyzer . despite a base pressure of @xmath39 torr ,
the samples surface life - time was only a few hours and a high intensity beam was required for quick measurements . as a consequence , the energy resolution in our experimental conditions was limited to @xmath40 mev . in fig .
[ fig1 ] , we present arpes data collected from clblco for the two samples : @xmath6 is presented in panels c , d , and e , while @xmath8 is depicted in panels f , g , and h. the data was collected at @xmath41 k and @xmath42 k for the @xmath6 and @xmath43 respectively .
all spectra are normalized by the measured detector efficiency . for each sample
, intensities along three cuts are shown .
the cuts are illustrated and numbered on the fermi surface ( fs ) drawing in the inset of fig .
[ fig1](b ) .
cuts numbered 1 and 2 are along @xmath44 ( @xmath45 direction ) .
these cuts allow better sensitivity to the gap size at the antinode . cut number 3 is along the diagonal line of the bz ( @xmath46 ) . in this configuration ,
a measurement of velocity in the nodal direction is possible .
the number on the bottom of each arpes panel indicates the cut from which data are collected . in figs .
[ fig1](c ) and [ fig1](f ) , spectra near the anti - node are plotted .
while @xmath6 shows high intensity spectra up to @xmath47 where no gap is visible , the @xmath8 sample shows a depletion of intensity close to @xmath48 , indicating a gap in the spectra at the antinode . for the @xmath6 , the gap ,
if one exists , is smaller than the experimental resolution . in figs .
[ fig1](d ) and [ fig1](g ) , we plot the intensity closer to the node . for both the @xmath6 and @xmath43 sample
, we clearly see the spectra crossing @xmath47 indicating a closed gap in the nodal region .
finally , in figs . [ fig1](e ) and [ fig1](h ) , both nodal cuts are seen , and again , the spectra cross @xmath47 , indicating an absence of a gap along the fermi arc for both samples .
the last panels also show a clear dispersion from which the nodal velocity is extracted .
-space at @xmath47 ( fs ) in clblco @xmath6 and @xmath8 sample , respectively .
the data were obtained at t=16 k for the @xmath6 sample and at t=11 k for the @xmath8 sample .
both samples were prepared with optimal doping and verified with a squid magnetometer .
the red curve is the fs of clblco obtained from tight binding fits to experimental data ( see text ) .
( c ) mdcs at zero binding energy along a nodal cut , for the @xmath6 ( black , fig .
[ fig1](e ) ) and @xmath49 ( red , fig .
[ fig1](h ) ) samples .
( d ) mdcs at zero binding energy along the antinodal cut , for the @xmath6 ( black , fig .
fig1(d ) ) and @xmath8 ( red , fig . [ fig1](g ) ) samples.,height=302 ] in fig .
[ fig2 ] , we show the fs in the first brillouin zone ( bz ) , for the two clblco samples : @xmath6 ( fig . [ fig2](a ) ) and @xmath43 ( fig .
[ fig2](b ) ) .
the fs was obtained by integrating 10 mev around the chemical potential .
the arpes intensity is displayed in a false color scale as a function of @xmath50 and @xmath51 . by comparing the shape of the fs
, we can see that the @xmath8 sample exhibits a fermi arc structure kanigel - arcs , which is typical for an antinodal gap . as for the @xmath6 sample ,
the arc is not present , and we observed strong intensity at the antinode , comparable to the intensity near the nodal region . unlike previously reported fs measurements of ybco @xcite , there is no apparent chain - like structure in the clblco fs , as expected .
the red line is a fit to a tight binding ( tb ) model up to three nearest neighbours hopping .
the fit parameters will be discussed below .
the fit for both fss gives the same size , as can be seen in fig.[fig2 ] .
in fact , the fs of a variety of samples was measured and found to be identical regardless of family ( @xmath10 ) or bulk oxygen level ( @xmath11 ) .
a clearer comparison of the fs size and doping between families can be obtained by examining the node - to - node and antinodal distances . in fig fig2(c ) , we show momentum distribution curves ( mdc ) at zero binding energy ( @xmath47 ) measured in a nodal cut ( `` cut 3 '' ) , for both samples .
the mdc for @xmath8 is sharper than for @xmath6 , but the peak - to - peak distance is equal for both mdcs .
similarly , figure [ fig2](d ) depicts an mdc measured in the antinode ( `` cut 2 '' ) at @xmath47 . here
, the mdc of @xmath6 is clearer than that of @xmath8 because of an open gap , but again the peak separation for both samples is identical .
we suspect that the bulk doping independence of the fs is due to the sample being cleaved on a charged plane , inducing surface charge reconstruction .
such behavior was previously reported from measurements of ybco @xcite . from the measured nodal peak - to - peak distance as a function of doping in ybco described in @xcite
, we can estimate the doping of our sample , which turns out to be @xmath52 .
this result is consistent with calculations based on the fs volume .
thus , we can conclude that the surface doping level of both samples is equal within the experimental error , and that the surface doping is on the edge of the superconducting dome on the overdoped side .
sample at @xmath53 as a function of the fermi surface angle @xmath54 , from the node ( top ) to the anti - node ( bottom ) .
the cuts are measured along the @xmath55 direction .
each curve is offset for clarity .
( b ) the same as ( a ) but for the @xmath6 sample.,height=302 ] to investigate the momentum dependence of the gap , we measured the dispersion along @xmath45 cuts between cut 1 and cut 2 for the @xmath56 and @xmath8 samples at a cold finger temperature of @xmath41 k and @xmath57 k , respectively . in fig .
[ fig3 ] , we plot symmetrized edc s at @xmath58 as a function of fs angle @xmath59 ( defined in the inset of fig fig1 ) .
for the @xmath8 sample ( fig .
[ fig3](a ) ) , one can see a zero - energy intensity peak close to the node ( @xmath60 ) .
in contrast , at an angle of @xmath61 and lower , we observe an opening of a gap , which grows up to @xmath62 mev at the antinode ( @xmath63 ) .
the angular dependence of the gap is shown in fig.1 of the supplementary material .
the gap value at the antinode is similar to optimally doped bi2212 gap - doping - bssco , mesot - bi2212gap and ybco @xcite . for the @xmath6 sample [ fig .
[ fig3](b ) ] , the situation is different .
close to the node , we observe a strong peak at zero energy ( @xmath64 ) . as we move to the antinode
, the intensity at zero energy is partly suppressed , but unlike the @xmath8 sample , there is no full depletion of spectral density at @xmath65 .
this indicates that a gap is not present in the @xmath6 sample , or that it is smaller than the experimental resolution ( @xmath66 mev ) .
a closed gap was measured with the same resolution for two more @xmath6 samples .
thus , we can safely say that @xmath67 .
last but not least , we compare the nodal velocity between families .
this study was performed on six @xmath6 and seven @xmath8 crystals .
the dispersion in the nodal direction , previously described in fig [ fig1](e , h ) , was measured for each sample in two branches with high statistics .
after an orientation procedure , which is described in the supplementary material , the peak positions in the mdc of each measured dispersion was extracted and plotted as a function of binding energy .
exemplary dispersions of two samples are shown in fig .
[ velocite ] .
an axis breaker is used in order to show the two branches .
the breaker emphasizes the differences between @xmath58 of the two samples , which in fact is very small .
two different linear regimes are observed .
the first regime involves low energies close to @xmath47 , between @xmath68 mev .
the second regime corresponds to high energies where @xmath69 mev .
the transition between these regimes is known as the kink and involves the electrons dispersion re - normalized due to correlations @xcite or coupling between electrons and low energy bosonic degrees of freedom lanzaranature01 .
the slope of the dispersion @xmath70 provides the velocity in the low ( @xmath71 ) and high - energy ( @xmath72 ) regimes .
the results are similar to other overdoped materials zhounature03 .
we did not find differences with statistical significance in @xmath71 between samples with different @xmath10 .
this is in agreement with previous work zhounature03 . as for @xmath72 ,
the results are summarized in the inset of fig .
[ velocite ] as histograms . for the @xmath6 family ,
the average high - energy velocity is @xmath73 ev@xmath74 , while for the @xmath8 it is @xmath75 ev@xmath74 . despite the velocity distribution overlap ,
the average velocities differ by 3.5@xmath76 , and hence are statistically different with 99.5% confidence . using these velocities we can now calculate all the tb coefficients by @xmath77 .
the unit cell parameter @xmath78 is nearly family independent @xcite .
the coefficient are presented in the supplementary material and are in agreement with previously published values @xcite@xcite .
, as a function of @xmath79 for @xmath6 ( black squares ) and @xmath8 ( red circles ) .
note the axis breaker .
the solid lines are a linear fit to the data in the @xmath69 mev range from which the high energy velocities @xmath80 are extracted .
inset : a histogram of high energy velocities obtained from a series of clblco samples with @xmath6 ( black , filled ) and @xmath8 ( red , crossed ) .
, height=264 ] from the data presented , we can draw several conclusions .
first , we discuss the ratio of velocities @xmath81 . despite the large error - bar ,
this ratio is very close to that expected from the ratio of the super - exchange @xmath31 between families .
this ratio is given by @xmath82 ( see fig . [ fig1 ] ) .
therefore , the @xmath36 relation is obeyed , and @xmath12 depends on orbital overlaps even when the measurements are done in the doped phase .
however , arpes measurements do not necessarily represent the bulk properties .
for example , the buckling angle might change close to the surface .
nevertheless , if such a thing happens in clblco , it might affect both families equally .
the fact that the ratio of @xmath31 measured magnetically agrees with the ratio of @xmath83 measure by arpes supports this notion .
second , we discuss the gap .
there are three possible scenarios that explain the difference in the gap size : i ) a scenario where disorder leads to broadening of the band structure features in @xmath6 which hide the gap .
however , high - resolution powder x - ray diffraction @xcite and nmr experiments @xcite indicate that @xmath6 samples are more ordered than @xmath8 ones .
ii ) a scenario where @xmath84 exist only below @xmath18 . it could be that in our experiment the surface of the @xmath8 sample is below @xmath18 , but the @xmath6 surface , is not since its @xmath18 is lower . in this case
only the @xmath8 sample will show a gap .
the problem with this scenario is the observation of a fermi arc in @xmath8 , which does not exist below @xmath18 in any other cuprate .
iii ) a scenario where both samples are above @xmath18 , but there is an intrinsic difference in their gap size . the problem here is again that in other materials there is no gap above @xmath18 in extreme overdoped samples @xcite .
further experiments are needed to clarify this point . in conclusion ,
we present the first arpes data from clblco .
we find that the surface doping is independent of the bulk doping or the ca to ba ratio .
we also demonstrate that the gap can be measured in this system .
the hopping parameter @xmath34 is larger for @xmath8 than for @xmath6 in the over - doped sides .
this suggests that @xmath14 is correlated with electron - orbital overlaps on neighboring sites .
this research was supported by the israeli science foundation ( isf ) and the joint german - israeli dip project . |
we have studied a class of cosmological models in which the universal scale factor grows as a power of the age of the universe ( @xmath9 ) and concluded that such models are not viable since constraints on the present age of the universe and from the magnitude - redshift relation favor @xmath10 , while those from the abundances of the light elements produced during primordial nucleosynthesis require that @xmath11 lie in a very narrow range around 0.55 @xcite .
successful primordial nucleosynthesis provides a very stringent constraint , requiring that a viable model simultaneously account for the observationally inferred primordial abundances of deuterium , helium-3 , helium-4 and lithium-7 .
for example , if the nucleosynthesis constraint is satisfied , the present universe would be very young ; @xmath12 for a hubble parameter @xmath13 ( or , requiring @xmath14 for @xmath15 ) .
recently , sethi @xcite noted that cosmologies where the scale factor grows linearly with time may produce the correct amount of he _ provided that _ the universal baryon fraction is sufficiently large . at first
this result might seem counter - intuitive since such a universe would have been very old at the time of big bang nucleosynthesis ( bbn ) suggesting that all neutrons have decayed and are unavailable to be incorporated in he . in fact , as sethi correctly pointed out , the expansion rate is so slow that the weak reactions remain in equilibrium sufficiently long to permit a simmering " synthesis of the required amount of he .
however , such an old universe also leaves more time to burn away d and 3 so that no astrophysically significant amounts can survive .
the observations of deuterium in high - redshift , low - metallicity qso absorbers @xcite , the observations of lithium in very old , very metal - poor halo stars ( the spite plateau " ) @xcite , and those of helium in low - metallicity extragalactic regions @xcite require an internally consistent primordial origin .
the sethi claim that deuterium could have a non - primordial origin is without basis as shown long ago by epstein , lattimer & schramm @xcite .
nevertheless , the paper of sethi @xcite prompted us to reinvestigate primordial nucleosynthesis in those power - law cosmologies which may produce interesting " amounts of he so as to study the predicted yields for d , 3 , and li .
preliminaries_. for a power - law cosmology it is assumed that the scale factor varies as a power of the age independent of the cosmological epoch : a / a_0 = ( t / t_0)^ = ( 1 + z)^-1 , [ abya0 ] where the subscript ` 0 ' refers throughout to the present time and ` @xmath16 ' is the redshift .
we may relate the present cosmic background radiation ( cbr ) temperature to that at any earlier epoch by @xmath17 , where @xmath18 accounts for any entropy production . for the models we consider ,
@xmath19 after electron - positron annihilation . the hubble parameter is then given by h==()^ .
[ hubble ] the second equality should be read with the understanding that it is not valid during the epoch of electron - positron annhilation due to the non - adiabatic nature of annhilations . power - law cosmologies with large @xmath11 share the common feature that the slow universal expansion rate permits neutrinos to remain in equilibrium until after electron - positron annihilation has ended so that neutrino and photon temperatures remain equal . in this case the entropy factor for @xmath20 in eq .
[ hubble ] is , = ( 29/43)^1/3 in contrast to the standard big bang nucleosynthesis ( sbbn ) value of @xmath21 . as increases , the expansion rate at a fixed temperature decreases due to the dominant effect of the @xmath22 power .
another useful way to view this is that at a fixed temperature , a power - law universe with a larger @xmath11 is older . as a consequence of the decreasing expansion rate ,
the reactions remain in equilibrium longer . in particular ,
as pointed out by sethi @xcite for the linear expansion model ( @xmath23 ) , the weak interactions remain in equilibrium to much lower temperatures than in the sbbn scenario , allowing neutrons and protons to maintain equilibrium at temperatures below 100 kev , as can be seen in fig .
[ equib ] . as is evident from fig .
[ equib ] , the he production rate below about @xmath24 is too slow to maintain nuclear statistical equilibrium .
however , the presence of neutrons in equilibrium and the enormous amount of time available for nucleosynthesis during neutron - proton equilibrium ( compared to sbbn ) make it possible to build up a significant abundance of he @xcite .
+ the above discussion is not restricted to @xmath23 , but applies for all values of @xmath11 which are sufficiently large ( so that the expansion rate is sufficiently small ) to allow neutrons to stay in equilibrium long enough to enable synthesis of he in sufficient amounts , as we show in fig .
[ iso - he4 ] .
although we explore a larger range in @xmath11 in this paper , we present detailed results for @xmath25 , a range consistent with the age and expansion rate of the universe , and we check these results for consistency with independent ( , non - bbn ) constraints on the baryon density . the iso - abundance contours in fig .
2 show clearly that as @xmath11 decreases towards 0.75 , a larger baryon density is required to produce the same abundance of he .
for example , although @xmath26 can be synthesized in the @xmath27 model , the density of baryons required is very large : @xmath28 .
these ` large @xmath29 ' models are constrained by dynamical estimates of the mass density , an issue we discuss later . + _
abundance_. in an earlier study @xcite , we showed that there is a very small region , centered on @xmath30 , for which the light elements can be produced in abundances similar to those predicted by sbbn .
but , this small window is closed by the snia magnitude - redshift data @xcite . here
we are concerned with larger values of @xmath11 and , correspondingly , larger baryon - to - photon ratios ( @xmath31 ) .
first we consider the nucleosynthesis of he in these models .
figure [ iso - he4 ] shows the connection between the baryon density ( @xmath32 , where @xmath33 ) and @xmath11 set by the requirement that the primordial helium mass fraction lie in the generous range @xmath34 .
we have included in fig .
[ iso - he4 ] the region investigated in ref .
@xcite , @xmath35 , as well . to understand the features in fig .
[ iso - he4 ] , we need to isolate the important factors controlling the synthesis of helium . in sbbn ,
the he abundance is essentially controlled by the number density ratio of neutrons to protons ( @xmath36 ) at the start of nucleosynthesis ( @xmath37 kev ) .
this ratio in turn is determined by ( 1 ) the @xmath38 ratio at ` freeze - out ' ( @xmath39 ) of the neutron - proton interconversion rates which may be approximated by @xmath40 , where @xmath41 is the neutron - proton mass difference and , ( 2 ) the time available for neutrons to decay after freeze - out , @xmath42 .
in contrast , for power - law cosmologies another factor comes into play
the time available for nucleosynthesis , @xmath43 , before the nuclear reactions freeze - out . for larger @xmath11 , the expansion rate of the universe ( at fixed temperature ) is smaller and the universe is older .
hence , for larger @xmath11 neutrons remain in equilibrium longer and the freeze - out temperature ( @xmath44 ) is smaller , so that @xmath45 is smaller .
however , the effect of the increase in @xmath43 as @xmath11 increases dominates that due to the change in @xmath44 . for @xmath46
the freeze - out temperature is around 4 mev whereas for @xmath30 , @xmath47 mev which implies a decrease in @xmath48 by a factor of about 2.5 . on the other hand ,
the age of the universe at @xmath49 kev ( about the temperature when sbbn ends ) is a factor of 25 larger for @xmath50 relative to that for @xmath46 .
thus , for the same @xmath31 , increasing @xmath11 from 0.50 to 0.55 has the effect of increasing the @xmath51he abundance because more time is available for nucleosynthesis .
but , since decreasing the baryon density decreases the nuclear reaction rates leading to a decrease in @xmath51he , we may understand the trend of the smaller baryon density requirement as @xmath11 increases from 0.50 to about 0.55 , even though the decrease in @xmath44 opposes this effect . the time - delay between ` freeze - out ' and bbn , @xmath52 , which has , until now , been much smaller than @xmath53 , becomes comparable to it at @xmath54 .
since a larger @xmath11 results in an older universe at a fixed temperature , @xmath52 increases with @xmath11 .
thus for @xmath55 , is increasingly suppressed ( exponentially ) as @xmath11 is increased .
the only way to compensate for this is by increasing @xmath56 ( since @xmath57 ) , which may be achieved by increasing the baryon density .
but since @xmath56 depends only logarithmically on the baryon density @xcite , this accounts for the exponential rise in the required value of @xmath58 as @xmath11 increases .
this trend can not continue indefinitely ; the curve must turn over for reasons we describe below . from fig .
[ iso - he4 ] , it is apparent that in the
large @xmath11 " range , the required value of @xmath59 decreases with increasing @xmath11 . in our previous analysis
@xcite of he nucleosynthesis which concentrated on @xmath11 in the vicinity of 0.55 , we implicitly assumed that the age of the universe at @xmath60 was not large enough for appreciable amounts of he to have been built up .
this assumption breaks down for large values of @xmath11 _ and _ @xmath31 . since d , 3 and 3
are not present in appreciable quantities , a large value of @xmath31 is needed to boost the he production rate .
now , the larger the value of @xmath11 , the longer neutrons remain in equilibrium , thus allowing more he to be slowly built up , with the neutrons incorporated in he being replaced via @xmath61 reactions . roughly speaking
, the required value of @xmath31 for a given @xmath11 is set by the condition : @xmath62_{t = t_f } \sim\;\ ; 0.24/t(t_f)\ , .
\label{req}\ ] ] the effects of @xmath11 on @xmath63 and @xmath31 on @xmath64 complement each other , giving rise to the trend shown by the he iso - abundance curves in fig .
[ iso - he4 ] for @xmath65 .
+ _ light element abundances in the linear expansion model_. we now turn to the production of deuterium and 3 . for large @xmath11 ( , @xmath23 )
, we expect the deuterium abundance to be insignificant since d can be efficiently burned to 3 during the long time available for nucleosynthesis .
the mean lifetime of deuterium against destructive collisions with protons at a low temperature of 10 kev is around 3 days ; at this temperature the @xmath23 universe is already 300 years old !
the fact that the timescales are so different allows us to derive analytical expressions for the deuterium , helium-3 and lithium-7 ( beryllium-7 ) mass fractions ( to be denoted by @xmath66 , @xmath67 and @xmath68 respectively ) . the generic equation for the rate of evolution of the mass fraction of nuclide
@xmath69 " can be parameterized as @xmath70 where `` prod '' and `` dest '' refer to the production and destruction rates of nuclide `` @xmath69 '' .
given that the universe remains at the same temperature for a very long time ( compared to the reaction time scales ) , it is not surprising that @xmath71 achieves its steady - state value at each temperature ( for a detailed discussion in the context of sbbn , see @xcite ) , @xmath72 we can write this explicitly for the simplest case deuterium : @xmath73 where the various @xmath74s represent the relevant deuterium creation ( @xmath75 and @xmath76 ) rates per target proton , and destruction ( @xmath773 and @xmath78 ) rates per target deuterium .
all of these rates can be obtained from ref .
once the reaction rates become smaller than the universal expansion rate ( say at some temperature @xmath79 ) , the abundances freeze out with values close to @xmath71 at the corresponding @xmath79 .
this is illustrated in fig .
[ analytic ] which clearly shows that the steady - state solution works very well .
we note here that the steady state ( dotted ) curves in fig .
[ analytic ] are not independent analytic derivations , but use the abundances of the various nuclei as calculated by the numerical code .
the figure intends to emphasize that nucleosynthesis in this ( linear expansion ) model can be well represented by the steady - state solutions in eq .
[ xdx3x7 ] . + in the expression for @xmath66 ( see eq .
[ xd ] ) , the @xmath80 reaction term dominates until about 20 kev after which the @xmath81 reaction makes the dominant contribution .
the final deuterium abundance is thus determined by the weak pp reaction ( @xmath76 ) , the effect of which can be seen in fig .
[ alpha1 ] as the very slow rise in @xmath66 between temperatures of 10 kev and 1 kev ( at which point the d abundance freezes out ) .
since both @xmath82he and @xmath83li freeze out much earlier , they do not get any significant boost from the weak pp reaction . from eq .
[ xd ] , @xmath66 , and thus @xmath67 ( @xmath82he is formed from @xmath84 ) , are proportional to @xmath85 , the neutron abundance .
one striking feature in fig .
[ alpha1 ] is the boost to the neutron abundance ( and hence the abundances of d and @xmath82he ) at temperatures around 40 kev .
the effect is subtle and may be missed in bbn codes with a limited nuclear reaction network .
the slow rate of expansion of the universe during nucleosynthesis facilitates the production of a relatively large metal " ( a@xmath86 ) abundance ( @xmath87 ) .
in particular , @xmath88c is produced in these models through the chain : @xmath89c+p@xmath90 @xmath88n+@xmath91 and the subsequent beta - decay of @xmath88n . in this environment @xmath88c+@xmath51he@xmath92@xmath93o+@xmath94 leads to the production of free neutrons .
the mass-7 abundance is entirely due to the production of @xmath83be through the reaction @xmath82he+@xmath51he@xmath92@xmath83be+@xmath91 .
@xmath83be decays to @xmath83li by electron capture once the universe has cooled sufficiently to permit the formation of atoms . once formed , it is difficult to destroy @xmath83be at temperatures @xmath95 kev .
in contrast , @xmath83li is very easily destroyed , specifically through its interaction with protons .
since the @xmath83be production ( and thus li ) follows the evolution of the @xmath82he abundance , and there is very little destruction of @xmath83be , @xmath83li also benefits from the boost to the neutron abundance described in the last paragraph .
this has the effect of boosting the li abundance from @xmath96 ( if this source of neutrons were not included ) to @xmath97 .
this is significant in that , at the level of a few parts in 10@xmath98 ( , @xcite ) , the primordial lithium abundance lies between these two estimates .
+ _ light element abundances vs. @xmath11_. having explored bbn in the linear model ( @xmath99 ) it is now important to ask how these results depend on @xmath11 .
it is clear from fig .
[ abund ] that nothing dramatically different happens as @xmath11 changes ; this is simply because the key physics remains the same . in preparing fig .
[ abund ] we adjust the value of @xmath31 ( baryon density ) for each choice of @xmath11 so that the primordial he mass fraction lies between 22% and 26% . as @xmath11 increases , the nuclei freeze out at lower temperatures since the expansion rate at the same temperature is lower for a larger @xmath11 .
the effect of this can be gauged by the behavior of @xmath66 , @xmath67 and @xmath68 with respect to temperature , as given by eq .
[ xdx3x7 ] . for deuterium
this implies a small increase with @xmath11 due to the pp ( weak ) reaction , which is also reflected in the behavior of @xmath82he for @xmath100 .
the fall of @xmath82he with increasing @xmath11 for @xmath101 is due to larger destruction of @xmath82he because of the increase in the time available for the nuclear reactions .
as already mentioned , the abundance of @xmath83be depends critically on the evolution of the @xmath82he abundance ; so while the mass-7 ( @xmath83be ) abundance increases appreciably with the increase in baryon density , it is relatively unaffected by a change in @xmath11 .
note that in those power - law models which can simultaneously reproduce an acceptable he abundance along with a consistent age and expansion rate , the corresponding baryon density must be very large , @xmath102 ( @xmath103 ; see fig . 2 ) .
most if not all of this range is far too large for consistency with independent ( non - bbn ) estimates of the universal density of baryons ( @xmath104 @xcite ) or , for that matter , the total matter density @xcite .
conservatively , clusters limit the total ( gravitating ) matter density to @xmath105 so that if there were no non - baryonic dark matter , @xmath106 ( @xmath107 ) for @xmath108 . however , if the x - ray emission from clusters is used to estimate the cluster baryon fraction ( see @xcite ) , the universal baryon density should be smaller than this very conservative estimate by a factor of 7 - 8 ( consistent with the upper bound from the baryon inventory of fukugita , hogan and peebles @xcite ) .
thus power - law cosmologies constrained to reproduce he ( only ) , an acceptable age and magnitude - redshift relation , and an acceptable baryon density , must have @xmath11 restricted to a very narrow range : @xmath109 .
furthermore , the baryon density in even this restricted range is large when compared with estimates @xcite of the baryon density from cluster x - rays .
finally , for @xmath11 in the narrow range of @xmath109 and @xmath110 , the other light element abundances are restricted to li / h @xmath111 , 3/h @xmath112 , and d / h @xmath113 . for deuterium and helium-3
this is in very strong disagreement ( by 8 to 13 orders of magnitude ! ) with observational data ( for a review see @xcite ) .
although the predicted @xmath83li abundance is comparable to that observed in the solar system , the local ism , and in stars , it is larger than the primordial abundance inferred from the halo stars @xcite , and marginally inconsistent with the observations of lithium in the ism of the lmc @xcite .
in response to the claim @xcite that a power - law universe expanding linearly with time could be consistent with the constraints on bbn , we have reexamined these models .
although it is true that observationally consistent amounts of he can be produced in these models , this is not the case for the other light elements d , 3 , li .
furthermore , consistency with he at @xmath99 requires a very high baryon density ( @xmath114 or @xmath115 ) , inconsistent with non - bbn estimates of the universal baryon density and , even with the total mass density .
we have also investigated bbn in power - law cosmologies with @xmath116 and have confirmed that although the correct he abundance can be produced , the yields of the other light elements d , 3 , and li are inconsistent with their inferred primordial abundances . in general ,
power - law cosmologies are unable to account simultaneously for the early evolution of the universe ( bbn ) ( which requires @xmath117 ) and for its presently observed expansion ( which requires @xmath118 ) .
m. kaplinghat , g. steigman , i. tkachev , t.p .
walker , phys .
d * 59 * ( 1999 ) 043514 .
m. sethi , a. batra , d. lohiya , phys .
d * 60 * ( 1999 ) 108301 .
r. i. epstein , j. m. lattimer , and d. n. schramm , nature , * 263 * ( 1976 ) 198 .
s. burles and d. tytler , ap.j . , * 499 * ( 1998 ) 699 ; _ ibid _ * 507 * ( 1998 ) 732 .
f. spite and m. spite , a.a .
, * 115 * ( 1982 ) 357 ; l. hobbs and j. thorburn , ap.j .
lett . , * 428 * ( 1994 ) l25 ; p. bonifacio and p. molaro , mnras . *
285 * ( 1997 ) 847 ; s. ryan , j. norris , and t. beers , astro - ph/9903059 .
k. a. olive and g. steigman , ap.j
* 97 * ( 1995 ) 49 ; k. a. olive , e. skillman , and g. steigman , ap.j . , * 483 * ( 1997 ) 788 ; y. i. izotov and t. x. thuan , ap.j .
* 500 * ( 1998 ) 188 .
f. pont , m. mayor , c. turon , and d. a. vandenberg , a.a . * 329 * ( 1998 ) 87 .
m. garnavich , ap.j .
lett . , * 493 * ( 1998 ) l53 ; s. perlmutter , nature * 391 * ( 1998 ) 51 . n. bahcall , l. m. lubin , and v. dorman , ap.j . ,
* 447 * ( 1995 ) l81 .
e. w. kolb , and m. s. turner , _ the early universe _ , addison - wesley publishing company ( 1990 ) r. esmailzadeh , g. d. starkman , and s. dimopoulos , ap.j .
, * 378 * ( 1991 ) 504 .
r. rood , g. steigman , and b. m. tinsley , ap.j .
, * 207 * ( 1976 ) l57 ; d. s. p. dearborn , g. steigman , and m. tosi , ap.j .
* 465 * ( 1996 ) 887 .
r. t. rood , t. m. bania , and t. l. wilson , nature * 355 * ( 1992 ) 618 ; d. s. balser , t. m. bania , r. t. rood , and t. l. wilson , ap.j . , * 483 * ( 1997 ) 320
. m. bureau , j. r. mould , and l. staveley - smith , ap.j . , * 463 * ( 1996 ) 60 ; t. kundi , et al . , ap .
j. * 482 * ( 1997 ) 75 ; g. a. tammann , and m. federspiel , in _ the extragalactic distance scale _
m. livio , m. donahue , and n. panagia ( cambridge : cambridge univ . press ) ( 1997 ) 137 ; j. l. tonry , j. p. blakeslee , e. a. ajhar , and a. dressler , ap . j. * 475 * ( 1997 ) 399 .
w. a. fowler , g. r. caughlan and b. a. zimmerman , annu .
astrophys . , * 13 * ( 1975 ) 69 .
m. persic and p. salucci , mnras , * 258 * ( 1992 ) 14p ; m. fukugita , c. j. hogan , and p. j. e. peebles , ap.j . ,
* 503 * ( 1998 ) 518 .
g. steigman , n. hata , and j. e. felten , ap.j .
* 510 * ( 1999 ) 564 .
k. a. olive , g. steigman , and t. p. walker , phys .
rep . to appear , astro - ph/9905320 ( 1999 ) .
g. steigman , ap.j .
* 457 * ( 1996 ) 737 . | we have recently considered cosmologies in which the universal scale factor varies as a power of the age of the universe and concluded that they can not satisfy the observational constraints on the present age , the magnitude - redshift relation for sn ia , and the primordial element ( d , 3 , he , and li ) abundances .
this claim has been challenged in a proposal that suggested a high baryon density model ( @xmath0 ) with an expansion factor varing linearly with time could be consistent with the observed abundance of primoridal helium-4 , while satisfying the age and magnitude - redshift constraints . in this paper
we further explore primordial nucleosynthesis in generic power - law cosmologies , including the linear case , concluding that models selected to satisfy the other observational constraints are incapable of accounting for _ all _ the light element abundances .
epsf # 1#23.6pt 3@xmath1 3@xmath2 he@xmath3 li@xmath4 li@xmath5 7@xmath6 4n@xmath7 10@xmath8 |
over the last years , the study of nonperturbative phenomena in string theory has enjoyed a lively interest .
research in this direction revealed that string field theory is the language in which , e.g. , the decay of unstable d - brane systems can be described most naturally ( for a review of the subject , see @xcite ) . here , the endpoint of the decay process corresponds to a solution of the equation of motion of the string field theory on the unstable brane . up to now
, however , no satisfactory solutions to this equation have been found even in the simplest setting , i.e. , witten s formulation of bosonic string field theory @xcite mainly , due to the following two difficulties : in this theory , the brst operator @xmath4 plays the role of the kinetic operator ; it involves matter as well as ghost fields which renders it impossible to study the equation of motion in each sector separately . as a loophole , rastelli , sen , and
zwiebach proposed to consider string field theory around the tachyon vacuum and to describe d - branes from this point of view @xcite .
it could be shown that , after a singular reparametrization of the worldsheet , the kinetic operator consists solely out of ghosts @xcite .
thus , the equation of motion factorizes into a matter and a ghost part , the matter part being simply the condition that the matter string field is a projector of the string field algebra . amongst other things ,
this discovery triggered an investigation of structural aspects of the string field algebra .
the second difficulty arises from the star product which is used to multiply string fields @xcite .
calculations in the customary discrete oscillator basis @xcite are notoriously difficult ; however , they may be facilitated by the fact that it is possible to choose a basis in the string field theory fock space in which the neumann matrices of the interaction vertex are diagonal @xcite .
in particular the analysis of algebraic properties of the string field algebra is simplified in this continuous moyal basis .
e.g. , it is possible to classify all twist - even projectors in certain conformal field theories which are diagonal in this basis @xcite ( see also @xcite ) .
this classification includes many states without immediate geometrical interpretation .
furthermore , due to its similarity with the moyal - weyl product , the algebraic approach to the star algebra could be useful for transferring solution - generating techniques from field theory to string field theory ( see , e.g. , @xcite ) .
alternatively , the star product can be computed via conformal field theory methods @xcite .
they lead to a more geometric understanding of states and their multiplication in string field theory ; in particular , surface states play an important role in the structural analysis of the star algebra .
these are states with field configurations arising from path integrations over fixed riemann surfaces whose boundary consists of a parametrized open string and a piece with open string boundary conditions .
a very fundamental result in this realm @xcite was that all such surfaces with the property that their boundaries touch the midpoint of the open string lead to projector functionals in the star algebra .
although this class of projectors is quite large , it may be rewarding to look for projectors without this singular property which could eventually lead to d - brane solutions with finite energy densities .
both methods obviously have their own advantages , and it usually depends on the respective case which one is given preference .
however , it seems that it should be possible to make rather strong statements concerning states which can be analyzed particularly well with both methods , i.e. , those which are diagonal in the moyal basis and simultaneously ( twist - even ) surface states .
the analysis of these states is the aim of the present note .
it turns out that a complete classification of such states is possible ; they comprise the butterfly family and a generalization of the wedge states ( with arbitrary angles ) . upholding the belief that possible solutions to the equation of motion of string field theory should have a geometrical interpretation as surface states , the additional demand that the states be simultaneously diagonal in the @xmath1-basis ( and in this sense particularly easy to handle algebraically ) apparently is very strict . in principle
, one could have imagined that this class included hitherto unknown states which could have served as candidate solutions to the equation of motion .
our result seems to imply that we have to look elsewhere for solutions describing the tachyon vacuum .
apart from that , we find that surface state projectors which do not satisfy the sufficient condition that their boundary touches the midpoint of the open string , are particularly rare at least in the subsector under consideration , we only encounter the identity state .
this result is obtained in three steps : first , we derive a consistency condition on @xmath1-diagonal surface states along the lines of earlier work @xcite .
this is done in terms of a fermionic first order system with weights 0 and 1 ( but is in principle possible in any other conformal field theory ) and is the subject of section [ sec : cond ] , where we also present the moyal basis for this system . from the technical point of view
, the consistency condition is a differential equation with deviating arguments on the map @xmath5 which defines a possible surface state .
second , we solve this differential equation ; the solution turns out to be a jacobi sine function parametrized by two complex parameters .
this is the task of section [ sec : solvcc ] .
third , we examine further constraints on these two parameters in section [ sec : kdiagss ] since not all solutions to the consistency condition are eligible as surface states .
this leads to the above - described result .
the necessary background on jacobian elliptic functions for this analysis is presented in appendix [ sec : jac ] . finally , we offer some concluding remarks in section [ sec : concl ] .
in this section , we briefly review the derivation of a consistency condition on twist - even surface states which are diagonal in the @xmath1-basis .
most of the material presented here is a summary of the relevant facts in @xcite which are included in this note in order to make the discussion self - contained .
the derivation starts from an oscillator representation of the surface state and leads to a requirement on the maps @xmath5 defining the surface state ; for consistency with the treatment in @xcite , the argument will be presented for a fermionic first order system with weights @xmath6 , i.e. , a twisted @xmath7 system .
naturally , the same condition on @xmath5 can be derived for any given conformal field theory . * continuous basis in the fermionic first order system . *
the fermionic first order system consists of grassmann - odd fields @xmath8 with weight 0 and @xmath9 with weight 1 . and @xmath10 , respectively .
these fields occur as twisted worldsheet fermions in string field theory for n=2 strings .
their conformal field theory coincides with the one common from the construction of solutions to the ghost part of the equations of motion in vacuum string field theory @xcite ; in this context , the fields are customarily denoted by @xmath11 and @xmath12 .
apart from that , the neumann coefficients @xmath13 are rescaled by a factor of 2 w.r.t .
] as fields of integral weight , both @xmath8 and @xmath9 are integer - moded ; the modes satisfy @xmath14 . in particular , the spin-0 field @xmath8 has a zero - mode on the sphere . in analogy to the ( untwisted ) @xmath7 system
there are thus two vacua at the same energy level : the bosonic @xmath15-invariant vacuum @xmath16 is annihilated by the virasoro modes @xmath17 and @xmath18 , @xmath19 ; its fermionic partner , @xmath20 , is annihilated by @xmath21 , @xmath22 . to get nonvanishing fermionic correlation functions , we need one @xmath8-insertion , i.e. , @xmath23 , @xmath24 . in terms of these oscillators and vacua , the interaction vertex for the @xmath25 system takes the form @xcite ( see also @xcite ) @xmath26
\label{eq : psivrtx}\ ] ] where the explicit form of the neumann coefficients @xmath13 ( @xmath27 ) can be found in @xcite .
the ( common ) eigenvectors of these neumann matrices are labeled by a parameter @xmath28 ; the coefficients of the right and left eigenvectors @xmath29 and @xmath30 are given by the generating functions @xmath31 with @xmath32 and @xmath33 .
one can show that @xmath34 which can be split into even and odd parts to give @xmath35 for @xmath36 .
thus , the continuous modes @xmath37 satisfy the anticommutation relations @xmath38 following from the anticommutation relations of @xmath39 and @xmath40 and the completeness relations of the eigenvectors @xmath41 and @xmath42 . the relations ( [ eq : contbas1 ] ) and ( [ eq : contbas2 ] ) can be inverted : @xmath43 the interaction vertex ( [ eq : psivrtx ] ) can be rewritten in the continuous basis ; one can take advantage of this fact , e.g. , in order to classify all ( twist - even ) projectors diagonal in the @xmath1-basis @xcite .
* diagonal squeezed states . * in this paper , we restrict the discussion to squeezed states with definite charge
w.r.t.the @xmath44 current @xmath45 of the first order system because only these lead to surface states ( recall that this is so since the virasoro generators are neutral ) . in the discrete basis ,
such a state takes the form @xmath46 \ , ; \label{eq : sqstdo}\ ] ] in the continuous basis , it features two @xmath1-integrations in the exponent .
the @xmath1-diagonal states are given by a squeezed state matrix @xmath47 which commutes with the neumann matrices @xmath13 ; in these cases , the exponent can be reduced to contain only one @xmath1-integration .
namely , for @xmath47 , the twist properties of @xmath48 , @xmath49 imply that the even and odd parts are also eigenvectors of @xmath47 , @xmath50 here , @xmath51 , @xmath52 , @xmath53 , and @xmath54 , denote the corresponding eigenvalues . and
@xmath54 have to be even functions of @xmath1 , and @xmath53 and @xmath52 have to be odd functions of @xmath1 . ] thus , eq .
( [ eq : cmpl ] ) guarantees that @xmath55 and similar relations hold for the other components . if one rewrites @xmath56 in terms of the continuously moded operators , the delta distributions on the right - hand side of eq .
( [ eq : vsv ] ) can be used to remove the @xmath57-integration .
therefore , @xmath58 \label{eq : sqstate}\ ] ] with @xmath59 .
* surface states . *
a surface state @xmath60 is determined by a map @xmath61 from the canonical upper unit half disk @xmath62 onto a riemann surface @xmath63 with boundary and the requirement that @xmath64 for all fock space states @xmath65 of weight @xmath66 in the boundary cft with corresponding operators @xmath67 .
the correlation function @xmath68 is evaluated on @xmath63 , and @xmath69 is the conformal transform of @xmath70 by the map @xmath71 .
now , let us evaluate the correlation function ( [ eq : surfstdef ] ) for a surface state of the form ( [ eq : sqstdo ] ) with @xmath72 , charges of the insertions of the correlation functions have to sum up to @xmath73 in order to give a nonvanishing result . ]
@xmath74 \sum_{k=-\infty}^\infty \frac{\ps^+_k}{z^k}\sum_{l=-\infty}^{-1 } \frac{\ps^-_l}{w^{l+1}}{\text{$|\;\!\!\!\downarrow\>$}}= \sum_{m , n>0 } s_{mn } z^n w^{m-1 } + \frac{1}{z - w}\frac{z}{w}\ , .\ ] ] on the other hand , according to eq .
( [ eq : surfstdef ] ) , this correlation function should be equal to @xmath75 and we can use @xmath15 invariance to fix @xmath76 , @xmath77 , and @xmath78 . then , the nonsingular part of the correlator becomes @xcite @xmath79 therefore , deriving w.r.t .
@xmath80 and choosing @xmath81 , @xmath82 this equation can be integrated to give @xmath83 for a candidate function @xmath71 for the map defining the surface state @xmath84 . here ,
the integration constant was chosen in such a way that for @xmath85 , we obtain the identity map @xmath86 .
obviously , the defining map @xmath5 is encoded in the coefficients @xmath47 via ( [ eq : szw ] ) ; we can extract a candidate function with the help of eq .
( [ eq : fs ] ) . for a diagonal surface state
, the same information can be equivalently obtained from the eigenvalues @xmath51 , @xmath52 , @xmath53 , and @xmath54 . for our purposes
we restrict to twist - even states , i.e.states with @xmath87 .
now , @xmath88 can be reconstructed from these eigenvalues using ( [ eq : orth ] ) , ( [ eq : eogenf1])([eq : eogenf4 ] ) , and ( [ eq : vsv ] ) , @xmath89\ , .
\label{eq : szw2}\ ] ] it should be noted that @xmath5 is purely given by the @xmath54-part , see ( [ eq : fs ] ) and @xmath90 as above , @xmath91 abbreviates @xmath92 .
obviously , @xmath71 is an _
function mapping the upper unit half - disk into some region of the upper half - plane .
* consistency condition for surface states .
* we will now derive a necessary condition on the maps @xmath5 defining a surface state .
this condition consists in a differential equation with deviating arguments , which , however , turns out to be exactly solvable .
in this way , we will show that all twist - even surface states belong to a two - parameter class of states .
not all functions in this two - parameter family define surface states ; further restrictions single out the butterfly family and the wedge states with arbitrary angles ( including the identity ) .
the starting point for the derivation of the condition is the observation that the @xmath54-part of eq .
( [ eq : szw2 ] ) is given by @xmath93 and the @xmath51-part can be determined via ( an integral of ) @xmath94 .
thus , the full integral ( [ eq : szw2 ] ) can be computed via @xmath5 .
together with ( [ eq : szw ] ) , this gives a restriction on @xmath5 . to begin with ,
note that ( [ eq : szw2 ] ) can be rewritten as @xmath95\ , , \label{eq : szwsplit}\ ] ] again with @xmath96 , @xmath97 , and @xmath98 obviously , @xmath99 .
next , it is easy to see that due to the form of ( [ eq : szw2 ] ) , only the @xmath51-part will contribute to @xmath100 ; in addition , one can recover @xmath101 fully from @xmath102 with some additive constant @xmath103 .
( [ eq : szwsplit ] ) ensures that the contributions of this constant to @xmath88 cancel . following this prescription
, we have to compute @xmath104 from ( [ eq : szw ] ) . a direct computation with our choice @xmath76 , @xmath77
, @xmath78 leads to a difference of singularities .
in order to avoid this one can compute @xmath105 , set @xmath81 , and integrate with respect to @xmath80 .
if one absorbs the integration constant of this latter integration into @xmath103 , the result is @xmath106 with @xmath107 putting everything together , we obtain from ( [ eq : szwsplit ] ) : @xmath108 this expression should be compared with eq .
( [ eq : szw ] ) .
finally , a somewhat messy calculation leads to the following result : @xmath109 this condition restricts the allowed maps @xmath5 .
a short computation shows that it is indeed satisfied for the generalized butterfly states with @xmath110 as well as the wedge states with @xmath111 . by a differentiation w.r.t .
@xmath91 and an integration over @xmath112 , eq .
( [ eq : surstcond ] ) can be transformed into @xmath113 the integration constant is fixed by the initial conditions @xmath114 , @xmath115 , and @xmath116 . in this formulation , the left - hand side of eq .
( [ eq : surstcond ] ) is a bosonic two - point function .
it is easy to see that condition ( [ eq : surstcond2 ] ) is a refined version of the surface state condition ( 3.40 ) in @xcite , cf . also eq .
( 3.30 ) in this reference . from the mathematical point of view
, this condition is a differential equation with deviating arguments .
in this section , we will solve the surface state condition ( [ eq : surstcond2 ] ) .
it will turn out that the general solution is given by the jacobi sine function .
however , not all maps in the resulting two - parameter family of solutions define admissible surface states .
* equivalent ode .
* the general technique will be to derive an ordinary differential equation from ( [ eq : surstcond2 ] ) sharing all solutions of the consistency equation .
it is possible to obtain two different differential equations with ( deviating ) arguments @xmath91 and @xmath117 from eq .
( [ eq : surstcond2 ] ) ; from the combination of both , we obtain the desired ordinary differential equation . on the one hand , by taking into account that @xmath118 is an odd function , the substitution @xmath119 in ( [ eq : surstcond2 ] ) yields @xmath120 the sum of eqs .
( [ eq : surstcond2 ] ) and ( [ eq : surstcond3 ] ) can be integrated with respect to @xmath91 and @xmath112 to give , respectively : @xmath121 now , the sum of eqs .
( [ eq : dewdahlp1 ] ) and ( [ eq : dewdahlp2 ] ) can be transformed into @xmath122 the limit @xmath123 in this equation leads to a further differential equation with deviating arguments : @xmath124 this is the first differential equation with arguments @xmath91 and @xmath117 . on the other hand , the limit @xmath125 in eq .
( [ eq : surstcond2 ] ) yields the second equation of this kind , @xmath126 inserting eq .
( [ eq : dewdaz2z_1 ] ) into eq .
( [ eq : dewdaz2z_2 ] ) , we obtain the desired ordinary differential equation : @xmath127 this is consistent for @xmath128 .
one can already read off at this point that given a solution @xmath118 to ( [ eq : eqvode ] ) , @xmath129 is also a solution for arbitrary @xmath130 ( with @xmath131 since @xmath115 ) obeying the same initial conditions .
we understand the right - hand side as an integration constant @xmath132 .
it is obvious that solutions to ( [ eq : surstcond2 ] ) are automatically solutions to eq .
( [ eq : eqvode ] ) .
indeed , the converse is also true ; as a physicists proof it may suffice to state that from both equations , the taylor coefficients of @xmath133 have to satisfy the relations @xmath134 in order to obtain these conditions from ( [ eq : surstcond2 ] ) , it is convenient to expand @xmath135 ( with @xmath136 ; such a @xmath137 exists as long as @xmath138 , the statement for @xmath139 follows from continuity ) w.r.t .
@xmath91 and @xmath137 .
note that this equivalence is not really necessary for our purposes since we will solve eq .
( [ eq : eqvode ] ) ; physical requirements will select two one - parameter families of functions out of the two - parameter family of solutions to eq .
( [ eq : eqvode ] ) for which it is easy to check that also the original consistency condition ( [ eq : surstcond2 ] ) is satisfied . * solving the equivalent ode . *
the ordinary differential equation can be integrated easily using standard methods : since eq .
( [ eq : eqvode ] ) does not contain @xmath91 explicitly , we can introduce @xmath140 , @xmath141 so that @xmath142 , and ( [ eq : eqvode ] ) reduces to @xmath143 in this paragraph , the primes denote derivatives w.r.t . @xmath144 . in terms of @xmath145 ,
the latter equation can be rewritten as @xmath146 with the obvious solution @xmath147 , where @xmath148 and @xmath149 are integration constants .
substituting back , we obtain @xmath150 the initial condition @xmath151 implies the choice of the plus sign and @xmath152 .
if we introduce @xmath153 this equation for @xmath149 translates into @xmath154 .
then , eq . ( [ eq : eqvodesolvhlp ] ) can be integrated , @xmath155 and this is with @xmath156 and @xmath157 the defining equation ( [ eq : sndef ] ) of the jacobi sine function with modulus @xmath158 .
thus , @xmath159 is the promised solution to eq .
( [ eq : eqvode ] ) . one can easily check that its fourier coefficients satisfy eqs .
( [ eq : fouriercond ] ) .
this proves that all twist - even @xmath1-diagonal surface states are of the form @xmath160 with some ( so far unrestricted ) complex constants @xmath161 .
one should note that this is already a fair amount of reduction of degrees of freedom : we started out from a twist - even @xmath1-diagonal squeezed state with two independent functions @xmath162 and @xmath163 and ended up with a two - parameter family of solutions to our consistency conditions .
we will see , however , that physical constraints even select two one - parameter solutions of the form ( [ eq : sol ] ) .
in this section , we will analyze which maps of the form ( [ eq : sol ] ) actually determine surface states .
namely in general , the maps @xmath5 defining a surface state are subject to several constraints which are not automatically satisfied by all solutions to our consistency condition .
this analysis will require some knowledge of jacobian elliptic functions ; their basic properties are summarized in appendix [ sec : jac ] . * conditions on @xmath71 .
* there are several conditions a map @xmath5 defining a surface state has to satisfy .
e.g. @xcite , 1 .
@xmath71 should be meromorphic , 2 .
it should map the real axis to the real axis , 3 .
it should map the unit upper half disk ( in the @xmath80-plane ) into some region of the upper half plane and 4 .
it should be one - to - one ( at least ) for @xmath164 .
the last condition guarantees that @xmath5 furnishes a good local coordinate around the puncture at @xmath165 on the surface @xmath63 .
although the computation of correlation functions requires @xmath5 only to be defined locally around the origin , the possibility to glue the surface necessitates a map from the full upper half disk into @xmath63 @xcite .
these requirements will suffice to reduce the number of possible surface states in the twist - even @xmath1-diagonal sector tremendously . as a doubly periodic function , @xmath5 in eq .
( [ eq : sol ] ) is meromorphic ; thus , condition ( i ) entails no further conditions on the candidate functions in eq .
( [ eq : sol ] ) . since @xmath166 for @xmath91 real , abbreviates @xmath92 ( as above ) .
note that the latter function maps the real line to the real line ( more explicitly , to the interval @xmath167 ) and the upper half plane to the region @xmath168 , @xmath169 . ]
requirement ( ii ) has the obvious solutions @xmath170 , @xmath171 ( where the signs can be chosen independently ) , i.e. @xmath172 and @xmath158 are restricted to be real or purely imaginary .
however , because of ( [ eq : snktmo ] ) , the right - hand side equals @xmath173 since eq .
( [ eq : sol ] ) features no periodicity in @xmath172 or @xmath158 one can read off that the only other solution to condition ( ii ) are functions with arbitrary @xmath174 , @xmath175 .
denotes the unit sphere in the complex @xmath158-plane .
note that due to the symmetry @xmath176 , we need not distinguish between the latter two cases . similarly , a distinction between @xmath172 and @xmath177 in eq .
( [ eq : sol ] ) is irrelevant since @xmath178 is odd in its first argument.[fn : ama ] ] we will now scrutinize properties ( iii ) and ( iv ) for these cases . * case 1 : @xmath158 , @xmath172 real . *
since @xmath179 maps the unit upper half disk into the region @xmath180 , @xmath169 , property ( iii ) requires that @xmath181 maps this region into the upper half plane again .
using eqs .
( [ eq : snadd ] ) , ( [ eq : sniu ] ) , ( [ eq : cniu ] ) and ( [ eq : dniu ] ) as well as @xmath182 , @xmath183 , we derive that @xmath184 here , @xmath185 denotes the complementary modulus . by inspection of ( [ eq : sndecomp ] )
, we find that the function on the left - hand side has positive imaginary part for , i.e. , @xmath186 $ ] and @xmath187 $ ] .
the situation is illustrated in figure [ fig : arkr](a ) .
obviously , property ( iii ) here leads to the requirement that @xmath188 the second condition in ( [ eq : sincond ] ) fixes @xmath189 , then , the first requires @xmath190 $ ] .. ] for @xmath189 , the jacobi sine reduces to the ordinary sine function ; therefore , this is the butterfly family with @xmath191\ , .
\label{eq : butfam}\ ] ] in particular , it contains the sliver state with @xmath192 for @xmath193 , the butterfly state with @xmath194 for @xmath195 , and the nothing state with @xmath196 for @xmath197 .
condition ( iv ) is automatically satisfied .
* case 2 : @xmath158 real , @xmath172 imaginary .
* again , we start with condition ( iii ) . for the analysis of this case , we can recycle eq .
( [ eq : sndecomp ] ) .
it is easy to see that @xmath118 maps the region where @xmath198 to the upper half plane .
this region is depicted in figure [ fig : aikr](a ) . denoting @xmath199 with @xmath200 real ,
we see that @xmath201 , i.e. the positive imaginary half - axis of @xmath91 extends along the negative @xmath202-axis in figure [ fig : aikr ] .
we read off that @xmath203 for @xmath204 , the jacobi sine reduces to the @xmath205-function , so that the solution may be rewritten as @xmath206\ , .
\label{eq : wdgfam}\ ] ] for @xmath207 with @xmath208 , this defines the customary wedge states , including the sliver state for @xmath209 and the identity state for @xmath210 with @xmath211 .
we have bounded @xmath200 from above in order to ensure condition ( iv ) ; for @xmath212 , the image of the unit disk already fills the upper half plane ; for @xmath213 , the map is no langer one - to - one . *
case 3 : @xmath158 imaginary , @xmath172 real .
* we first scrutinize condition ( iii ) .
an imaginary value of the elliptic modulus , @xmath214 , can be mapped to a real value @xmath215 . introducing the abbreviations @xmath216 , @xmath217 , @xmath218 , and @xmath219
, we can use eqs .
( [ eq : snik ] ) , ( [ eq : snadd ] ) , and ( [ eq : dnadd ] ) to find the following decomposition of @xmath5 into real and imaginary parts : @xmath220 with some real and positive definite function @xmath221 . for real @xmath172 ,
the imaginary part is positive exactly where @xmath222 is positive .
this region is depicted in figure [ fig : ki](a ) . just as above
, one can read off that @xmath223 the latter condition fixes @xmath189 which was already dealt with in case 1 .
there are no new solutions in this situation .
* case 4 : @xmath158 , @xmath172 imaginary .
* with the same notation as above , we may read off from eq . ( [ eq : kidecomp ] ) that the imaginary part of @xmath118 is positive exactly where @xmath224 due to eqs .
( [ eq : sn2cn2])([eq : cn2dn2 ] ) , the expression in brackets equals @xmath225 .
thus , @xmath118 maps the region where @xmath226 to the upper half plane .
this situation is illustrated in figure [ fig : ki](b ) ; obviously , there are no values of @xmath227 such that @xmath5 maps the unit upper half disk completely into the upper half plane again . * case 5 : @xmath228 , @xmath229 . * the last case is the situation where @xmath158 takes values in the unit circle .
the analysis in this case is a slightly more involved than before .
we restrict the discussion to the case @xmath230 ; the other case with @xmath231 can be treated analogously .
we already saw in eqs .
( [ eq : solreal ] ) and ( [ eq : solreal2 ] ) that with @xmath229 , @xmath232 for @xmath233 ; similarly , it is trivial to see that @xmath234 for @xmath235 .
since furthermore , for @xmath204 , @xmath236 , eq . ( [ eq : snadd ] ) yields the following imaginary part : @xmath237 where the elliptic modulus is always @xmath238 . because of ( [ eq : snxre ] ) and ( [ eq : sniyim ] ) , the denominator is positive ; thus , the imaginary part is positive exactly where @xmath239 is . a careful analysis using the zeroes and singularities given in appendix [ sec : jac ] and the conjugation properties ( [ eq : kpconj ] ) and ( [ eq : kconj ] ) shows that the singularities of the jacobi sine function bound the real values of @xmath240 for which the imaginary part of @xmath241 is positive .
the intervals for which this is so are ( for @xmath242 ) : @xmath243 as above , the elliptic modulus is always @xmath238 .
since condition ( iii ) requires the upper bound of these intervals to be infinite this case only leads ( asymptotically ) to the solution @xmath244 ( i.e. , the sliver , which was already discussed in case 1 ) .
this concludes the proof that there are no twist - even @xmath1-diagonal surface states apart from the butterfly family ( [ eq : butfam ] ) and a trivial extension of the wedge states admitting arbitrary angles , cf .
( [ eq : wdgfam ] ) .
in this paper , we have derived and analyzed a consistency condition on twist - even @xmath1-diagonal surface states .
the general solution was found to be a two - parameter family of ( candidate ) maps defining a surface state , @xmath245 .
this family contains the well - known butterfly family as well as an extension of the family of wedge states to arbitrary angles ; and the conditions that the meromorphic function @xmath5 map the real line to itself and the unit upper half disk one - to - one to some region of the upper half plane were enough to show that these are in fact the only admissible maps to define a surface state .
in particular , this result demonstrates that surface state projectors which do not satisfy the sufficient condition that their boundary touches the midpoint of the open string ( i.e. , @xmath246 in upper half plane coordinates ) are particularly rare at least in the subsector of twist - even @xmath1-diagonal surface state projectors , we only found the identity state .
all other projectors share this singular property .
furthermore , this result completes the classification of @xmath1-diagonal projectors ( in a fermionic first order system ) commenced in @xcite .
besides , it also proves the conjecture made in @xcite that the projectors @xmath247 defined by @xmath248 for @xmath249 are not contained in the @xmath250 subalgebra .
hitherto , this conjecture had only been checked numerically .
it is conceivable that the strategy pursued here can be extended to the general case without the restriction to the twist - even subsector .
the hope that underlies this approach is that a better knowledge of the algebraic structures of the star algebra will help to pave the way to solving the equation of motion of string field theory .
i would like to thank alexander kling , yuji okawa , martin schnabl , wati taylor , david tong , and barton zwiebach for discussions .
furthermore , i would like to thank matthias ihl for encouragement as well as a collaboration at an intermediate stage .
this work was supported by a fellowship within the postdoc program of the german academic exchange service ( daad ) .
in this appendix , we present the definitions and properties of the jacobian elliptic functions used throughout the text .
a good reference for some of the formulas is @xcite .
* definitions . *
the jacobian elliptic functions ( of which we need here only @xmath178 , @xmath251 , and @xmath252 ) are doubly - periodic functions defined as the ( analytic continuation of the ) inverse of elliptic integrals , @xmath253 here , @xmath158 denotes the modulus , @xmath185 is the complementary modulus .
note that a number of incompatible conventions for the modulus are in common use , e.g. , @xmath158 is often replaced by its square . if the modulus is obvious , it is mostly omitted ( as above ) ; otherwise , we denote it explicitly by @xmath254 etc . for @xmath189 , @xmath178 and @xmath251
reduce to the ordinary trigonometric functions , @xmath255 and @xmath256 ( and @xmath257 ) . for @xmath204 , @xmath258 and @xmath259 .
* complete elliptic integrals . *
the periodicity and zeroes of the jacobian elliptic functions can be expressed in terms of the complete elliptic integral of the first kind , @xmath260 where the modulus @xmath158 is often omitted .
this is the value @xmath261 where the jacobi sine function equals 1 .
it is customary to set @xmath262 in general , we have @xmath263 where the bar denotes complex conjugation .
this ensures that the period parallelogram of the jacobian elliptic functions specified in the next paragraph is a rectangle .
the elliptic integral of the first kind has a branch cut from 1 to @xmath264 .
particular values are @xmath265 and @xmath266 . for the analysis in section [ sec : kdiagss ]
, we derive some further properties for @xmath267 : setting @xmath268 and being careful with the branchcuts , we can check that @xmath269 and @xmath270 therefore , we obtain the conjugation properties @xmath271 and @xmath272 * symmetries , periodicities , zeroes , and poles . *
the jacobi sine function is an odd function of its first argument , the jacobi cosine function as well as the delta function are even w.r.t . to @xmath261 .
all three functions are even in @xmath158 . for @xmath273 ,
the power series expansion takes the form : @xmath274 a closed form for the taylor coefficients is unknown .
the jacobian elliptic functions are not periodic in @xmath158 ; the periods , zeroes , and poles in @xmath261 are located at : here , it is understood that @xmath275 .
one can read off from @xmath276 and @xmath277 , @xmath278 that the jacobi sine function changes sign only along the real axis whereas the jacobi cosine function does so along both axes .
99 w. taylor and b. zwiebach , _ d - branes , tachyons , and string field theory _ , arxiv : hep - th/0311017 .
e. witten , _ noncommutative geometry and string field theory _
b * 268 * ( 1986 ) 253 .
l. rastelli , a. sen and b. zwiebach , _ string field theory around the tachyon vacuum _ , adv .
* 5 * ( 2002 ) 353 [ arxiv : hep - th/0012251 ] .
l. rastelli , a. sen and b. zwiebach , _ classical solutions in string field theory around the tachyon vacuum _ , adv .
theor . math .
phys . * 5 * ( 2002 ) 393 [ arxiv : hep - th/0102112 ] .
h. hata and t. kawano , _
open string states around a classical solution in vacuum string field theory _
, jhep * 0111 * ( 2001 ) 038 [ arxiv : hep - th/0108150 ] .
d. gaiotto , l. rastelli , a. sen and b. zwiebach , _ ghost structure and closed strings in vacuum string field theory _ , adv .
theor . math .
* 6 * ( 2003 ) 403 [ arxiv : hep - th/0111129 ] . k. okuyama , _ ghost kinetic operator of vacuum string field theory _ ,
jhep * 0201 * ( 2002 ) 027 [ arxiv : hep - th/0201015 ] ; d. j. gross and a. jevicki , _ operator formulation of interacting string field theory _ , nucl .
b * 283 * ( 1987 ) 1 .
d. j. gross and a. jevicki , _ operator formulation of interacting string field theory
. 2 _ , nucl .
b * 287 * ( 1987 ) 225 .
m. r. douglas , h. liu , g. moore and b. zwiebach , _ open string star as a continuous moyal product _
, jhep * 0204 * ( 2002 ) 022 [ arxiv : hep - th/0202087 ] .
d. m. belov , _ diagonal representation of open string star and moyal product _
, arxiv : hep - th/0204164 . i. ya .
arefeva and a. a. giryavets , _ open superstring star as a continuous moyal product _ , jhep * 0212 * ( 2002 ) 074 [ arxiv : hep - th/0204239 ]
. t. g. erler , _ moyal formulation of witten s star product in the fermionic ghost sector _
, arxiv : hep - th/0205107 .
d. m. belov and a. konechny , _ on continuous moyal product structure in string field theory _
, jhep * 0210 * ( 2002 ) 049 [ arxiv : hep - th/0207174 ] .
d. m. belov and c. lovelace , _
witten s vertex made simple _
d * 68 * ( 2003 ) 066003 [ arxiv : hep - th/0304158 ] ; d. m. belov , _
witten s ghost vertex made simple ( bc and bosonized ghosts ) _ , phys .
d * 69 * ( 2004 ) 126001 [ arxiv : hep - th/0308147 ] . c. maccaferri and d. mamone ,
_ star democracy in open string field theory _ , jhep * 0309 * ( 2003 )
049 [ arxiv : hep - th/0306252 ] .
a. kling and s. uhlmann , _ string field theory vertices for fermions of integral weight _
, jhep * 0307 * ( 2003 ) 061 [ arxiv : hep - th/0306254 ] .
m. ihl , a. kling and s. uhlmann , _ string field theory projectors for fermions of integral weight _
, jhep * 0403 * ( 2004 ) 002 [ arxiv : hep - th/0312314 ] . i. bars , _ map of witten s * to moyal s * _ , phys . lett .
b * 517 * ( 2001 ) 436 [ arxiv : hep - th/0106157 ] .
i. bars and y. matsuo , _ computing in string field theory using the moyal star product _ , phys .
d * 66 * ( 2002 ) 066003 [ arxiv : hep - th/0204260 ] .
e. fuchs , m. kroyter and a. marcus , _ squeezed state projectors in string field theory _ ,
jhep * 0209 * ( 2002 ) 022 [ arxiv : hep - th/0207001 ] .
o. lechtenfeld , a. d. popov and s. uhlmann , _ exact solutions of berkovits string field theory _ , nucl .
b * 637 * ( 2002 ) 119 [ arxiv : hep - th/0204155 ] .
a. kling , o. lechtenfeld , a. d. popov and s. uhlmann , _ on nonperturbative solutions of superstring field theory _
b * 551 * ( 2003 ) 193 [ arxiv : hep - th/0209186 ] .
a. kling , o. lechtenfeld , a. d. popov and s. uhlmann , _ solving string field equations : new uses for old tools _ , fortschr .
phys . * 51 * ( 2003 ) 775 [ arxiv : hep - th/0212335 ] .
a. leclair , m. e. peskin and c. r. preitschopf , _ string field theory on the conformal plane .
1 . kinematical principles _ , nucl . phys .
b * 317 * ( 1989 ) 411 .
a. leclair , m. e. peskin and c. r. preitschopf , _ string field theory on the conformal plane .
2 . generalized gluing _
b * 317 * ( 1989 ) 464 .
d. gaiotto , l. rastelli , a. sen and b. zwiebach , _
star algebra projectors _ , jhep * 0204 * ( 2002 ) 060 [ arxiv : hep - th/0202151 ] .
t. okuda , _ the equality of solutions in vacuum string field theory _ , nucl .
b * 641 * ( 2002 ) 393 [ arxiv : hep - th/0201149 ] .
i. bars , i. kishimoto and y. matsuo , _ fermionic ghosts in moyal string field theory _ ,
jhep * 0307 * ( 2003 ) 027 [ arxiv : hep - th/0304005 ] .
m. schnabl , _ anomalous reparametrizations and butterfly states in string field theory _ ,
b * 649 * ( 2003 ) 101 [ arxiv : hep - th/0202139 ] .
l. rastelli , a. sen and b. zwiebach , _ vacuum string field theory _ , arxiv : hep - th/0106010 . i. s. gradshteyn , i. m. ryzhik , _ table of integrals , series , and products _ , 6th edition , academic press ( 2000 ) , sections 8.11 - 8.15 . | we classify all twist - even squeezed states in string field theory which are diagonal in the @xmath0-basis and simultaneously surface states . for this purpose
, we derive a consistency condition for the maps defining @xmath1-diagonal surface states .
it restricts these maps to a two - parameter family of jacobi sine functions .
not all of them are admissible maps for surface states ; standard requirements single out two one - parameter families containing the generalized butterfly states and the wedge states .
hep - th/0408245 + mit ctp3535 + 2.0 cm * a note on @xmath1-diagonal surface states * sebastian uhlmann@xmath2 + _ center for theoretical physics + massachusetts institute of technology + cambridge , ma 02139 , usa _
+ august , 2004 6.5truein ' '' '' width 5.cm @xmath3 [email protected] |
studies of large samples of distant galaxies are fundamental to provide a deeper insight in the formation and evolution of galaxies and systems of galaxies , such as clusters and groups .
using measurements of galaxy clustering at @xmath13 one can test the predictions of cosmological models of structure formation and evolution @xcite . in the last years
diverse photometric selection techniques have been developed in order to select high redshift galaxies .
in particular the lyman break technique is an ideal method for selecting a large number of distant galaxies and for studying the large - scale distribution and properties of star - forming systems at high redshifts from multicolour optical data @xcite .
this technique requires a high rest - frame far ultraviolet luminosity , and so will preferentially select galaxies with recent or unobscured active star - formation @xcite .
a new near - infrared selection technique has been developed in recent years to select samples of galaxies at high redshifts .
the distant red galaxies ( drgs , hereafter ) colour cut criterion ( @xmath0 , vega system ) @xcite is expected to select galaxies with prominent rest frame optical breaks , caused by the 3625 balmer - break or the 4000 ca ii h+k break . the balmer discontinuity at 3625 is strongest in a - type stars and the 4000 break is characteristic of coolers stars with types later than g0 and strong in giant and supergiant stars @xcite .
van dokkum et al . (
2003 ) found that the drg criterion selects galaxies with rest - frame optical colours similar to those of normal nearby galaxies . using deep mid - infrared observations with irac on the spitzer space telescope
, @xcite found that 70% of the drgs are best described by dust - reddened star forming models and 30% are very well fit with old stellar population , passive star evolution galaxy models .
by comparison of the stellar populations of dgrs to those obtain in lyman - break galaxies ( lbgs ) , @xcite found that the average mass - to light ratios ( @xmath14 ) of the drgs are about three times higher than the lbgs sample , indicating that dgrs may represent massive and old galaxies , similar to those found in the local universe .
a similar study by @xcite comparing lbgs and drgs at similar redshifts and rest - frame @xmath15-band luminosities , shows that drgs are older , more massive and more obscured for any given star formation history than lbgs .
recent works by @xcite and @xcite show that this single near ir colour cut selects a rather heterogeneous sample of galaxies , from distant luminous massive systems , to a significant fraction of less luminous dusty star - burst and galaxies with mixed morphology at redshifts @xmath16 , with extended tails at @xmath17 and @xmath18 . at low redshifts ,
qso environments show similar characteristics of those found in normal galaxies @xcite .
moreover , @xcite found that the cross - correlation function between low ( @xmath19 ) qsos and galaxies , is consistent with the auto - correlation function of galaxies selected from the apm galaxy survey .
unlike radio - loud qsos , @xcite found that radio - quiet qsos at @xmath20 are rarely found in high density environments as rich as abell class 1 .
recently , @xcite studying the environment of @xmath21 luminous ( @xmath2222 ) quasars in the sdss area , found that they are located in higher local overdensity regions than are typical @xmath23 galaxies .
however , the results for qso environment at higher redshifts ( @xmath1 ) are contradictory .
@xcite found that some of the @xmath24 qsos are located in environments as rich as those of abell class 1 clusters .
@xcite reported a significant excess of faint galaxies in the fields of @xmath25 quasars .
no excess galaxy population associated with radio - quiet qsos are reported by @xcite at @xmath26 .
@xcite found an anti - correlation between radio - quiet qsos and galaxies at redshifts @xmath27 .
@xcite studied the clustering of galaxies around a sample of @xmath28 qsos selected from the sdss and deep2 surveys .
they found , from a two - point cross - correlation analysis , that the local environment of qsos is consistent with the mean environment of the full deep2 galaxy population and that they cluster similar to the blue , star - forming galaxies rather than the red galaxies .
their results imply that high redshift qsos do not reside in particularly massive dark matter halos . using the drgs criterion , @xcite reported a discovery of proto - clusters candidates around 6 high redshifts radio galaxies at @xmath29 on the basis of excess of @xmath30 colour selected galaxies .
recently , a similar colour criterion was adopted to identify and to study the environment of very distant objects .
this method uses a colour cut @xmath31 which select objects at @xmath32 .
an over - density of galaxies around the most distant radio loud qso ( sdss j0836 + 0054 at @xmath33 ) was reported by @xcite , who found a surface density six times higher than the number expected from deep fields at similar redshifts .
@xcite found an excess of @xmath31 galaxies around one of the most distant qsos at @xmath34 . in this paper
we investigate the population and colour distribution of drgs , selected at bright near ir magnitudes ( @xmath35 , vega system ) around high redshift qsos ( @xmath1 ) and optical obscured sources ( @xmath36 ) in the flamingos extragalactic survey ( flamex , @xcite ) .
this paper is organised as follows : section 2 describes the sample analysed , the drg number counts are analysed in section 3 . in section 4
we study the colour distribution of galaxies in different environments .
we investigate the qso drgs and obscured drgs cross - correlation analysis in section 5 .
finally we discuss our results in section 6 . in this work
we assume a standard @xmath37cdm model universe with cosmological parameters , @xmath38=0.3 , @xmath39=0.7 and a hubble constant of @xmath40100 km s@xmath8mpc@xmath8 .
all magnitudes are expressed in the vega system .
the flamingos extragalactic survey ( flamex , @xcite ) is a wide - area , near - infrared imaging survey in @xmath41 and @xmath42 bands within the noao deep wide - field survey ( ndwfs ) regions @xcite .
this paper uses catalogues of the first data release dr1 of the northern part of the survey ( botes field ) that covers 4.7 deg@xmath5 in both @xmath41 and @xmath42 bands . using monte carlo simulations @xcite found that more than 90 % of the survey region is complete to @xmath43 19.2 , with 50 % being complete to @xmath44 .
a detailed description of the observing strategy and data reduction can be found in @xcite .
the catalogues were made using @xmath42selected objects for each survey subset with the sextractor package @xcite , using dual image mode to measure the @xmath41band magnitudes within the same regions .
detections in the different bands were matched if the centroids were within 1 @xmath45 of each other .
@xmath30 colours were calculated using 4@xmath45 diameter aperture photometry , sufficiently large to avoid intrinsic aberrations that cause the psf to vary significantly across the field and to produce robust photometry across entire subfields . in order to reject spurious objects lying in the edges of the images , we selected objects that not overlap with another object using the sextractor parameters @xmath46 and @xmath47 @xmath48 0.7 , in both @xmath41 and @xmath42 bands . for two different reasons , no star - galaxy separation was performed .
first , the variable psf in the flamex area precluded use of structural information to separate stars from extended sources and second , drgs are of such small apparent size that considerable fraction of them would possibly be misclassified as stars and rejected from the final catalogues .
as noted by @xcite , and as can be seen in the colour - magnitude diagram ( figure 1 ) , the @xmath30 colours alone provides a simple means of performing star - galaxy separation .
the two horizontal sequences at @xmath30=0.4 and @xmath30=0.8 correspond to galactic stars with types later than g5 and earlier than k5 , respectively @xcite .
red galaxies with @xmath492.3 are well separated from stars .
we also show in figure 1 the drgs criterion adopted ( dashed lines ) .
our final catalogue consist of 7131 drgs in a contiguous area of 4.7 degree@xmath5 , representing the largest sample of drgs selected at bright magnitudes @xmath35 . vs @xmath42 ( vega ) colour - magnitude diagram for the flamex field ( botes area ) .
right panel : colour distribution . dashed lines show the drgs criterion adopted.,width=302 ] the qso sample analysed in this work was obtained using the ned database .
we selected qsos in the redshift range @xmath1 in the flamex area .
the sample consists of twelve qsos selected from the sloan digital sky survey ( sdss , data release 1 to 4 ) @xcite and one x
ray selected agn @xcite ( see table 2 ) .
we also choose a sample of optically obscured galaxies discovered with the spitzer space telescope , which are either optically very faint ( @xmath5024.5 ) or invisible ( @xmath5026 ) with no counterparts in the noao deep wide - field survey regions .
the main goal of this survey is to select optically very faint galaxies but bright enough in the mid - infrared for redshift determination with the infrared spectrograph on spitzer ( irs ) , which would not have been identified in previous optical studies .
sources with @xmath2 mag were identified with the multi - band imaging photometer ( mips ) on the spitzer space telescope to a limiting 24 @xmath3 m flux density above 0.75 mjy .
114 sources of the total sample met this optical criterion .
out of these , 17 sources have no optical or have very faint optical counterparts . rejecting sources lying in the edges of the images , where noise level are high and signal to noise values are low
, we finally selected 6 sources located within the flamex area , with a mean redshift distribution @xmath36 .
the redshift range of the selected sources are similar to that of the redshift distribution of drgs with @xmath51 ( vega magnitudes ) found by @xcite .
redshifts from this sample were derived primarily from strong silicate absorption features @xcite .
we also selected another two sources with feasible redshift determinations derived by a weak silicate absorption feature , based on fits of redshifted spectral templates of local ultra - luminous infrared galaxies ( ulirgs ) from @xcite .
full details of the sample selection and redshift determination is given in @xcite .
as noted by @xcite template fits suggest that most of the optically obscured sources are dominated by an agn component , i.e. a steeply rising spectra and varying levels of silicate absorption , similar to the infrared spectra of known local agns . by comparing bolometric luminosities of submillimeter selected galaxies , such as those detected in scuba or mambo suryeys , @xcite found that the obscured source sample seems to represent sources that have a hotter dust component than typical submillimeter - selected galaxies , assuming that mid - infrared luminosities scale to bolometric luminosities similar than in template sources which fit the mid - infrared spectra . using this scaling ,
the obscured sources have implied infrared luminosities between 6@xmath52 l@xmath53 and 6@xmath54 l@xmath53 , similar to bright qsos at @xmath16 @xcite .
magnitude for the drg sample ( filled circles ) .
the error bars plotted represent poissonian errors ( 3 @xmath55 ) .
we include the drg number counts obtained in the literature .
open triangles represent data from the ukidss ultra deep survey early data release ( foucaud et al .
2006 , private communication ) and open circles are data obtained in the goods - music sample , @xcite.,width=302 ]
we have computed the differential number counts of drgs per unit area as a function of @xmath42 magnitude which are shown in figure 2 .
we use sextractor @xmath56 magnitudes for this calculation and estimate error bars using 3 @xmath55 uncertainties estimated using poissonian errors of the raw galaxy counts ( see table 1 ) .
we compare our determinations with @xmath57band number counts for drgs obtained by @xcite in the goods music sample and with estimates obtained in the ukidss ultra deep survey early data release ( foucaud et al .
2006 , private communication ) .
we have done this comparison using vega magnitudes , assuming a transformation @xmath58 @xcite . for @xmath5917.5
our counts are systematically higher than those obtain by foucaud et al .
this discrepancy can be explained by cosmic variance because of the small area ( 0.62 deg@xmath5 ) of the early data release of the ukidss ultra deep survey .
+ [ t0 ] table 1 .
drg number counts as a function of @xmath42 band .
the columns give the central bin magnitude , the raw counts , counts in units of arcmin@xmath60mag@xmath8 and 3 @xmath55 uncertainties estimated using poissonian errors . | we study the distant red galaxy ( drg , @xmath0 ) neighbour population of quasi stellar objects ( qsos ) selected from the sloan digital sky survey ( sdss ) in the redshift range @xmath1 .
we perform a similar analysis for optically obscured agns ( i.e. with a limiting magnitude @xmath2 ) detected in the mid - infrared ( 24 @xmath3 m ) with the spitzer space telescope and a mean redshift @xmath4 in the flamingos extragalactic survey ( flamex ) . both qsos and obscured agn target samples cover 4.7 deg@xmath5 in the same region of the sky .
we find a significant difference in the environment of these two target samples .
neighbouring galaxies close to qsos tend to be bluer than galaxies in optically obscured source environments .
we also present results on the cross
correlation function of drgs around qsos and optically faint mid - infrared sources .
the corresponding correlation length obtained for the qso sample targets is @xmath6=@xmath7 mpc h@xmath8 and a slope of @xmath9=@xmath10 . for the optically obscured galaxy sample
we find @xmath6=@xmath11 mpc h@xmath8 and a slope of @xmath9=@xmath12 .
these results indicate that optically faint obscured sources are located in denser environment of evolved red galaxies compare to qsos .
cosmology : large - scale structure of universe galaxies : galaxies : high - redshift quasars : general |
for a biological sample , the dna copy number of a genomic region is the number of copies of the dna in that region within the genome of the sample , relative to either a single control sample or a pool of population reference samples .
dna copy number variants ( cnvs ) are genomic regions where copy number differs among individuals .
such variation in copy number constitutes a common type of population - level genetic polymorphism .
see @xcite , @xcite , @xcite and @xcite for detailed discussions on cnv in the human population . on another front
, the genomes of tumor cells often undergo somatic structural mutations such as deletions and duplications that affect copy number .
this results in copy number differences between tumor cells and normal cells within the same individual .
these changes are often termed copy number aberrations or copy number alternations ( cna ) .
there is significant scientific interest in finding cnvs in normal individuals and cnas in tumors , both of which entail locating the boundaries of the regions in the genome that have undergone copy number change ( i.e. , the breakpoints ) , and estimating the copy numbers within these regions . in this article
, we use next - generation sequencing data for copy number estimation .
microarrays have become a commonly used platform for high - throughput measurement of copy number .
there are many computational methods that estimate copy number using the relative amount of dna hybridization to an array .
see @xcite , @xcite and @xcite for a general review of existing methods for array - based data
. however , the precision of breakpoint estimates with array - based technology is limited by its ability to measure genomic distances between probes , which currently averages about 1000 bases ( 1 kb ) on most arrays .
hence , the lower limit in the length of detectable cnv events is about 1 kb . with sequencing capacity growing and its cost dropping dramatically , massively parallel sequencing is now an appealing method for measuring dna copy number . in these newer sequencing technologies
, a large number of short reads ( 36100 bp ) are sequenced in parallel from the fragmentation of sample dna .
then each read is mapped to a reference genome .
the basic rationale is that _ coverage _ , defined as the number of reads mapped to a region of the reference genome , reflects the copy number of that region in the sample , but with many systematic biases and much variability across the genome .
@xcite was one of the first to use genome - wide sequencing to detect cna events .
the reader is also referred to @xcite for a review of recent studies in cnv / cna detection using sequencing data .
more details of the data , with an illustrative example ( figure [ figillusreads ] ) , are given in section [ secdataexistingmethod ] .
in the shift from array - based to sequencing - based copy number profiling , the main statistical challenge arises from the fundamental change in the type of data observed .
array - based data are represented by a large but fixed number of continuous valued random variables that are approximately normal after appropriate preprocessing , and cnv / cna signals based on array data can be modeled as shifts in mean .
sequencing - based data , as we will discuss further in section [ secdataexistingmethod ] , are realizations of point processes , where cnv / cna signals are represented by shifts in intensity of the process .
while one can apply a normal approximation to the large number of discrete events in sequencing data , hence translating the problem into the familiar array - based setting , this approach is inefficient and imprecise .
a more direct model of the point process is preferred .
this type of data calls for a new statistical model , new test statistics , and , due to the quick growth of sequencing capacity , new and highly efficient computing implementation . in copy number profiling it is important to assess the confidence in the estimated copy numbers . with the exception of @xcite ,
existing segmentation methods , both for array data and for sequencing data , give a hard segmentation and do not quantify the uncertainty in their change - point estimates .
some methods , such as @xcite and @xcite , provide confidence assessments for the called cnv or cna regions , in the form of false discovery rates or @xmath0-values , thus inherently casting the problem in a hypothesis testing framework .
however , for the analysis of complex regions with nested changes , such as those in tumor data , confidence intervals on the copy number , from an estimation perspective , are often more useful .
intuitively , the copy number estimate is less reliable for a region near a change point than for a region far away from any change points .
also , copy number estimates are more reliable for regions with high coverage than for regions with low coverage , since coverage directly affects the number of observations used for estimation .
this latter point makes confidence intervals particularly important for interpretation of results derived from short read sequencing data , where coverage can be highly uneven across the genome . in this paper
, we take a bayesian approach with noninformative priors to compute point - wise confidence intervals , as described in section [ secbayesianci ] .
the proposed methods are based on a simple and flexible inhomogeneous poisson process model for sequenced reads .
we derive the score and generalized likelihood ratio statistics for this model to detect regions where the read intensity shifts in the target sample , as compared to a reference .
we construct a modified bayes information criterion ( mbic ) to select the appropriate number of change points and propose bayesian point - wise confidence intervals as a way to assess the confidence in the copy number estimates . as a proof of concept ,
we apply seqcbs , our sequencing - based cnv / cna detection algorithm , to a number of actual data sets and found it to have good concordance with array - based results .
we also conduct a spike - in study and compare the proposed method to segseq , a method proposed by @xcite .
the methods developed in this paper have been implemented in an open - source r - package , seqcbs , available from cran ` http://cran.r - project .
org / web / packages / seqcbs / index.html ` .
in a general next - generation genome sequencing / resequencing pipeline , shown in figure [ figseqpipe ] , the dna in the sample is randomly fragmented , and a short sequence of the ends of the fragments is `` read '' by the sequencer .
after the bases in the reads are called , the reads are mapped to the reference genome .
there are many different approaches to the preparation of the dna library prior to the sequencing step , some involving amplification by polymerase chain reaction , which lead to different distribution of reads along the reference genome .
when a region of the genome is duplicated , fragments from this region have a higher representation , and thus its clones are more likely to be read by the sequencer .
hence , when mapped to reference genome , this duplicated region has a higher read intensity .
similarly , a deletion manifests as a decrease in read intensity . since reads are contiguous fixed length sequences , it suffices to keep track of the reference mapping location of one of the bases within the read .
customarily , the reference mapping location of the 5@xmath1 end of the read is stored and reported .
this yields a point process with the reference genome as the event space.=-1 as noted in previous studies , sequencing coverage is dependent on characteristics of the local dna sequence , and fluctuates even when there are no changes in copy number , as shown in @xcite . just as adjusting for probe - effects
is important for interpretation of microarray data , adjusting for these baseline fluctuations in depth of coverage is important for sequencing data .
the bottom panels of figure [ figseqarrayhcc1954 ] show the varying depth of coverage for chromosomes 8 and 11 in the sequencing of a normal human sample , hcc1954 .
many factors cause the inhomogeneity of depth of coverage .
for example , regions of the genome that contain more g / c bases are typically more difficult to fragment in an experiment .
this results in lower depth of coverage in such regions . some regions of the genome are highly repetitive .
it is challenging to map reads from repetitive regions correctly onto the reference genome and , hence , some of the reads are inevitably discarded as unmappable , resulting in loss of coverage in that region , even though no actual deletion has occurred .
some ongoing efforts on the analysis of sequencing data involve modeling the effects of measurable quantities , such as gc content and mappability , on baseline depth .
@xcite demonstrated that read counts in sequencing are highly dependent on gc content and mappability , and discussed a method to account for such systematic biases .
@xcite investigated the relationship between gc content and read count on the illumina sequencing platform with a single position model , and identified a family of unimodal curves that describes the effect of gc content on read count .
we take the approach of empirically controlling for the baseline fluctuations by comparing the sample of interest to a control sample that was prepared and sequenced by the same protocol . in the context of tumor cna detection ,
the control is preferably a matched normal sample , for it eliminates the discovery of germline copy number variants and allows one to focus on somatic cna regions of the specific tumor genome .
if a perfectly matched sample is not possible , a carefully chosen control or a pool of controls , with sequencing performed on the same platform with the same experimental protocol , would work for our method as well since almost all of the normal human genome are identical . as a simple and illustrative example of the data
, we generated points according to a nonhomogeneous poisson process .
figure [ figillusreads ] shows the point processes and the underlying @xmath2 function , defined as the probability that a read at genomic position @xmath3 is from the case / tumor sample , conditional on the existence of a read at position @xmath3 .
the model is discussed in more detail in section [ secthemodel ] .
the @xmath4-values for the points are jittered for graphical clarity . .
] existing methods on cnv and cna detection with sequencing data generally follow the change - point paradigm , which is natural since copy number changes reflect actual breakpoints along chromosomes .
@xcite proposed the algorithm segseq that segments the genomes of a tumor and a matched normal sample by using a sliding fixed size window , reducing the data to the ratio of read counts for each window .
@xcite proposed cnv - seq that detects cnv regions between two individuals based on binning the read counts and then applying methods developed for array data .
@xcite designed a method named event - wise testing ( ewt ) that detects cnv events using a fixed - window scan on the gc content adjusted read counts .
@xcite proposed a method called cnaseg that uses read counts in windows of predefined size , and discovers cnv using a hidden markov model segmentation .
as for single sample cnv detection method , @xcite constructed a computational algorithm that normalizes read counts by gc content and estimates the absolute copy number .
these existing methods approach this statistical problem by binning or imposing fixed local windows .
some methods utilize the log ratio of read counts in the bin or window as a test statistic , thereby reducing the data to the familiar representation of array - based cnv / cna detection , with @xcite being an exception in that it uses the difference in tumor - normal window read counts in their hmm segmentation .
there are a number of downsides to the binning or local window approach .
first , due to the inhomogeneity of reads , certain bins will receive much larger number of reads overall than other bins , and the optimal window size varies across the genome .
if the number of reads in a bin is not large enough , the normal approximations that are employed in many of these methods break down .
second , by binning or fixed - size window sliding , the estimated cnv / cna boundaries can be imprecise if the actual breakpoints are not close to the bin or window boundary .
this problem can be somewhat mitigated by refining the boundary after the change point is called , as done in segseq . in this paper
, we propose a unified model , one that detects the change points , estimates their locations , and assesses their uncertainties simultaneously . to illustrate and evaluate our method , we apply it to real and spiked - in data based on a pair of nci-60 tumor / normal cell lines , hcc1954 and bl1954 .
the data for these samples were produced and investigated by @xcite .
the whole - genome shotgun sequencing was performed on the illumina platform and the reads are 36 bp long .
after read and mapping quality exclusions , 7.72 million and 6.65 million reads were used for the tumor ( hcc1954 ) and normal ( bl1954 ) samples , respectively .
newer sequencing platforms produce much more massive data sets .
we start with a statistical model for the sequenced reads .
let @xmath5 and @xmath6 be the number of reads whose first base maps to the left of base location @xmath3 of a given chromosome for the case and control samples , respectively .
we can view these count processes as realizations of two nonhomogeneous poisson processes ( nhpp ) , one each for the case and control samples , @xmath7 \\[-8pt ] \nonumber \ { y_{t}\ } & \sim&\operatorname{nhpp}(\lambda_{t}).\end{aligned}\ ] ] the scale @xmath3 is in base pairs .
the scenario where two or more reads are mapped to the same genomic position is allowed by letting @xmath8 and @xmath9 take values larger than 1 and assuming that the observed process is binned at the integers .
we propose a change - point model on the conditional probability of an event at position @xmath3 being from @xmath10 , given that there is such an event from either @xmath10 or @xmath11 , namely , @xmath12 an example of data according to this model is shown in figure [ figillusreads ] .
the change - point model assumption can be equivalently expressed as @xmath13 where @xmath14 $ ] is piecewise constant with change points @xmath15 .
of course , we require the collection of change points to lie within the observation window : @xmath16 this model does not force the overall intensity of case and control reads to be the same .
the intensity function @xmath9 reflects the inhomogeneity of the control reads .
one interpretation of the model is that , apart from constant shifts , the fluctuation of coverage in the case sample is the same as that in the control sample .
this is reasonable if the case and control samples are prepared and sequenced by the same laboratory protocol and mapped by the same procedure , as we discussed in section [ secdataexistingmethod ] .
the model would not be valid if the intensity functions for samples have significant differences caused by nonmatching protocols or experimental biases .
let @xmath17 , @xmath18 be the event locations for processes @xmath10 and @xmath19 , respectively .
that is , @xmath20 and @xmath21 are the mapped positions of the reads from the case and control samples .
let @xmath22 be the total number of reads from the case and control samples combined .
we combine the read positions from the case and control processes and keep them ordered in the genome position , obtaining combined read positions @xmath23 and indicators of whether each event is a realization of the case process or the control process @xmath24 : @xmath25 for any read @xmath26 in the combined process , we will sometimes use the term `` success '' to mean that @xmath27 , that is , that the read is from the case process .
notice that the collection of change - point locations that can be inferred with the data is precisely @xmath28 , since we do not have data points to make inference in favor of or against any change points in between observations .
this means that estimating the copy number between two genome positions is equivalent to doing so for the closest pair of reads that span the two genome positions of interest .
namely , there is a one - to - one correspondence between the set of possible change points on @xmath29 and the set of change points @xmath30 defined on the indices @xmath31 through the following : @xmath32 the above statement can be made formal through the equivariance principle .
consider the sample space @xmath33 $ ] of any change point , any monotonically increasing function @xmath34\rightarrow[0,t]$ ] , and its natural vector extension @xmath35 .
a change - point estimator @xmath36 is monotone transform equivariant if for all monotonically increasing functions @xmath34\rightarrow[0,t ] $ ] , we have @xmath37 the following theorem shows that any breakpoint estimator @xmath38 satisfying the equivariance condition can be decomposed into a simpler form . [ thmequivthm ] let @xmath39 , which takes values in @xmath40 , be an estimator of the breakpoints .
then @xmath36 is monotone transform equivariant if and only if @xmath41 , where @xmath42 taking integer values @xmath43 does not depend on @xmath40 . for ease of notation , we let @xmath44 be the natural extension of @xmath45 .
suppose @xmath46 , where @xmath42 .
note that @xmath47 is invariant to all monotone transformations of the arrival times , hence so is @xmath48
. therefore , @xmath49 . in the other direction ,
since @xmath50 and @xmath51 contain the same information as @xmath52 , we must have @xmath53 .
suppose that @xmath54 depend on @xmath40 in a nontrivial way but @xmath36 satisfies the monotone transform equivariance condition .
this means that there exist @xmath55 such that @xmath56 .
but since @xmath40 and @xmath57 are both increasing finite sequences on @xmath58 $ ] with the same number of elements , we must have some @xmath59 that @xmath60 .
note that @xmath61 induces @xmath62 .
however , @xmath63 .
hence , the equivariance property holds if and only if @xmath48 is only a function of @xmath47 .
theorem [ thmequivthm ] implies that any breakpoint estimation procedure , that is , monotone transform equivariant uses the estimator @xmath48 of integer breakpoints based on @xmath47 , and that the actual read position @xmath40 merely serves as a genomic scale lookup table .
hence , we can define our change - point model on the indices @xmath31 for the read counts , and use the conditional likelihood which depends only on @xmath64 but not on the event positions @xmath65 : @xmath66 for the rest of this section , we will exclusively work with equation ( [ eqmodelspec , indexscale ] ) .
the mapping positions @xmath67 will re - enter our analysis when we compute confidence intervals for the copy number estimates , in section [ secbayesianci ] .
our statistical problem is hence two - fold .
first , given @xmath68 , we need to estimate the change points @xmath69 .
second , we need a method to select model complexity , as dictated by @xmath68 .
we start by considering the following simplified problem : given a single interval spanning reads @xmath26 to @xmath70 in the combined process , we want to test whether the success probability inside this interval , @xmath71 , is different from the overall success probability , @xmath0 .
the null model @xmath72 states that @xmath73 .
we derive two statistics to test this hypothesis .
the first is adopted from the conditional score statistic for a general exponential family model where the signal is represented by a kernel function , as discussed in @xcite , @xmath74 where @xmath75 .
this statistic is simply the difference between the number of observed and expected case events under the null model .
its variance at the null is @xmath76 and is used to standardize @xmath77 for comparison between regions of different sizes .
the standardized score statistic @xmath78 is intuitive and simple , and would be approximately standard normal if @xmath79 were large .
however , the normal approximation is not accurate if the number of reads that map to the region is low . to attain higher accuracy for regions with low read count ,
observe that @xmath80 is a binomial random variable , and use an exact binomial generalized likelihood ratio ( glr ) statistic , @xmath81 where the null model with one overall success probability parameter @xmath0 is compared with the alternative model with one parameter @xmath71 for inside the @xmath82 $ ] interval and another parameter @xmath83 for outside the interval . from the binomial log - likelihood function one obtains @xmath84}\biggl\ { z_{k}\log\biggl(\frac { \hat{p}_{ij}}{\hat{p}}\biggr)+(1-z_{k})\log\biggl(\frac { 1-\hat{p}_{ij}}{1-\hat{p}}\biggr)\biggr\}\\ & & { } + \sum_{k\notin[i , j]}\biggl\ { z_{k}\log\biggl(\frac{\hat { p}_{0}}{\hat{p}}\biggr)+(1-z_{k})\log\biggl(\frac{1-\hat { p}_{0}}{1-\hat{p}}\biggr)\biggr\ } , \label{eqbinomglr}\end{aligned}\ ] ] where @xmath85 , @xmath86 , @xmath87 are maximum likelihood estimates of success probabilities @xmath88}z_{k}/(j - i+1 ) , \\
\hat{p}_{0}&=&\sum_{k\notin[i , j]}z_{k}/(m - j+i-1).\end{aligned}\ ] ] the glr and score statistics allow us to measure how distinct a specific interval @xmath82 $ ] is compared to the entire chromosome .
for the more general problem in which @xmath89 is not given but only one such pair exists , we compute the statistic for all unique pairs of @xmath89 to find the most significantly distinct interval .
this operation is @xmath90 and to improve efficiency , we have implemented a search - refinement scheme called iterative grid scan in our software .
it works by identifying larger interesting intervals on a coarse grid and then iteratively improving the interval boundary estimates .
the computational complexity is roughly @xmath91 and hence scales easily .
a similar idea was studied in @xcite . in the general model with multiple unknown change points
, one could theoretically estimate all change points simultaneously by searching through all possible combinations of @xmath92 .
but this is a combinatorial problem where even the best dynamic programming solution [ @xcite ; @xcite ; @xcite ] would not scale well for a data set containing millions of reads .
thus , we adapted circular binary segmentation [ @xcite ; @xcite ] to our change - point model as a greedy alternative .
in short , we find the most significant region @xmath89 over the entire chromosome , which divides the chromosome in to 3 regions ( or two , if one of the change points lies on the edge ) .
then we further scan each of the regions , yielding a candidate subinterval in each region . at each step , we add the most significant change point(s ) over all of the regions to the collection of change - point calls .
model complexity grows as we introduce more change points .
this brings us to the issue of model selection : we need a method to choose an appropriate number of change points @xmath68 .
@xcite proposed a solution to this problem for gaussian change - point models with shifts in mean . like the gaussian model
, the poisson change - point model has irregularities that make classic measures such as the aic and the bic inappropriate .
an extension of @xcite gives a modified bayes information criterion ( mbic ) for our model , derived as a large sample approximation to the bayes factor in the spirit of @xcite : @xmath93 where @xmath94 is the number of unique values in @xmath95 .
the first term of mbic is the generalized log - likelihood ratio for the model with @xmath68 change points versus the null model with no change points . in our context
, @xmath68 ideally reflects the number of biological breakpoints that yield the copy number variants .
the remaining terms can be interpreted as a `` penalty '' for model complexity .
these penalty terms differ from the penalty term in the classic bic of @xcite due to nondifferentiability of the likelihood function in the change - point parameters @xmath96 and also due to the fact that the range of values for @xmath69 grow with the number of observations @xmath97 . for more details on the interpretation of the terms in the mbic , see @xcite .
finally , we report the segmentation with @xmath98 change points .
as noted in the , it is particularly important for sequencing data to assess the uncertainty in the relative read intensity function at each genomic position .
we approach this problem by constructing approximate bayesian confidence intervals .
suppose @xmath24 are independent realizations of bernoulli random variables with success probabilities @xmath99 .
consider first the one change - point model ( which can be seen as a local part of a multiple change - point model ) , where @xmath100 without loss of generality , we may take @xmath101 .
assume a uniform prior for @xmath102 on this discrete set .
let @xmath103 be the number of successes up to and including the @xmath3th realization , @xmath104 our goal is to construct confidence bands for @xmath105 at each @xmath106 .
assume a @xmath107 prior for @xmath83 and @xmath108 .
if we knew @xmath102 , then the posterior distribution of @xmath83 and @xmath108 is @xmath109 now , without knowing the actual @xmath110 , we compute the posterior distribution of @xmath105 as @xmath111 as before , the first part of the summation term is a beta distribution , @xmath112 and for the second term , we define the likelihood of the change point at @xmath26 as @xmath113 and observe that with the uniform prior on @xmath102 , @xmath114 where @xmath115 and @xmath116 and @xmath117 are with respect to the prior distributions of @xmath83 and @xmath108 . with @xmath118 priors on @xmath83 and @xmath108 , we can find the closed form expression of @xmath119 : @xmath120 \\[-8pt ] \nonumber & = & \frac{b(\alpha+s _
{ i},\beta+ i - s _ { i})b(\alpha + s_{m}-s _ { i},\beta+m- i - s_{m}+s _ { i})}{b(\alpha,\beta ) ^{2}}\\ & = & \frac{\gamma(\alpha+s _ { i})\gamma(\beta+
i - s _ { i})}{\gamma(\alpha+\beta+ i)}\nonumber\\ & & { } \times\frac{\gamma(\alpha+s_{m}-s _ { i})\gamma(\beta+m- i - s_{m}+s _ { i})}{\gamma(\alpha+\beta+m- i)}\frac{\gamma ( \alpha+\beta)^{2}}{\gamma(\alpha)^{2}\gamma ( \beta)^{2}}.\nonumber\end{aligned}\ ] ] hence , we can compute , without knowing the actual value of @xmath102 , @xmath121 observe that the posterior distribution is a mixture of @xmath122 distributions . in theory , we could compute weights @xmath123 for all positions @xmath26 and then numerically compute @xmath124 quantiles of the posterior beta mixture distribution to obtain the bayesian confidence intervals . however , in practice , one can approximate the sum in ( [ newlab ] ) by @xmath125,\ ] ] for some small @xmath126 , hence ignoring the highly unlikely locations for the change points .
empirically , we use @xmath127 .
it is easy to see that the sequence of log likelihood ratios for alternative change points , @xmath128 , form random walks with negative drift as @xmath26 moves away from the true change point @xmath102 [ @xcite ] .
the negative drift depends on the true @xmath129 , @xmath130 and is larger in absolute magnitude when the difference between @xmath83 and @xmath108 is larger . with @xmath102 unknown , since @xmath131 can be made arbitrarily close to 1 for @xmath132
, one can make the same random walk construction for @xmath133 bounded away by @xmath134 from @xmath36 , as done in @xcite .
this implies that , for any @xmath126 , one may find a constant @xmath135 such that for any @xmath26 at least @xmath135 steps away from @xmath36 , @xmath136 with probability approaching 1 .
hence , it is reasonable to use a small cutoff to produce a close approximation to the posterior distribution .
the extension of this construction to multiple change points is straightforward .
it entails augmenting the mixture components of one change point with that of its neighboring change points .
this gives a computationally efficient way of approximating the bayesian confidence interval using , typically , a few hundred mixture components , which has been implemented in ` seqcbs ` .
there is also an extensive body of literature on constructing confidence intervals and confidence sets for estimators of the change point @xmath102 .
we refer interested readers to @xcite for discussion and efficiency comparison of various confidence sets in change - point problems .
we first applied the proposed method to a matched pair of tumor and normal nci-60 cell lines , hcc1954 and bl1954 .
@xcite conducted the sequencing of these samples using the illumina platform .
for comparison with array - based copy number profiles on the same samples , we obtained array data on hcc1954 and bl1954 from the nci-60 database at http://www.sanger.ac.uk / genetics / cgp / nci60/. we applied the cbs algorithm [ @xcite , @xcite ] with modified bic stopping algorithm [ @xcite ] to estimate relative copy numbers based on the array data . [ cols="^,^ " , ] figure [ figperfoverall ] summarizes the performance comparison at default settings for a number of spike - in signal lengths .
the horizontal lines are mean recall and precision rates for the methods . we see that seqcbs , used with either the score test statistic or the glr statistic , offers significant improvement over the existing method in both precision and recall .
the performances of the score and glr statistics are very similar , as their recall and precision curves almost overlap .
the improvement in precision can be largely attributed to the fact that mbic provides a good estimate of model complexity , as can be seen in figure [ fig6](a ) .
@ld1.3d1.3d1.3d1.3d1.3d1.3d1.3d1.3d1.3@ + length & 2.49 & 2.68 & 2.91 & 3.10 & 3.18 & 3.29 & 3.51 & 3.73 & 3.95 + + segseq & + default & 0.066 & 0.146 & 0.394 & 0.514 & 0.464 & 0.476 & 0.502 & 0.438 & 0.424 + w250 & 0.23 & 0.422 & 0.67 & 0.662 & 0.59 & 0.654 & 0.64 & 0.564 &
0.516 + w750 & & & 0.148 & 0.206 & 0.248 & 0.36 & 0.408 & 0.384 & 0.346 + a500 & & 0.148 & 0.394 & 0.514 & 0.464 & 0.476 & 0.502 & 0.438 & 0.424 + a2000 & & 0.146 & 0.394 & 0.514 & 0.464 & 0.476 & 0.502 & 0.438 & 0.424 + b25 & & 0.182 & 0.404 & 0.532 & 0.484 & 0.502 & 0.506 & 0.452 & 0.458 + b5 & & 0.126 & 0.382 & 0.476 & 0.432 & 0.442 & 0.476 & 0.428 & 0.364 + seqcbs & + scr - def & 0.49 & 0.714 & 0.95 & 0.988 & 0.95 & 0.936 & 0.968 & 0.878 & 0.782 + bin - def & 0.492 & 0.718 & 0.948 & 0.99 & 0.956 & 0.936 & 0.968 & 0.876 & 0.81 + scr - g5 & 0.496 & 0.71 & 0.922 & 0.99 & 0.956 & 0.956 & 0.978 & 0.922 & 0.844 + bin - g5 & 0.496 & 0.712 & 0.928 & 0.99 & 0.958 & 0.962 & 0.98 & 0.946 & 0.844 + scr - g15 & 0.494 & 0.708 & 0.926 & 0.974 & 0.942 & 0.938 & 0.968 & 0.89 & 0.736 + bin - g15 & 0.496 & 0.716 & 0.93 & 0.976 & 0.946 & 0.96 & 0.972 & 0.91 & 0.748 + + segseq & + default & 0.049 & 0.105 & 0.235 & 0.305 & 0.284 & 0.263 & 0.276 & 0.237 & 0.255 + w250 & 0.174 & 0.317 & 0.490 & 0.472 & 0.467 & 0.478 & 0.442 & 0.405 &
0.399 + w750 & & & 0.107 & 0.137 & 0.165 & 0.227 & 0.242 & 0.232 & 0.212 + a500 & & 0.097 & 0.235 & 0.305 & 0.284 & 0.263 & 0.276 & 0.237 & 0.255 + a2000 & & 0.101 & 0.235 & 0.305 & 0.284 & 0.263 & 0.276 & 0.237 & 0.255 + b25 & & 0.101 & 0.203 & 0.278 & 0.254 & 0.243 & 0.246 & 0.219 & 0.240 + b5 & & 0.104 & 0.278 & 0.361 & 0.323 & 0.308 & 0.327 & 0.295 & 0.271 + seqcbs & + scr - def & 0.980 & 0.997 & 0.985 & 0.988 & 0.985 & 0.944 & 0.968 & 0.878 & 0.839 + bin - def & 0.984 & 0.997 & 0.988 & 0.990 & 0.992 & 0.947 & 0.968 & 0.876 & 0.884 + scr - g5 & 0.984 & 0.997 & 0.956 & 0.990 & 0.992 & 0.980 & 0.994 & 0.945 & 0.942 + bin - g5 & 0.984 & 0.994 & 0.959 & 0.990 & 0.994 & 0.990 & 0.996 & 0.977 & 0.942 + scr - g15 & 0.980 & 0.994 & 0.953 & 0.944 & 0.961 & 0.949 & 0.964 & 0.876 & 0.710 + bin - g15 & 0.984 & 0.994 & 0.953 & 0.946 & 0.973 & 0.984 & 0.972 & 0.910 & 0.733 + we studied the performance sensitivity on tuning parameters .
segseq allows three tuning parameters : local window size ( w ) , number of false positive candidates for initialization ( a ) , and number of false positive segments for termination ( b ) .
the proposed method has a step size parameter ( g ) that controls the trade - off between speed and accuracy in our iterative grid scan component , and hence influences performance .
we varied these parameters and recorded the performance measures in table [ tabperformancemeasures ] .
it appears that local window size ( w ) is an important tuning parameter for segseq , and in scenarios with relatively short signal length , a smaller w@xmath137250 provides significant improvement in its performance .
this echoes with our previous discussion that methods using a single fixed window size would perform less well when the signals are not of the corresponding length .
some of the parameter combinations for segseq result in program running errors in some scenarios , and are marked as na .
the step size parameter ( g ) in seqcbs , in constrast , controls the rate at which coarse segment candidates are refined and the rate at which the program descends into searching smaller local change points , rather than defining a fixed window size .
a smaller step size typically yields slightly better performance .
however , the proposed method is not nearly as sensitive to its tuning parameters .
we also conducted a timing experiment to provide the reader with a sense of the required computational resources to derive the solution .
our proposed method compares favorably with segseq as seen in figure [ fig6](b ) .
the glr statistic is slightly more complex to compute than the score statistic , as is reflected in the timing experiment .
however , copy number profiling is inherently a highly parallelizable computing problem : one may distribute the task for each chromosome among a multi - cpu computing grid , hence dramatically reducing the amount of time required for this analysis .
we proposed an approach based on nonhomogeneous poisson processes to directly model next - generation dna sequencing data , and formulated a change - point model to conduct copy number profiling . the model yields simple score and generalized likelihood ratio statistics , as well as a modified bayes information criterion for model selection .
the proposed method has been applied to real sequencing data and its performance compares favorably to an existing method in a spike - in simulation study .
statistical inference , in the form of confidence estimates , is very important for sequencing - based data , since , unlike arrays , the effective sample size ( i.e. , coverage ) for estimating copy number varies substantially across the genome . in this paper
, we derived a procedure to compute bayesian confidence intervals on the estimated copy number .
other types of inference , such as @xmath0-values or confidence intervals on the estimated change points , may also be useful .
@xcite compares different types of confidence intervals on the change points , and the methods there can be directly applied to this problem .
the reader is referred to @xcite and @xcite for existing methods on significance evaluation .
some sequencing experiments produce paired end reads , where two short reads are performed on the two ends of a longer fragment of dna
. the pairing information can be quite useful in the profiling of structural genomic changes .
it will be important to extend the approach in this paper to handle this more complex data type .
a limitation of the proposed method and the existing methods is that they do not handle allele - specific copy number variants .
it is possible to extend our model to accommodate this need . with deep sequencing
, one may assess whether each loci in a cnv is heterozygous , and estimate the degree to which each allele contributes to the gain or loss of copy number , by considering the number of reads covering the locus with the major allele versus those with the minor allele .
this is particularly helpful for detecting deletion .
furthermore , in the context of assessing the allele - specific copy number , existing snp arrays have the advantage that the assay targets specific sites for that problem , whereas to obtain sufficient evidence of allele - specific copy number variants with sequencing , a much greater coverage would be required since the overwhelming majority of reads would land in nonallelic genomic regions .
spatial models that borrow information across adjacent variant sites , such as @xcite and @xcite , would be helpful for improving power .
recently , there has been increased attention to the problem of simultaneous segmentation of multiple samples [ @xcite ; @xcite ; @xcite ; @xcite ] .
one may also wish to extend this method to the multi - sample setting , where in addition to modeling challenges , one also needs to address more sources of systematic biases , such as batch effects and carry - over problems .
computational challenges remain in this field . with sequencing capacity growing at record speed ,
even basic operations on the data set are resource - consuming .
it is pertinent to develop faster and more parallelizable solutions to the copy number profiling problem .
we thank h. p. ji , g. walther and d. o. siegmund for their inputs . | we propose a flexible change - point model for inhomogeneous poisson processes , which arise naturally from next - generation dna sequencing , and derive score and generalized likelihood statistics for shifts in intensity functions .
we construct a modified bayesian information criterion ( mbic ) to guide model selection , and point - wise approximate bayesian confidence intervals for assessing the confidence in the segmentation .
the model is applied to dna copy number profiling with sequencing data and evaluated on simulated spike - in and real data sets . . |
a possibly degenerate circuit is a point configuration @xmath0 of cardinality @xmath1 which span a sublattice @xmath2 of rank @xmath3
. that is , such that the newton polytope @xmath4 is of full dimension .
a polynomial system @xmath5 is said to be supported on a circuit @xmath6 if each polynomial occurring in @xmath7 is supported on @xmath6 .
polynomial systems supported on circuits have recently been been studied in the context of , e.g. , real algebraic geometry @xcite , complexity theory @xcite , and amoeba theory @xcite .
the name `` circuit '' originate from matroid theory ; see @xcite and @xcite for further background .
the aim of this article is to describe geometrical and topological properties of coamoebas of polynomials supported on circuits .
such an investigation is motivated not only by the vast number of applications of circuits in different areas of geometry , but also since circuits provide an ideal testing ground for open problems in coamoeba theory .
this paper is organized as follows . in section [ sec : coamoebas ] we will give a brief overview of coamoeba theory . in section [ sec : realpoints ] we will discuss the relation between real polynomials and the coamoeba of the @xmath6-discriminant .
the main results of this paper are contained in sections [ sec : configurationspace][sec : systems ] , each of which can be read as a standalone text . in section [ sec : configurationspace ] we will give a complete description of the space of coamoebas .
that is , we will describe how the topology of the coamoeba @xmath8 depends on the coefficients of @xmath9 . describing
the space of amoebas is the topic of the articles @xcite and @xcite , and to fully appreciate our result one should consider these spaces simultaneously , see e.g. figure [ fig : configurationspace ] .
the geometry of the space of coamoebas is closely related to the @xmath6-discriminantal variety , see theorems [ thm : configurationspaceinteriorpoint ] and [ thm : configurationspacevertex ] . in section [ sec : maximalarea ] we will prove that the area of a planar circuit coamoeba is bounded from above by @xmath10 .
that is , a planar circuit coamoeba covers at most half of the torus @xmath11 .
furthermore , we will prove that a circuit admits a coamoeba of maximal area if and only if it admits an equimodular triangulation .
note that we calculate area without multiplicities , in contrast to @xcite .
however , the relation between ( co)amoebas of maximal area and harnack curves is made visible also in this setting . in section [ sec : criticalpoints ] we will prove that , under certain assumptions on @xmath6 , the critical points of @xmath7 are projected by the componentwise argument mapping into distinct connected component of the complement of the coamoeba @xmath8 . furthermore , this projection gives a bijective relation between the set of critical points and the set of connected components of the complement of the closed coamoeba .
this settles a conjecture used in in @xcite when computing monodromy in the context of dimer models and mirror symmetry . in section [ sec :
systems ] we will consider bivariate systems supported on a circuit . if such a system is real , then it admits at most three roots in @xmath12 .
the main contribution of this section is that we offer a new approach to fewnomial theory . using our method
, we will prove that if @xmath13 is a simplex , then , for each @xmath14 , a complex bivariate system supported on @xmath6 has at most two roots in the _ sector _ @xmath15 .
let @xmath6 denote a point configuration @xmath16 , where @xmath17 . by abuse of notation , we identify @xmath6 with the @xmath18-matrix @xmath19 the _ codimension _ of @xmath6 is the integer @xmath20 .
a _ circuit _ is a point configuration of codimension one .
a circuit is said to be nondegenerate if it is not a pyramid over a circuit of smaller dimension .
that is , if all maximal minors of the matrix @xmath6 are nonvanishing .
we will partition the set of circuits into two classes ; _ simplex circuits _ , for which @xmath13 is a simplex , and _ vertex circuits _ , for which @xmath21 .
we associate to @xmath6 the family @xmath22 consisting of all polynomials @xmath23 where @xmath7 is identified with the point @xmath24 . by slight abuse of notation
, we will denote by @xmath25 the monomial function @xmath26 .
we denote the algebraic set defined by @xmath9 by @xmath27 .
the _ coamoeba _
@xmath8 is the image of @xmath28 under the componentwise argument mapping @xmath29 defined by @xmath30 where @xmath31 denotes the real @xmath3-torus .
it is sometimes beneficial to consider the multivalued argument mapping , which gives the coamoeba as a multiply periodic subset of @xmath32 .
coameobas were introduced by passare and tsikh as a dual object , in an imprecise sense , of the amoeba @xmath33 .
we will say that a point @xmath34 is a _ critical point _ of @xmath9
if it solves the system @xmath35 if in addition @xmath36 then @xmath37 will be called a _ singular point _ of @xmath9 .
the @xmath6-discriminant @xmath38 is an irreducible polynomial with domain @xmath39 which vanishes if and only if @xmath9 has a singular point in @xmath40 @xcite .
a _ gale dual _ of @xmath6 is an integer matrix @xmath41 whose columns span the right @xmath42-kernel of @xmath6 .
that is , @xmath41 is an integer @xmath43-matrix , of full rank , such that its maximal minors are relatively prime .
a gale dual is unique up to the action of @xmath44 .
the rows @xmath45 of @xmath41 are indexed by the points @xmath46 . to each gale
dual we associate a zonotope @xmath47 we will say that a triangulation @xmath48 of @xmath13 is a triangulation of @xmath6 if @xmath49 .
such a triangulation is said to be _ equimodular _ if all maximal simplices has equal volume .
let @xmath50 be a _ height function _ @xmath51 .
the function @xmath50 induces a triangulation @xmath52 of @xmath6 in the following manner .
let @xmath53 denote the polytope in @xmath54 with vertices @xmath55 .
the lower facets of @xmath53 are the facets whose outward normal vector has negative last coordinate . then , @xmath52 is the triangulation of @xmath6 whose maximal simplices are the images of the lower facets of @xmath53 under the projection onto the first @xmath3 coordinates .
a triangulation @xmath48 of @xmath6 is said to be _ coherent _ if there exists a height function @xmath50 such that @xmath56 .
if @xmath6 is a circuit then @xmath41 is a column vector , unique up to sign .
hence , the zonotope @xmath57 is an interval .
let @xmath58 , with associated matrix @xmath59 , and let @xmath60 . if @xmath6 is a nondegenerate circuit , so that @xmath61 for all @xmath62 , then @xmath13 admits exactly two coherent triangulations with vertices in @xmath6 @xcite .
denote these two triangulations by @xmath63 for @xmath64 .
each simplex @xmath65 occurs in exactly one of the triangulations @xmath63 .
hence , there is a well - defined assignment of signs @xmath66 , where @xmath67 , such that @xmath68 here , we have identified a triangulation with its set of maximal simplices . as shown in ( * ? ? ?
* chp . 7 and chp
. 9 ) and @xcite , a gale dual of @xmath6 is given by @xmath69 thus , the zonotope @xmath57 is an interval of length @xmath70 .
the @xmath6-discriminant @xmath71 has @xmath72 homogeneities , one for each row of the matrix @xmath6 .
each gale dual correspond to a dehomogenization of @xmath71 . to be specific , introducing the variables @xmath73
there is a laurent monomial @xmath74 and a polynomial @xmath75 such that @xmath76 we will say that @xmath77 is the _ reduced form _ of @xmath71 .
such a reduction yields a projection @xmath78 , and we will say that @xmath79 is the reduced family associated to @xmath6 , and that @xmath80 is the reduced form of @xmath9 .
let @xmath81 , so that @xmath39 is the family of quadratic univariate polynomials @xmath82 consider the gale dual @xmath83 , and introduce the variable @xmath84 . in this case
the @xmath6-discriminant @xmath85 is well - known , and we find that @xmath86 the projection @xmath87 correspond to performing the change of variables @xmath88 , and multiplying @xmath7 by @xmath89 , after with we obtain the reduced family consisting of all polynomials of the form @xmath90 let @xmath91 denote a subset of @xmath6 .
the _ truncated polynomial _
@xmath92 is the image of @xmath9 under the projection @xmath93 .
of particular interest is the case when @xmath94 for some face @xmath95 of the newton polytope @xmath13 ( denoted by @xmath96 ) .
we will write @xmath97 .
it was shown in @xcite that @xmath98 let @xmath99 denote the set of edges of @xmath13 , then the _ shell _ of the coamoeba is defined by @xmath100 as an edge @xmath95 is one dimensional , the shell @xmath101 is a hyperplane arrangement .
its importance can be seen in that each full - dimensional cell of @xmath101 contain at most one connected component of the complement of @xmath102 , see @xcite .
the coamoeba of @xmath103 , as described in @xcite and @xcite , can be seen in figure [ fig : simplexcoamoeba ] , where it is drawn in the fundamental domains @xmath104 ^ 2 $ ] and @xmath105 ^ 2 $ ] .
the shell @xmath101 consist of the hyperplane arrangement drawn in black . in this case
, it is equal to the boundary of @xmath8 . the newton polytope @xmath13 and its outward normal vectors are drawn in the rightmost picture . if @xmath101 is given orientations in accordance with the outward normal vectors of @xmath13 , then the interior of the coamoeba consist of the oriented cells . in two fundamental regions , and the newton polytope @xmath13.,title="fig:",width=113 ] in two fundamental regions , and the newton polytope @xmath13.,title="fig:",width=113 ] in two fundamental regions , and the newton polytope @xmath13.,title="fig:",width=113 ] acting on @xmath6 by an _ integer affine transformation _ is equivalent to performing a monomial change of variables and multiplying @xmath9 by a laurent monomial .
such an action induces a linear transformation of the coamoeba @xmath8 , when viewed in @xmath32 @xcite .
we will repeatedly use this fact to impose assumptions on @xmath6 , e.g. , that it contains the origin .
the polynomial @xmath9 is said to be _ colopsided _ at a point @xmath106 if there exist a phase @xmath107 such that @xmath108 with at least one of the inequalities being strict .
the motivation for this definition is as follows .
if @xmath9 is colopsided at @xmath109 , then @xmath110 since at least one term of the sum is strictly positive . hence , colopsidedness at @xmath109 implies that @xmath111 .
the colopsided coamoeba , denoted @xmath112 , is defined as the set of all @xmath109 such that does not hold for any phase @xmath107 @xcite . hence , @xmath113 .
each monomial @xmath25 defines an affinity ( i.e. , a group homomorphism composed with a translation ) @xmath114 by @xmath26 .
we thus obtain unique affinities @xmath115 and @xmath116 such that the following diagram of short exact sequences commutes : ( o1 ) 0 ; ( r1 ) [ right of = o1 ] @xmath117 ; ( c1 ) [ right of = r1 ] @xmath40 ; ( t1 ) [ right of = c1 ] @xmath118 ; ( e1 ) [ right of = t1 ] @xmath119 ; ( o2 ) [ below of = o1 ] @xmath119 ; ( r2 ) [ right of = o2 ] @xmath120 ; ( c2 ) [ right of = r2 ] @xmath121 ; ( t2 ) [ right of = c2 ] @xmath122 ; ( e2 ) [ right of = t2 ] @xmath119 . ; ( o1 ) to node ( r1 ) ; ( r1 ) to node ( c1 ) ; ( c1 ) to node ( t1 ) ; ( t1 ) to node ( e1 ) ; ( o2 ) to node ( r2 ) ; ( r2 ) to node ( c2 ) ; ( c2 ) to node ( t2 ) ; ( t2 ) to node ( e2 ) ; ( r1 ) to node [ swap ] @xmath115 ( r2 ) ; ( c1 ) to node [ swap ] @xmath123 ( c2 ) ; ( t1 ) to node [ swap ] @xmath116 ( t2 ) ; notice that @xmath124 .
we denote by @xmath125 the vector with components @xmath126 .
assume that @xmath9 contains the constant monomial @xmath127 , and consider the map @xmath128 defined by @xmath129 where @xmath130 denotes the componentwise principal argument map .
it was shown in @xcite that the map @xmath131 induces a map @xmath132 which in turn induces a bijection between the set of connected components of the complement of @xmath133 and the finite set in the right hand side of .
the map @xmath131 is known as the _ order map _ of the lopsided coamoeba .
the requirement that @xmath9 contains the monomial @xmath127 is related to the choice of branch cut of the function @xmath134 ; in order to obtain a well - defined map , we need the right hand side of to be discontinuous only for @xmath109 such that two components of @xmath135 are antipodal , see @xcite . if @xmath9 does not contain the constant monomial @xmath127 , then one should fix a point @xmath46 and multiply the vector @xmath135 by the scalar @xmath136 before taking principal arguments .
it is shown in ( * ? ? ?
4.3 ) that the obtained map is independent of the choice of @xmath137 .
if @xmath138 , then we can choose @xmath107 such that @xmath139 that is , the boundary of @xmath112 is contained in the hyperplane arrangement consisting of all @xmath109 such that two components of @xmath135 are antipodal .
it has been conjectured that the number of connected components of the complement of @xmath102 is at most @xmath140 .
a proof in arbitrary dimension has been proposed by nisse in @xcite , and an independent proof in the case @xmath141 was given in @xcite . that the number of connected components of the complement of @xmath133 is at most @xmath140 follows from the theory of mellin
barnes integral representations of @xmath6-hypergeometric functions , see @xcite and @xcite .
a finite set @xmath142 which is in a bijective correspondence with the set of connected components of the complement of @xmath102 by inclusion , will be said to be an _
index set _ of the coamoeba complement .
this notation will be slightly abused ; a set @xmath143 of cardinality @xmath140 will be said to be an index set of the coamoeba if each connected component of its complement contains exactly one element of @xmath143 .
the term `` lopsided '' was first used by purbhoo in @xcite , denoting the corresponding condition to for amoebas : the polynomial @xmath9 is said to be _ lopsided _ at a point @xmath144 if there is a @xmath46 such that the moduli @xmath145 is greater than the sum of the remaining modulis . as a comparison ,
note that the polynomial @xmath9 is colopsided at @xmath146 if and only if the greatest intermediate angle of the components of @xmath135 is greater than the sum of the remaining intermediate angles .
we will say that @xmath9 is _ real at @xmath109 _ , if there is a real subvector space @xmath147 such that @xmath148 for all @xmath149 .
if such a @xmath109 exist then @xmath9 is real , that is , after a change of variables and multiplication with a laurent monomial @xmath150 . in this section
, we will study the function @xmath151 from the viewpoint of real polynomials .
our main result is the following characterization of the coamoeba of the @xmath6-discriminant of a circuit .
[ pro : discriminantmonomialarguments ] let @xmath6 be a nondegenerate circuit , and let @xmath152 be as in .
then , @xmath153 if and only if after possibly multiplying @xmath9 with a constant , there is a @xmath106 such that @xmath154 for all @xmath62 .
if @xmath6 is a circuit and @xmath41 is a gale dual of @xmath6 then the horn
kapranov parametrization of the reduced discriminant @xmath77 can be lifted to a parametrization of the discriminant surface @xmath71 as @xmath155 taking componentwise arguments , we obtain a simple proof proposition [ pro : discriminantmonomialarguments ] . in particular , the proposition can be interpreted as a coamoeba version of the horn
kapranov parametrization valid for circuits .
our proof of proposition [ pro : discriminantmonomialarguments ] will be more involved , however , for our purposes the lemmas contained in this section are of equal importance .
[ lemma : realpointsduallattice ] assume that the polynomial @xmath9 is real at @xmath156 . then , @xmath9 is real at @xmath106 if and only if @xmath157 , where @xmath158 is the dual lattice of @xmath2 .
after translating @xmath109 and multiplying @xmath9 with a laurent monomial , we can assume that @xmath159 , that @xmath160 , and that @xmath9 contains the monomial @xmath127 .
that is , all coefficients of @xmath9 are real , in particular proving _
if_-part of the statement . to show the _ only if_-part ,
notice first that @xmath161 implies that @xmath162 contains both the origin and @xmath127 .
that is , @xmath163 .
furthermore , @xmath164 only if for each @xmath165 there is a @xmath166 such that @xmath167 , which concludes the proof .
the @xmath6-discriminant @xmath71 related to a circuit has been described in ( * ? ? ?
* chp . 9 , pro . 1.8 ) where the formula @xmath168 was obtained . in particular
, @xmath71 is a binomial .
as the zonotope @xmath57 is a symmetric interval of length @xmath169 , the image of the map @xmath131 is of cardinality @xmath140 unless @xmath170 in particular , the complement of @xmath102 has the maximal number of connected components ( i.e. , @xmath140-many ) unless the equivalence holds .
[ lem : simplexequalarguments ] for each @xmath171 , there are exactly @xmath172-many points @xmath173 such that @xmath174 by applying an integer affine transformation , the statement follows from the case when @xmath175 consist of the vertices of the standard simplex .
[ lem : uniquelastarguments ] fix @xmath176 . for each @xmath109 fulfilling ,
let @xmath177 be defined by the condition that if @xmath178 then @xmath179 assume that @xmath180 .
then , the numbers @xmath181 are distinct .
we can assume that @xmath182 and that @xmath183 .
assume that @xmath184 .
then , @xmath185 by translating , we can assume that @xmath186 , and hence , since @xmath127 is a monomial of @xmath9 , that all coefficients are real .
consider the lattice @xmath158 consisting of all points @xmath187 such that @xmath9 is real at @xmath109 .
since @xmath180 , lemma [ lemma : realpointsduallattice ] shows that @xmath188 .
however , we find that @xmath189 and hence @xmath190 .
this implies that @xmath191 , and hence @xmath192 in @xmath31 .
assume first that there is a @xmath109 as in the statement of the proposition , where we can assume that @xmath193 .
then , @xmath194 .
it follows that the monomials @xmath195 have equal signs .
therefor , @xmath71 vanishes for @xmath196 , implying that @xmath153 . for the converse ,
fix @xmath197 , and reduce @xmath9 by requiring that @xmath198 for @xmath199 .
let @xmath143 denote the set of points @xmath200 such that @xmath154 for @xmath199 , which by lemma [ lem : simplexequalarguments ] has cardinality @xmath201 .
by lemma [ lem : uniquelastarguments ] , the set @xmath143 is in a bijective correspondence with values of @xmath202 such that @xmath203 .
therefor , we find that @xmath71 vanishes at @xmath204 for each @xmath205 . however , the discriminant @xmath71 specializes , up to a constant , to the binomial @xmath206 which has exactly @xmath201-many solutions in @xmath121 of distinct arguments .
hence , since @xmath207 by assumption , and comparing the number of solutions , it holds that @xmath203 for one of the points @xmath208 .
let @xmath209 denote the set of all @xmath9 such that the number of connected components of the complement of @xmath102 is @xmath210 . describing the sets @xmath211
is known as the problem of describing the _ space of coamoebas _ of @xmath39 . in this section , we will give explicit descriptions of the sets @xmath211 in the case when @xmath6 is a circuit .
as a first observation we note that the image of the map @xmath131 is at least of cardinality @xmath212 , implying that @xmath213 and in particular @xmath214 for @xmath215 .
hence , it suffices for us to give an explicit description of the set @xmath216 .
our main result is the following two theorems , highlighting also the difference between vertex circuits and simplex circuits .
note that @xmath71 is a real polynomial @xcite .
[ thm : configurationspaceinteriorpoint ] assume that @xmath6 is a nondegenerate simplex circuit , with @xmath217 as an interior point .
choose @xmath41 such that @xmath218 , and let @xmath71 be as in .
then , @xmath219 if and only if @xmath220 and @xmath221 [ thm : configurationspacevertex ] assume that @xmath6 is a vertex circuit . then , @xmath219 if and only if @xmath220 .
the article @xcite describes the _ space of amoebas _ in the case when @xmath6 is a simplex circuit in dimension at least two . in this case , the number of connected components of the amoeba complement is either equal to the number of vertices of @xmath13 or one greater .
one implication of ( * ? ? ?
* thm . 4.4 and thm .
5.4 ) is that , if the amoeba complement has the minimal number of connected components , then @xmath222 furthermore , this set intersect @xmath216 only in the discriminant locus @xmath207 .
the space of amoebas in the case when @xmath6 is a simplex circuit in dimension @xmath223 has been studied in @xcite , and is a more delicate problem . on the other hand ,
if @xmath6 is a vertex circuit , then each @xmath224 is maximally sparse and hence has a solid amoeba .
that is , the components of the complement of the amoeba is in a bijective correspondance with the vertices of the newton polytope @xmath13 .
in particular , the number of connected components of the amoeba complement does not depend on @xmath9 . from theorems [ thm : configurationspaceinteriorpoint ] and
[ thm : configurationspacevertex ] we see that a similar discrepancy between simplex circuits and vertex circuits occur for coamoebas .
[ ex : configurationspaces ] the reduced family @xmath225 was considered in ( * ? ? ?
* ex . 6 , p. 59 ) , where the study of the space of amoebas was initiated .
we have drawn the space of amoebas and coamoebas jointly in the left picture in figure [ fig : configurationspace ] .
the blue region , whose boundary is a hypocycloid , marks values of @xmath226 for which the amoeba complement has no bounded component .
the set @xmath216 is seen in orange .
the red dots is the discriminant locus @xmath227 , which is contained in the circle @xmath228 corresponding an equality in .
[ ex : configurationspaces2 ] the reduced family @xmath229 is a vertex circuit . in this case , the topology of the amoeba does not depend on the coefficient @xmath226 .
the space of coamoebas is drawn in the right picture in figure [ fig : configurationspace ] .
the set @xmath216 comprises the three orange lines emerging from the origin .
the red dots is the discriminant locus @xmath227 .
it might seem like the set @xmath230 is disconnected , however this a consequence of that we consider @xmath9 in reduced form . in @xmath39
the set @xmath230 is connected , though not simply connected . and
[ ex : configurationspaces2].,title="fig:",width=113 ] @xmath231 and [ ex : configurationspaces2].,title="fig:",width=113 ] impose the assumptions of theorem [ thm : configurationspaceinteriorpoint ] .
then , @xmath232 and in particular @xmath233 . by lemma [ lem : simplexequalarguments ]
there is a set @xmath143 of cardinality @xmath140 consisting of all points @xmath109 such that @xmath234 in particular , @xmath9 is colopsided at @xmath208 unless @xmath235 .
it was shown in @xcite that , if @xmath236 , then @xmath143 is an index set for the complement of @xmath102 .
in fact , @xmath143 is an index set of the complement of @xmath102 for arbitrary @xmath9 .
[ pro : testingtheta ] let @xmath6 be a simplex circuit .
assume that @xmath220 , i.e. , that there exists a @xmath208 with @xmath237 .
then , the complement of @xmath102 has @xmath140-many connected components if and only if it contains @xmath109 .
we can assume that @xmath193 .
to prove the _
if_-part , assume that @xmath238 for some connected component @xmath239 of the complement of @xmath102 .
we wish to show that @xmath9 is never colopsided in @xmath239 , for this implies that the complement of @xmath102 has @xmath140-many connected components .
assume to the contrary that there exist a point @xmath240 such that @xmath9 is colopsided at @xmath241 .
then , @xmath242 for some integer @xmath243 , with @xmath244 , see .
let @xmath245 .
then @xmath246 is colopsided at @xmath119 for @xmath247 . by continuity of roots , for @xmath248
sufficiently small , the points @xmath119 and @xmath241 are contained in the same connected component of the complement of @xmath249 .
hence , by ( * ? ? ?
3.9 ) , they are contained in the same connected component of the complement of @xmath250 .
however , @xmath251 contradicting that @xmath252 is constant on connected components of the complement of @xmath253 . to prove the _ only if_-part ,
assume that there exists a connected component @xmath239 of the complement of @xmath102 in which @xmath9 is never colopsided .
we wish to prove that @xmath254 .
as @xmath246 is colopsided at @xmath119 for @xmath248 sufficiently small , we find that @xmath255 . indeed ,
if this was not the case , then the complement of @xmath249 has @xmath256-many connected components , a contradiction .
as @xmath257 , and by ( * ? ? ?
2.3 ) , there exists a disc @xmath258 around @xmath119 such that @xmath259 furthermore , @xmath260 , since @xmath261 .
let @xmath262 .
as @xmath9 is a real polynomial , conjugation yields that @xmath263 .
however , @xmath264 is convex , implying that @xmath238 . if @xmath265 then the image of @xmath131 is of cardinality @xmath140 , and hence @xmath266 .
thus , we only need to consider @xmath267 , where we can assume that @xmath268 for all @xmath62 .
in particular , @xmath9 is a real polynomial . by proposition [ pro : testingtheta ]
, it holds that the complement of @xmath102 has @xmath140-many connected components if and only if it contains @xmath119 . keeping @xmath269 and @xmath270 fixed ,
let us consider @xmath9 as a function of @xmath271 . as @xmath9 is a real polynomial , it restricts to a map @xmath272 , whose image depends nontrivially on @xmath271 .
notice that @xmath273 if and only if @xmath274 contains the origin . since @xmath275 for @xmath276 , and since @xmath217 is an interior point of @xmath6 , the map @xmath9 takes the boundary of @xmath277 to @xmath278 . in particular , if @xmath279 , then @xmath280 .
the boundary of the set of all @xmath271 for which @xmath279 is the set of all values of @xmath271 for which @xmath281 .
furthermore , @xmath282 holds if and only if there exists an @xmath283 such that @xmath284 , while @xmath285 in a neighborhood of @xmath286 , implying that @xmath286 is a critical point of @xmath9 .
that is , @xmath287 since @xmath71 is a binomial , there is exactly one such value of @xmath271 .
finally , we note that @xmath288 if @xmath289 , which finishes the proof .
if @xmath290 , then the image of @xmath131 is of cardinality @xmath140 and hence @xmath266 .
assume that @xmath153 , and that @xmath291 for all @xmath62 .
it holds that @xmath292 since there exists two adjacent vertices @xmath293 and @xmath294 of @xmath6 such that @xmath295 and @xmath296 .
let , @xmath297 for @xmath298 sufficiently small , continuity of roots implies that there is a connected component @xmath299 in which @xmath246 is never colopsided .
however , by choosing the sign of @xmath300 , we can force @xmath301 .
this implies that the coamoeba @xmath302 has @xmath256-many connected components of its complement , a contradiction .
in this section , we will prove the following bound .
[ thm : areabound ] let @xmath6 be a planar circuit , and let @xmath224
. then @xmath303 .
furthermore , we provide the following classification of for which circuits the bound of theorem [ thm : areabound ] is sharp .
[ thm : unimodularmaximalthm ] let @xmath6 be a planar circuit . then there exist a polynomial @xmath224 such that @xmath304 if and only if @xmath6 admits an equimodular triangulation .
[ ex : maximalareasquare ] let @xmath305 for @xmath306 .
notice that @xmath6 admits a unimodular triangulation .
the shell @xmath101 consist of the families @xmath307 and @xmath308 for @xmath309 .
hence , the shell @xmath101 divides @xmath11 into four regions of equal area .
exactly two of these regions are contained in the coamoeba , which implies that @xmath310 .
see the left picture of figure [ fig : maximalareaexamples ] .
[ ex : maximalareainterior ] let @xmath311 for @xmath306 .
also in this case @xmath6 admits a unimodular triangulation .
notice that @xmath153 .
the coamoeba of the trinomial @xmath312 has three components of its complement , of which @xmath9 is colopsided in two .
we have that @xmath313 .
thus , if the complement of @xmath102 has two connected components , i.e. if @xmath314 , then one of the three components of the complement of @xmath315 is contained in @xmath102 , which in turn implies that @xmath304 .
see the right picture of figure [ fig : maximalareaexamples ] . and
[ ex : maximalareainterior].,title="fig:",height=113 ] and [ ex : maximalareainterior].,title="fig:",height=113 ] let us compare our results to the corresponding results of planar circuit amoebas .
it was shown in ( * ? ? ?
12 , p. 30 ) that the sharp upper bound on the number of connected components of the amoeba complement is @xmath316 . in @xcite ,
a bound on the area of the amoeba was given as @xmath317 , and it was shown that maximal area was obtained for harnack curves .
for coamoebas , to roles of the integers @xmath140 and @xmath316 are reversed .
the upper bound on the number of connected components of the coamoeba complement is given by @xmath140 .
while , at least for codimension @xmath318 , the maximal area of the coameoba is @xmath319 .
note also that the coamoebas of examples [ ex : maximalareasquare ] and [ ex : maximalareainterior ] are both coamoebas of harnack curves . consider a bivariate trinomial @xmath9 , with one marked monomial .
let @xmath320 denote the quadruple of polynomials obtained by flipping signs of the unmarked monomials .
furthermore , let @xmath321 which is a hyperplane arrangement in @xmath322 ( or @xmath11 ) .
let @xmath323 denote the set of all intersection points of distinct hyperplanes in @xmath324 .
[ pro : unioncovers ] let @xmath7 be a bivariate trinomial .
then , the union @xmath325 covers @xmath322 . to be specific ,
@xmath323 is covered thrice , @xmath326 is covered twice , and @xmath327 is covered once . after applying an integer affine transformations ,
we reduce to the case when @xmath6 consist of the vertices of the standard simplex .
this case that follows from the description in @xcite and @xcite , see also figure [ fig : simplexcoamoeba ] . if @xmath7 is a bivariate trinomial , then @xmath328
. the coamoebas appearing in the union @xmath329 , when considered in @xmath322 , are merely translations of eachother .
hence , they have equal area .
as they cover the torus once a.e . , and @xmath330 , the result follows .
notice that a bivariate trinomial is not supported on a circuit , but on the vertex set of a simplex .
let @xmath331 denote the image of @xmath9 under the projection @xmath332 .
as shown in @xcite the family of trinomials @xmath331 , @xmath333 , contains all necessary information about the lopsided coamoeba @xmath112 .
[ lem : twocolopsidedtrinomial ] let @xmath6 be a planar circuit , and let @xmath224 .
assume that @xmath334 is generic in the sense that no two components of @xmath135 are antipodal , and assume furthermore that @xmath9 is _ not _ colopsided at @xmath109 .
then , exactly two of the trinomials @xmath335 are colopsided at @xmath109 .
fix an arbitrary point @xmath336 , and let @xmath147 denote the real subvector space containing @xmath337 .
as @xmath9 is not colopsided at @xmath109 , both half spaces relative @xmath162 contains at least one component of @xmath135 .
there is no restriction to assume that the upper half space contains the two components @xmath338 and @xmath339 , and that the latter is of greatest angular distance from @xmath337 .
then , @xmath340 is colopsided at @xmath109 . furthermore , we find that @xmath341 is not colopsided at @xmath109 , for if it where then so would @xmath9 . as @xmath342 and @xmath343 ,
there is at least one trinomial obtained from @xmath9 containing @xmath294 which is not colopsided at @xmath109 , and at least one which is colopsided at @xmath109 .
as @xmath294 was arbitrary , it follows that exactly two of the trinomials @xmath344 are colopsided at @xmath109 , and exactly two are not . by containment
, it holds that @xmath345 , and thus it suffices to calculate the area of @xmath112 . by ( * ? ? ?
3.4 ) , we have that @xmath346 for a generic point @xmath347 , lemma [ lem : twocolopsidedtrinomial ] gives that @xmath109 ( and , in fact , a small neighborhood of @xmath109 ) is contained in the interior of exactly two out of the four coamoebas in the right hand side of .
hence , @xmath348 to prove the _ if _ part , we will prove that @xmath6 admits an equimodular triangulation only if , after applying an integer affine transformation , it is equal to the point configuration of either example [ ex : maximalareasquare ] or example [ ex : maximalareainterior ] .
assume that @xmath349 , and @xmath350 are vertices of @xmath13 .
after applying an integer affine transformation , we can assume that @xmath351 , that @xmath352 with @xmath353 , and that @xmath354 .
notice that such a transformation rescales @xmath6 , though it does not affect the area of the coamoeba @xmath8 @xcite .
let @xmath355 . if @xmath6 is a vertex circuit , then each triangulation of @xmath6 consist of two simplices , which are of equal area by assumption . comparing the areas of the subsimplices of @xmath6
, we obtain the relations @xmath356 in @xmath243 , this system has @xmath357 as the only nontrivial solution , and we conclude that @xmath6 is the unit square , up to integer affine transformations . if @xmath6 is a simplex circuit , then @xmath6 has one triangulation with three simplices of equal area . comparing areas
, we obtain the relations @xmath358 thus , @xmath359 and @xmath360 , and we conclude that @xmath6 is the simplex from example [ ex : maximalareainterior ] , up to integer affine transformations . to prove the _ only if_-statement , consider @xmath224 .
let @xmath361 be such that the line segment @xmath362 $ ] is interior to @xmath13 . applying an integer affine transformation
, we can assume that @xmath362\subset\rr\e_1 $ ] , and that @xmath350 and @xmath363 lies in the upper and lower half space respectively .
then , the hyperplane arrangement @xmath364 consist of @xmath365$]-many lines , each parallel to the @xmath366-axis .
if @xmath367 and @xmath368 , then @xmath339 and @xmath369 takes @xmath370 respectively @xmath371 turns around the origin when @xmath109 traverses once a line of @xmath372 .
notice that @xmath373 , as @xmath337 and @xmath338 are antipodal for @xmath374 .
that is , for such @xmath109 , @xmath375 is contained in a real subvector space @xmath376 .
assume that @xmath9 is colopsided for some @xmath377 , so that in particular @xmath378 . if @xmath379 ,
then at exactly one of the points @xmath339 and @xmath369 is contained in @xmath380 , for otherwise @xmath9 would not be colopsided at @xmath109 . by wiggling @xmath109 in @xmath372
we can assume that @xmath381 . under this assumption , we find that @xmath382 .
thus , there is a neighborhood @xmath383 which is separated from @xmath102 . as @xmath384 , the intersection @xmath385 has positive area , implying that @xmath386 .
thus , if @xmath9 is such that @xmath387 , then @xmath9 can never be colopsided in @xmath372 . in particular , for @xmath377 such that @xmath388 , it must hold @xmath389 , and vice versa . as there are @xmath390 points of the first kind , and @xmath391 points of the second kind , it holds that @xmath392 .
hence , the simplices with vertices @xmath393 and @xmath394 have equal area . if @xmath6 is a vertex cicuit , this suffices in order to conclude that @xmath6 admits an equimodular triangulation .
if @xmath6 is a simplex circuit , then we can assume that @xmath294 is an interior point of @xmath13 .
repeating the argument for either @xmath395 or @xmath396 yields that @xmath6 has a triangulation with three triangles of equal area .
that is , it admits an equimodular triangulation .
let @xmath397 denote the critical points of @xmath9 , that is , the variety defined by .
let @xmath398 denote the coamoeba of @xmath397 .
we will say that @xmath143 is the set of _ critical arguments _ of @xmath9 . in this section we will prove that , under certain assumptions on @xmath6 , the set @xmath143 is an index set of the coamoeba complement . that it is necessary to impose assumptions on @xmath6 is related to the fact that an integer affine transformation acts nontrivially on the set of critical points @xmath397 .
let @xmath6 be a circuit , with the elements @xmath165 ordered so that it has a gale dual @xmath399 such that @xmath400 and that @xmath401 .
that is , @xmath402 has only positive entries , while @xmath403 has only negative entries .
we have that @xmath404 .
let @xmath405 denote the corresponding decomposition of the matrix @xmath6 .
we will say that @xmath6 is in _ orthogonal form _ if @xmath406 where @xmath407 is an @xmath408-matrix and @xmath409 is an @xmath410-matrix . in particular
, the newton polytopes @xmath411 and @xmath412 has @xmath413 as a relatively interior point , and as their only intersection point . with @xmath6 in the form , we can act by integer affine transformations affecting @xmath407 and @xmath409 separately .
therefor , if @xmath6 is in orthagonal form , then we can assume that @xmath414 where @xmath415 are positive integers , and hence @xmath416 has only positive coordinates .
we will say that @xmath6 is in _ special orthogonal form _ if holds .
the main result of this section is the following lemma and theorem .
[ lem : orthogonalform ] each circuit @xmath6 can be put in ( special ) orthogonal form by applying an integer affine transformation .
[ thm : criticalpoints ] let @xmath6 be a circuit in special orthogonal form .
then , for each @xmath224 , the set of critical arguments is an index set of the complement of @xmath102 .
the conditions of theorem [ thm : criticalpoints ] can be relaxed in small dimensions . when @xmath223 , it is enough to require that @xmath413 is an interior point of @xmath13 . when @xmath141 , for generic @xmath9 , it is enough to require that each quadrant @xmath417 fulfills that @xmath418 has nonempty intersection with @xmath6 .
let @xmath419 be a basis for the left kernel @xmath420 , and let @xmath421 be a basis for the left kernel @xmath422 .
multiplying @xmath6 from the left by @xmath423 it takes the desired form .
we need only to show that @xmath424 .
notice that @xmath425 , since @xmath6 is assumed to be of full dimension .
assume that there is a linear combination @xmath426 then , since @xmath41 is a gale dual of @xmath6 , @xmath427 and hence @xmath428 .
this implies that @xmath429 , and hence @xmath430 .
thus , @xmath431 for all @xmath432 by linear independence of the vectors @xmath433 .
then , linear independence of the vectors @xmath434 imply that @xmath435 .
we find that @xmath436 hence , for each @xmath208 , it holds that @xmath437 in particular , @xmath9 is colopsided at @xmath109 unless , after a rotation , @xmath438 for all @xmath62 . in the latter case
, we refer to theorems [ thm : configurationspaceinteriorpoint ] and [ thm : configurationspacevertex ] . to see that the points @xmath439 for which @xmath9 is colopsided at @xmath109 are contained in distinct connected components of the complement of @xmath133 , consider a line segment @xmath162 in @xmath32 with endpoints in @xmath143 .
then , not all identities of can hold identically along @xmath162 .
since the argument of each monomial is linear in @xmath109 , this implies that for a pair such that the identity in does not hold identically along @xmath162 , there is an intermediate point @xmath440 for which the corresponding monomials are antipodal , and hence @xmath441 .
in this section we will consider a system @xmath442 of two bivariate polynomials .
we will write @xmath5 for the system .
the system is said to be generic if it has finitely many roots in @xmath443 , and it is said to be supported on a circuit @xmath6 if the supports of @xmath444 and @xmath445 are contained in , but not necessarily equal to , @xmath6 .
that is , we allow coefficients in @xmath446 rather than @xmath121 . by the bernstein
kushnirenko theorem , a generic system @xmath5 has at most @xmath140-many roots in @xmath443 . however ,
if @xmath9 is real , then fewnomial theory states that a generic system @xmath7 has at most three roots in @xmath447 .
we will solve the complexified fewnomial problem , i.e. , for @xmath7 with complex coefficients we will bound the number of roots in each sector @xmath15 .
our intention is to offer a new approach to fewnomial theory .
we will restrict to the case of simplex circuits , for the following two reasons .
firstly , it allows for a simpler exposition .
secondly , for vertex circuits our method recovers the known ( sharp ) bound , while for simplex circuits we obtain a sharpening of the fewnomial bound .
[ thm : fewnomialsimplexcircuits ] let @xmath5 be a generic system of two bivariate polynomials supported on a planar simplex circuit @xmath448 .
then , each sector @xmath15 contains at most two solutions of @xmath5 . a generic system @xmath7 is , by taking appropriate linear combinations , equivalent to a system of two trinomials whose support intersect in a dupleton .
that is , we can assume that @xmath7 is in the form @xmath449 with coefficients in @xmath121 .
we will use the notation @xmath450 where @xmath59 denotes the support of @xmath451 ( notice that this differs from the notation used in previous sections ) . notice that we can identify a system @xmath7 in the form with its corresponding vector in @xmath452 .
when reducing @xmath7 to the form by taking linear combinations , there is a choice of which monomials to eliminate in @xmath444 and @xmath445 respectively . in order for the arguments of the roots of @xmath453 to depend continously on the coefficients , we need to be careful with which choice to make .
[ lem : argumentsvarycontinuously ] let @xmath162 denote the line through @xmath454 and @xmath350 , and let @xmath455 be a compact path in @xmath452 . if @xmath162 intersect the interior of @xmath13 , then the arguments of the solutions to @xmath5 vary continuously along @xmath455 .
we first claim that our assumptions imply that the normal fans of @xmath411 and @xmath412 has no coinciding one dimensional cones .
indeed , these fans has a coinciding one dimensional cone if and only if the newton polytopes @xmath411 and @xmath412 has facets @xmath458 and @xmath459 with a common outward normal vector @xmath460 .
as @xmath6 is a circuit , it holds that @xmath461 \subset \ell$ ] . since the normal vector @xmath460 is common for @xmath411 and @xmath412 , we find that @xmath458 ( and @xmath459 ) is a facet of @xmath13 .
but then @xmath162 contains a facet of @xmath13 , and hence it can not intersect the interior of @xmath13 , a contradiction . consider a point @xmath224 . since the normal fans @xmath411 and @xmath412 has no coinciding one dimensional cones , the intersection of the amoebas @xmath462 and @xmath463 is bounded ( this follows , e.g. , from the fact the amoeba has finite hausdorff distance from the archimedean tropical variety , see @xcite ) .
thus , the amoeba @xmath33 is bounded , say that @xmath464 where @xmath465 denotes the disk of radii @xmath466 centered around the origin . by continuity of roots ,
@xmath467 for all @xmath468 in some neighborhood @xmath469 of @xmath9 .
the compactness of @xmath455 implies our result . in order for the assumptions of lemma [ lem : argumentsvarycontinuously ] to be fulfilled , for a simplex circuit @xmath6 , we need that @xmath293 and @xmath294 are vertices of @xmath470 , see figure [ fig : systeminteriorpolytopes ] . if @xmath451 is nonreal , then the fiber in @xmath471 over a point @xmath472 is a singleton . hence ,
if the number of roots of @xmath5 in @xmath15 is greater than one , then both @xmath444 and @xmath445 are real at @xmath109 .
the implication of proposition [ pro : nonrealsystem ] is that the complexified fewnomial problem reduces to the real fewnomial problem .
however , our approach is dependent on allowing coefficients to be nonreal .
in fact , we will consider a partially complexified problem , allowing @xmath473 but requiring @xmath474 .
we define the colopsided coamoeba of the system @xmath7 by @xmath475 where the last equality follows from ( * ? ? ?
that is , @xmath9 is said to be colopsided at @xmath109 if either @xmath444 or @xmath445 is colopsided at @xmath109 .
we will say that @xmath9 is real at @xmath109 if both @xmath444 and @xmath445 are real at @xmath109 .
[ lem : binomialcoamoebaincomplement ] assume that @xmath9 is nonreal .
let @xmath476 be a binomial constructed by choosing two monomials from , possibly alternating signs .
if @xmath477 and @xmath478 are of opposite signs , then @xmath479 .
if @xmath477 and @xmath478 are of equal signs , then @xmath479 _ except _ for @xmath480 . if , for @xmath481 , two components of @xmath482 is contained in a real subvector space @xmath147 , then either @xmath444 is colopsided at @xmath109 or @xmath483 .
however , the latter implies that two components of @xmath484 are contained in @xmath162 . repeating the argument yields that either @xmath9 is real , or it is colopsided at @xmath109 .
thus , the only binomials we need to consider is @xmath485 . for each @xmath486 the vectors @xmath482 and @xmath484 differ in sign in their first component , and hence at least one is colopsided at @xmath109 , unless @xmath9 is real . for each @xmath487
, the vectors @xmath482 and @xmath484 differ in signs in the the last component only if @xmath477 and @xmath478 differ in signs .
if this is the case , then at least one is colopsided at @xmath109 unless @xmath9 is real .
[ lem : shellpointsofsystemiscontainedinclosure ] let @xmath488 for truncated binomials @xmath489 and @xmath490 of @xmath444 and @xmath445 respectively .
if the newton polytopes ( i.e. , line segments ) of @xmath489 and @xmath490 are nonparallel , then @xmath384 .
if @xmath444 and @xmath445 are both real at @xmath109 , then @xmath491 . if @xmath444 is nonreal at @xmath109 , then for a sufficiently small neighborhood @xmath492 , it holds that @xmath493 where @xmath460 is a normal vector of @xmath494 .
since connected components of the complement of @xmath495 are convex , either @xmath495 intersect @xmath496 in @xmath383 , or the boundary of @xmath495 is contained in the line @xmath497 .
as the boundary of @xmath495 contains @xmath498 , it holds in the latter case that @xmath499 , which in turn implies that @xmath460 is a normal vector of @xmath500 , contradicting our assumptions .
we conclude that @xmath501 .
since this holds for any sufficiently small neighborhood @xmath383 , the result follows .
[ ex : lopsidedsystem ] consider the system @xmath502 we have that @xmath503 .
hence @xmath504 divides @xmath11 into three cells .
the lopsided coamoeba @xmath112 , and the hyperplane arrangement @xmath504 , can be seen in figure [ fig : systeminteriorexplained2 ] . in the first two picture ,
the generic respectively real situation when @xmath477 and @xmath478 differs in signs . in last two pictures , the generic
respectively real situation when @xmath477 and @xmath478 have equal signs . in the generic case , the lopsided coamoeba @xmath112 consist of three polygons .
when deforming from the generic to the real case , we observe the following behavior . some polygons of @xmath112
deform into single points - by necessity points contained in the lattice @xmath505 .
some pairs of polytopes of @xmath112 deforms to nonconvex polygons , typically with a single intersection point .
our proof of theorem [ thm : fewnomialsimplexcircuits ] is based on the observation that , when deforming from a generic to a real system , at most two polytopes of @xmath506 deforms a nonconvex polygon intersecting @xmath504 .
let us consider the auxiliary binomials @xmath507 the vectors @xmath508 and @xmath509 span the simplex @xmath13 , hence the hyperplane arrangement @xmath510 divides @xmath11 into @xmath140-many parallelograms with the points @xmath511 as their centers of mass .
if @xmath9 is nonreal , then lemma [ lem : binomialcoamoebaincomplement ] shows that @xmath512 , and lemma [ lem : shellpointsofsystemiscontainedinclosure ] shows that @xmath513 . by lemma [ lem : argumentsvarycontinuously ]
we find that @xmath112 has at most @xmath140-many connected components .
hence , @xmath112 has at exactly one connected component in each of the cells of @xmath504 , and the number of roots of @xmath5 projected by the argument map into each such component is exactly one .
consider now the real case when @xmath477 and @xmath478 differs in signs .
then , at least one of @xmath444 and @xmath445 are colopsided at the intersection points @xmath514 . thus , if @xmath15 contains a root of @xmath5 , then a sufficiently small neighborhood @xmath383 intersect at most two of the cells of the hyperplane arrangement @xmath504 . hence , using lemma [ lem : argumentsvarycontinuously ] and wiggling the arguments of coefficients of @xmath9 by @xmath300 , @xmath383 intersect at most two of the polygons of @xmath515 .
hence , there can be at most two roots contained in @xmath15 .
consider now the case when @xmath9 real with @xmath477 and @xmath478 of equal signs . in this case
, a point @xmath488 can be contained in @xmath112 .
see the left picture of figure [ fig : systeminteriorproof ] , where the hyperplane arrangement @xmath504 is given in black , and the shells @xmath516 and @xmath517 are given in red and blue respectively , with indicated orientation .
wiggling the arguments of coefficient @xmath518 and/or @xmath519 by @xmath300 , we claim the we obtain a situation as in the right picture of figure [ fig : systeminteriorproof ] .
that is , at most two polygons of @xmath515 will intersect a small neighborhood @xmath383 of @xmath109 .
let us prove this last claim .
let @xmath9 be generic , with @xmath477 and @xmath478 real and of equal signs .
the hyperplanes @xmath520 and @xmath498 ( locally ) divides the plane into four regions .
we can assume that @xmath521 .
then , @xmath520 consist of all @xmath109 such that @xmath522 , and @xmath520 consist of all @xmath109 such that @xmath523 .
thus , locally , the cells of @xmath504 can be indexed by the signs of the imaginary parts of @xmath524 and @xmath525 .
assume that @xmath526 .
then neither @xmath444 nor @xmath445 is colopsided at @xmath527 .
observe that @xmath528 , since @xmath477 and @xmath478 has equal sign .
we find that @xmath529 where the first and the last equality holds since neither @xmath444 nor @xmath445 is colopsided at @xmath527 .
this implies that polygons of @xmath112 intersecting a small neighbourhood of @xmath109 are necessarily contained in the cells of @xmath504 which corresponds to that the imaginary parts of @xmath530 and @xmath531 have equal signs .
as there are two such cells , we find that there are at most two polygons of @xmath112 intersecting a small neighbourhood of @xmath109 .
close to a point of @xmath514 when @xmath477 and @xmath478 have equal in signs and @xmath518 and @xmath519 are real . to the right : the same picture after wiggling the argument of @xmath518 or @xmath519.,title="fig:",height=113 ] close to a point of @xmath514 when @xmath477 and @xmath478 have equal in signs and @xmath518 and @xmath519 are real . to the right : the same picture after wiggling the argument of @xmath518 or @xmath519.,title="fig:",height=113 ] bihan , f. , rojas , j. m. , and stella , c. e. , faster real feasibility via circuit discriminants , _
issac 2009proceedings of the 2009 international symposium on symbolic and algebraic computation _
, 3946 , _ acm , new york _ , 2009 . | we study coamoebas of polynomials supported on circuits .
our results include an explicit description of the space of coamoebas , a relation between connected components of the coamoeba complement and critical points of the polynomial , an upper bound on the area of a planar coamoeba , and a recovered bound on the number of positive solutions of a fewnomial system . |
highly correlated electron systems are characterized by a strong repulsion between electrons on the same lattice site , effectively restricting the dynamics to the fock subspace of states without double occupancy of sites .
the prototype model for such systems is the anderson impurity model , which consists of an electron in a localized level @xmath0 ( called d - level in the following ) with on - site repulsion @xmath1 , hybridizing via a transition matrix element @xmath2 with one or several degenerate conduction electron bands or channels @xcite .
depending on the number of channels @xmath3 , the model exhibits the single- or the multi - channel kondo effect , where at temperatures @xmath4 below the kondo temperature @xmath5 the local electron spin is screened ( @xmath6 ) or overscreened ( @xmath7 ) by the conduction electrons , leading to fermi liquid ( fl ) or to non - fermi liquid ( nfl ) behavior @xcite with characteristic low - temperature singularities , respectively . as perhaps the simplest model to investigate the salient features of correlations induced by short - range repulsion ,
the anderson model plays a central role for the description of strongly correlated electron systems : in the limit of large spatial dimensions @xcite strongly correlated lattice systems reduce in general to a single anderson impurity hybridizing with a continuum of conduction electron states whose properties are determined from a self - consistency condition imposed by the translational invariance of the system @xcite .
quantum impurity models have received further interest due to their relevance for mesoscopic systems like single electron transistors or defects in quantum point contacts .
the above - mentioned systems call for the development of accurate and flexible theoretical methods , applicable to situations where exact solution methods are not available
. we here present a general , well - controlled auxiliary boson technique which correctly describes the fl as well as the nfl case of the generalized su(n)@xmath8su(m ) anderson impurity model . as a standard diagram technique it has the potential
to be generalized for correlated lattice problems as well as for non - equilibrium situations in mesoscopic systems .
the auxiliary particle method @xcite is a powerful tool to implement the effective restriction to the sector of fock space with no double occupancy imposed by a large on - site repulsion @xmath1 .
the creation operator for an electron with spin @xmath9 in the @xmath10-level is written in terms of fermionic operators @xmath11 and bosonic operators @xmath12 as @xmath13 .
this representation is exact , if the constraint that the total number operator of auxiliary fermions @xmath11 and bosons @xmath12 is equal to unity is obeyed . @xmath14 and @xmath15 may be envisaged as creating the three allowed states of the impurity : singly occupied with spin @xmath9 or empty . in view of the possibility of both fl and nfl behavior in quantum impurity systems mentioned in the introduction it is useful
to introduce @xmath3 degenerate channels for the conduction electron operators @xmath16 , labeled @xmath17 , in such a way that in the limit of impurity occupation number @xmath18 ( kondo limit ) the @xmath3-channel kondo model is recovered , i.e. the model obeys an su(m ) channel symmetry .
the slave bosons then form an su(m ) multiplet @xmath19 which transforms according to the conjugate representation of su(m ) , so that @xmath20 is a conserved quantum number .
generalizing , in addition , to arbitrary spin degeneracy @xmath21 , @xmath22 , one obtains the su(n)@xmath8su(m ) anderson impurity model in pseudoparticle representation @xmath23 where the local operator constraint @xmath24 must be fulfilled at all times .
the system described by the auxiliary particle hamiltonian ( [ sbhamilton ] ) is invariant under simultaneous , local @xmath25 gauge transformations , @xmath26 , @xmath27 , with @xmath28 an arbitrary , time dependent phase . while the gauge symmetry guarantees the conservation of the local , integer charge @xmath29 , it does not single out any particular @xmath29 , like @xmath30 . in order to effect the projection onto the @xmath30 sector of fock space
, one may use the following procedure @xcite : consider first the grand - canonical ensemble with respect to @xmath29 and the associated chemical potential @xmath31 .
the expectation value in the @xmath30 subspace of any physical operator @xmath32 acting on the impurity states is then obtained as @xmath33 _ g } { \frac{\partial } { \partial \zeta } \mbox{tr } \bigl [ e^{-\beta ( h+\lambda q ) } \bigr ] _ g } = \lim _ { \lambda \rightarrow\infty}\frac{\langle \hat a\rangle _ g } { \langle q \rangle _ g}\ , \label{projection}\ ] ] where the index @xmath34 denotes the grand canonical ensemble and @xmath35 is the fugacity @xmath36 . in the second equality of eq .
( [ projection ] ) we have used the fact that any physical operator @xmath32 acting on the impurity is composed of the impurity electron operators @xmath37 , @xmath38 , and thus annihilates the states in the @xmath39 sector , @xmath40 .
it is obvious that the grand - canonical expectation value involved in eq .
( [ projection ] ) may be factorized into auxiliary particle propagators using wick s theorem , thus allowing for the application of standard diagrammatic techniques .
the local u(1 ) gauge symmetry , which must not be broken according to elitzur s theorem @xcite , precludes the existence of an auxiliary bose condensate .
the latter would imply a spurious phase transition at finite @xmath4 in quantum impurity models and should , therefore , be avoided in any approximation .
it is important to note that , in general , @xmath41 plays the role of a time dependent gauge field . in eq .
( [ projection ] ) a time independent gauge for @xmath41 has been chosen . in this way
, the projection is only performed at one instant of time , explicitly exploiting the conservation of the local charge @xmath29 .
thus , choosing the time independent gauge means that in the subsequent development of the theory , the @xmath29 conservation must be implemented exactly .
this is achieved in a systematic way by means of conserving approximations @xcite , i.e. by deriving all self - energies and vertices by functional derivation from one common luttinger - ward functional @xmath42 of the fully renormalized green s functions , @xmath43 this amounts to calculating all quantities of the theory in a self - consistent way , but has the great advantage that gauge field fluctuations need not be considered .
the projection onto the physical subspace , eq .
( [ projection ] ) , implies that the pseudofermion and slave boson green s functions @xmath44 , @xmath45 are definied as the usual time - ordered , grand canonical expectation values of a pair of creation and annihilation operators , however evaluated in the limit @xmath46 .
it follows that the traces involved in @xmath44 , @xmath45 are taken purely over the the @xmath39 sector of fock space , and thus the backward - in - time contribution to the auxiliary particle propagators vanishes .
consequently , the auxiliary particle propagators are formally identical to the core hole propagators appearing in the well - known x - ray problem @xcite , and the long - time behavior of @xmath44 ( @xmath45 ) is determined by the orthogonality catastrophe @xcite of the overlap of the fermi sea without impurity ( @xmath39 ) and the fully interacting conduction electron sea in the presence of a pseudofermion ( slave boson ) ( @xmath30 ) .
it may be shown that the auxiliary particle spectral functions have threshold behavior with vanishing spectral weight at @xmath47 for energies
@xmath48 below a threshold @xmath49 , and power law behavior above @xmath49 , @xmath50 . for the single - channel anderson model , which is known to have a fl ground state , the threshold exponents may be deduced from an analysis in terms of scattering phase shifts , using the friedel sum rule , since in the spin screened fl state the impurity acts as a pure potential scatterer @xcite , @xmath51 these results have been confirmed by numerical renormalization group ( nrg ) calculations @xcite and by use of the bethe ansatz solution in connection with boundary conformal field theory ( cft ) @xcite . on the contrary , in the nfl case of the multi - channel kondo model
the threshold exponents have been deduced by a cft solution @xcite as @xmath52 since the dependence of @xmath53 , @xmath54 on the impurity occupation number @xmath55 shown above originates from pure potential scattering , it is characteristic for the fl case .
the auxiliary particle threshold exponents are , therefore , indicators for fl or nfl behavior in quantum impurity models of the anderson type .
equivalently to the method shown in eq.([projection ] ) the projection onto the @xmath30 subspace can be represented as an integration over the parameter @xmath56 , @xmath57 \int \frac{d\lambda}{2\pi t } \ , { \rm e } ^{-\beta [ i\lambda ( q-1)+s\ { f,\bar f;b,\bar b;c,\bar c \}]}\ , \langle \hat a\rangle _ g \ , , \label{sproj}\end{aligned}\ ] ] where @xmath58 is the action of the anderson model corresponding to eq.([sbhamilton ] ) .
this representation opens up the possibility of constructing an approximate projection scheme @xcite : performing the @xmath41-integration in saddlepoint approximation ( @xmath41spa ) , i.e.evaluating eq.([sproj ] ) at the stationary point of the grand canonical action @xmath59 , is equivalent to determining @xmath60 as a ( real ) thermodynamical chemical potential @xmath61 , fixing @xmath62 .
the @xmath41spa respects several of the exact properties discussed in sections 2.2 and 2.3 : ( 1 ) although the threshold property of the auxiliary spectral function no longer holds , their infrared powerlaw behavior is preserved . within nca ( section 3.1 )
it can be shown explicitly that the @xmath41spa does not change the exponents @xmath53 , @xmath54 , and the same may be expected to hold for higher order self - consistent approximations ( section 3.3 ) .
( 2 ) since the boson field is treated as a purely fluctuating field , bose condensation does not occur in the @xmath41spa .
technically , this is achieved in that the boson spectral function @xmath63 , not restricted by the threshold property in this approximation , acquires weight at negative freqencies , which is negative because of the stability requirement @xmath64 for bosonic functions .
the latter is seen in fig.[saddlep ] .
obviously , the saddlepoint projection is immediately generalizable to lattice problems , since the _ average _ local charge @xmath65 and thus @xmath61 are space independent .
the conserving formulation discussed in section 2.2 precludes mean field approximations which break the @xmath25 gauge symmetry , like slave boson mean field theory .
although the latter can in some cases successfully describe the low @xmath4 behavior of models with a fl ground state , it leads to a spurious phase transition at finite @xmath4 and , in particular , fails to describe nfl systems .
rather , the approximation should be generated from a luttinger - ward functional @xmath42 . using the hybridization @xmath2 as a small parameter
, one may generate successively more complex approximations .
the lowest order conserving approximation generated in this way is the non - crossing approximation ( nca ) @xcite , defined by the first diagram in fig .
[ ctma ] , labeled `` nca '' . the nca is successful in describing anderson type models at temperatures above and around the kondo temperature @xmath5 , and even reproduces the threshold exponents eq .
( [ alpha_fb2 ] ) for the nfl case of the anderson impurity model . however , it fails to describe the fl regime at low temperatures .
this may be traced back to the failure to capture the spin - screened kondo singlet ground state of the model , since coherent spin flip scattering is not included in nca , as seen below . in order to eliminate the shortcomings of the nca mentioned above
, we may use as a guiding principle to look for contributions to the vertex functions which renormalize the auxiliary particle threshold exponents to their correct values , since this is a necessary condition for the description of fl and nfl behavior , as discussed in section 2.3 . as shown by power counting arguments @xcite , there are no corrections to the nca exponents in any finite order of perturbation theory .
thus , any renormalization of the nca exponents must be due to singularities arising from an infinite resummation of terms .
in general , the existence of collective excitations leads to a singular behavior of the corresponding two particle vertex function . in view of the tendency of kondo systems to form a collective spin singlet state
, we expect a singularity in the spin singlet channel of the pseudofermion
conduction electron vertex function .
it is then natural to perform a partial resummation of those contributions which , at each order in the hybridization @xmath2 , contain the maximum number of spin flip processes
. this amounts to calculating the conduction electron
pseudofermion vertex function in the `` ladder '' or t - matrix approximation , @xmath66 , where the irreducible vertex is given by @xmath67 .
the bethe salpeter equation for @xmath66 reads ( fig .
[ tmat ] ) , @xmath68where @xmath9 , @xmath69 , @xmath70 , @xmath71 represent spin indices and @xmath20 a channel index .
a similar integral equation holds for the charge fluctuation t - matrix @xmath72 ; it is obtained from @xmath66 by interchanging @xmath73 and @xmath74 .
inserting nca green s functions for the intermediate state propagators of eq .
( [ cftmateq ] ) , we find at low temperatures and in the kondo regime @xmath75 a pole in the singlet channel of @xmath66 at an energy of the c - f pair which scales with @xmath5 , indicating the onset of spin singlet formation . in order to find a gauge invariant approximation incorporation multiple spin flip processes as discussed in the previous section , we have to find the luttinger - ward functional , which by second functional derivation w.r.t .
@xmath76 and @xmath44 or @xmath45 generates the @xmath4-matrices @xmath66 and @xmath72 , respectively .
this generating functional @xmath42 is shown diagrammatically in fig.[ctma ] .
the auxiliary particle self - energies are obtained in the conserving scheme as the functional derivatives of @xmath42 with respect to @xmath44 or @xmath45 , respectively ( eq . ( [ fderiv ] ) ) , and are , in turn , nonlinear functionals of the full , renormalized auxiliary particle propagators .
this defines a set of self - consistency equations , which we term conserving t - matrix approximation ( ctma ) .
we have solved the ctma equations numerically for a wide range of impurity occupation numbers @xmath55 from the kondo to the empty impurity regime both for the single - channel and for the two - channel anderson model down to temperatures of the order of at least @xmath77 .
the solution of the ctma equations forces the t - matrices to have vanishing spectral weight at negative com frequencies @xmath78 .
indeed , the numerical evaluation shows that the poles of @xmath66 and @xmath72 are shifted to @xmath79 by self - consistency , where they merge with the continuous spectral weight present for @xmath80 , and thus renormalize the threshold exponents of the auxiliary spectral functions . for @xmath81 , @xmath6 the threshold exponents
@xmath53 , @xmath54 extracted from the numerical solutions are shown in fig .
[ exponents ] . in the kondo limit of the multi channel case ( @xmath82 , @xmath83 ) the ctma solutions are found not to alter the nca values and reproduce the the correct threshold exponents , @xmath84 , @xmath85 .
the good agreement of the ctma exponents with their exact values over the complete range of @xmath55 for the single channel model and in the kondo regime of the multi
channel model may be taken as evidence that the t - matrix approximation correctly describes both the fl and the non fl regimes of the su(n)@xmath8su(m ) anderson model ( n=2 , m=1,2,4 ) .
the static spin susceptibility @xmath86 of the single- and of the two - channel anderson model in the kondo regime calculated within ctma as the derivative of the magnetization @xmath87 with respect to a magnetic field @xmath88 is shown in fig .
good quantitative agreement with exact solutions is found for @xmath81 , @xmath6 ( fl ) . for @xmath81 , @xmath89 ( nfl ) ctma correctly reproduces the exact @xcite logarithmic temperature dependence below the kondo scale @xmath5 .
in contrast , the nca solution recovers the logarithmic behavior only far below @xmath5 .
we have reviewed a novel technique to describe correlated quantum impurity systems with strong onsite repulsion , which is based on a conserving formulation of the auxiliary boson method .
the conserving scheme allows to implement the conservation of the local charge @xmath29 without taking into account time dependent fluctuations of the gauge field @xmath41 . by including the leading infrared singular contributions ( spin flip and charge fluctuation processes ) , physical quantities , like the magnetic susceptibility
, are correctly described both in the fermi and in the non fermi liquid regime , over the complete temperature range , including the crossover to the correlated many
body state at the lowest temperatures . as a standard diagram technique
this method has the potential to be applicable to problems of correlated systems on a lattice as well as to mesoscopic systems out of equilibrium via the keldysh technique .
we wish to thank s. bcker , t.a .
costi , k. haule , s. kirchner , a. rosch , a. ruckenstein and th .
schauerte for stimulating discussions .
this work is supported by dfg through sfb 195 and by the hochleistungsrechenzentrum stuttgart . | we review a recently developed method , based on a pseudoparticle representation of correlated electrons , to describe both fermi liquid and non - fermi liquid behavior in quantum impurity systems .
the role of the projection onto the physical hilbert space and the impossibility of slave boson condensation are discussed . by summing the leading coherent spin and
charge fluctuation processes in a fully self - consistent and gauge invariant way one obtains the correct infrared behavior of the pseudoparticles .
the temperature dependence of the spin susceptibility for the single channel and two - channel anderson models is calculated and found to agree well with exact results . |
in the present paper we consider an extension of the standard model ( sm ) based on the randall - sundrum model with two branes stabilized by a bulk scalar field @xcite , which is necessary for the model to be phenomenologically acceptable .
a characteristic feature of this extension is the presence of a massive scalar radion field together with its kaluza - klein ( kk ) tower .
these fields have the same quantum numbers as the neutral higgs field .
thus , the radion field and its excitations can mix with the higgs field , if they are coupled .
originally , a higgs - radion coupling in the unstabilized randall - sundrum model arising due to a higgs - curvature term on the brane was put forward in @xcite .
then , such a coupling and the resulting higgs - radion mixing in the case of the stabilized model were discussed in paper @xcite without taking into account the kk tower of higher scalar excitations .
the phenomenology of the higgs - radion mixing originating from the higgs - curvature term was also considered in view of the discovery of the higgs - like boson at the lhc @xcite ; various assumptions about the masses and the mixings of the scalar states have been analyzed in papers @xcite@xcite .
in particular , it was shown that the light radion - dominated state with mass below or above the observed 125-gev boson is still not completely excluded by all the electroweak precision constraints and the lhc data . here
, we discuss a different mechanism of higgs - radion mixing immanent in stabilized brane - world models , where a higgs - radion coupling naturally arises due to spontaneous symmetry breaking on the brane involving the stabilizing scalar field .
this approach takes into account the influence of the kk tower of higher scalar excitations on the parameters of the higgs - radion mixing , which turns out to be of importance . in principle , the most general mechanism of higgs - radion mixing can include both possible higgs - radion couplings .
however , after the spontaneous symmetry breaking on the brane the higgs - curvature term also gives rise to a brane - localized curvature term that affects the mass spectrum and the couplings to matter of the graviton kk modes @xcite ; this effect should be taken into account when examining the four - dimensional effective theory . for this reason , here we restrict ourselves to the new mechanism of higgs - radion mixing , which has the advantage that it modifies only the scalar sector of the model and leaves intact the masses and the coupling constants of the graviton kk excitations .
this also allows one to isolate the effects due only to the new mechanism of the mixing .
a stabilized brane - world model in five - dimensional space - time @xmath0 with coordinates @xmath1 , @xmath2 , with the coordinate @xmath3 parameterizing the fifth dimension , is defined by the action @xmath4 where @xmath5 and @xmath6 are given by @xmath7 here the signature of the metric @xmath8 is chosen to be @xmath9 , @xmath10 is the fundamental five - dimensional energy scale , @xmath11 is a bulk scalar field potential , @xmath12 are brane scalar field potentials , @xmath13 , and @xmath14 denotes the metric induced on the branes .
the space of extra dimension is the orbifold @xmath15 , which is realized as the circle of circumference @xmath16 with the points @xmath17 and @xmath18 identified .
correspondingly , the metric @xmath8 and the scalar field @xmath19 satisfy the orbifold symmetry conditions @xmath20 the branes are located at the fixed points of the orbifold , @xmath21 and @xmath22 , and it is assumed that the sm fields with lagrangian @xmath23 live on the brane at @xmath22
. for this form of the action , the vacuum solution for gravity and the stabilizing scalar field and the vacuum solution for the sm fields are independent .
if we consider this brane - world model to be an indivisible theory , it is reasonable to believe that there should be a common interconnected vacuum solution for all these fields . to this end , let us modify action ( [ actionsm ] ) so that it contains a quartic interaction of the stabilizing scalar field and of the higgs field , which would connect the stabilization of the extra dimension size and the sm spontaneous symmetry breaking : @xmath24 where the lagrangian @xmath25 is the sm lagrangian without the higgs potential , which is replaced by the interaction lagrangian @xmath26 @xmath27 being a positive dimensionless parameter .
a similar interaction lagrangian quadratic in the stabilizing scalar field was discussed in @xcite .
the background solutions for the metric and the scalar field , which preserve the poincar invariance in any four - dimensional subspace @xmath28 , look like @xmath29 with @xmath30 denoting the flat minkowski metric , whereas the background ( vacuum ) solution for the higgs field is standard , @xmath31 with all the other sm fields being equal to zero .
if one substitutes this ansatz into the equations corresponding to action ( [ actiondw ] ) , one gets a relation between the vacuum value of the higgs field and the value of the field @xmath19 on the brane at @xmath32 , @xmath33 this means that in such a scenario the higgs field vacuum expectation value , being proportional to the value of the stabilizing scalar field on the tev brane , arises dynamically as a result of the gravitational bulk stabilization .
the gravitational background solution follows from the well - known system of nonlinear differential equations for functions @xmath34 @xcite , @xmath35 where the first equation is the equation for the scalar field and the other two follow from the einstein equations . here
@xmath36 .
suppose we have a solution @xmath37 to this system for an appropriate choice of the parameters of the potentials such that the interbrane distance is stabilized and equal to @xmath38 .
this means that the vacuum energy of the scalar field has a minimum for this value of the interbrane distance .
now the linearized theory is obtained by representing the metric , the scalar and the higgs field in the unitary gauge as @xmath39 substituting this representation into action ( [ actiondw ] ) and keeping the terms of the second order in @xmath40 , @xmath41 and @xmath42 . the lagrangian of this action is the standard free lagrangian of the sm ( i.e. , the masses of all the sm fields , except the higgs field , are expressed in the same way as usual in terms of the vacuum value of the higgs field and the coupling constants ) together with the standard second variation lagrangian of the stabilized randall - sundrum model @xcite supplemented by an interaction term of fields @xmath41 and @xmath42 .
the part of the lagrangian relevant to the higgs - radion mixing is @xmath43\delta(y - l).\ ] ] the bulk scalar field @xmath41 can be expanded in kk modes . in the case under consideration
we can , in the standard way , find the equations for the mass spectrum and the wave functions of the kk excitations of the scalar fields just by replacing @xmath44 in the formulas of paper @xcite ( we note that in this paper the signature of the metric was @xmath45 , and for this reason some formulas of that paper my differ in sign from the corresponding formulas in the present paper ) .
namely , in this paper it was shown that it was most convenient to describe the scalar degrees of freedom of stabilized brane - world models by the field @xmath46 related to the field @xmath41 by the gauge condition @xmath47 the equations of motion and the boundary conditions on the branes for the field @xmath48 define wave functions @xmath49 , corresponding to mass eigenvalues @xmath50 .
expanding the field @xmath41 in these modes , substituting this expansion into the second variation lagrangian and integrating over the extra dimension coordinate @xmath17 , we get a four - dimensional lagrangian , in which there is an interaction between the modes and the higgs field coming from the term @xmath51 using gauge condition ( [ gauge_f ] ) , the mode decomposition of @xmath52 , @xmath53 and the boundary condition for the mode wave function @xmath54 at @xmath22 @xcite rewritten in terms of eq .
( [ sigma_f_tems ] ) , @xmath55 the field @xmath41 can be expressed through @xmath54 as follows : @xmath56 where we have taken into account that @xmath57 in order to have galilean coordinates on the brane at @xmath58 @xcite .
thus , the couplings of the modes to the higgs field can be written as @xmath59 where we have introduced dimensionless quantities @xmath60 which are supposed to be positive .
these couplings are proportional to the squared masses of the modes and can be rather large for large @xmath61 . below we will see that it is the dimensionless quantities @xmath62 that are of importance .
they are proportional to the values of the wave functions of the modes on the brane and it is convenient to express them in terms of the ratios of the latter and @xmath63 : @xmath64 these ratios must go to zero for large @xmath61 in order for the cumulative effect of the radion kk tower to be finite . in paper
@xcite the scalar wave functions @xmath49 have been found explicitly in a stabilized brane - world model , where the warp factor is approximately equal to that of the unstabilized randall - sundrum model , and one can check that the quantities @xmath62 are indeed positive and the ratios @xmath65 really fall off fairly quickly for large @xmath61 in this case .
it is rather difficult to find them explicitly in the stabilized brane - world models , where the warp factor is different from the exponential of a linear function , as it is in the randall - sundrum model .
a study of such stabilized brane - world models has been carried out in papers @xcite , and it was found that they may also solve the hierarchy problem of the gravitational interaction , giving rise to the masses of kk excitations in the @xmath66 energy range .
however , the corresponding equations can not be solved exactly and should be studied numerically for each set of the fundamental parameters of the model , which is a very complicated and laborious task . here
, we will not carry out such calculations for a specific stabilized brane - world model , but rather give a qualitative description of the phenomena due to the interaction of the higgs field with the radion and its kk tower , choosing the masses and the coupling constants in a consistent manner and taking into account the present - day theoretical and experimental restrictions on their values .
the part of the extended sm lagrangian containing the scalar fields linearly or quadratically looks like @xmath67 where @xmath68 is the trace of the sm energy - momentum tensor and @xmath69 is the conformal anomaly of massless vector fields explicitly given by @xmath70 with @xmath71 being the well - known qcd and qed @xmath71 functions .
the coupling of the scalar modes to the trace of the energy - momentum tensor comes from the standard interaction lagrangian for the metric fluctuations @xcite , which results in the expression for the parameters @xmath72 in terms of the scalar mode wave functions : @xmath73 in the case of the lowest scalar mode , the radion , this parameter is usually denoted as @xmath74 where @xmath75 is supposed to be of the order of the fundamental energy scale @xmath10 . below we will use @xmath75 as a natural scale for the interactions of the radion and for those induced by its kk tower . for this reason , it is also convenient to express @xmath72 in terms of @xmath75 and @xmath65 defined in ( [ ratios ] ) as @xmath76 in what follows , we consider the phenomenology of the higgs boson and the radion in the energy range much lower than the masses of the radion excitations . in this case
, we can pass to a low - energy approximation for this lagrangian by dropping the kinetic terms of the radion excitation fields and integrating them out , which gives the following effective lagrangian for the interactions of the higgs and radion fields with the sm fields @xmath77 where the new parameters are defined in terms of the old ones as follows : @xmath78 we have already mentioned that these series should be convergent .
the positive parameter @xmath79 can be of the order of @xmath80 , because the mass term @xmath81 should be positive , and , within the approach under consideration , the coupling constant of the higgs boson self - interaction @xmath82 can be larger than in the regular sm case .
possible restrictions on the parameters @xmath83 will be discussed below .
next we turn to the mass matrix of the fields @xmath42 and @xmath84 , which looks like @xmath85 this matrix can be diagonalized by a rotation matrix @xmath86 where the rotation angle @xmath87 is defined by the relation @xmath88 it is not difficult to see that it is sufficient to take this angle only in the interval @xmath89 , which is one full period of the function @xmath90 .
the observable masses of the mass eigenstates are given by the expressions @xmath91 which are written in the form that explicitly shows that the smaller mass becomes smaller and the larger one becomes larger due to the mixing .
the latter , in particular , means that the observable masses @xmath92 and @xmath93 can not be exactly equal in the presence of the mixing and the mixing angle @xmath87 is negative for @xmath94 in accordance with the observation in paper @xcite . the original unobservable parameters of the matrix @xmath95 can be expressed in terms of the observable parameters @xmath93 , @xmath92 and @xmath87 .
the correspondence between these sets of parameters is one - to - one , if one puts the mixing angle @xmath96 for @xmath97 .
then the following formula for the parameter @xmath63 can be easily obtained : @xmath98 this formula , together with the second formula in ( [ parameter_c ] ) , gives an expression for the parameter @xmath83 in terms of the physical masses and the mixing angle @xmath99 which also depends on the sum of the wave function ratios @xmath100 .
these ratios are , of course , model dependent , and , although they should fall off with @xmath61 in order for the sum to be convergent , one can not exclude that , in certain models , several first ratios can be of the order of unity .
thus , one can conservatively estimate this sum to be also of the order of unity , for the radion mass in the interval @xmath101 and the mass of its first kk excitation of the order of 1 tev . ] which gives the estimate for the parameter @xmath83 @xmath102 which will be used in the subsequent calculations .
the physical mass eigenstate fields @xmath103 are easily expressed in terms of matrix @xmath104 and the original fields as @xmath105 the field @xmath106 is called the higgs - dominated field , and the field @xmath107 is called the radion - dominated field because @xmath108 in the interval @xmath109 ( we recall that @xmath110 for @xmath94 ) . the interaction of these fields with the fields of the standard model is given by the following effective lagrangian : @xmath111 the effective four - dimensional interaction lagrangian ( [ h - r - lagrangian ] ) expressed in terms of the physical higgs - dominated @xmath106 and radion - dominated @xmath107 fields contains their interactions with the sm fields and involves only five parameters in addition to those of the sm : the masses of the higgs - dominated and radion - dominated fields @xmath112 and @xmath113 , the mixing angle @xmath87 , the ( inverse ) coupling constant of the radion to the trace of the energy - momentum tensor of the sm fields @xmath75 and the parameter @xmath83 that accommodates the contributions of the integrated - out heavy scalar modes .
we would like to note here that , according to formulas ( [ restriction ] ) and ( [ restriction1 ] ) , the parameter @xmath83 is proportional to its maximal value @xmath114 and therefore depends on the masses @xmath112 , @xmath113 and the mixing angle @xmath87 , which is a particular feature of the model .
it is interesting to examine what occurs to lagrangian ( [ h - r - lagrangian ] ) if the fundamental energy scale @xmath10 goes to infinity . in the model under consideration , it is impossible just to take the limit @xmath115 because there exists relation ( [ relation ] ) for the model parameters , which includes the vacuum value of the higgs field . in the stabilized brane - world model discussed in @xcite ,
the radion mass was found to be proportional to @xmath116 .
thus , in order to keep this mass fixed , we have to take the limit @xmath117 and @xmath118 so that @xmath119 .
one can see that in this case @xmath63 , which is proportional to @xmath120 , does not vanish , whereas @xmath75 defined in ( [ bees ] ) and ( [ lambda_r ] ) goes to infinity . as a result , in lagrangian ( [ h - r - lagrangian ] ) the terms with the energy - momentum tensor that are proportional to @xmath121 vanish , but all the other terms remain because the mixing angle must not be equal to zero .
the situation in this case is rather similar to the one in the sm extended by an extra singlet scalar @xcite .
however , there are differences due to extra interactions between the higgs field and the singlet scalar in these models , arising from the scalar field potentials that are absent in our model .
we would also like to note here that , if we formally put the parameters @xmath63 , @xmath83 and @xmath87 equal to zero , i.e. , consider the case of the zero mixing , lagrangian ( [ h - r - lagrangian ] ) becomes just the sm higgs lagrangian plus the usual lagrangian of the radion interaction with the sm fields via the trace of the sm energy - momentum tensor . in the case of a nonzero mixing , there are additional terms in this lagrangian that may lead to certain changes in the collider phenomenology of the higgs boson and the radion .
these issues will be discussed in the next section .
the effective lagrangian ( [ h - r - lagrangian ] ) describes the interactions of the higgs - dominated @xmath106 and the radion - dominated @xmath107 fields with the standard model fields and with each other . a few natural questions arise .
what model parameter regions are allowed by the present - day experimental data such as lep and tevatron searches , higgs - like 125-gev boson discovery and the higgs signal strength measurements at various signatures at the lhc , and the limits from searches for the second higgs - like boson at the lhc ? is an interpretation of the observed boson as a radion - dominated state is still possible for some model parameters ?
detailed answers to such questions require a corresponding detailed phenomenological analysis , which will be considered in a separate study . in this paper
we present constraints coming from the discovery of a 125-gev higgs - like boson at the lhc , measurements of the signal strengths at various production channels and decay modes , and searches for heavier higgs bosons .
the feynman rules needed for our study can be easily derived from lagrangian ( [ h - r - lagrangian ] ) and are given in the table in the appendix .
we have omitted the terms coming from off - shell fermions since the corresponding contributions to physical processes are exactly canceled out by additional diagrams with contact four - point vertices coming from the first - order expansions in gauge couplings of the sm energy - momentum tensor trace as was proved in @xcite .
the feynman rules have been implemented as a `` new model '' in a special version of the comphep code @xcite which also includes a routine for @xmath122 analysis of the higgs signal strength in a way similar to those used in papers @xcite .
the next - to - next - to - leading order ( nnlo ) corrections are taken into account in the comphep computations and in the corresponding analysis by multiplying the involved vertices by correction factors for each model parameter point such that the partial and the total sm higgs decay widths and the higgs production cross sections in gluon - gluon fusion ( ggf ) are exactly equal to those presented by the lhc higgs cross section working group @xcite . because of the similarity of the higgs boson and the radion production and decay amplitudes including loops @xcite , the same correction factors have been used for the higgs- and the radion - dominated states .
we examine two possible scenarios , where the observed 125-gev boson is either the higgs - dominated state or the radion - dominated state . in order to understand how much room is left for the radion - dominated state ,
we first demonstrate various decay and production properties of this state . in fig .
[ fig_1 ] , the decay branching ratios for the radion - dominated boson are shown as functions of its mass .
radion - dominated state branching ratios as functions of @xmath113 ( @xmath123 , @xmath124 ) .
the left plot corresponds to @xmath125 and @xmath126 ; the right plot corresponds to @xmath127 , and @xmath83 is equal to its maximum value @xmath114.,width=326 ] radion - dominated state branching ratios as functions of @xmath113 ( @xmath123 , @xmath124 ) .
the left plot corresponds to @xmath125 and @xmath126 ; the right plot corresponds to @xmath127 , and @xmath83 is equal to its maximum value @xmath114.,width=326 ] the left plot shows the well - known branching behavior @xcite of the radion without mixing with the higgs boson and interacting with the sm fields only via the trace of the energy - momentum tensor .
as is well known , the main decay mode of the light radion is the decay to two gluons due to the anomaly enhancement .
however , once the mixing with the higgs boson is included , the picture might be changed drastically because of the appearance of higgs - like couplings for the radion - dominated state proportional to @xmath128 and because of the influence of the higher scalar modes encoded in the parameter @xmath83 . in fig .
[ fig_2 ] , the behavior of the upper level of this parameter as a function of the radion mass and the mixing angle is illustrated .
we recall that for @xmath129 the radion - dominated state is heavier than the higgs - dominated state , @xmath130 , and vice versa ( @xmath110 for @xmath131 ) , and therefore , for @xmath132 and for @xmath133 , the areas shown as zero plates are not allowed . if one takes a rather large mixing angle , for example , @xmath127 , the @xmath134 decay mode is dominating as shown in fig .
[ fig_1 ] on the right plot .
because of a compensation in the gluon - gluon - radion vertex between the trace anomaly part proportional to @xmath135 and the mentioned higgs - like part proportional to @xmath136 the vertex can be very small for some particular regions of the parameter space .
this leads to an interesting feature shown in fig .
[ fig_1 ] that the decay branching ratio for the radion - dominated state to two gluons may be smaller than the branching ratio to two photons
. two - dimensional plot of the upper level @xmath114 of the parameter @xmath83 as a function of @xmath137.,width=326 ] because of the mentioned compensations between various parts of the interaction vertex of the radion - dominated state with gluons , the behavior of the total width and the production cross sections for @xmath138 and @xmath139 modes in gluon - gluon fusion also have some minima , as one can see from figs .
[ fig_3 ] and [ fig_4 ] . from the cross sections
one gets the corresponding signal strengths for the radion - dominated state divided by the sm higgs cross sections . in our analysis
we use the higgs signal strengths for all the channels , as given recently by the cms @xcite and atlas @xcite collaborations . as a result of the standard @xmath122 analysis
, one gets the regions in the parameter space ( the mass of the radion - dominated state and the mixing angle with the higgs boson ) which are still allowed .
the regions allowed either by considering only the @xmath138 mode , and the @xmath138 together with the @xmath139 modes are given in the left and the right plots of fig .
[ fig_5 ] , correspondingly , for a not - too - large mass range for the radion - dominated state .
for both cases , all the main production processes [ ggf , vector boson fusion ( vbf ) , and associated production with vector bosons ( vh ) and with the top quarks ( tth ) ] are taken into account .
the interference effects of the higgs- and radion - dominated states are also taken into account , which is especially important in the case of close resonances .
the contours in all figures correspond to 65% , 90% and 99% best - fit confidence level ( cl ) regions with @xmath140 less than 2.10 , 4.61 and 9.21 , respectively , which is the standard presentation of two - parameter fits and was used in earlier papers ( see , e.g. , @xcite ) .
thus , the dark shaded area corresponds to 65% cl of the fit , the medium shaded area corresponds to 90% cl of the fit , and the light shaded area corresponds to 99% cl of the fit .
three - dimensional plots of the total width as a function of @xmath137 for the lhc at @xmath141 and @xmath123 , @xmath124 , @xmath142 .
the left plot corresponds to the higgs - dominated state @xmath106 , and the right plot corresponds to the radion - dominated state @xmath107.,width=326 ] three - dimensional plots of the total width as a function of @xmath137 for the lhc at @xmath141 and @xmath123 , @xmath124 , @xmath142 .
the left plot corresponds to the higgs - dominated state @xmath106 , and the right plot corresponds to the radion - dominated state @xmath107.,width=326 ] three - dimensional plots of the partial cross section as a function of @xmath137 for the lhc at @xmath141 and @xmath123 , @xmath124 , @xmath142 .
the left plot corresponds to the @xmath143 channel , and the right plot corresponds to the @xmath144 channel.,width=326 ] three - dimensional plots of the partial cross section as a function of @xmath137 for the lhc at @xmath141 and @xmath123 , @xmath124 , @xmath142 .
the left plot corresponds to the @xmath143 channel , and the right plot corresponds to the @xmath144 channel.,width=326 ] exclusion contours for the partial @xmath122 fit in the @xmath137 plane for the lhc at @xmath141 and @xmath123 , @xmath124 , @xmath142 .
the dark , medium and light shaded areas correspond to 65% , 90% and 99% cl of the fit , respectively .
the left plot corresponds to the @xmath145 channel , and the right plot corresponds to the combined fit of the @xmath145 and @xmath146 channels.,width=326 ] exclusion contours for the partial @xmath122 fit in the @xmath137 plane for the lhc at @xmath141 and @xmath123 , @xmath124 , @xmath142 .
the dark , medium and light shaded areas correspond to 65% , 90% and 99% cl of the fit , respectively .
the left plot corresponds to the @xmath145 channel , and the right plot corresponds to the combined fit of the @xmath145 and @xmath146 channels.,width=326 ] in fig .
[ fig_5 ] , one can see that the @xmath138 mode alone allows the presence of the radion - dominated state for masses both below and above the 125 gev .
the @xmath139 mode gives some additional restrictions on the parameter space especially in the mass region closer to the z - boson pair threshold , where the cross section is increased .
when all the modes are taken into account , the allowed parameter space region is reduced further , which is shown in fig .
[ fig_6 ] . for larger values of the radion mass ,
the influence of this parameter on 125-gev signal strength is very small , resulting in the allowed area shown in fig .
[ fig_7 ] .
however , in the case of a large resonance mass region , one has to take into account the exclusion limits given very recently by the cms @xcite and atlas @xcite collaborations , coming from searches for heavy higgs bosons .
if the radion - dominated state has a large mass , it would contribute as a resonance in this region . from the comparison of the radion cross section computation and the experimental limits by cms and atlas by performing the corresponding @xmath122 fit ,
one gets a rather large allowed area in the @xmath137 parameter plane , as shown in fig .
[ fig_8 ] . when taken together ,
fits of the data from both the signal strengths and from the direct searches for heavy higgs resonances significantly restrict the allowed region for the heavy radion mass to a small area at very high masses , as demonstrated in fig .
[ fig_9 ] . obviously , if one considers larger values of the parameter @xmath75 , the cross section of the radion - dominated state gets smaller and the allowed region for such a state is increased .
this is demonstrated in fig .
[ fig_10 ] , where the parameter @xmath75 is chosen to be 5 gev with all the other parameters being the same as for the considered 3-tev case .
now let us consider briefly the case where the radion - dominated state has mass 125 gev , and , using the same analysis strategy , find the regions for the mass of the higgs - dominated state and the mixing angle allowed by two signal strengths .
for this case , we have carried out all the corresponding calculations and have drawn the plots similar to those in figs .
[ fig_2][fig_7 ] , which we will not present here for the sake of brevity .
we only dwell upon the result of the @xmath122 analysis that is presented in fig .
[ fig_11 ] .
as one can see , such a scenario is strongly disfavored in comparison with the previous case .
the mass of the higgs - dominated state might still be very close to the mass of the radion - dominated state in a wide range of mixing angle .
such a possibility exists in both scenarios .
in the present paper we have considered the higgs - radion mixing arising in stabilized brane - world models due to merging the mechanism of stabilization of the extra dimension size and the higgs mechanism of spontaneous symmetry breaking on the tev brane .
we have discussed phenomenological restrictions on model parameters coming from coupling measurements and searches for heavy bosons at the lhc .
this mixing is , of course , similar in many aspects to the one arising due to the higgs - curvature term on the brane .
however , an important difference is the presence of an extra coupling at low energies of the higgs - dominated field to the trace of the sm energy - momentum tensor originating from the coupling of this field to the heavy scalar states of the radion kk tower . in order to study the physical consequences of the higgs - radion mixing in stabilized brane - world models , we derived the effective lagrangian and gave a qualitative description of the phenomena , taking consistent values for the masses , the coupling constants , and the mixing angle , which are useful for comparing the results of our calculations with the experimental data at the lhc .
it turned out that , though the interaction of an individual higher excited scalar state with the higgs field may be weak , their cumulative effect on the higgs - radion mixing may be observable .
if they give noticeable contributions to the parameter @xmath83 , it leads to certain changes in the collider phenomenology of the higgs boson .
a similar contribution of the directly unobservable higher tensor kk modes to scattering processes was discussed in @xcite . in the framework of this model ,
we have studied two _ a priori _ possible scenarios : the scalar state discovered at 125 gev is either a higgs - dominated state or a radion - dominated state . in our analysis , we used measurements of the 125-gev higgs signal strengths by the atlas and cms collaborations and exclusion limits obtained in searches for heavy higgs - like states .
our results show that the interpretation of the 125-gev scalar state as a higgs - dominated state is the preferred one , although the radion component in this state can be rather large . depending on the value of the radion coupling constant @xmath75
, the allowed regions for the mass of the radion - dominated state have been found .
it turns out that the radion - dominated state can either have a mass close to 125 gev or a mass close to the tev range .
the allowed regions somewhat increase with the growth of the radion coupling constant @xmath75 .
we have also shown that the interpretation of the 125-gev scalar state as a radion - dominated state is not completely excluded by the two leading signal strength measurements , though in this case the restrictions on the allowed masses of the higgs - dominated state are very stringent
. the mass of the higgs - dominated state can be close to 125 gev , which is in accordance with our analysis of this state at 125 gev .
thus , in the considered model the presence of two nearly degenerate states close to 125 gev is a very probable scenario .
the work was supported by grant no .
14 - 12 - 00363 of the russian science foundation .
[ cols="<,<",options="header " , ] in table 1 , the constants and the functions of the momenta are explicitly given by the following expressions : @xmath147 , @xmath148 , @xmath149 , @xmath150 , @xmath151 , @xmath152 , @xmath153 , @xmath154 , @xmath155 ^ 2\ ; , & z \ge 1 \\ -\frac{1}{4 } \left [ \log \frac{1+\sqrt{1-z}}{1-\sqrt{1-z } } - i \pi \right ] ^2 \ ; , & z < 1 .
\end{array } \right .\ ] ] k. -m .
cheung , phys .
d * 63 * , 056007 ( 2001 ) [ hep - ph/0009232 ] ; v. khachatryan _ et al . _
[ cms collaboration ] , eur .
j. c * 75 * ( 2015 ) 5 , 212 [ arxiv:1412.8662 [ hep - ex ] ] .
g. aad _ et al .
_ [ atlas collaboration ] , arxiv:1507.04548 [ hep - ex ] . | we consider a quartic interaction of the higgs and goldberger - wise fields , which connects the mechanism of the extra dimension size stabilization with spontaneous symmetry breaking on our brane and gives rise to a coupling of the higgs field to the radion and its kk tower .
we estimate a possible influence of this coupling on the higgs - radion mixing and study restrictions on model parameters from the lhc data . *
higgs - radion mixing in stabilized brane world models
* + edward e. boos , viacheslav e. bunichev , maxim a. perfilov , mikhail n. smolyakov , + igor p. volobuev + skobeltsyn institute of nuclear physics , lomonosov moscow state university + 119991 moscow , russia + |
galaxies with strong balmer absorption lines in their spectra ( h@xmath1-strong , or h@xmath1s galaxies ) represent a short - lived but potentially important phase in galaxy evolution .
such lines indicate a stellar population dominated by a - stars which are normally either absent or overwhelmed by the much brighter ob stars . there is likely a variety of physical mechanisms that can lead to such a stellar population ( e.g. dressler et al .
1982 ; couch et al . 1987 ;
poggianti et al . 1999 ; balogh et al . 1999 ) , but most invoke a major transformation from one galaxy type to another .
although such galaxies are very rare in the local universe , their short lifetime means they could potentially represent an important phase in the evolution of most normal galaxies . while high resolution imaging has demonstrated that most bright , nearby h@xmath1s galaxies are spheroidal , often with signs of interaction ( yang et al .
2004 ; couch et al . 2004 ; balogh et al .
2004 ) , the dynamics have been extremely difficult to observe and understand .
norton et al .
( 2001 ) have obtained longslit spectra of galaxies in the zabludoff et al .
( 1996 ) sample , and find evidence that most of the galaxies are in the process of transforming from rotationally - supported , gas - rich galaxies to pressure - supported , gas - poor galaxies . however , longslit spectroscopy mixes spatial and spectral resolution , and a better understanding of the dynamics can be obtained from integral field spectroscopy .
this allows us to identify the physical locations of the gas , old ( k ) stars , young ob - stars and ( the short - lived ) massive a - stars , and to decouple their dynamics from their spatial distribution . in this paper
we demonstrate the feasibility of using an integral field unit ( ifu ) to study the dynamics of h@xmath1s galaxies .
we have selected one field galaxy from the sloan digital sky survey ( sdss ) and used the gmos - south ifu in nod & shuffle mode to study the rest - frame optical spectra . using this technique we can investigate the spatial distribution of star formation through the [ oii ] emission and the distribution of the young a - stars through the ( much stronger ) stronger h@xmath1 and h@xmath7 absorption lines . in 2
we present the data reduction and analysis .
the results are presented in 3 .
finally we summarise our results and present the implications in 4 .
we use a cosmology with @xmath8=0.3 , @xmath90.7 and a hubble constant of 70@xmath6mpc@xmath10 . in this cosmology ,
2kpc subtends 1@xmath11 on the sky at @xmath12 .
the target is selected from the sdss dr1 ( abazajain et al .
2001 ) , and was identified as a h@xmath1s galaxy by goto et al .
this field galaxy has a redshift @xmath13 , with j2000 coordinates ra : 10:13:45.39 , dec : + 01:16:13.66 .
absorption lines in the sdss spectra were fit with a double gaussian profile model to account for narrower emission filling within the absorption line , and all galaxies with rest - frame equivalent widths ( ews ) @xmath14 were identified as h@xmath1s . for this pilot study , we chose a relatively bright example , with @xmath15 mag ( which corresponds to @xmath16 ) . sdssj101345.39 has strong absorption lines and weak ( but non - negligible ) emission lines ( fig . 2 ) .
post - starburst galaxies with strong h@xmath1 absorption , but with non - negligible [ oii ] or h@xmath17 emission lines are usually referred to as e(a ) galaxies .
however as balogh et al .
( 2004 ) show , e(a ) galaxies with w@xmath18([oii])@xmath1910 and w@xmath0(h@xmath17)@xmath1910 appear to have the same properties of strict k+a galaxies ( which are h@xmath1s but have no detectable [ oii ] or h@xmath17 emission ) .
k+a galaxies are predominantly bulge dominated with little sign of spiral structure and signatures of recent ( substantial ) bursts of star - formation ( balogh et al .
2004 ) . in this study
, we have selected one galaxy which falls in this _
`` k+a plus weak emission '' _ category in order to compare and contrast the dynamics and spatial distribution of the gas with that of the young stellar component . using the sdss _ gri_-band imaging
we construct a true colour image of sdssj101345.39 .
each image has an exposure time of 60 seconds and was obtained in @xmath201.2@xmath11 seeing .
the final images have a plate scale of 0.4@xmath11/pixel .
the high surface brightness material in sdssj101345.39 looks like a bulge dominated galaxy , however the lower surface brightness structure has a disturbed morphology ( fig . 1 ) of diffuse material , distributed asymmetrically toward the north , which may be indicative of a tidal interaction or merger .
the gmos - ifu field of view covers the central nucleus ( where most of the stars are located ) , but the much lower surface brightness material is located outside the ifu field and extends over a @xmath21 radius .
the sdss spectroscopy of sdssj101345.39 shows strong h@[email protected] and h@[email protected] absorption lines , as well as emission lines . from the original sdss spectra , we measure w@xmath0(h@xmath1)=7.5 in absorption , and w@xmath0(o[ii])=4.1 in emission .
the galaxy also shows h@[email protected] and [ nii]@xmath226583.0 emission ; although the measured equivalent width of h@xmath17 is 7.9 , this is compromised by the strong underlying balmer absorption .
taking into account the underlying absorption , the emission line ratios may be indicative of weak agn activity in the galaxy ( although as we show in 3 , most of the emission is from a resolved component , and we therefore attach a cautionary note that , from our current data , the evidence for an agn is tentative ) .
new observations of sdssj101345.39 were taken with the gmos - south ifu in nod & shuffle mode on 2004 february 28@xmath23 u.t . during science verification time for a total of 3 hours in @xmath24 seeing and photometric conditions . using the nod & shuffle mode we chopped away from the target by @xmath25 every 30 seconds . in this configuration
the ifu uses a fiber fed system to reformat the @xmath26 field onto two long slits . using the @xmath27-band filter in conjunction with the b600 grating results in two tiers of spectra being recorded . the spectral resolution in this configuration
is @xmath281700 .
the [ o ii](@xmath293726.1,3728.8 ) emission line and h@xmath7 and h@xmath1 absorption features all fall in regions of low sky emission and absorption . in the nod & shuffle mode ,
the object and background regions are observed alternately through the same fibres by nodding the telescope . in between each observation
the charge is shuffled on the ccd by a number of rows corresponding to the the centre - to - centre spacing of the blocks of 50 fibres into which each slit is divided . for the nod & shuffle mode , each alternate block is masked off so that it receives no light from the sky but acts simply as an image store . the slit
to field mapping was arranged so that the resultant half of the total object and background field ( 7@xmath305@xmath31 + 3.5@xmath305@xmath31 ) formed a contiguous sub - field of 5@xmath305@xmath31 . since each alternate block
is masked , each exposure can be stored in different regions of the ccd without contamination from any other region of the sky .
the sequence of object and background exposures can be repeated as often as desired with the photoelectrons from each exposure being stored in their own unique regions of the detector . at the end of the sequence , the ccd is read , incurring a read - noise penalty only once ( see glazebrook & bland - hawthorn 2001 ) for further details of this general approach ) . for each fiber ( and corresponding spectrum ) , we identify the corresponding sky spectrum in the shuffled position and subtract them to achieve poisson limited sky subtraction .
the gmos data reduction pipeline was used to extract and wavelength calibrate the spectra of each ifu element .
the variations in fiber - to - fiber response were removed using twilight flat - fields and the wavelength calibration was achieved using a cuar arc lamp .
the wavelength coverage of the final data is 40805400 .
no flux standards were observed , as the spectral features of interest are narrow enough that the standard sensitivity is sufficient . since the datacube has a large wavelength coverage we correct for the parallactic angle by modelling a 2-hour observation running from -1 to + 1 hour angle and corresponding minimum and maximum airmasses of 1.17 and 1.21 . the average atmospheric dispersion between 400 nm and 540 nm is 0.67@xmath31 at an angle roughly north south .
after building the datacube , we use idl to model and correct this aberration using a linear interpolation at each slice of the datacube along the wavelength axis .
we show the reduced , sky - subtracted spectrum , collapsed over the central two arcseconds of the galaxy in fig . 4 and identify the strongest features .
to investigate the spatial distribution of the gas and stars in the galaxy we begin by extracting narrow - band slices from the datacube around the emission and absorption lines .
we fit and subtract the continuum around the line of interest using a 3@xmath32 clip to be sure that neighbouring emission and absorption lines are omitted .
we then extract the narrow - band images from the datacube by collapsing each spectrum over the feature of interest .
we also extract the continuum regions by median filtering each spectral pixel in the datacube between 49005300 . in fig .
3 we show the spatial distribution of the relatively older stars ( as traced by the continuum light between 4900 and 5300 ) , the gas ( as traced through the [ oii ] emission line ) and the young , massive stars ( traced by the h@xmath7 and h@xmath1 absorption lines ) . fig .
3 shows that , while both the [ oii ] emission and balmer absorption lines are extended , they do not have the same spatial distribution .
in particular , the centroid of the [ oii ] emission is offset @xmath33 kpc to the se from the older stars and the h@xmath1 and h@xmath7 absorption lines .
this difference is shown in fig . 5 , where we split the datacube into nine spatial bins .
it is clear that the shape of the continuum is not constant across the galaxy .
this is confirmed by comparing with the colour gradients in the sdss imaging . by binning the imaging data into @xmath34 pixels ( 1.2@xmath35 bins , matching the same spatial bins as in fig .
5 ) we investigate the colour gradients across the galaxy ( which can be seen in the inset panel of fig . 1 ) . in each panel of fig . 5
we show the corresponding _ g - r _ and _
u - r _ colours determined from the sdss imaging data .
both the spectroscopic and imaging data show the colour gradient across the galaxy with panels [ 1,2&4],[3,5&7 ] and [ 6,8&9 ] having average _ g - r _ colours of @xmath36,@xmath37 and @xmath38 and _ u - r _ colours of @xmath39 , @xmath40 and @xmath41 respectively confirming that changing shape of the continuum seen in the ifu data is real and not simply an artifact of the observations .
the spatial distribution of the young ob stars , as traced by the rest - frame uv continuum , is consistent with that of the [ oii ] emission , and offset by @xmath422kpc in projection from the older stellar population ( indicated also by the colour gradient in the imaging ) .
the colour gradient is in approximately the same direction as the surface brightness asymmetry , and suggests that the galaxy consists of at least two components , possibly as the result of a recent merger .
to investigate the dynamics in more detail we return to the datacube and fit the emission and absorption lines on a pixel - by - pixel basis .
the [ oii ] emission line doublet and underlying ( rest frame uv ) continuum was fitted using a @xmath43 minimisation procedure .
the spectra were averaged over a @xmath44 spatial pixel region , except where the signal was too low to give a significant detection of the line , in which case the smoothing area was increased to @xmath45 pixels . in regions where this averaging process still failed to give an adequate @xmath43 ( i.e. the inclusion of an emission line component does not improve the fit ) , no fit was made .
we required a minimum @xmath43 of 25 ( s / n of 5 ) to detect the line , and allow the signal to drop by a @xmath43 of 9 to calculate the error in the velocity .
this corresponds to a formal 3@xmath32 error . in fig .
6 we show the [ o ii ] velocity structure .
this shows a velocity shear of @xmath3 across @xmath206kpc in projection .
the velocity field of the gas does not resemble that of a disk .
however , if one of the progenitors was a gas rich disk , then the observed shear may be arise due residual motion of a gas disk following a merger . assuming this is the case , we estimate that the mass of the gas - rich progenitor was ( @xmath46 ) @xmath201.2@xmath4710@xmath48@xmath49 .
this mass estimate should be viewed as a lower limit to the progenitors mass since we may be seeing only a residual part of the motion that the disk had before the encounter . to investigate the ews of the short - lived a- stars we turn to the h@xmath7 and h@xmath1 absorption lines . using a similar procedure as used to fit the [ oii ] emission features we simultaneously fit the h@xmath7 and h@xmath1 absorption features ( with a fixed separation but variable intensity and width ) and show the resulting two dimensional distribution of ews in fig .
we estimate the 1@xmath32 uncertainties by perturbing all of the parameters of the best fit such that @xmath50 . from the distribution of ews
it is clear that the a- star population is most prominent at the same location as the old stellar population , but also has a large contribution near the peak in the [ oii ] distribution . whilst the two peaks in the distribution of ews in the galaxy are statistically significant , we note that systematic uncertainties such as emission line filling of the h@xmath7 and h@xmath1 absorption lines may cause the ew to appear artificially low .
we therefore attach little significance to the ` double peaked ' distribution of ews , but rather note that the distribution of a - stars in the galaxy is widespread .
this is particularly important for understanding the formation of the a - star population since it suggests that the a - stars are not formed from gas funnelled into the nucleus but rather from material which is much more widespread . to obtain the velocity structure of the a - type stars , the spectra from the 9 spatial bins in fig .
5 are cross - correlated with a template star of type a2v .
the absorption line velocities in individual panels are shown in fig
. 5 , with the zero - point defined such that the a - stars in the middle panel have zero velocity . the a - stars velocity field appears to have a @xmath51 shear in the same direction of the gas , although the lines have a greater intrinsic width and therefore the large error bars preclude any stronger conclusion .
we can obtain an estimate of the age of the a - star population by comparing the average galaxy spectrum with the spectral library of jacoby et al .
the strength of the k - line at @xmath52 is especially sensitive to the spectral type of the early a - type stars ( rose 1985 ) .
based on the relative strength of the k - line and the neighbouring h+h@xmath53 line at @xmath54 the galaxy spectrum best resembles that of an a2@xmath55 star ( hence the choice of the a2 stellar template for cross correlation above ) . taking the effective temperature of an a2 dwarf to be 9100 k ( di benedetto 1998 ) , and interpolating the theoretical isochrones of bertelli et al .
( 1994 ) we arrive at an estimate of 0.44 gyr for a ms turnoff at a2 .
likewise , the ages for a1 and a3 dwarfs are estimated to be 0.39 and 0.47 gyr respectively although stellar mixes and metallicity makes this estimate uncertain .
we have selected one field galaxy from the sdss which shows strong balmer absorption lines ( which are indicative of a strong a- star population ) , but has only a small ongoing star - formation rate . in this pilot study , we have demonstrated the ability of ifu observations to disentangle the spatial and velocity distributions of the stellar and gaseous components through the measurement of both emission and absorption features . 1 .
the sdss images show a bright nucleus and more diffuse material distributed asymmetrically toward the north ( fig . 1 ) .
the disturbed morphology of the field galaxy sdssj101345.39 is suggestive of an interaction ( or merger ) between two galaxies . 2 .
the continuum from the older stellar population is spatially offset by @xmath202kpc from the young , hot ob stars ( traced from the rest - frame uv continuum ) and is easily seen from the changing shape of the continuum across the spatial domain of the datacube in the ifu observation ( fig . 5 ) .
this colour gradient is also reflected by the strong colour gradient seen in the sdss imaging .
3 . by extracting narrow - band slices of the emission and absorption lines in the the gmos ifu datacube
, we find that the gas ( traced through the [ oii ] emission line ) is spatially extended with an elongated morphology , approximately coincident with the young , hot , ob star population ( fig . 3 ) .
4 . the @xmath4 and @xmath5 absorption lines are widely distributed ( fig .
the peaks of these absorption lines are spatially co - incident with the older stellar population and the centre of the gas .
the a - star population are not confined to the nuclear regions .
their spatial distribution is different from that of the [ oii ] emission , but the two appear related .
5 . the gas has a velocity shear of @xmath3 across 6kpc in projection ( fig .
6 ) , but the velocity field does not resemble that of a disk .
however , assuming that the velocity shear arises due to residual motion of a gas disk which has been disturbed as the result of an interaction , we estimate that the gas rich progenitor had a dynamical mass of at least @xmath56 .
this mass estimate should be viewed as a lower limit since some of the angular momentum of the gas - rich progenitor may have been lost during the interaction .
the a - stars do not show the same amplitude of velocity motion as the gas , however , there is evidence of a 40@xmath6 shear in approximately the same direction as the velocity motion of the gas ( although the large error bars preclude any strong conclusions ) .
our observations suggest that the activity in sdssj102145.39 may be the result of a strong tidal interaction between a passive ( possibly elliptical ) galaxy ( giving rise to the old stellar population ) , and a gas - rich , star - forming ( spiral or irregular ? ) galaxy ( giving rise to the spatially offset of hot stars and nebular emission ) .
the interaction between the two components is likely to have been responsible for the production of the a - star population in a burst of star formation . by comparing the galaxy spectrum with spectral libraries and stellar population models ( jocoby et al .
1984 , vazdekis & arimoto 1999 ) we estimate an burst age of @xmath200.5gyr although we caution that systematic uncertainties such as hidden emission , differing metallicities and pollution of k- type stars make this estimate uncertain .
unfortunately the wavelength coverage of our gmos data do not extend to mgii or h@xmath17 , making it difficult to constrain the underlying old population or dust absorption .
future observations at these wavelengths would allow us to trace the dynamics and metallicities of the old stars and further constrain the burst age , as well as locate the position of any possible agn ( which may be important for understanding black hole growth in these galaxies ) .
we can estimate the mass of the underlying old population from the @xmath57-band magnitude ( k=13.2+/-0.001 , balogh et al .
2004 ) and assuming a canonical value of 1.0 for the stellar mass - to - light ratio ( bruzual & charlot 2003 ) .
we estimate that the total stellar mass is @xmath58 .
it is interesting to contrast this with the mass of the young a - star population , and the current star formation rate .
if we assume a salpeter imf ( salpeter 1955 ) , we can estimate the mass of the burst population by comparing the equivalent width of h@xmath1 ( 7.5 ) with the models of shioya et al .
( 2004 ) . in order to produce absorption of this strength
, their models require a burst population of at least @xmath59 by mass ( ie . , @xmath60 ) .
we note that this agrees well with the lower limit on the dynamical mass that we estimated from the velocity shear in the [ oii ] emission ( we note that it is difficult to place an upper limit on the mass of the gas rich progenitor ) .
if the burst mass fraction was skewed towards higher mass stars , the burst mass could be a factor @xmath61 lower .
the current residual star formation rate inferred from the [ oii ] flux is @xmath62 ( kennicutt 1998 ) , so that in 0.5gyr , a mass of only @xmath63 would be converted into stars .
thus it seems that the residual star formation is at least 1/100@xmath64 of the rate during the burst , while the mass involved in the burst is about 1/10@xmath64 of the mass of the old stellar population .
the spectra , morphological information and kinematics of sdssj121345.39 appears very similar to local field e+a galaxy ngc2865 ( hau , carter and balcells 1999 ) in which a 0.41.2gyr old burst has occurred due to the merger of a gas - rich ( sb or sc ) spiral and an elliptical galaxy .
a quantitative discussion of the star - formation truncation mechanism will have to wait until we have a statistically significant sample , but we can outline the possibilities which cause this process .
competing models for e+a galaxy formation predict very different distributions of stellar populations following the interaction . in sph simulations for mergers between spiral and elliptical galaxies
the gas tends to sink to the galaxy centre on timescales much less than the starburst duration , and therefore suggest that the gas and a - stars should be centrally concentrated and segregated away from the older stellar population ( e.g. , hernquist & weil 1992 ) ( although these models do not take into account the hot ism in the elliptical ) . however , other models suggest that the gas in the progenitor should be clumpy ( in molecular clouds ) rather than distributed as a smooth medium , and the result of the merger is to scatter these star forming regions across the galaxy ( compressing / squeezing the gas and causing the a - star starburst as the interaction occurs ) . as the gas clouds scatter either in the spiral arms of the progenitor or in the new icm , they evaporate and the star formation is truncated as the density of the clouds falls ( eg.mahos & hernquist , 1996 , kojima & noguchi 1997 ) . in our pilot study
we have shown that the a - stars are widely distributed across the galaxy , which coupled with the spatially extended morphology of the gas suggests a clear preference for the latter model .
although these observations show a clear preference for the _ squeezed cloud _ model it is not yet clear that this is true for the h@xmath1s galaxy population as a whole . the next step is to generate a statistically useful sample of such galaxies , in a variety of environments , to be able to generalise our conclusions and investigate the star - formation rate of the present day universe .
this is important for constraining current models for galaxy formation in which these h@xmath1s galaxies may represent an important phase in the evolution of todays local luminous spheroidal galaxies .
we are very grateful to the gemini staff for accepting this programme for science verification time .
we would like to thank chris simpson and ian smail for useful discussions .
ams acknowledges financial support from pparc .
rgb holds a pparc senior fellowship . | in this paper we present the three dimensional dynamics of the galaxy sdssj101345.39 + 011613.66 , selected for its unusually strong balmer absorption lines ( w@xmath0(h@xmath1)=7.5 ) . using the gmos - south ifu in nod & shuffle mode
we have mapped the continuum and optical absorption lines of this @xmath2=0.1055 field galaxy .
this galaxy has a disturbed morphology , with a halo of diffuse material distributed asymmetrically toward the north . using the [ oii ] emission line ( w@xmath0[oii]=4.1 ) we find that the gas and hot ob stars are offset from the older stars in the system .
the gas also has a spatially extended and elongated morphology with a velocity gradient of @xmath3 across 6kpc in projection . using the strong @xmath4 and @xmath5 absorption lines
we find that the a- stars are widely distributed across the system and are not centrally concentrated arguing that the a - star population has formed in molecular clouds outside the nucleus . by cross correlating the spectra from the datacube with an a - star template we find evidence that the a - star population has a 40@xmath6 shear in the same direction as the gas .
the disturbed morphology , strong colour gradients and strong h@xmath1 and h@xmath7 absorption lines in sdssj101345.39 argue that this is a recent tidal interaction / merger between a passive elliptical and star - forming galaxy .
although based on a single object , these results show that we can spatially resolve and constrain the dynamics of this short lived ( yet important ) phase of galaxy formation in which the evolutionary process take galaxies from star - forming to their quiescent end products . |
heavy ion collisions is a very active field in particle physics . at present
the maximal energy is reached at cern where lead ions are accelerated to an energy of 160 gev / c per nucleon .
the next machine will be the rhic collider , expected to operate in 1999 , in which gold or lead ions will reach a c. of m. energy of @xmath0 200 gev .
one decade later an energy of @xmath0 5.5 tev will be reached at cern - lhc .
the aim of these programs it to produce and detect a new state of matter called quark matter ( qm ) or quark - gluon plasma ( qgp ) . since ordinary matter
is composed of quark and gluons it is natural to think that by heating and/or compressing it , a new phase of deconfined quarks and gluons can be reached .
this idea is quite old and was first explored in the framework of the bag model .
however , it only gained full credibility when lattice calculations showed that such a phase transition does occur in statistical qcd .
5 truemm in the case of a pure yang - mills theory ( i.e. qcd without quarks ) the phase transition is of first order and takes place at a critical temperature @xmath1 150 mev . in the big bang theory the same transition occurred , in the opposite direction ( i.e. from qgp to ordinary nuclear matter ) , when the universe was only a few microseconds old .
schematically , the situation is as follows . for @xmath2
one heats the system and the temperature increases whereas the ratio @xmath3 , where @xmath4 is the energy density , remains constant . at @xmath5
one can heat the system but the temperature does not increase ( only the latent heat does ) .
finally , for @xmath6 the temperature increases again and @xmath7 remains constant .
this new constant is much larger than the corresponding one for @xmath2 .
this is due to the larger number of degrees of freedom of a qgp as compared to a pion gas . using the value of the new constant and the value of @xmath8 one finds that the phase transition takes place when the density of the system is an order of magnitude larger than that of ordinary nuclear matter .
it is also found that there exists a restoration of chiral symmetry which takes place at the same temperature @xmath8 .
when quarks are introduced , the calculations are more complicated and the order of the phase transition depends on the number of light quark flavors
. 5 truemm although qgp could be produced in the case of neutron stars we are interested here in its production in heavy ion collisions .
the first remark is that , in this case , we are a priori very far from the conditions of thermodynamical equilibrium assumed in the lattice calculations in statistical qcd . however , as far as the energy density is concerned , it is plausible that values of the order required by theory ( i.e. 1 gev/@xmath9 ) can be reached in central @xmath10-@xmath10 collisions at present cern energies . in the bjorken scenario ,
the energy density is given by = [ dn / dy]_y^ * ~0 < e_t > r_p_b^2 _ 0 , [ 1e ] where @xmath11 is the number of secondaries per unit rapidity , @xmath12 is the average transverse energy per particle , and @xmath13 is the plasma formation time .
this time is very poorly known and there is no clear consensus among experts in the field . using @xmath14 1 fm together with the experimental values of the other quantities in ( [ 1e ] ) ,
one gets @xmath15 gev/@xmath9 .
5 truemm even if qgp is produced , the phenomenon is a transient one , and one has to find experimental signals of its formation .
although some signals regarding the bulk properties of the production have been proposed in the past ( i.e. behavior of @xmath16 versus @xmath17 ) , it seems at present that the best signals correspond to rear events or to electromagnetic processes .
i will only discuss here two of these signals : strangeness enhancement , and @xmath18 suppression .
due to chiral symmetry restoration it is expected that the ratio of strange over non - strange particles will increase due to a reduction of the production threshold in dense hadronic matter . the suppression of @xmath18 production , proposed by matsui and satz , is a consequence of color screening in a plasma ( debye screening ) - similar to the screening of electric charge in an ordinary plasma .
this screening produces a modification of the potential of the type v(r ) = v_0(r ) ( - r / r_d(t ) ) [ 2e ] where @xmath19 is the debye radius - which decreases when @xmath20 increases .
when @xmath21 is smaller than the binding radius , the corresponding bound state can not exist .
of course , at sufficiently high @xmath20 , there will be a transient melting of all hadrons .
the interest of the @xmath18 ( as well as other charmonium and bottonium states ) is twofold .
first , lattice qcd calculations show that the melting of @xmath18 occurs at temperatures @xmath22 only slightly above the critical temperature and , second , @xmath23 and @xmath24 being very rear will combine with light quarks to produce open charmed mesons and therefore , once melted , the @xmath18 will not be formed again .
the two phenomena described above have been observed experimentally in a very clear way .
indeed it has been found that the values of the ratios , , , , [ 3ea ] increase between @xmath25 or @xmath26 and central @xmath27 collisions .
the na35 collaboration claims that there is no increase between @xmath25 and @xmath26 , but errors are still large .
@xmath18 suppression has been observed by measuring the ratio @xmath18 over drell - yan ( dy ) .
it is found that this ratio decreases continuously from @xmath25 to @xmath26 to @xmath27 collisions .
furthermore , for a given @xmath27 system , the ratio decreases with increasing centrality .
this effect is very strong in @xmath10-@xmath10 collisions ( see section 2.6 and fig . 8) .
5 truemm there are claims in the literature that strangeness enhancement and particularly the enhancement of multi - strange antibaryons indicates the formation of qgp .
nevertheless , it has also been possible to explain these data in the framework of string models - at the cost of introducing some new ingredients such as string fusion and final state rescattering of secondaries ( see section 2.5 ) . likewise ,
alternative explanations of the observed @xmath18 suppression have been proposed ( see 2.6 ) . in the following
i will briefly review an independent string model : the dual parton model ( dpm ) and will discuss to what extent the above signals can be understood in this framework , what are the required modifications of the model and what these heavy ion experiments teach us regarding the mechanism of multiparticle production . 5 truemm
the dual parton model ( dpm ) is a dynamical model aimed at describing the mechanism of multiparticle production in hadronic and nuclear collisions .
a very similar model , the quark gluon string model ( qgsm ) has been introduced by the itep group .
these models are based on the large-@xmath28 expansion of non - perturbative qcd .
a comprehensive review can be found in ref .
@xcite . a short description of the physical basis of the model and some of its applications to hadronic collisions can be found in @xcite . here
i will concentrate on the generalization of the model to nuclear collisions , and , in particular , on the eventual modifications of the model required to explain the cern heavy ion data . in @xmath25 collisions
the main contribution consists in a color exchange producing two strings of type diquark quark ( see fig .
the hadronic spectra of each string is obtained by a convolution of the momentum distribution and fragmentation functions of the constituents at the string ends . due to energy - momentum conservation ,
the longitudinal momentum fractions of the constituents at the string ends in each hemisphere add up to unity .
there are also more complicated configurations , corresponding to higher order terms in the large-@xmath28 expansion , involving 4 , 6 , @xmath29 etc strings , with sea quarks and antiquarks at their ends .
these configurations correspond to multiple inelastic scattering in an @xmath30-matrix approach . in @xmath25 collisions , these higher order configurations give small contributions at moderate energies , but their contribution becomes increasingly important with increasing energy . in proton - nucleus and nucleus - nucleus collisions , these higher - order contributions are enhanced by trivial combinatorial factors ( proportional to @xmath31 in @xmath26 collisions ) .
more precisely , in @xmath26 collisions the dominant graph , involving only one struck nucleon of @xmath32 and @xmath33 spectator nucleons , is the same two string diagram of fig .
1 . a double inelastic collision , involving two struck nucleons of @xmath32 and @xmath34 spectators , corresponds to a four string graph as shown in fig . 2 , and so on .
the weights of these contributions can not be computed in qcd and are taken from the glauber - gribov approach .
one assumption of dpm , which is not motivated by the large-@xmath28 expansion , is that strings are independent and particles emitted in different strings are uncorrelated .
this assumption works well in @xmath25 , @xmath26 and peripheral @xmath27 collisions .
however , we shall see that for central @xmath27 collisions the cern data require the introduction of final state interaction of secondaries .
apart from this assumption the model has a sound theoretical basis and has great predictive power .
the latter is due to the fact that strings are universal . the momentum distribution and fragmentation functions , allowing to compute the properties of each individual string , are the same in all hadronic and nuclear collisions .
moreover , the momentum distribution functions are determined from regge intercepts .
for instance , that of a valence quark in a pion is given by _
q_v^(x_q_v ) x_q_v^-1/2 ( 1 - x_q_v ) ^-1/2 .
[ 4ea ] the behaviour for @xmath35 is the well known @xmath36 behaviour associated to an ordinary regge trajectory of intercept 1/2 .
since @xmath37 , the behaviour for @xmath38 is the same as for @xmath39 , and therefore one has the symmetric behaviour ( [ 4ea ] ) . in this way
the end point behaviour is under control .
of course , we can multiply ( [ 4ea ] ) by a smooth function of @xmath40 .
however , in view of the singular behaviour at the end points , this will have little influence on the numerical results . for a valence quark in a proton , one has instead _
q_v^p(x ) x^-1/2 ( 1 - x)^1.5 .
[ 3e ] here the behaviour at @xmath41 is obtained by taking the diquark momentum fraction to zero .
this behaviour is controlled by an exotic @xmath42 regge trajectory of intercept close to - 3/2 . as a consequence of ( [ 3e ] )
the valence quark in a proton is slow in average whereas the diquark is fast @xmath43 .
its fragmentation into a leading baryon accounts , in this way , for the leading particle effect ( see section 2.3 ) . in the first applications of dpm
there was no attempt to compute the fragmentation functions . only their universality was used .
for instance in studying @xmath26 collisions , one can obtain the fragmentation function from a fit to @xmath25 data ( or the @xmath26 data for a given @xmath32 ) .
the model then predicts both the @xmath32 dependence and the energy dependence .
an important progress in the model happened when kaidalov showed @xcite that the fragmentation functions can also be determined to a large extent from regge arguments . in this way
the model becomes very predictive and the spectra of different hadrons can be computed basically without free parameters - except absolute normalizations . in order to illustrate this point ,
i show in figs .
3 and 4 the leading proton and leading @xmath44 @xmath45 distribution in @xmath25 collisions @xcite .
the shape of the leading proton spectrum is practically fixed theoretically ( here we show only the non - diffractive part but the diffractive one can also be computed in a similar way @xcite ) . moreover ,
the different shapes of the proton and @xmath44 distributions is entirely due to the difference between @xmath46 , which appears in the power of @xmath47 for the leading proton , and @xmath48 , which controls the @xmath49 behaviour of the @xmath44-distribution .
likewise , the sharp decrease of @xmath50 and @xmath51 distributions , also shown in figs .
3 and 4 , is entirely controlled by known regge intercepts @xcite , @xcite .
the generalization of the model to nucleus - nucleus collisions is rather straightforward . for simplicity
i consider here @xmath52 collisions . in the approximation of only two strings per nucleon - nucleon collision , the rapidity distribution of secondaries
is given by @xmath53\ ] ] + 2 ( |n - |n_a ) n^q_s-|q_s(y ) .
[ 4e ] here @xmath54 are the rapidity distributions of the individual strings , @xmath55 is the average number of wounded nucleons of @xmath32 and @xmath56 is the average number of nucleon - nucleon collisions .
both @xmath55 and @xmath56 can be computed in the glauber model .
for instance for an average collision ( i.e. integrated over impact parameter ) , one has note that the total number of strings is @xmath57 , i.e. two strings per inelastic nucleon - nucleon collision .
the interpretation of ( [ 4e ] ) is obvious . with
@xmath55 struck nucleons , we have at our disposal @xmath55 diquarks of projectile and target ( @xmath58 and @xmath59 , respectively ) and as many valence quarks .
this accounts for the first term of ( [ 4e ] ) .
the remaining strings : @xmath60 have to be stretched by sea quarks and antiquarks , because the available valence constituents are all included in the first term ; this accounts for the second term of ( [ 4e ] ) .
of course we should combine the valence and sea constituents of the projectile with those of the target in all possible ways .
however , for linear quantities such as multiplicities , each ordering gives practically the same result .
we can see from ( [ 4e ] ) and ( [ 5e ] ) that , if all strings would have the same plateau height ( i.e. the same value of @xmath61 ) , the plateau height in an average @xmath52 collision would increase like @xmath62 . however , at present energies , the plateau height of the @xmath63-@xmath64 , strings is smaller than that of @xmath65-@xmath66 ones , and the first term of ( [ 4e ] ) dominates . at higher energies
the contribution of the sea strings becomes increasingly important .
therefore , in order to make predictions for rhic and lhc we have to introduce the multistring configurations in each nucleon - nucleon collision .
if their average number is @xmath67 ( this number can be computed in a generalized eikonal model ; one gets @xmath68 at @xmath69 and @xmath70 at @xmath71 tev ) the total number of strings is @xmath72 , and eq .
( [ 4e ] ) is changed into @xmath73 . + ( 2 |k - 2 )
n^q_s-|q_s ] + ( |n - |n_a ) 2|k n^q_s-|q_s .
[ 6e ] the predictions @xcite for average and central ( @xmath74 ) @xmath75 and @xmath10-@xmath10 collisions at rhic and lhc energies are given in table 1 .
.the number of charged secondaries per unit rapidity at @xmath76 for average ( @xmath77 ) and central collision ( @xmath78 ) at rhic and lhc [ 6 ] .
[ cols="^,^,^,^",options="header " , ] we turn next to @xmath18 suppression @xcite . for all @xmath26 and @xmath27 data @xcite
involving a light projectile ( up to @xmath30 ) there is an alternative explanation to debye color screening proposed in @xcite .
this explanation is based on the following mechanism @xcite . in the present experimental conditions the @xmath18
is produced outside ( behind ) the nucleus .
however , the @xmath80 pair is produced inside the nucleus and can interact with nucleons in its path through the nucleus .
this interaction can modify its wave function in such a way that it has a vanishing projection into the @xmath18 bound state .
a single absorptive cross - section , @xmath81 6.2 mb , allows to describe all data except the @xmath10 @xmath10 ones .
this is illustrated in fig .
the data @xcite for @xmath10 @xmath10 , also displayed in this figure , show clearly that nuclear absorption alone does not work .
another `` conventional '' mechanism is the final state interaction of the @xmath18 bound state with other secondaries produced at comparable velocity ( co - movers ) producing open charm : + j / d + |d + [ 14e ] this idea was introduced in the literature a long time ago @xcite .
however , recent theoretical calculations have shown that the @xmath82 and @xmath83 cross - sections increase very slowly from threshold ( in contrast to @xmath84 which has a very rapid increase ) @xcite . therefore in the case of the @xmath18 the co - mover cross - section
is expected to be very small ( not larger than 0.5 mb ) . because of that the idea of @xmath18 suppression via interaction with co - movers has been progressively abandoned . in a recent paper @xcite we have shown that introducing the final state interaction ( [ 14e ] ) with a cross - section of 0.4 mb
, one can explain the recent @xmath10 @xmath10 data .
the results are shown in fig .
a similar result has been obtained independently in @xcite .
one should note , however , that when introducing the interaction with co - movers one has to reduce the value of the cross - section for nuclear absorption . in this way
the @xmath18 yield in @xmath26 collisions does not fall with increasing @xmath32 as steeply as the data seem to indicate ( the `` effective '' absorptive cross - section resulting from the combined effect of nuclear absorption and interaction with co - movers is 5 mb while experimental data require 6.2 @xmath85 0.7 mb ) .
moreover the @xmath18 yield in @xmath86 falls somewhat faster than indicated by experiment . at present , however , the data are not precise enough to rule out such an explanation and further study is needed to distinguish it from a qgp scenario @xcite and/or other collective effects such as string percolation @xcite .
there are , of course , many other interesting results obtained at cern which i have not mentioned . in particular bose - einstein correlation measurements ( interferometry ) indicate that the size of the interaction region is larger than the geometrical size of the colliding system . again , this requires the existence of final state interaction of secondaries . as discussed in section 2.5 ,
there are hints in the data in favor of string fusion , although there is no compelling evidence yet . nevertheless , it is interesting to examine its possible consequences in cosmic ray physics . in a recent paper
the santiago group @xcite discusses two such consequences : the rise of the average shower depth of maximum for cosmic rays in the energy range @xmath87 to @xmath88 ev and the possibility that string fusion acts as a hadronic accelerator . when the energy increases , string fusion becomes more important and produces a decrease of the average multiplicity .
taking a fixed chemical composition ( i.e. independent of energy ) , consisting of 90 @xmath89 of iron and 10 @xmath89 of protons , one gets an average multiplicity , which , in the range @xmath90 to @xmath88 ev , is identical to the one obtained without string fusion and a chemical composition of primaries changing with energy ( from 75 @xmath89 fe and 25 @xmath89 p at @xmath87 ev to 50 @xmath89 fe and 50 @xmath89 p at @xmath88 ev ) .
this is shown in fig .
. another interesting consequence of string fusion is that it produces secondaries outside the kinematical range of a nucleon - nucleon interaction .
the authors can explain in this way the so - called cumulative effect .
moreover , they claim that these two effects can change the profile of extensive air showers and in this way alter conclusions on the primary composition and energy . for example , provided their composition could be determined by an independent method , cosmic ray events with energies around 3.10@xmath91 ev would contain , in the presence of string fusion , particles with @xmath92 or 3 ( or more in the presence of string percolation @xcite ) , and would thus correspond to an energy for the primary upto @xmath93 times smaller than the one without string fusion .
such a reduction in energy could make these events compatible with the cut - off due to the scattering of cosmic rays with the microwave background .
the qgp search has gained full credibility with the lattice qcd results .
experimental data are of very high quality but give only some hints of qgp formation . future data and especially future machines will be decisive in this search . in the mean time
, we have gained new insight in the dynamics of multiparticle production allowing to improve current models ( strong baryon stopping hinting at diquark breaking , string fusion with interesting consequences for cosmic ray physics and final state interaction of secondaries ) .
11 a. capella , u. sukhatme , c .- i . tan and j. tran thanh van , phys .
236 ( 1994 ) 225 .
a. capella , proceedings 22nd international cosmic ray conference , dublin , ireland , 1991 .
a. b. kaidalov , yad .
( 1987 ) 1452 .
a. capella , a. kaidalov , a. kouider akil , c. merino and j. tran thanh van , z. phys .
c70 ( 1996 ) 507 . a. b. kaidalov and o. i. piskunova , sov .
( 1985 ) 816 .
a. capella , c. merino and j. tran thanh van , phys .
b265 ( 1991 ) 415 . b. z. kopeliovich and b. g. zakharov , z. phys .
c43 ( 1989 ) 241 .
a. capella and b. z. kopeliovich , phys .
b381 ( 1996 ) 325 .
a. blankenbecler , a. capella , c. pajares , a. v. ramallo and j. tran thanh van , phys .
b107 ( 1981 ) 106 .
c. pajares and a. v. ramallo , phys .
d31 ( 1985 ) 2800 .
na35 collaboration , h. strbel et al .
a525 ( 1991 ) 59c .
na49 collaboration in proceedings xxxi rencontres de moriond , les arcs ( france ) , 1996 ed .
j. tran thanh van ; see also na44 collaboration , ibid .
n. armesto , m.a .
braun , e. g. ferreiro and c. pajares , phys .
b344 ( 1995 ) 301 .
references to earlier papers on string fusion can be found in g. gustafson , nucl .
a566 ( 1994 ) 233c .
rqmd : h. sorge , r. matiello , a. von keetz , h. stcker , w. greiner , z. phys .
c47 ( 1990 ) 629 ; h. sorge , m. berenguer , h. stcker , w. greiner , phys .
b289 ( 1992 ) 6 ; th .
schnfeld et al .
a544 ( 1992 ) 439c .
j. ranft , a. capella , j. tran thanh van , phys .
b320 ( 1994 ) 346 ; h. j. mhring , j. ranft , a. capella , j. tran thanh van , phys .
d47 ( 1993 ) 4146 ( the calculations in these papers are based on the dpmjet and dtnuc codes ) .
a. capella , phys .
b364 ( 1995 ) 175 .
a. capella , preprint lpthe orsay 96 - 30 , hep - ph 96 - 05216 , phys .
b in press .
na35 collaboration , t. alber et al , z. phys .
c64 ( 1994 ) 195 .
na49 collaboration , proceedings qm 96 , heidelberg .
t. matsui and h. satz , phys .
b178 ( 1986 ) 416 .
na38 collaobration , c. baglin et al .
, phys . lett .
b201 ( 1989 ) 471 ; phys .
b255 ( 1991 ) 459 .
r. salmeron , nucl .
a566 ( 1994 ) 199c .
a. capella , j. a. casado , c. pajares , a.v .
ramallo and j. tran thanh van , phys .
b206 ( 1988 ) 354 .
a. capella , c. merino , c. pajares , a. v. ramallo and j. tran thanh van , phys .
lett . b230 ( 1989 ) 149 .
c. gerschel and j. hfner , phys .
b207 , 253 ( 1988 ) ; z. phys .
c56 ( 1992 ) 71 .
na50 collaboration , p. bordal et al .
, proceedings xxxi rencontres de moriond 1996 ibid .
m. gonin et al . , proceedings quark matter 96 , ibid .
blaizot and j.y .
ollitrault , phys .
d39 ( 1989 ) 232 .
s. gavin , m. gyulassy and a. jackson , phys .
lett . 207 ( 1988 ) 194 .
j. ftacnik , p. lichard and j. pitsut , phys .
b207 ( 1988 ) 194 .
r. vogt , m. parakash , p. koch and t. h. hansson , phys .
b207 ( 1988 ) 263 .
s. gavin and r. vogt , nucl .
b345 ( 1990 ) 1104 .
s. gavin , h. satz , r. thews and r. vogt , z. phys .
c61 ( 1994 ) 351 .
g. bhanot and m. e. peskin , nucl .
b156 ( 1979 ) 365 .
a. kaidalov and p. volkovitsky , phys .
69 ( 1992 ) 3155 .
a. kaidalov , proceedings xxviii rencontres de moriond ( 1993 ) , ed .
j. tran thanh van .
m. luke et al , phys .
b288 ( 1992 ) 355 .
d. kharzeev and h. satz , phys .
b306 ( 1994 ) 155 .
a. capella , a. kaidalov , a. kouider akil and c. gerschel , lpthe 96 - 55 , hep - ph/9607265 .
s. gavin et r. vogt , proceedings qm 96 , ibid and preprint lbl-37980 ( 1996 ) .
j. p. blaizot and j. y. ollitrault , proceedings quark matter 96 , ibid ; phys .
rev . lett .
77 ( 1996 ) 1703 . c. y. wong
, proceedings quark matter 96 , ibid .
d. kharzeev and h. satz , proceedings qm96 , ibid .
n. armesto , m. a. braun , e. g. ferreiro and c. pajares , phys .
77 ( 1996 ) 3736 .
n. armesto , m. a. braun , e. g. ferreiro , c. pajares and yu m. shabelski , preprints us - ft/6 - 96 ( phys .
b to appear ) and us - ft/11 - 96 ( astropart .
phys . to appear ) .
a. e. brenner et al . , phys .
d26 ( 1982 ) 1497 . | in the first part , i give a brief description of the quark - gluon plasma search at cern and of some experimental results . in the second part
, i review a dynamical model of nucleus - nucleus interactions and propose a physical interpretation of those results .
= cmmib10 9=="0920 |
entanglement is probably the most important resource in quantum information processing : quantum teleportation @xcite , quantum cryptography @xcite , quantum computation @xcite , quantum error correction @xcite are only some examples of its ubiquitous crucial role . in all the instances mentioned above its main use
is transmitting , processing , correcting quantum information . in this paper
we would like to analyse whether it is useful in one particular scenario of transmission of classical information along a noisy channel : the depolarising channel .
it has been proved that when multiple uses of the channel are allowed entangled states can maximise distinguishability in a particular case of noisy channel : the two - pauli channel @xcite .
it is however common belief that , due to their fragility in the presence of noise , entangled states can not improve the transmission of classical information when used as signal states , at least when the noise is isotropic - in the sense that will be explained in the following sections - as in the case of the depolarising channel .
although there is numerical evidence in this direction @xcite no explicit proof has been produced so far , at least at the best of the authors knowledge .
the scope of this paper is precisely to prove that the mutual information can not be increased when signals of entangled states of two qubits are used with a memoryless depolarising channel .
in the following we will consider a memoryless quantum channel acting on individual qubits .
the channel is described by
action " operators @xcite @xmath0 satisfying @xmath1 such that if we send through the channel a qubit in a state described by the density operator @xmath2 the output qubit state is given by the map @xmath3 by definition a channel is memoryless when its action on arbitrary signals @xmath4 , consisting of @xmath5 qubits ( including entangled ones ) , is given by @xmath6 for the depolarising channel , on which we will concentrate our attention , the action operators take the following simple form @xmath7 here @xmath8 are the pauli matrices , which are transformed under the action of the channel as @xmath9 as we can see , the depolarising channel can be specified by the parameter @xmath10 , whose meaning will be clear in the following section , or equivalently by the error probability @xmath11 .
this gives us a complete description of the channel for any input state composed of an arbitrary number of qubits .
in the simplest scenario the transmitter can send one qubit at a time along the channel . in this case the codewords will be restricted to be the tensor product of the states of the individual qubits .
quantum mechanics however allows also the possibility to entangle multiple uses of the channel . for this more general strategy
it has been shown that the amount of reliable information which can be transmitted per use of the channel is given by @xcite @xmath12 where @xmath13 with @xmath14 is the input ensemble of states @xmath4 of @xmath5 generally entangled qubits and @xmath15 is the mutual information @xmath16 where the index @xmath5 stands for the number of uses of the channel .
here @xmath17 is the von neumann entropy , @xmath18 are the density matrices of the outputs and @xmath19 .
logarithms are taken to base 2 .
the advantage of the expression ( [ def - i ] ) is that it includes an optimisation over all possible povms at the output , including collective ones .
therefore no explicit maximisation procedure for the decoding at the output of the channel is needed .
the interest for the possibility of using entangled states as channel input is that it can not generally be excluded that @xmath15 is superadditive for entangled inputs , i.e. we might have @xmath20 and therefore @xmath21 . in this scenario
the classical capacity @xmath22 of the channel is defined as @xmath23 for the depolarising channel a lower bound on @xmath22 is given by the one - shot capacity @xmath24 ( see @xcite ) , while upper bounds are given in @xcite . in this paper
we will not attempt to find the classical capacity of the depolarising channel , as this would imply analysing signals with any degree of entanglement between @xmath5 qubits and performing the limit @xmath25 .
we will restrict ourselves to the simplest non - trivial case , namely @xmath26 , and we will find the maximal mutual information @xmath27 .
the question we want to address is : is it possible to increase the mutual information by entangling the two qubits , i.e. is @xmath28 for the depolarising channel ? to answer this question with ``
no '' we will show that due to the isotropy of the depolarising channel the mutual information for orthogonal entangled states of two qubits depends only on the degree of entanglement , and in particular that it is a decreasing function of the entanglement .
we anticipate that , as we will show later , the mutual information can not be increased by an alphabet of non - orthogonal states . to simplify our analysis we express the signal states in the bloch vector " representation @xcite @xmath29 where as before @xmath30 are the pauli operators of the two qubits . using eq .
( [ shrink ] ) the output of the channel can be written as @xmath31
this shows that the output states are linked to the input signals by an isotropic shrink of the bloch vectors @xmath32 by a factor @xmath10 and of the tensor @xmath33 by a factor @xmath34 .
we will now consider the following general signal state @xmath35 in the above equation @xmath36 $ ] parametrises the degree of entanglement between the two qubits that carry the signal , which we will measure with the usual entropy of entanglement @xmath37 @xcite , where @xmath38 denotes the reduced density operator of one qubit ( note that @xmath39 is the same for both qubits as the trace can be taken indifferently over any of the two ) .
for the state ( [ signals ] ) we have @xmath40 .
we point out that this is the most general choice for a signal state .
this can be easily verified observing that any pure state @xmath41 of two qubits can be decomposed in the schmidt basis as follows @xcite @xmath42 where @xmath43 are real and @xmath44 , while @xmath45 and @xmath46 represent orthogonal bases for the first and the second qubit , respectively .
these two bases will not in general have the same orientation in the bloch vector representation , but for our purpose this is not an impediment because of the isotropy of the depolarising channel . in other words , since there is no privileged direction for the choice of basis @xmath47 - they do not even have to be the same for each of the two qubits - we are free to let them coincide with the two respective schmidt bases .
after the action of the depolarising channel , the density operator corresponding to the input state ( [ signals ] ) takes the form @xmath48 in the basis @xmath49 the output @xmath50 reads @xmath51 the corresponding eigenvalues are @xmath52 they depend on the degree of entanglement , but are independent of the choice of basis . for maximisation of the second term of the mutual information @xmath53 , given in eq .
( [ def - i ] ) , it is sufficient to optimise independently terms of the form @xmath54 remember that the _ a priori _ probabilities add to one , so neither the number of input states nor their probabilities enter in the maximisation of the second term . as we will show later , the first term in ( [ def - i ] ) and therefore the mutual information is maximised for a set of orthogonal and equally probable states .
we can search the extremum of ( [ mutual ] ) analytically leading to the requirement @xmath55 for @xmath56 there are two solutions , namely @xmath57 which corresponds to non - entangled input states and turns out to be the maximum , and @xmath58 which means maximal entanglement of the inputs and corresponds to the minimum of @xmath53 .
the explicit expression for the maximal mutual information corresponding to four equiprobable orthogonal non - entangled input states is given by @xmath59 notice that this is equivalent to twice the one - shot capacity .
so we have shown that in this case entanglement does not help to increase the mutual information @xcite .
as promised , we now justify our choice of orthogonal states by proving that the mutual information can not be increased by using non - orthogonal alphabets .
we were able to maximise the two parts of the mutual information , defined in equation ( [ def - i ] ) , independently of each other for the following reasons .
note first that @xmath60 is maximal for @xmath61 .
for the depolarising channel this form is achieved by any set of four orthogonal equiprobable input states .
the minimum of the term @xmath62 in the mutual information , however , does not depend on the orthogonality of the states , as only the eigenvalues of the output states determine the extremum .
we noticed that each @xmath63 can be minimised independently for each input state @xmath4 . as we have shown before , @xmath63 is minimal , and reaches the same minimum value , for any choice of input states of non - entangled qubits .
note that due to these reasons the same mutual information could be also reached by using a larger alphabet of non - orthogonal non - entangled states and adjusted probabilities , but it can never be improved beyond the maximum value given in ( [ maxinfo ] ) .
as an illustration , in figure [ fig1 ] we show the mutual information for the following set of equally probable orthogonal states @xmath64 and @xmath65 for @xmath66 . as we can see , the mutual information is a decreasing function of the degree of entanglement . in figure [ fig2 ]
we report the mutual information as a function of @xmath10 for uncorrelated states and maximally entangled states . as proved above
, we can see that uncorrelated signals lead to a higher mutual information for any channel with @xmath67 .
to summarise , we have asked whether classical communication through a depolarising channel can be improved by entangling two uses of the channel .
our analytical results show that this is not the case : the mutual information is maximised when using orthogonal equiprobable non - entangled states .
the generalisation to more than two qubits remains an open problem @xcite .
we would like to thank c.h .
bennett , c. fuchs , r. jozsa , g. mahler and j. schlienz for helpful discussions .
in particular , we thank a. peres for constructive critics .
this work was supported in part by the european tmr research network erb 4061pl95 - 1412 the physics of quantum information " , by ministero delluniversit e della ricerca scientifica e tecnologica under the project `` amplificazione e rivelazione di radiazione quantistica '' and by deutsche forschungsgemeinschaft under grant sfb 407 .
part of this work was completed during the 1998 workshops on quantum information organised by isi foundation - elsag - bailey and by the benasque center for physics .
bennett , g. brassard , c. crpeau , r. jozsa , a. peres and w. wootters , _ phys .
lett . _ * 70 * , 1895 ( 1993 ) .
ekert , _ phys .
lett . _ * 67 * , 66 ( 1991 ) .
r. jozsa , in _ `` geometric issues in the foundations of science '' _ , eds .
s. hugett , l. mason , k.p .
tod , s.t .
tsou and n.m.j .
woodhouse , oxford university press 1997 , ( preprint quant - ph/9707034 ) .
j. preskill , _ proc .
. london _ * 454 * , 385 ( 1998 ) .
bennett , c. fuchs and j. smolin , in _ quantum communication , computing and measurement _ , eds .
o. hirota , a.s .
holevo and c.m .
caves ( plenum press , ny , 1997 ) , p.79 ( preprint quant - ph/9611006 ) . k. kraus , _ states , effects , and operations : fundamental notions of quantum theory _
( springer , berlin , 1983 ) . b. schumacher and m.d .
westmoreland , _ phys .
_ a * 56 * , 131 ( 1997 ) .
holevo , ieee trans .
inf . theory ( preprint quant - ph/9611023 ) .
bennett , d.p .
divincenzo and j. smolin , _ phys .
lett . _ * 78 * , 3217 ( 1997 ) ; c.h .
et al . _ , quant - ph/9904023 .
j. schlienz and g. mahler _ phys .
rev . _ a * 52 * , 4396 ( 1995 ) .
for a recent review see m.b .
plenio and v. vedral , _ contemp . phys . _ * 39 * , 431 ( 1998 ) .
a. peres , _ quantum theory : concepts and methods _ ( kluwer academic , dordrecht , 1993 ) .
it was pointed out that entanglement is useful in super - dense coding ( c.h .
bennett and s.j .
wiesner , _ phys .
lett . _ * 69 * , 2881 ( 1992 ) ) .
notice that our scenario is different from super - dense coding because in our case both qubits are transmitted through the noisy channel . notice that our line of argument is based on the schmidt decomposition and therefore it can not be directly generalised to more than two uses of the channel . | we analyse the role of entanglement for transmission of classical information through a memoryless depolarising channel . using the isotropic character of this channel we prove analytically that the mutual information can not be increased by encoding classical bits into entangled states of two qubits . 6.5 in 8.5 in -0.2 in -023 in |
it is now widely accepted that bipolar outflows from active galactic nuclei ( agn ) largely offset the radiative losses of the hot plasma that forms `` cooling flows '' systems in which there is a central depression in the temperature of the gravitationaly trapped plasma @xcite . notwithstanding
this concensus about the importance of agn for cooling - flow evolution , considerable controversy surrounds the mechanisms by which agn heat cooling flows .
observations , first by rosat @xcite and later by chandra @xcite of regions of depressed x - ray emission and ( usually ) enhanced synchrotron emission , have played a crucial role in overturning the paradigm of distributed mass dropout that dominated the field in the 1980s and 1990s .
these regions have been called `` bubbles '' and considerable effort , both numerically @xcite and theoretically @xcite has been devoted to understanding the way bubbles interact with the surrounding , higher - density plasma .
since lines of sight through a bubble also pass through large volumes of x - ray emitting plasma in front of and behind a bubble , even a weak depression in the x - ray surface brightness at the location of the bubble may be indicative of a large drop in emissivity inside the bubble .
in fact @xcite report that the data for the perseus cluster are consistent with zero emissivity inside one of the bubbles .
consequently , it is generally assumed that inside bubbles the density of thermal plasma is so low that the plasma emits negligibly in x - rays .
this material is generally assumed to be matter shot out from the nucleus in a high - velocity jet , which was thermalized to a high temperature in the hotspot that marks the interface between the jet and shocked ambient plasma .
given that the plasma is close to an ideal gas , the thermal energy inside a bubble is @xmath4 , where @xmath3 is the ratio of principal specific heats . by taking the bubble to be in pressure balance with the x - ray emitting plasma immediately outside it , @xmath1 can be determined from the x - ray data , while @xmath2 follows from the geometry . hence the energy _ inside _ bubbles can be reliably estimated from observational data . during inflation of a bubble work
is inevitably done on the surroundings plasma .
the _ minimum _
work that must be done at this stage is the work that would be done if the bubble expanded reversibly , which is @xmath5 , bringing the minimum energy associated with a bubble to @xmath6 for a non - relativistic plasma ( @xmath7 ) .
however , the inflation of bubbles is probably anything but reversible , and in this case , by clausius inequality the work done will exceed @xmath5 .
we use high - quality hydrodynamical simulations of bubble formation to examine how much work is actually done by the jet , and conclude that it exceeds @xmath8 by a substantial factor .
the framework within which we simulate cluster heating has been described elsewhere @xcite .
we use the adaptive - grid eulerian hydrocode enzo @xcite with a piece - wise - parabolic riemann solver .
the plasma is assumed to be an ideal , monatomic , non - relativistic gas .
radiative losses are calculated in the optically - thin limit using the line and continuum cooling rates of @xcite .
the initial conditions are modelled on the current configuration of the hydra cluster as determined from chandra data by @xcite .
jets are launched by depositing mass , momentum and energy in an injection disk of radius @xmath9 @xcite .
the grid cells are cubical and @xmath10 on a side at the centre , growing in stages by a factor of two in side length up to @xmath11 at the edge of the grid , which is a cube @xmath12 on a side .
periodic boundary conditions are imposed on this outer boundary .
in contrast to many other simulations of cluster heating ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ?
* ) there is no inner boundary .
we have analyzed the energetics of all the simulations described in @xcite .
these differed in the amount of time the cluster cooled passively before jet igniton , and in the power and duration of the jets , as well as in whether there was more than one episode of jet ignition .
however , we found similar energetics of bubbles in all simulations , so here we present only results from simulation s2 , in which the cluster cooled passively for @xmath13 before a pair of jets with combined power @xmath14 ignited and ran for @xmath15 . ]
in a series of snapshots bubbles were identified as regions in which the density was a factor @xmath16 smaller than the mean density at that radius , and the volume @xmath2 of the bubbles determined as a function of @xmath16 .
clearly @xmath2 is zero for @xmath17 and is comparable to the total available volume for @xmath18 .
for @xmath19 there is a plateau in @xmath2 that we identify with the bubbles , and we present results obtained by adopting @xmath20 . in fig .
[ timefig ] the full curve shows as a function of time since jet ignition the energy injected into the simulation , while the dashed curve shows @xmath21 for the bubbles . the energy associated with the bubbles peaks at @xmath22 , while the energy injected is @xmath23 .
thus less than a fifth of the injected energy is ever associated with identifiable bubbles , and at most epochs the fraction is less . to answer the question `` where _ is _ the energy ? ''
we have analysed the volume within @xmath24 of the cluster centre in the following way .
we grouped the computational cells into approximately 40 spherical shells and in each shell @xmath25 calculated @xmath26,\ ] ] where @xmath27 is the volume of the @xmath28th cell and @xmath29 is an energy density within that cell .
the energy densities considered were @xmath30 , where @xmath31 is the particle density , @xmath32 , where @xmath33 is the matter density and @xmath34 is the gravitational potential , and @xmath35 , where @xmath36 is the flow speed . in this way we obtain measures @xmath37 , @xmath38 and @xmath39 of the changes since jet ignition in internal , potential and kinetic energy within each shell .
[ radfig ] gives the radial distribution of energy differences at four times , @xmath40 , @xmath41 , @xmath42 and @xmath43 after jet ignition . at @xmath15
( top left panel ) the jets have just switched off and the curves are structured by two distinct processes . at @xmath44 the jets are dominant , producing sharp peaks in internal ( full curve ) , potential ( short - dashed curve ) , and kinetic energy ( dotted ) .
interior to the peak in potential energy @xmath38 at @xmath45 , there is a region in which @xmath46 because uplift has reduced the mass density there . in this region
the internal energy is only very slightly depressed because the drop in @xmath33 is compensated by a rise in @xmath47 .
beyond @xmath48 the curves for @xmath49 and @xmath38 are structured by radiative cooling , which causes a slow decline in @xmath47 and a flow of material from large to small radius that causes @xmath38 to be positive at @xmath50 and negative further out .
.energy changes ( in units of @xmath51 ) in the radius ranges @xmath52 ( columns 2 to 5 ) and @xmath53 ( columns 6 to 9 ) .
the values of @xmath54 are 15 , 37 , 58 and @xmath55 , while those of @xmath56 are 35 , 75 , 100 and @xmath57 .
@xmath58.[etable ] [ cols="<,>,>,>,>,>,>,>,>",options="header " , ] the long - dashed curve in the top left panel shows the radiated energy @xmath59 : unlike the other curves , this is not calculated by differencing energy densities at two epochs but by integrating the x - ray emissivity @xmath60 through the volume of each cell and with respect to time since jet ignition .
the top right panel of fig .
[ radfig ] shows the disposition of energy @xmath61 after jet ignition .
compared to the top left panel for @xmath62 , the long - dashed curve of the radiative losses has risen at all radii , and the peaks in @xmath37 , @xmath38 and @xmath39 have moved outwards with the rising jet debris . now the curves for @xmath37 , @xmath38 and @xmath63 are structured by the jet for @xmath64 and by cooling further out . at small radii
the peak in @xmath37 is now preceeded by a pronounced trough , reflecting the development of an under - pressured region behind the zone filled by material that is being carried up by its momentum , and decelerating in the cluster s gravitational field . at @xmath65 cooling continues to drive an inward flow that makes the negativity of @xmath37 and @xmath38 at large radii more pronounced .
the bottom left panel of fig .
[ radfig ] , for @xmath66 after jet ignition , shows a new feature : the curves for @xmath37 and especially @xmath38 have developed peaks at @xmath45 , interior to the troughs that preceed the main peaks .
this new region of enhanced potential energy reflects material that has fallen down after being uplifted and is now at a radius lower than the radius it would have in an equilibrium model , given its specific entropy .
thus this peak reflects a particular phase of the gravity waves that are strongly excited by the jets .
the bottom - right panel , for @xmath67 is very similar to that for @xmath68 . at each panel of fig .
[ radfig ] one can identify two characteristic radii , namely the radius @xmath56 that marks the end of the sharp drop in curves for @xmath37 and @xmath38 that the shock produces , and the radius @xmath54 at which these curves cross zero on the other side of the peak .
for the time of each panel in fig .
[ radfig ] , table [ etable ] gives energies , in units of @xmath51 , contained in the regions @xmath69 and @xmath70 .
we see that at @xmath15 , as the jet shuts off , the total energy inside @xmath54 has increased by an amount @xmath71 , that greatly exceeds the total injected energy , @xmath72 and is more than an order of magnitude greater than the energy then associated with bubbles ( @xmath73 from fig .
[ timefig ] ) .
this energy increase is largely caused by a negative value of @xmath74 following uplift of material to achieve this value , the jet has not in fact supplied @xmath75 of energy , which would suffice to lift the uplifted material to @xmath76 , but only the smaller amount required to move material out of the cavity zone
. this explains why @xmath77 for this region exceeds the the jet energy . in the dense region outside @xmath54 and inside @xmath78 , @xmath38 has increased by twice the jet energy , reflecting ( i ) the arrival from below of gravitationally bound uplifted material , and ( ii ) the arrival from above of material that has cooled .
the internal energy of the gas in this region has risen by 75% of @xmath79 . at this stage
the bulk kinetic energy alone is equal to the energy associated with bubbles . at later times
we see that within @xmath54 , @xmath37 and @xmath39 continue to increase , while @xmath38 becomes less negative . in the dense shell
further out @xmath39 continually decreases , while @xmath37 slowly rises .
the total energy change @xmath77 in the dense shell is always falling as material is driven in by the jet - induced flow on one side and gravity on the other , bringing with it large ( negative ) potential energies . in this region cooling also attenuates the rise in @xmath37 that would otherwise occur .
we have shown that in the simulations the minimum energy associated with bubbles , @xmath80 , is smaller than the injected energy by at least a factor of nearly 6 .
why is this so , and can we expect the result to extend to real cooling flows , or does it merely reflect limitations of the simulations ?
the factor @xmath81 allows for the energy @xmath82 that is actually in the bubble , and an allowance @xmath5 for the work that must be done to create a cavity in the ambient medium . as explained in the introduction , this is the _ minimum _ allowance , and is the actual requirement only if the cavity is created reversibly and at constant pressure . in this case
the displaced material would be neither heated nor accelerated , and the work @xmath5 would be done on the mechanism ( piston etc ) that held the pressure constant . in reality the creation of the cavity is strongly irreversible , so the displaced material is heated and accelerated . in as much as real jets probably involve larger velocities than those ( @xmath83 ) characteristic of the simulations , the simulations may be expected to under- rather than over - estimate the irreversibility of bubble inflation . the internal energy of displaced material is changed by of order itself , and its original value was equal to the bubble energy . moreover
the material evacuated from the cavity itself displaces material , so the mass of the material that has to be moved and heated exceeds the mass originally associated with the cavity by a factor @xmath84 , say .
doubling the internal energy of the shifted material , would involve supplying @xmath85 of energy over and above what is required in the reversible case . in the simulations the timescale for bubble inflation is of order the sound - crossing time at the ambient sound speed , and it is widely assumed that the same is true in real clusters . consequently , the shifted material inevitably gains an amount of kinetic energy that is comparable to the original internal energy , @xmath86 , as table [ etable ] confirms to be the case .
these two estimates of energy that must be supplied should be added to the energy @xmath87 required in the reversible case , where @xmath3 is the ratio of principal specific heats of the material in the cavity . in our simulations @xmath88 , so the actual energy required is larger than the minimum energy by a factor @xmath89 .
if the cavity contains relativistic plasma , as is frequently assumed , and the ambient plasma is non - relativistic , the energy that must be supplied is larger than the minimum energy by a factor @xmath90 .
the energetics of the simulations implies that the shifted mass is @xmath91 times the mass originally in the bubble in reality @xmath28 is probably not so large because diffusivity will lead to @xmath5 being underestimated , both in the simulations and in real cooling flows .
our simulations do not explicitly include thermal conductivity , but any numerical hydrodynamic scheme has significant diffusivity on the scale of the grid ( or the mean - free path in the case of smooth - particle hydrodynamics ) by virtue of assigning single values to the dynamical variables at each grid point , when in reality the material in a cell would contain hot and cold regions . as a consequence , energy can cross a cell in one timestep when hot matter enters on one side at the start of a timestep , and warmed matter leaves on the opposite side at the start of the next timestep .
there is no concensus regarding the appropriate values of the kinetic transport coefficients of intracluster gas [ see @xcite for conflicting positions ] , and even less understanding of the effectiveness of turbulent transport .
however , turbulence stretches and distorts bubbles , and when their narrowest dimension becomes comparable to the grid spacing , diffusivity will eliminate them . thus kinetic and turbulent diffusivity cause bubbles to shrink on a dynamical timescale in the simulations , as they probably do in reality ] the broken curve in fig .
[ timefig ] illustrates this phenomenon at two points : between @xmath92 and @xmath93 , @xmath5 dips while the jet is still firing , and then at @xmath94 it falls continuously after the jet has switched off .
the dip at early times reflects shrinkage in the volume of low - density material by @xmath95 percent following the emergence of turbulent eddies around the jet ; as the jet establishes itself at @xmath96 , the cavities shift from being nearly hemispherical to elongated ( fig .
[ cavfig ] ) .
the steep velocity gradient perpendicular to the jet axis then induces large - scale turbulence , which draws cool material into the cavity .
the ensuing fragmentation of the cavity leads to the decline in cavity volume that causes the dip in the @xmath5 curve of fig .
[ timefig ] . at @xmath97 a new regime
is entered in which lobes swell around the now more slowly advancing hotspots .
the drop in the @xmath5 curve at @xmath94 is also driven by turbulent breakup .
how realistic the steepness of this drop is , we do not know .
observationally it is inevitable that the volume of cavities will be underestimated because it has been determined from x - ray surface brightness , rather than from a knowledge of the plasma density at each point in three dimensions , as can be done in the simulations
. clearly such underestimation of the volume will lead to a corresponding underestimation of the power input by the jets .
it would be interesting to compare estimates of cavity volumes obtained from x - rays with ones obtained from high - resolution radio maps , which might be less sensitive to projection effects .
on the other hand , the x - ray cavities sometimes do not coincide with the radio lobes the hydra cluster provides an example of this phenomenon @xcite .
this discussion leads to the conclusion that taking the energy injected by jets to be @xmath21 probably does lead to its being underestimated by a substantial factor , which we have estimated to be @xmath98 .
our estimate of this factor may be on the high side if the simulations over - estimate the diffusivity of the plasma that arises from both turbulent mixing and kinetic processes , but may be on the low side because our jet speeds are relatively low and cavity volumes are likely to be underestimated from x - ray surface brightnesses .
the fact that most of the injected energy is never in cavities , has important implications for the time- and space - dependence of cluster heating .
as table [ etable ] shows , much of this energy is stored as potential energy the energy required to lift low - entropy gas above its natural position in the cluster atmosphere .
this energy subsequently heats the plasma through the dissipation of the non - linear gravity waves to which it gives rise .
the timescale of this release can be very long , especially in the case of an unusually violent outburst of the agn , which drives cool plasma to great heights , at which the dynamical time can be long @xcite .
several studies ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) have considered the implications of cooling flows for the phenomenolgy of agn and radio sources under the assumption that the injected energy is @xmath99 .
the studies conclude that heating by agn can balance radiative cooling provided the duty cycle of the jets is high . with our increased estimate of the energy injected in a given episode of bubble inflation , the required duty cycles will be shorter . assuming a 10 percent efficiency in the conversion of rest mass to jet power , @xcite
conclude that jet powers are smaller than the rate of energy release from bondi accretion of hot plasma by a factor @xmath100 ( their fig .
4 ) . prima facie
, our increase in the required jet power would bring it up to the rate of release of energy by bondi accretion .
is this conclusion problematic , given that only a fraction of the plasma that passes over the bondi accretion radius can be swallowed by the black hole ?
the rate at which material passes the bondi radius provides a useful measure of the mean jet power , but not of the instantaneous power , because inside the bondi radius an accretion disc is likely to provide a reservoir of material , and jet outbursts are thought to be brief episodes in the evolution of this disc @xcite .
thus on the timescale of bubble inflation , jet power can safely exceed the mean power available by bondi accretion , even if the black hole s only mass supply is from virial - temperature gas ( which is by no means certain ) .
estimates of the power available from bondi accretion are most usefully compared with the radiative losses of the cooling flow @xcite .
this work has been supported by st johns college , oxford | we examine the distribution of injected energy in three - dimensional , adaptive - grid simulations of the heating of cooling flows .
we show that less than 10 percent of the injected energy goes into bubbles .
consequently , the energy input from the nucleus is underestimated by a factor of order 6 when it is taken to be given by @xmath0 , where @xmath1 and @xmath2 are the pressure and volume of the bubble , and @xmath3 the ratio of principal specific heats .
[ firstpage ] cooling flows galaxies : clusters : general galaxies : formation |
for the @xmath0 system , there is evidence in lattice qcd @xcite , at least for static quarks , that the hamiltonian , is @xmath1 with the potential being given by @xmath2 , where @xmath3 and @xmath4 are the two possible meson - meson potentials and @xmath5 is the tetraquark sector potential .
we call @xmath6 the triple flip - flop potential . in the tetraquark configuration ,
a single connected string is formed .
this string is formed from five segments of string as can be seen in fig .
[ potentials ] ( see @xcite for a detailed explanation ) .
we have studied the colour field configurations for the tetraquark .
the results are shown on fig .
[ chromofields ] .
it is hard to solve this hamiltonian , so we will instead solve a simplified potential model which contains the essential ingredients of the triple flip - flop potential , but depends on a smaller number of variables .
we will do the following approximations .
first we will consider that all the quarks and antiquarks have an equal mass , while neglecting the exchange effects .
secondly , we will consider only the confining ( linear ) part of the potential and that the quarks are non - relativistic .
finally , we will also reduce the number of variables of the model . and
[ potentials ] , title="fig : " ] and @xmath7 . [ potentials ] , title="fig : " ]
is the intra - diquark distance and @xmath8 the inter - diquark distance .
[ chromofields ] , title="fig : " ] is the intra - diquark distance and @xmath8 the inter - diquark distance .
[ chromofields ] , title="fig : " ] is the intra - diquark distance and @xmath8 the inter - diquark distance .
[ chromofields ] , title="fig : " ] is the intra - diquark distance and @xmath8 the inter - diquark distance .
[ chromofields ] , title="fig : " ] is the intra - diquark distance and @xmath8 the inter - diquark distance .
[ chromofields ] , title="fig : " ] is the intra - diquark distance and @xmath8 the inter - diquark distance .
[ chromofields ] , title="fig : " ] is the intra - diquark distance and @xmath8 the inter - diquark distance .
[ chromofields ] , title="fig : " ] is the intra - diquark distance and @xmath8 the inter - diquark distance .
[ chromofields ] , title="fig : " ]
we now simplify the triple flip - flop potential , in order to reduce it to a two - dimension problem , similar to the one of a cherry in a broken glass . to simplify the triple flip - flop potential , we assume that the quark - quark distance @xmath9 and the antiquark - antiquark distance @xmath10 are equal @xmath11 .
thus , we only need to consider the variable @xmath12 and @xmath13 in our model . in this way , only one meson - meson branch is relevant , and the simplified flip - flop potential is given by @xmath14 , where @xmath15 and @xmath16 this potential is depicted in the right side of fig .
[ potentials ] . in this way
, we obtain the schrdinger equation , for this model : @xmath17 \psi(r , \rho ) = e \psi(r , \rho ) \ , .
\label{schro}\ ] ] and what we call the outgoing spherical wave method ( useful to study resonances ) .
since the hamiltonian , is invariant for independent rotations of @xmath12 and @xmath7 , we have two conserved angular momenta @xmath18 and @xmath19 .
so , the wavefunction can be written as @xmath20 and the schrdinger equation([schro ] ) becomes @xmath21 we now place the system in a lattice and discretize the second derivatives the usual way .
dirichlet boundary conditions are used at the origin , for the wavefunction to be regular and at the box boundary for convenience .
we then evaluate the eigenstates of the system .
only a few of them are bound or semi - localized states .
they can be identified by calculating @xmath22 and comparing it to the box size . in the following results , we consider only @xmath23 . for @xmath24 ,
only continuum states are found .
however , for @xmath25 , we are able to found some semi - localized states , such as the one depicted on the left side of fig .
[ finitediff_res ] . on the center , we can see a true bound state found for @xmath26 . .
center : bound state for @xmath26 .
right : `` phase shifts '' obtained from the finite differences .
[ finitediff_res ] , title="fig : " ] .
center : bound state for @xmath26 .
right : `` phase shifts '' obtained from the finite differences . [ finitediff_res ] , title="fig : " ] . center : bound state for @xmath26 .
right : `` phase shifts '' obtained from the finite differences .
[ finitediff_res ] , title="fig : " ] we try to calculate the phase shifts using the finite differences results , by calculating the functions @xmath27 where the @xmath28 are the eigenfunctions of the hamiltonian @xmath29 and , then by fitting the long range limit @xmath30 .
we obtain the results in the right side of fig .
[ finitediff_res ] .
as we can see there , the results become irregular when a second channel is open , due the interference of the different channels , and so we ca nt use this method to calculate the phase shifts .
for that , we had to resort to a scattering theory approach .
by expanding the wavefunction as @xmath31 we obtain the coupled channel schrdinger equation @xmath32 where @xmath33 , @xmath34 are the eigenvalues of @xmath35 and @xmath36 . considering the scattering from the channel @xmath37 to the channel @xmath38 , we have asymptotically @xmath39 , if the channel @xmath38 is open , otherwise vanishing .
the @xmath40 can be computed by considering the outgoing solutions of the eq .
( [ schro_cc ] ) , @xmath41 .
we obtain @xcite @xmath42 also we can obtain the phase shifts by studying the asymptotic behaviour of the solutions .
we can describe the scattering process with four quantum numbers : the scattering angular momentum @xmath43 , the confined angular momentum @xmath44 and the initial and final states radial number in the confined coordinate @xmath45 , @xmath46 and @xmath47 .
we now compute the phase shifts , in order to search for resonances in our simplified flip - flop model .
the energies in the plots will be given in terms of @xmath48 , which is the energy scale of the problem . in fig .
[ deltas ] we see the phase shifts for different values of @xmath44 , from the radial ground state @xmath49 .
as can be seen , no resonance is formed for @xmath24 , but for @xmath50 and @xmath51 one resonance is formed since the phase shift has a jump of @xmath52 .
this is somewhat expected , since the centrifugal barrier has the effect of maintaining the diquarks separated , while increasing the stability of the system .
we also see a bound state for @xmath26 , which agrees with the results obtained with the finite differences method and presented in fig .
[ finitediff_res ] .
the phase shifts for scattering for @xmath53 and different @xmath44 and for @xmath24 and different values of @xmath54 are also presented . and
@xmath55 , with @xmath49 ( left ) and @xmath53 ( center ) .
in the right , phase shifts , for @xmath24 and different @xmath54 . [ deltas ] , title="fig : " ] and @xmath55 , with @xmath49 ( left ) and @xmath53 ( center ) . in the right , phase shifts , for @xmath24 and different @xmath54 .
[ deltas ] , title="fig : " ] and @xmath55 , with @xmath49 ( left ) and @xmath53 ( center ) . in the right , phase shifts , for @xmath24 and different @xmath54 .
[ deltas ] , title="fig : " ] if we include the relativistic corrections for the reduced mass of the system , the @xmath56 becomes @xmath57 .
greater reduced masses could increase the binding of the system . which is in fact what we observe in fig .
[ redmass ] . for @xmath24 we can see a resonance that is destroyed for masses as low as @xmath58 , while for @xmath50 the resonance is always present .
when @xmath51 for sufficiently low masses we can observe a bound state and even a resonance in addition to it . , @xmath59 and @xmath60 .
[ redmass ] , title="fig : " ] , @xmath59 and @xmath60 .
[ redmass ] , title="fig : " ] , @xmath59 and @xmath60 .
[ redmass ] , title="fig : " ] finally we can compute the decay widths of the resonances by using the expression @xmath61 when @xmath62 is @xmath63 .
the results are given in table [ widthtables ] .
the width is larger for large angular momentum and small quark mass . using this results to estimate the physical scale of our decay widths , we obtain , in the @xmath50 state , a decay width @xmath64 for light quarks ( @xmath65 gev ) . for the charm quark
we obtain @xmath66 and for the bottom quark , the estimated width is @xmath67 . in the @xmath51 state
the estimated widths are @xmath68 , @xmath69 and @xmath70 , while in the @xmath26 state , the values are @xmath71 , @xmath72 and @xmath73 .
.left : decay widths as a function of @xmath44 .
right : decay widths with the reduced mass correction . [ cols="^,^,^",options="header " , ]
we have constructed a simplified model for the tetraquark . with this model ,
we have observed the formation of a resonance for non null @xmath44 , and even a bound state for @xmath26 .
so , we conclude that a higher angular momentum favours the formation of resonances and bound states , in agreement with karliner and lipkin @xcite . extrapolating this results to the real world
, it seems plausible the existence of tetraquarks with high angular momentum . the formation of resonances and bound states is also favoured by low quark masses in our model .
this happens since a low quark mass produces a higher binding energy which increasing the meson masses and so favouring the localization of resonances and bound states .
p. bicudo and m. cardoso , phys .
b * 674 * , 98 ( 2009 ) [ arxiv:0812.0777 [ physics.comp-ph ] ] .
c. ay , j. -m .
richard , j. h. rubinstein , phys .
lett . * b674 * , 227 - 231 ( 2009 ) .
[ arxiv:0901.3022 [ math - ph ] ] . | tetraquarks are studied with an approximation which reduces the internal degrees of freedom of the system , using both finite differences and scattering theory .
the existence of bound states and resonances is inspected and the masses and widths of the found resonances are calculated . |
lcccc observation date(s ) & 2016 jul 18 , 2016 aug 02 & 2016 jul 17 , 2016 aug 02 & 2008 aug 17,19 & 2008 aug 11 + julian day(s ) & 2457587.7 , 2457603.4 & 2457586.7 , 2457603.5 & 2454695.8 , 2454700.7 & 2454689.8 + configuration(s ) & c36 - 4 ,
c36 - 5 & c36 - 4 , c36 - 5 & very extended & very extended + project code & 2015.a.00022.t & 2015.a.00022.t & 2008a - s008 & 2008a - s008 + time on source ( min ) & 26 , 26 & 27 , 27 & 150 , 211 & 226 + number of antennas & 40 , 39 & 38 , 37 & 8 , 6 & 8 + fwhp primary beam ( @xmath3 ) & 20 & 17 & 56 & 37 + baseband freqs .
( ghz ) & 280.1 , 282.0 , 292.1 , 294.0 & 337.1 , 339.0 , 349.1 , 351.0 & 220.1 , 230.1 & 336.5 , 346.5 + spectral windows & 4 & 4 & 48 & 48 + channel width ( mhz , km s@xmath4 ) & 0.9766 , 1.0 & 0.9766 , 0.85 & 0.8125 , 1.1 & 0.8125 , 0.71 + total bandwidth ( ghz ) & 7.5 & 7.5 & 3.9 & 3.9 + continuum bandwidth ( ghz ) & 0.28 & 0.38 & 1.2 & 0.9 + proj .
baseline range ( k@xmath5 ) & 13 - 1247 & 14 - 1424 & 16 - 380 & 24 - 588 + bandpass calibrator & j1924 - 2914 , j1517 - 2422 & j1924 - 2914 & j1924 - 2914 , 3c454.3 & j1924 - 2914 , 3c454.3 + gain calibrator & j1717 - 3342 & j1717 - 3342 & j1733 - 1304 & j1733 - 1304 + flux calibrator & j1733 - 1304 & j1733 - 1304 & callisto & callisto + robust clean parameter & @xmath6 & + 0.5 & 0.0 & 0.0 + resolution ( @xmath7 ( p.a.@xmath8 ) ) & @xmath9 ( @xmath10 ) & @xmath11 ( @xmath12 ) & @xmath13 ( @xmath14 ) & @xmath15 ( @xmath16 ) + rms noise ( mjy beam@xmath4 ) & 2 & 3 & 7 & 20 over the past decade , millimeter interferometers have found many examples of `` massive protoclusters '' , loosely defined as three or more continuum sources clustered within @xmath1710000 au that appear to trace massive protostars ( e.g. * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) .
the cluster members typically span a diversity of evolutionary stages ranging from ultracompact hii regions and hot molecular cores to cold dust sources .
the cluster member separations are well matched to the trapezium stars @xcite , suggesting that we are seeing the formation phase of the central massive stars of future ob clusters .
high resolution studies of these deeply embedded protoclusters at wavelengths that can penetrate the high dust column should provide clues to the accretion mechanisms that ultimately produce massive stars .
one such proto - trapezium is ngc 6334i , which is located in the northern end of the galactic `` mini - starburst '' ngc 6334 @xcite .
the bolometric luminosity of this protocluster is @xmath18 @xmath19@xcite using the new maser parallax distance of [email protected] kpc for the neighboring source i(n ) @xcite .
we first resolved ngc 6334i with the submillimeter array ( sma ) at 1.3 mm @xcite , into four sources : the well - known ultracompact ( uc ) hii region ( mm3=ngc 6334f ) , two line - rich hot cores ( mm1 and mm2 , * ? ? ?
* ; * ? ? ?
* ) , and a dust core mm4 .
the latter three objects are so deeply embedded that they remain undetected at infrared wavelengths at least as long as 18@xmath20 m @xcite .
the hot core mm1 lies at the origin of a high - velocity bipolar molecular outflow from the region ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , indicating active accretion .
we observed ngc 6334i with the atacama large millimeter / submillimeter array ( alma ) in cycle 2 at band 6 ( 1.3 mm ) with an angular resolution of @xmath21 ( 220 au ) . the continuum from the hot core mm1
was resolved into seven components within a projected radius of @xmath221000 au and with brightness temperatures ranging from 100 to 260 k , consistent with the gas temperatures implied by the presence of a molecular line forest and copious h@xmath23o maser emission @xcite .
the luminosities inferred from the brightness temperatures and sizes suggest that mm1 is the dominant member of the protocluster in terms of bolometric luminosity .
thus , alma s initial view of ngc 6334i mm1 suggested a `` hot multi - core '' of gas and dust containing multiple massive protostars undergoing accretion .
however , the anomalously steep 1.5 cm to 1.3 mm spectral energy distributions that we reported for several components of mm1 motivated further investigation . in this letter , we perform a detailed comparison of alma images from 2015 and 2016 with sma images obtained 7 - 8 years earlier , which reveals an extraordinary increase in the continuum emission from mm1 while the other protocluster members remained constant .
we analyze several years of single - dish maser spectra and find flaring by a factor of @xmath24 in multiple transitions beginning in early 2015 . finally , using 2016 alma target - of - opportunity observations
, we characterize the outburst in terms of luminosity , duration and location , highlighting its potential impact on the study of episodic accretion in massive star formation .
this letter includes analysis of new alma cycle 3 observations at 1.1 and 0.87 mm acquired in 2016 via director s discretionary time , along with older observations using the sma in 2008 at 1.3 and 0.87 mm .
the observing parameters for these data are presented in table [ obs ] .
the cycle 3 alma data were calibrated using version c3r4b of the pipeline .
the sma data were calibrated in miriad and imaged in casa .
because of the strong hot cores in ngc 6334i ( mm1 and mm2 , * ? ? ?
* ) , identifying line - free channels suitable for continuum imaging is challenging .
we followed the procedure described in @xcite and estimate that the resulting images suffer @xmath25 residual line contamination toward the hot cores .
the line - free continuum bandwidths employed are given in table [ obs ] .
the estimated flux scale uncertainties are 5% for alma and 10% for sma .
the continuum data were iteratively self - calibrated , using the bright emission as the initial model .
multi - scale clean was used on the alma data with scales of 0 , 5 , and 15 times the pixel size ( @xmath26 ) .
robust visibility weighting was employed to promote matched angular resolution in the alma data ; the beamsizes achieved are given in table [ obs ] .
all images were corrected for primary beam attenuation .
our earlier ( cycle 2 ) alma observations of this field at 1.3 mm are also used in this letter .
they are more fully described in @xcite , but we include their essential parameters here : observation date : 2015 august 29 ( julian day 2457264.5 ) , mean frequency : 230.0 ghz , line - free bandwidth : 0.4 ghz , resolution : @xmath27 ( 220 au ) , rms : 1.2 mjy beam@xmath4 .
ngc 6334i is one of several targets in a long - term maser monitoring program at the hartebeesthoek radio observatory ( hartrao ) using the 26 m telescope ( see , e.g. , * ? ? ?
spectra of the 22.2 ghz h@xmath23o and 6.7 ghz ch@xmath28oh maser transitions were recorded in dual circular polarization with resolutions of 7.81 khz ( 0.11 km s@xmath4 ) and 0.98 khz ( 0.044 km s@xmath4 ) , respectively , and typical rms sensitivities of 2 jy and 1 jy .
the cadence of observations is typically every 10 - 15 days .
flux calibration is performed by observing hydra a and 3c123 assuming the flux scale of @xcite .
the telescope beam size is @xmath29 and @xmath30 at the respective frequencies .
we present the alma 1.3 mm image from @xcite in figure [ fivepanel]a , alongside an sma image observed seven years earlier ( fig .
[ fivepanel]b ) .
when we convolve the alma image to the resolution of the sma image , the result ( fig . [ fivepanel]c ) shows a significant difference in the morphology and the total flux density of mm1 ( [email protected] vs. [email protected] jy ) between the two epochs .
interestingly , the other sources in the protocluster ( mm2 , mm3 , and mm4 ) do not show significant differences . as described in
[ obstext ] , residual line emission contaminating the channels chosen to make the alma continuum image can not be the cause of such a large difference in flux density .
furthermore , the spectrum of the mm2 hot core is similarly line - rich @xcite so a line contamination effect should manifest in that source as well . to exclude @xmath31-sampling effects as the culprit
, we simulated the 2008 sma observations using the 2015 1.3 mm alma image ( fig [ fivepanel]a ) as the sky model and applying the identical @xmath31-coverage and flagging of our sma observations ( fig [ fivepanel]b ) using the casa simobserve task . to account for the small difference in frequency between the alma and sma 1.3 mm tunings , we first scaled the sky model image using a spectral index ( @xmath32 ) image constructed from the contemporaneous 2016 1.1 and 0.87 mm alma images .
this technique corrects for the distinct differences in @xmath33 among the dust sources ( average values : mm1=2.9 , mm2=3.2 , mm4=2.4 ) , which would otherwise cause a systematic error up to 6% . the resulting image ( fig .
[ fivepanel]d ) demonstrates what the sma would have seen had it observed the region simultaneously with the 2015 alma observations .
we find that the sma could have recovered @xmath34 ( 9.4 jy ) of the mm1 emission recovered by alma , meaning that the bulk of the difference between the 2008 and 2015 images can not be due to @xmath31-sampling .
furthermore , fig [ fivepanel]e , which shows the difference between the simulated 2015 sma image and the actual 2008 sma image , demonstrates that mm1 exhibits an excess of @xmath35 jy , while mm2 - 4 are consistent with being unchanged ( @xmath36 ) .
we conclude that mm1 has flared by a factor of [email protected] at 1.3 mm during the seven - year interval between the observations , where the uncertainty includes the combined uncertainties in the flux scales and image measurements .
this dramatic increase in flux density explains the anomalously steep 1.5 cm to 1.3 mm spectral energy distributions reported by @xcite for several of the subcomponents of mm1 , and further indicates that the millimeter flaring happened after the 2011 may vla observations .
the 2016 0.87 mm alma image of ngc 6334i is shown in figure [ fivepanel2]a .
the morphology of this image matches well with the 1.3 mm alma image observed one year earlier .
performing the same analysis on our 2016 alma and 2008 sma data at 0.87 mm ( fig .
[ fivepanel2]a - e ) yields an increase in the flux density of mm1 from [email protected] to [email protected] jy .
the sma could have recovered @xmath37 ( 27.8 jy ) of the mm1 emission , yielding a comparable increase factor of [email protected] . combining the two bands ,
the mean factor is [email protected] ; however , the brightening factor could have evolved over 2015 - 2016 . in any case , in both bands
this increase represents only @xmath1730% of the total single - dish flux density from ngc 6334i measured in an @xmath38 beam @xcite ; thus , it would not have been so obvious in non - interferometric observations . in the highest - resolution image of the excess emission ( 0.87 mm , fig .
[ difference ] ) , the hypercompact ( hc ) hii region mm1b is the apparent center of activity .
however , the emission is extended and elongated ( @xmath39 au ) , consistent with simultaneous increases in four other subcomponents of mm1 ( a , c , d , and g ) . the positions of these four subcomponents form an elongated symmetric pattern relative to mm1b , with a major axis at position angle + 140.5 .
this pattern , combined with their simultaneous brightening , suggests that all of the subcomponents may not be independent protostars as originally interpreted by @xcite .
e ) overlaid with 2011 vla water maser spots and 7 mm continuum contours ( 0.24 and 0.6 mjy beam@xmath4 ; brogan et al .
the sma 0.87 mm beam is shown in the lower right .
letters mark the 1.3 mm continuum components modeled in @xcite.[difference ] ] strong flaring in 10 of 15 maser transitions from three species ( h@xmath23o , ch@xmath28oh , and oh ) toward ngc 6334i began in early 2015 ( macleod , g. et al . 2017 , in preparation ) . in fig .
[ masers ] we show light curves for the 22.2 ghz h@xmath23o and 6.7 ghz class ii ch@xmath28oh maser lines , each for the component at @xmath40 km s@xmath4 , corresponding to the mean lsr velocity of the thermal gas @xcite . both masers had increased by a factor of 10 by mid - may 2015 .
the h@xmath23o maser at this velocity ultimately increased by a factor of 40 compared to our 2011 may vla data @xcite , an increase that is reminiscent of past flares in orion - kl @xcite .
masers in star - forming regions are believed to occur behind @xmath41-shocks @xcite , suggesting that an energetic event has occurred .
serendipitously , the 2015 august 1.3 mm alma observations occurred when both masers reached their initial peak . while mm2 and mm3 contain 6.7 ghz masers ( e.g. , * ? ? ?
* ; * ? ? ?
* ) , mm1 has never been detected in this line , nor in the 12.2 ghz class ii maser line , in three decades of interferometric observations @xcite . because this maser requires far - infrared continuum to pump the torsionally excited states @xcite ,
the increase in continuum from mm1 could have initiated unprecedented maser activity in mm1 and/or enhanced the existing emission in mm2 and mm3 .
follow - up vla observations are underway to pinpoint the flaring masers .
km s@xmath4 , is chosen for each maser species .
the water maser measurements began on 2013 january 26 .
the methanol masers in ngc 6334 i(n ) ( located @xmath42 north , * ? ? ?
* ; * ? ? ?
* ) are also within the main beam of the 26 m dish at 6.7 ghz , but they appear at a different velocity ( @xmath432.6 km s@xmath4 ) . [ masers ] ] to estimate the increase in luminosity implied by the continuum flaring of mm1 , we first measure the increase in 0.87 mm brightness temperature ( @xmath44 ) averaged over the region of brightening shown in fig .
[ difference ] . in an aperture of diameter @xmath45 ( @xmath46 au ) , @xmath44
has increased by a factor of [email protected] ( from 33 to 96 k ) . given the uncertain geometry
, we consider this change in the context of a simple model of a central protostar surrounded by a thick spherical shell of dust .
since our image resolves this structure , the dust temperature ( @xmath47 ) will equal @xmath48 , where @xmath49 is the optical depth at 0.87 mm .
the two temperatures should vary proportionally regardless of @xmath49 since the column density and emissivity of the grains ( and hence @xmath49 ) can not have changed substantially in such a short period , except in the hot innermost regions where grains are destroyed ( see [ disc ] ) .
thus , we can infer a similar increase factor of 2.9 for @xmath47 at the effective emitting surface .
radiative transfer models of dust envelopes surrounding outbursting protostars indicate that @xmath47 scales as the 1/4 power of the central luminosity @xcite ; thus , we compute a luminosity increase factor of @xmath50 . we can estimate a lower limit to the outburst luminosity as the sum of the brightness temperature luminosity estimates for the five mm1 components affected ( a , b , c , d , and g ) reported by @xcite , which is @xmath51 @xmath19 . applying @xmath52 ,
the pre - outburst luminosity is then @xmath53 @xmath19 . assuming the outburst began simultaneously with the maser burst , its duration is @xmath54 years , so the total energy released ( so far ) is @xmath55 erg .
while unprecedented for a massive star forming core , a rapid quadrupling of the ( sub)millimeter continuum flux density from a deeply embedded protostellar system like mm1 is within the realm of theoretical expectation due to episodic accretion , a phenomenon that is increasingly recognized as being essential to star formation @xcite .
although dramatic , the @xmath5670 luminosity increase estimated for mm1 is only about double that of the first identified outbursting class 0 low - mass protostar hops 383 in orion ( @xmath5630 - 50 , * ? ? ?
* ) , whose submillimeter flux density increased by a factor @xmath57 .
based on continuum imaging alone , the mm1 increase occurred sometime during the 4.3 year interval between the vla observations and the alma 1.3 mm observations .
since the initiation of the maser flare lies within this window , it likely further constrains the rise time . if the maser and continuum flares began simultaneously , then the rise time was less than a few months , compared to the @xmath2218 month constraint for hops 383 from its mid - infrared measurements . @xcite simulate the effect of accretion outbursts on the envelope surrounding a low - mass protostar . a @xmath56100 increase in luminosity yields approximately a @xmath565 increase in the 0.87 mm flux density after 208 days , which is somewhat faster and larger than the factor of 4.2 seen in mm1 at 445 days after the maser flare .
the discrepancy may be due to longer timescales in a higher mass system where the dust will be heated to much higher temperatures in the inner optically thick zone .
indeed , the dust in the innermost region will be evaporated as the radius of the dust sublimation temperature ( @xmath171500 k ; e.g. , * ? ? ?
* ) moves outward , thereby enriching the gas phase @xcite .
the timescale to transfer heat from the surviving dust to the gas will be longer than the initial dust heating time @xcite .
modeling the rich , severely blended line emission from complex organic species in order to measure the ongoing changes in gas temperature and chemical abundance compared to the pre - burst state will be a challenging but important avenue of study .
the primary mechanism that could account for such a large outburst is a rapid increase in the accretion rate due either to disk fragmentation @xcite or an encounter @xcite .
a more extreme possibility is a protostellar merger , though for massive protostars this phenomenon would typically produce a much larger energy release than observed @xcite . in the theory of competitive accretion @xcite , multiple seeds accrete gas simultaneously with higher rates occurring at the center of the protocluster .
simulations of such clustered environments find that encounter - induced boosting of the disk accretion rate is possible @xcite , including up to @xmath58 @xcite . while there are several massive protostars with candidate disks ( see the list in * ? ? ?
* ) , the first evidence for a disk - mediated accretion burst from a high - mass young stellar object was recently reported using infrared observations toward s255ir nirs3 @xcite .
it is similarly massive but less deeply embedded than mm1 , enabling better constrained luminosity measurements . while its luminosity increase factor ( @xmath565.5 ) is lower than mm1 , its initial luminosity ( @xmath59 @xmath19 ) is significantly larger .
nevertheless , the two events are quantitatively similar in terms of the rapid onset , sustained duration ( @xmath60 yr ) , and total energy , which points toward a common origin .
both objects appear to drive bipolar outflows , which nominally require central disks .
intriguingly , the position angle of the elongated mm1 excess emission is exactly perpendicular to the large - scale bipolar outflow ( + 50 , * ? ? ? * ) and the quasi - symmetry of the excess with respect to mm1b suggests a nearly edge - on , thick disk - like structure .
thus , mm1a , c , d , and g may trace clumps in a fragmented disk heated by an accretion event onto mm1b , which would explain their co - brightening .
however , in this scenario , whether any of these components also contain central protostars is uncertain from the present data , since their high @xmath44 could be consistent with external heating by a @xmath61 @xmath19 central source after accounting for a modest beaming factor ( @xmath622 ) from the `` flashlight effect '' @xcite , an anisotropic radiation field favoring the poles that develops in simulations of massive accretion disks ( e.g. * ? ? ?
* ; * ? ? ?
the large , sustained increase in luminosity is analogous to fu ori outbursts @xcite , whose decay periods range from 10 to several 100 years @xcite , during which time an elevated quasi - steady accretion rate can persist ( e.g. , * ? ? ? * ) .
it is also consistent with hops 383 , which shows no fading after 6 years @xcite . in recent numerical hydrodynamic simulations of massive protostellar accretion
including gas self - gravity and radiative feedback @xcite , the central protostar undergoes a burst in accretion rate from 10@xmath63 to 10@xmath4 @xmath64 yr@xmath4 as a massive ( 0.55 @xmath64 ) gas fragment approaches and enters the protostellar sink cell . spaced by @xmath172000 years , these largest events generate luminosity bursts of @xmath5650 - 100 lasting for several years , two characteristics analogous to the mm1 event . to test this scenario
, it will be crucial to monitor the evolution of mm1 in future years in both dust and free - free emission , as well as to discern if there has been any change to the outflow properties due to the outburst . toward this end
, we have higher - resolution alma observations in the cycle 4 queue , which should also help elucidate the nature of the mm1 subcomponents .
the national radio astronomy observatory is a facility of the national science foundation operated under agreement by the associated universities , inc .
the dunlap institute is funded through an endowment established by the david dunlap family and the university of toronto .
this letter makes use of the following alma data : ads / jao.alma#2013.1.00600.s , 2015.a.00022.t .
alma is a partnership of eso ( representing its member states ) , nsf ( usa ) and nins ( japan ) , together with nrc ( canada ) and nsc and asiaa ( taiwan ) and kasi ( republic of korea ) , in cooperation with the republic of chile .
the joint alma observatory is operated by eso , aui / nrao and naoj .
cjc acknowledges support from the stfc ( grant number st / m001296/1 ) .
this research used nasa s astrophysics data system bibliographic services . | based on sub - arcsecond atacama large millimeter / submillimeter array ( alma ) and submillimeter array ( sma ) 1.3 mm continuum images of the massive protocluster ngc 6334i obtained in 2015 and 2008 , we find that the dust emission from mm1 has increased by a factor of [email protected] during the intervening years , and undergone a significant change in morphology . the continuum emission from the other cluster members ( mm2 , mm4 and the uchii region mm3=ngc
6334f ) has remained constant .
long term single - dish maser monitoring at hartrao finds that multiple maser species toward ngc 6334i flared beginning in early 2015 , a few months before our alma observation , and some persist in that state .
new alma images obtained in 2016 july - august at 1.1 and 0.87 mm confirm the changes with respect to sma 0.87 mm images from 2008 , and indicate that the ( sub)millimeter flaring has continued for at least a year .
the excess continuum emission , centered on the hypercompact hii region mm1b , is extended and elongated ( @xmath1 au ) with multiple peaks , suggestive of general heating of the surrounding subcomponents of mm1 , some of which may trace clumps in a fragmented disk rather than separate protostars . in either case , these remarkable increases in maser and dust emission provide direct observational evidence of a sudden accretion event in the growth of a massive protostar yielding a sustained luminosity surge by a factor of @xmath2 , analogous to the largest events in simulations by @xcite .
this target provides an excellent opportunity to assess the impact of such a rare event on a protocluster over many years . |
coupled oscillator theory is now a pervasive part of the theoretical neuroscientist s toolkit for studying the dynamics of models of biological neural networks .
undoubtedly this technique originally arose in the broader scientific community through a fascination with understanding synchronisation in networks of interacting heterogeneous oscillators , and can be traced back to the work of huygens on an odd kind of sympathy `` between coupled pendulum clocks '' @xcite .
subsequently the theory has been developed and applied to the interaction between organ pipes @xcite , phase - locking phenomena in electronic circuits @xcite , the analysis of brain rhythms @xcite , chemical oscillations @xcite , cardiac pacemakers @xcite , circadian rhythms @xcite , flashing fireflies @xcite , coupled josephson junctions @xcite , rhythmic applause @xcite , animal flocking @xcite , fish schooling @xcite , and behaviours in social networks @xcite . for a recent overview of the application of coupled phase oscillator theory to areas as diverse as vehicle coordination , electric power networks , and clock synchronisation in decentralised networks
see the recent survey article by drfler and bullo @xcite . given the widespread nature of oscillations in neural systems it is no surprise that the science of oscillators has found such ready application in neuroscience @xcite .
this has proven especially fruitful for shedding light on the functional role that oscillations can play in feature binding @xcite , cognition @xcite , memory processes @xcite , odour perception @xcite , information transfer mechanisms @xcite , inter - limb coordination @xcite , and the generation of rhythmic motor output @xcite .
neural oscillations also play an important role in many neurological disorders , such as excessive synchronisation during seizure activity in epilepsy @xcite , tremor in patients with parkinson s disease @xcite or disruption of cortical phase synchronisation in schizophrenia @xcite . as such
it has proven highly beneficial to develop methods for the control of ( de)synchronisation in oscillatory networks , as exemplified by the work of tass _ et al_. @xcite for therapeutic brain stimulation techniques . from a transformative technological perspective
, oscillatory activity is increasingly being used to control external devices in brain - computer interfaces , in which subjects can control an external device by changing the amplitude of a particular brain rhythm @xcite .
neural oscillations can emerge in a variety of ways , including intrinsic mechanisms within individual neurons or by interactions between neurons . at the single neuron level
, sub - threshold oscillations can be seen in membrane voltage as well as rhythmic patterns of action potentials .
both can be modelled using the hodgkin - huxley conductance formalism , and analysed mathematically with dynamical systems techniques to shed light on the mechanisms that underly various forms of rhythmic behaviour , including tonic spiking and bursting ( see e.g. @xcite ) .
the high dimensionality of biophysically realistic single neuron models has also encouraged the use of reduction techniques , such as the separation of time - scales recently reviewed in @xcite , or the use of phenomenological models , such as fitzhugh - nagumo ( fhn ) @xcite , to regain some level of mathematical tractability .
this has proven especially useful when studying the response of single neurons to forcing @xcite , itself a precursor to understanding how networks of interacting neurons can behave .
when mediated by synaptic interactions , the repetitive firing of pre - synaptic neurons can cause oscillatory activation of post - synaptic neurons . at the level of neural ensembles ,
synchronised activity of large numbers of neurons gives rise to macroscopic oscillations , which can be recorded with a micro - electrode embedded within neuronal tissue as a voltage change referred to as a local field potential ( lfp ) .
these oscillations were first observed outside the brain by hans berger in 1924 @xcite in electroencephalogram ( eeg ) recordings , and have given rise to the modern classification of brain rhythms into frequency bands for alpha activity ( 8 - 13 hz ) ( recorded from the occipital lobe during relaxed wakefulness ) , delta ( 1 - 4 hz ) , theta ( 4 - 8 hz ) , beta ( 13 - 30 hz ) and gamma ( 30 - 70 hz ) .
the latter rhythm is often associated with cognitive processing , and it is now common to link large scale neural oscillations with cognitive states , such as awareness and consciousness .
for example , from a practical perspective the monitoring of brain states via eeg is used to determine depth of anaesthesia @xcite .
such macroscopic signals can also arise from interactions between different brain areas , the thalamo - cortical loop being a classic example @xcite .
neural mass models ( describing the coarse grained activity of large populations of neurons and synapses ) have proven especially useful in understanding eeg rhythms @xcite , as well as in augmenting the dynamic causal modelling framework ( driven by large scale neuroimaging data ) for understanding how event - related responses result from the dynamics of coupled neural populations @xcite .
one very influential mathematical technique for analysing networks of neural oscillators , whether they be built from single neuron or neural mass models , has been that of weakly coupled oscillator theory , as comprehensively described by hoppensteadt and izhikevich @xcite . in the limit of weak coupling between limit cycle oscillators invariant manifold theory @xcite and averaging theory @xcite
can be used to reduce the dynamics to a set of phase equations in which the relative phase between oscillators is the relevant dynamical variable .
this approach has been applied to neural behaviour ranging from that seen in small rhythmic networks @xcite up to the whole brain @xcite . despite the powerful tools and wide - spread use afforded by this formalism ,
it does have a number of limitations ( such as assuming the persistence of the limit cycle under coupling ) and it is well to remember that there are other tools from the mathematical sciences relevant to understanding network behaviour . in this review
we wrap the weakly coupled oscillator formalism in a variety of other techniques ranging from symmetric bifurcation theory and groupoid formalisms through to more `` physics - based '' approaches for obtaining reduced models of large networks .
this highlights the regimes where the standard formalism is applicable , and provides a set of complementary tools when it does not .
these are especially useful when investigating systems with strong coupling , or ones for which the rate of attraction to a limit cycle is slow . in [ sec :
neuronoscillators ] we review some of the key mathematical models of oscillators in neuroscience , ranging from single neuron to neural mass , as well as introduce the standard machinery for describing synaptic and gap junction coupling .
we then present in
[ sec : collectivedyns ] an overview of some of the more powerful mathematical approaches to understanding the collective behaviour in coupled oscillator networks , mainly drawn from the theory of symmetric dynamics .
we touch upon the master stability function approach and the groupoid formalism for handling coupled cell systems . in [ sec : coupledlimitcycles ] we review some special cases where it is either possible to say something about the stability of the globally synchronous state in a general setting , or that of phase - locked states for strongly coupled networks of integrate - and - fire neurons .
the challenge of the general case is laid out in [ sec : reduced ] , where we advocate the use of phase - amplitude coordinates as a starting point for either direct network analysis or network reduction . to highlight the importance of dynamics off cycle we discuss the phenomenon of shear - induced chaos . in the same section
we review the reduction to the standard phase - only description of an oscillator , covering the well known notions of isochrons and phase response curves .
the construction of phase interaction functions for weakly coupled phase oscillator networks is covered in [ sec : weakcoupling ] , together with tools for analysing phase - locked states .
moreover , we go beyond standard approaches and describe the emergence of turbulent states in continuum models with non - local coupling .
another example of something more complicated than a periodic attractor is that of a heteroclinic attractor , and these are the subject of [ sec : heteroclinic ] .
the subtleties of phase reduction in the presence of stochastic forcing are outlined in [
sec : stochastic ] .
the search for reduced descriptions of very large networks is the topic of
[ sec : macroscopic ] , where we cover recent results for winfree networks that provide an exact mean - field description in terms of a complex order parameter .
this approach makes use of the ott - antonsen ansatz that has also found application to chimera states , and which we discuss in a neural context . in
[ sec : applications ] we briefly review some examples where the mathematics of this review have been applied , and finally in [ sec : discussion ] we discuss some of the many open challenges in the field of neural network dynamics .
we will assume the reader has familiarity with the following : * the basics of nonlinear differential equation descriptions of dynamical systems such as linear stability and phase - plane analysis .
* ideas from the qualitative theory of differential equations / dynamical systems such as asymptotic stability , attractors and limit cycles .
* generic codimension - one bifurcation theory of equilibria ( saddle - node , hopf ) and of periodic orbits ( saddle - node of limit cycles , heteroclinic , torus , flip ) .
there are a number of texts that cover this material very well in the context of neuroscience modelling , for example @xcite . at the end
we include a glossary of some abbreviations that are introduced in the text .
nonlinear ionic currents , mediated by voltage - gated ion channels , play a key role in generating membrane potential oscillations and action potentials .
there are many ordinary differential equation ( ode ) models for voltage oscillations , ranging from biophysically detailed conductance - based models through to simple integrate - and - fire ( if ) caricatures .
this style of modelling has also proved fruitful at the population level , for tracking the averaged activity of a near synchronous state . in all these cases bifurcation analysis
is especially useful for classifying the types of oscillatory ( and possibly resonant ) behaviour that are possible .
here we give a brief overview of some of the key oscillator models encountered in computational neuroscience , as well as models for electrical and chemical coupling necessary to build networks .
the work of hodgkin and huxley in elucidating the mechanism of action potentials in the squid giant axon is one of the major breakthroughs of dynamical modelling in physiology @xcite , and see @xcite for a review .
their work underpins all modern electrophysiological models , exploiting the observation that cell membranes behave much like electrical circuits .
the basic circuit elements are 1 ) the phospholipid bilayer , which is analogous to a capacitor in that it accumulates ionic charge as the electrical potential across the membrane changes ; 2 ) the ionic permeabilities of the membrane , which are analogous to resistors in an electronic circuit ; and 3 ) the electrochemical driving forces , which are analogous to batteries driving the ionic currents .
these ionic currents are arranged in a parallel circuit .
thus the electrical behaviour of cells is based upon the transfer and storage of ions such as k@xmath0 and na@xmath0 .
our goal here is to illustrate , by exploiting specific models of excitable membrane , some of the concepts and techniques which can be used to understand , predict , and interpret the excitable and oscillatory behaviours that are commonly observed in single cell electrophysiological recordings .
we begin with the mathematical description of the hodgkin - huxley model .
the standard dynamical system for describing a neuron as a spatially isopotential cell with constant membrane potential @xmath1 is based upon conservation of electric charge , so that @xmath2 where @xmath3 is the cell capacitance , @xmath4 the applied current and @xmath5 represents the sum of individual ionic currents : @xmath6 in the hodgkin - huxley ( hh ) model the membrane current arises mainly through the conduction of sodium and potassium ions through voltage dependent channels in the membrane .
the contribution from other ionic currents is assumed to obey ohm s law ( and is called the leak current ) .
the @xmath7 , @xmath8 and @xmath9 are conductances ( conductance=1/resistance ) .
the great insight of hodgkin and huxley was to realise that @xmath7 depends upon four activation gates : @xmath10 , whereas @xmath8 depends upon three activation gates and one inactivation gate : @xmath11 . here
the gating variables all obey equations of the form @xmath12 the conductance variables described by @xmath13 take values between @xmath14 and @xmath15 and approach the asymptotic values @xmath16 with time constants @xmath17 .
these six functions are obtained from fits with experimental data .
it is common practice to write @xmath18 , @xmath19 , where @xmath20 and @xmath21 have the interpretation of opening and closing channel transition rates respectively .
the details of the hh model are provided in appendix a for completeness .
a numerical bifurcation diagram of the model in response to constant current injection is shown in fig . [ fig : hhbif ] , illustrating that oscillations can emerge in a hopf bifurcation with increasing drive . .
the green circles show the amplitude of a stable limit cycle and the blue circles indicate unstable limit cycle behaviour .
the solid red line shows the stable fixed point and the black line shows the unstable fixed point .
details of the model are in appendix a. [ fig : hhbif ] , width=288 ] the mathematical forms chosen by hodgkin and huxley for the functions @xmath22 and @xmath23 , @xmath24 , are all transcendental functions .
both this and the high dimensionality of the model make analysis difficult . however , considerable simplification is attained with the following observations : ( i ) @xmath25 is small for all @xmath1 so that the variable @xmath26 rapidly approaches its equilibrium value @xmath27 , and ( ii ) the equations for @xmath28 and @xmath29 have similar time - courses and may be _ slaved _ together .
this has been put on a more formal footing by abbott and kepler @xcite , to obtain a reduced planar model for @xmath30 obtained from the full hodgkin - huxley model under the replacement @xmath31 and @xmath32 for @xmath33 with a prescribed choice of @xmath34 .
the phase - plane and nullclines for this system are shown in fig . [
fig : hhr ] . and
green for @xmath34 ) of the reduced hh neuron mode , obtained using the reduction technique of abbott and kepler @xcite , for the oscillatory regime ( @xmath35 ) capable of supporting a periodic train of spikes .
the periodic orbit is shown in blue .
[ fig : hhr ] , width=288 ] for zero external input the fixed point is stable and the neuron is said to be _
excitable_. when a positive external current is applied the low - voltage portion of the @xmath1 nullcline moves up whilst the high - voltage part remains relatively unchanged . for sufficiently large constant external input
the intersection of the two nullclines falls within the portion of the @xmath1 nullcline with positive slope . in this case
the fixed point is unstable and the system may support a limit cycle .
the system is said to be oscillatory as it may produce a train of action potentials .
action potentials may also be induced in the absence of an external current for synaptic stimuli of sufficient strength and duration .
this simple planar model captures all of the essential features of the original hh model yet is much easier to understand from a geometric perspective .
indeed the model is highly reminiscent of the famous fhn model , in which the voltage nullcline is taken to be a cubic function .
both models show the onset of repetitive firing at a non - zero frequency as observed in the hh model ( when an excitable state loses stability via a subcritical hopf bifurcation ) . however , unlike real cortical neurons they can not fire at arbitrarily low frequency .
this brings us to consider modifications of the original hh formalism to accommodate bifurcation mechanisms from excitable to oscillatory behaviours that can respect this experimental observation .
many of the properties of real cortical neurons can be captured by making the equation for the recovery variable of the fhn equations quadratic ( instead of linear ) .
we are thus led to the cortical model of wilson @xcite : @xmath36 where @xmath37 and @xmath38 . here
@xmath39 is like the membrane potential @xmath1 , and @xmath40 plays the role of a gating variable .
in addition to the single fixed point of the fhn model it is possible to have another pair of fixed points , as shown in fig .
[ fig : cortical ] ( right ) . as
@xmath4 increases two fixed points can annihilate at a _
saddle node on an invariant curve _
( snic ) bifurcation at @xmath41 @xcite . in the neighbourhood of this global bifurcation the firing
frequency scales like @xmath42 .
for large enough @xmath4 there is only one fixed point on the middle branch of the cubic , as illustrated in fig .
[ fig : cortical ] ( left ) .
in this instance an oscillatory solution occurs via the same mechanism as for the fhn model . , @xmath43 , @xmath44,@xmath45 .
the voltage nullcline is shown in red and that of the recovery variable in green .
left : @xmath46 , showing a stable fixed point ( black filled circle ) , a saddle ( grey filled circle ) and an unstable fixed point ( white filled circle ) .
right : @xmath47 , where there is an unstable fixed point ( white filled circle ) with a stable limit cycle ( in blue ) for @xmath48 .
[ fig : cortical ] , width=384 ] a snic bifurcation is not the only way to achieve a low firing rate in a single neurone model .
it is also possible to achieve this via a homoclinic bifurcation , as is possible in the morris - lecar ( ml ) model @xcite .
this was originally developed as a model for barnacle muscle fiber .
morris and lecar introduced a set of coupled ordinary differential equations ( odes ) incorporating two ionic currents : an outward going , non - inactivating potassium current and an inward going , non - inactivating calcium current .
assuming that the calcium currents operate on a much faster time scale than the potassium current one they formulated the following two dimensional system : @xmath49 with @xmath50)$ ] , @xmath51)$ ] , and @xmath52 $ ] . here
@xmath40 represents the fraction of k@xmath0 channels open , and the ca@xmath53 channels respond to @xmath1 so rapidly that we assume instantaneous activation . here
@xmath54 is the leakage conductance , @xmath55 are the potassium and calcium conductances , @xmath56 are corresponding reversal potentials , @xmath57 , @xmath58 are voltage - dependent gating functions and @xmath59 is a voltage - dependent rate . referring to fig .
[ fig : ml ] , as @xmath4 decreases the periodic orbit grows in amplitude , it comes closer to a saddle point and slows down such that near the homoclinic bifurcation , where the orbit collides with the saddle at @xmath41 , the frequency of oscillation scales as @xmath60 . with @xmath61 , @xmath62 , @xmath63 , @xmath64 , @xmath65 , @xmath66 , @xmath67 , @xmath68 , @xmath69 , @xmath70 , @xmath71 and @xmath72 .
the voltage nullcline is shown in red and that of the gating variable in green .
the filled black circle indicates a stable fixed point , the grey filled circle a saddle and the filled white circle an unstable fixed point .
the periodic orbit is shown in blue .
[ fig : ml ] , width=288 ] although conductance - based models like that of hodgkin and huxley provide a level of detail that helps us to understand how the kinetics of channels ( with averaged activation and inactivation variables ) can underlie action - potential generation their high dimensionality is a barrier to studies at the network level .
the goal of a network - level analysis is to predict emergent computational properties in populations and recurrent networks of neurons from the properties of their component cells .
thus simpler ( lower dimensional and hopefully mathematically tractable ) models are more appealing - especially if they fit single neuron data .
a one - dimensional nonlinear if model takes the form @xmath73 such that @xmath74 is reset to @xmath75 just after reaching the value @xmath76 .
in other words we seek a piecewise discontinuous solution @xmath74 of ( [ eq : one ] ) such that @xmath77 firing times are defined iteratively according to @xmath78 real cortical data can be very accurately fit using : @xmath79 with @xmath80 mv , @xmath81 ms , @xmath82 mv and @xmath83 mv @xcite .
there are many varieties of nonlinear if model , with the quadratic one @xcite being well known as a precursor for the planar izhikevich spiking model @xcite , itself capable of generating a wide variety of firing patterns , including bursting and chattering as well as regular spiking . for a more thorough discussion of if models and the challenges of analysing non - smooth dynamical systems we refer the reader to @xcite . at a synapse
presynaptic firing results in the release of neurotransmitters that cause a change in the membrane conductance of the postsynaptic neuron .
this postsynaptic current may be written @xmath84 where @xmath85 is the voltage of the postsynaptic neuron , @xmath86 is the membrane reversal potential and @xmath87 is a constant conductance .
the variable @xmath88 corresponds to the probability that a synaptic receptor channel is in an open conducting state .
this probability depends on the presence and concentration of neurotransmitter released by the presynaptic neuron .
the sign of @xmath86 relative to the resting potential ( which without loss of generality we may set to zero ) determines whether the synapse is excitatory ( @xmath89 ) or inhibitory ( @xmath90 ) .
the effect of some synapses can be described with a function that fits the shape of the postsynaptic response due to the arrival of action potential at the pre - synaptic release site .
a postsynaptic potential ( psp ) @xmath91 would then be given by @xmath92 , @xmath93 where @xmath94 is the arrival time of a pre - synaptic action potential and @xmath95 fits the shape of a realistic psp ( with @xmath96 for @xmath97 ) .
a common ( normalised ) choice for @xmath95 is a difference of exponentials : @xmath98 or the alpha function @xmath99 obtained from ( [ eq : doe ] ) in the limit @xmath100 .
the psp arising from a train of action potentials is given by @xmath101 where @xmath102 denotes the arrival time of the @xmath26th action potential at a synapse .
interestingly even purely inhibitory synaptic interactions between non - oscillatory neurons can create oscillations at the network level , and can give rise to central pattern generators of half - center " type @xcite . to see this we need only consider a pair of ( reduced ) hodgkin - huxley neurons with mutual reciprocal inhibition mediated by an @xmath103-function synapse with a negative reversal potential .
the phenomenon of anode break excitation ( whereby a neuron fires an action potential in response to termination of a hyperpolarising current ) can underlie a natural anti - phase rhythm , and is best understood in terms of the phase plane shown in fig .
[ fig : hhr ] . in this case inhibition will effectively move the voltage nullcline down , and the system will equilibrate to a new hyperpolarised state . upon release of inhibition
the fixed point will move to a higher value , though to reach this new state the trajectory must jump to the right hand branch ( since now @xmath104 ) . an example is shown in fig .
[ fig : halfcenter ] .
-function synapse and a negative reversal potential .
the inset shows the voltage traces ( solid and dashed lines ) for the two neurons .
the solid blue line in the @xmath30 space shows the common orbit of the two neurons ( though each neuron travels half a period out of phase with the other ) .
parameters : @xmath105 ms , @xmath106 mv , @xmath107 ms @xmath108 .
spike times determined by a voltage threshold at @xmath109 mv .
[ fig : halfcenter ] , width=288 ] gap junctions differ from chemical synapses in that they allow for direct communication between cells .
they are typically formed from the juxtaposition of two hemichannels ( connexin proteins ) and allow the free movement of ions or molecules across the intercellular space separating the plasma membrane of one cell from another .
as well as being found in the neocortex , they occur in many other brain regions , including the hippocampus , inferior olivary nucleus in the brain stem , the spinal cord , and the thalamus @xcite . without the need for receptors to recognise chemical messengers , gap junctions are much faster than chemical synapses at relaying signals .
the synaptic delay for a chemical synapse is typically in the range @xmath15@xmath110ms , while the synaptic delay for an electrical synapse may be only about @xmath111ms .
it is common to view the gap - junction as nothing more than a channel that conducts current according to a simple ohmic model . for two neurons with voltages @xmath112 and @xmath113 the current flowing into cell @xmath114 from cell @xmath115
is given by @xmath116 , where @xmath117 is the constant strength of the gap junction conductance .
they are believed to promote synchrony between oscillators ( e.g. see @xcite ) , though the story is more subtle than this as we shall discuss in
[ sec : coupledlimitcycles ] . as well as supporting oscillations at the single neuron level ,
brain tissue can also generate oscillations at the tissue level . rather than model
this using networks built from single neuron models , it is has proven especially useful to develop low dimensional models to mimic the collection of thousands of near identical interconnected neurons with a preference to operate in synchrony .
these are often referred to as neural mass models , with state variables that track coarse grained measures of the average membrane potential , firing rates or synaptic activity .
they have proven especially useful in the description of human eeg power spectra @xcite , as well as resting brain state activity @xcite and mesoscopic brain oscillations @xcite . in many neural population models , such as the well known wilson - cowan model @xcite , it is assumed that the interactions are mediated by firing rates rather than action potentials ( spikes ) _ per se_. to see how this might arise we rewrite ( [ eqstt ] ) in the equivalent form using a sum of dirac @xmath118-functions @xmath119 identifying the right hand side of ( [ eq : ss ] ) as a train of pre - synaptic spikes motivates the form of a phenomenological rate model in the form @xmath120 with @xmath121 identified as a firing rate and @xmath122 identified as the differential operator @xmath123 . at the network level it is then common practice to close this system of equations by specifying @xmath121 to be a function of pre - synaptic activity .
a classic example is the jansen - rit model @xcite , which describes a network of interacting pyramidal neurons ( p ) , inhibitory interneurons ( i ) and excitatory interneurons ( e ) , and has been used to model both normal and epileptic patterns of cortical activity @xcite .
this can be written in the form @xmath124 which is a realisation of the structure suggested by ( [ eq : rate ] ) , with the choice @xmath125 and @xmath126 for @xmath127 . here
@xmath128 is an external input . when this is a constant we obtain the bifurcation diagram shown in fig .
[ fig : jr ] .
oscillations emerge via hopf bifurcations and it is possible for a pair of stable periodic orbits to coexist .
one of these has a frequency in the alpha band and the other is characterised by a lower frequency and higher amplitude . recently
a network of such modules , operating in the alpha range and with additive noise , has been used to investigate mechanisms of cross - frequency coupling between brain areas @xcite .
neural mass models have also previously been used to model brain resonance phenomena @xcite , for modelling of epileptic seizures @xcite , and are very popular in the neuroimaging community for model driven eeg / fmri fusion @xcite .
s@xmath129 , @xmath130 s@xmath129 , @xmath131 mv , @xmath132 mv , @xmath133 s@xmath129 , @xmath134 mv , @xmath135 mv@xmath129 , @xmath136 , @xmath137 , @xmath138 .
solid red ( black ) lines represent stable ( unstable ) fixed points .
green ( blue ) points denote the amplitude of stable ( unstable ) periodic orbits that emerge via hopf bifurcations .
the inset shows the co - existence of two stable periodic orbits at @xmath139 hz .
[ fig : jr ] , width=288 ] now that we have introduced some oscillator models for neurons and neural populations it is appropriate to consider the set of tools for analysing their behaviour at the network level .
we give a brief overview of some dynamical systems approaches , concepts and techniques that can be used to understand collective behaviour that spontaneously appears in coupled dynamical system models used for neuroscience modelling .
we do not give a complete review of this area but try to highlight some of the approaches and how they interact ; some examples of applications of these to neural systems are given in later chapters .
a set of @xmath140 coupled nonlinear systems represented by odes can be written @xmath141 where @xmath142 represent the state space of the individual systems whose evolution @xmath143 is affected both by the current state of the system and by the states of those coupled to that system . in the mathematics literature , the @xmath144 are often called `` cells '' though we note that the @xmath144 may include degrees of freedom in the coupling as well as variables such as membrane potential that reflect the state of the `` biological cell '' .
note there is potential for notational confusion here : to clarify this we write @xmath145 one of the most important observations concerning the collective dynamics of coupled nonlinear systems relates to whether the collection behaves as one or not - whether there is an attracting synchronous state , or whether more complex spatio - temporal patterns such as clustering appear .
there is a very large literature , even restricting to the case of applications of synchrony , and one where we can not hope to do the whole area justice .
we refer in particular to @xcite .
various applications of synchrony of neural models are discussed , for example , in @xcite while there is a large literature ( e.g. @xcite ) discussing the role of synchrony in neural function .
other work looks for example at synchronisation of groups of networks @xcite and indeed synchrony can be measured experimentally @xcite in groups of neurons using dynamic patch clamping .
we discuss some of the types of behaviour that can emerge in the collective dynamics and the response of partial synchronised states to external forcing .
clearly , any system of @xmath140 coupled dynamical systems can be written in the form @xmath146 where each system is parametrised by @xmath142 ; @xmath147 corresponds to decoupling the systems and the functions @xmath143 represent drive from other systems on the @xmath114th system .
many approaches start with the less general case @xmath148 which can be justified for system where the collective interaction between systems can be broken into `` pairwise '' interactions @xmath149 that are summed according to some linear weights @xmath150 ( some of which may be zero ) that represent the strength of the couplings and @xmath151 the strength of coupling of the network .
note that there is clearly no unique way to write the system in this form ; more specifically , one can without loss of generality choose @xmath152 , @xmath153 by suitable choice of @xmath149 . on the other hand it can be useful to be able to e.g. modulate the strength of the coupling across the whole network independently of changing individual coupling strengths .
similarly it is often useful to specialise to a case such as @xmath154 and have similar interactions .
we focus now on the dynamics of pairwise coupled networks such as ( [ eq : pairwisecoupledcells ] ) as this form is assumed in most cases . under an additional assumption that the coupling between the oscillators is of the same type and either present or absent , one can consider uniformly weighted coupling of the form @xmath155 where @xmath117 is fixed and @xmath156 is an adjacency matrix of the graph of interactions , i.e. @xmath157 if there is a link from @xmath115 to @xmath114 and @xmath14 otherwise , or more generally @xmath158 where @xmath159 represents the strength of coupling and @xmath156 the adjacency matrix .
it is clear that the coupling structure as represented in @xmath156 will influence the possible dynamics on the network and to make further progress it is useful to restrict to particular network structures .
some important literature on network structures is reviewed for example in @xcite , while @xcite reviews work on synchrony in complex networks up to the time of their publication ; recent work includes for example @xcite .
for a recent review of the application of complex networks to neurological disorders in the brain , see @xcite .
it is interesting to try and understand the effect of network structure on synchrony , so we briefly outline some basic measures of network structure .
the _ in - degree _ of the node @xmath114 is the number of incoming connections ( i.e. @xmath160 ) , while the _ out - degree _ is the number of outgoing connections ( i.e. @xmath161 ) and the distribution of these degrees is often used to characterise a large graph .
scale - free _ network is a large network where the distribution of in ( or out ) degrees scales as a polynomial .
this can be contrasted with highly structured homogeneous networks ( for example on a lattice ) where the degree may be the same at each node .
other properties commonly examined include the clustering properties and path lengths within the graph .
there are also various measures of centrality that help one to determine the most important nodes in a graph - for example the _ betweenness centrality _ is a measure of centrality that is the probability that a given node is on the shortest path between two uniformly randomly chosen nodes @xcite . as expected , the more central nodes are typically most important if one wishes to achieve synchrony in a network .
other basic topological properties of networks that are relevant to their dynamics include , for example , the following , most of which are mentioned in @xcite : the network is _ undirected _ if @xmath162 for all @xmath163 , otherwise it is _
directed_. we say nodes @xmath115 and @xmath114 in the network @xmath156 are _ path - connected _ if for some @xmath29 there is a path from @xmath115 to @xmath114 , i.e. @xmath164 for some @xmath29 .
the network is _ strongly connected _
if for each @xmath163 is path connected in both directions while it is _ weakly connected _ if we replace @xmath156 by @xmath165 ( i.e. we make the network undirected ) and the latter network is strongly connected .
n.b . there may be some confusion in that strong and weak connectivity are properties of a directed network - while strong and weak coupling are properties of the coupling strengths for a given network !
the _ diameter _ of a network is the maximal length of a shortest path between two points on varying the endpoints .
other properties of the adjacency matrix are discussed for example in @xcite where spectral properties of graph laplacians are linked to the problem of determining stability of synchronised states .
other work we mention is that of pecora _
et al_. @xcite on synchronisation in coupled oscillator arrays ( and see [ sec : msf ] ) , while @xcite explores the recurrent appearance of synchrony in networks of pulse - coupled oscillators ( and see [ sec : pulse ] ) .
finally , we mention _ network motifs _ - these are subgraphs that are `` more prevalent '' than others within some class of graphs .
more precisely , given a network one can look at the frequency with which a small subgraph appears relative to some standard class of graphs ( for example random graphs ) and if a certain subgraph appears more often than expected , this characterises an important property of the graph @xcite .
such analysis has been used in systems biology ( such as transcription or protein interaction networks ) and has been applied to study the structure in neural systems ( see for example @xcite ) and the implications of this for the dynamics .
the have also been used to organise the analysis of the dynamics of small assemblies of coupled cells , see for example @xcite .
continuing with systems of the form ( [ eq : coupledcells ] ) or ( [ eq : pairwisecoupledcells ] ) , if the coupling parameter @xmath151 is , in some sense , small we refer to the system as `` weakly coupled '' .
mathematically , the weak coupling approximation is very helpful because it allows one to use various types of perturbation theory to investigate the dynamics @xcite . for coupling of limit cycle
oscillators it allows one to greatly reduce the dimension of phase space . nonetheless , many dynamical effects ( e.g. `` oscillator death '' where the oscillations in one or more oscillators are completely suppressed by the action of the network @xcite ) can not occur in the weak coupling limit and moreover real biological systems that often have `` strong coupling ''
. we will return to this topic to discuss oscillator behaviour .
one can sometimes use additional structure such as weak dissipation and weak coupling of the oscillators to perform a semi - analytic reduction to phase oscillators ; see for example @xcite .
if one has a notion of synchrony between the systems of ( [ eq : coupledcells ] ) , it is possible to discuss _ clustering _ according to mutual synchrony .
caution needs to be exercised whenever discussing synchrony - there are many distinct notions of synchrony which may be appropriate in different contexts and , in particular , synchrony is typically a property of a particular solution at a particular point in time rather than a property of the system as a whole .
an important case of synchrony is _ exact synchrony _ : we say @xmath166 and @xmath167
are exactly synchronised if @xmath168 for all @xmath169 .
_ generalised synchrony _ is , as the name suggests , much more general and corresponds to there simply being a functional relationship of the form @xmath170 . for oscillators
, phases can be used to define additional notions such as _ phase _ and _ frequency synchrony _ : see [ subsec : locking ] .
although we focus mostly on individual units with simple ( periodic ) dynamics , if the units have more complex dynamics ( such as `` chaotic oscillators '' ) one can understand synchrony of the cells by analysis of the linearised differences between oscillators , and there is a sizeable literature on this ; see the review @xcite , or @xcite for clusters in a system of globally coupled bistable oscillators . in the case of two linearly
coupled identical chaotic oscillators @xmath171 where @xmath172 , if the individual oscillator @xmath173 has a chaotic attractor @xmath174 then the coupled system will have an attracting exactly synchronised attractor @xmath175 only if the coupling @xmath151 is sufficiently large in relation to the maximal liapunov exponent of the synchronous state @xcite .
an area of intense interest is the role of time delay in the collective dynamics of coupled systems . in the neural context
it is natural to include propagation delay between neurons explicity , for example in models such as @xmath176 where the delay time @xmath177 represents the time of propagation of the signal from one neuron to another .
this presents a number of serious mathematical challenges , not least due to the fact that delay dynamical systems are infinite dimensional : one must specify the initial condition over a time interval @xmath178 $ ] in order to have any chance of uniquely defining the future dynamics ; this means that for a state to be linearly stable , an infinite number of eigenvalues need to have real part less than zero .
nonetheless , much can be learned about stability , control and bifurcation of dynamically synchronous states in the presence of delay ; for example @xcite , and the volume @xcite includes a number of contributions by authors working in this area .
there are also well - developed numerical tools such as dde - biftool @xcite that allow continuation , stability and bifurcation analysis of coupled systems with delays .
for an application of these techniques to the study of a wilson - cowan neural population model with two delays we refer the reader to @xcite .
although no system is ever truly symmetric , in practise many models have a high degree of symmetry . neurons , but of the order of @xmath179 types http://neuromorpho.org/neuromorpho/index.jsp meaning there is a very high replication of cells that are only different by their location and exact morphology .
] indeed many real world networks that have _ grown _ ( e.g. giving rise to tree - like structures ) are expected to be well approximated by models that have large symmetry groups @xcite .
symmetric ( more precisely , equivariant ) dynamics provides a number of powerful mathematical tools that one can use to understand emergent properties of systems of the form @xmath180 with @xmath181 .
we say ( [ eq : ode ] ) is _ equivariant _ under the action of a group @xmath182 if and only if @xmath183 for any @xmath184 and @xmath185 .
there is a well developed theory of dynamics with symmetry ; in particular see @xcite .
these give methods that help in a number of ways : * * description : * one can identify symmetries of networks and dynamic states to help classify and differentiate between them * * bifurcation : * there is a well - developed theory of bifurcation with symmetry to help understand the emergence of dynamically interesting ( symmetry broken ) states from higher symmetry states * * stability : * bifurcation with symmetry often gives predictions about possible bifurcation scenarios that includes information about stability * * generic dynamics : * symmetries and invariant subspaces can provide a powerful structure with which one can understand more complex attractors such as heteroclinic cycles * * design : * one can use symmetries to systematically build models and test hypotheses the types of symmetries that are often most relevant for mathematical modelling of finite networks of neurons are the permutation groups , i.e. the symmetric groups and their subgroups .
nonetheless , continuum models of neural systems may have continuous symmetries that influence the dynamics and can be used as a tool to understand the dynamics ; see for example @xcite .
we review some aspects of the equivariant dynamics that have proven useful in coupled systems that are relevant to neural dynamics - see for example @xcite . in doing so
we mostly discuss dynamics that respects some symmetry group of permutations of the systems .
the full permutation symmetry group ( or simply , the symmetric group ) on @xmath140 objects , @xmath186 , is defined to be the set of all possible permutations of @xmath140 objects .
formally it is the set of permutations @xmath187 ( invertible maps of this set ) . to determine effects of the symmetry , not only the group must be known but also its _ action _ on phase space .
if this action is linear then it is a _ representation _ of the group .
the representation of the symmetry group is critical to the structure of the stability , bifurcations and generic dynamics that are equivariant with the symmetry .
for example , if each system is characterised by a single real variable , one can view the action of the permutations on @xmath188 as a _ permutation matrix _
@xmath189_{ij } = \left\ { \begin{array}{cl } 1 & \mbox { if } i=\sigma(j)\\ 0 & \mbox { otherwise}\end{array}\right . , \ ] ] for each @xmath190 ; note that @xmath191 for any @xmath192 .
table [ tab : symms ] lists some commonly considered examples of symmetry groups used in coupled oscillator network models .
.some permutation symmetry groups that have been considered as examples of symmetries of coupled oscillator networks . [ cols="<,^,<",options="header " , ] we now review some techniques of reduction which can be employed to study the dynamics of when @xmath193 so that the perturbations may take the dynamics away from the limit cycle . in doing
so we will reduce for example to an ode for @xmath194 taken modulo @xmath94 .
clearly any solution of an ode must be continuous in @xmath169 and typically @xmath194 will be unbounded in @xmath169 growing at a rate that corresponds to the frequency of the oscillator . strictly speaking ,
the coordinate we are referring to in this case is on the _ lift _ of the circle @xmath195 to a covering space @xmath196 , and for any phase @xmath197 there are infinitely many lifts to @xmath196 given by @xmath198 for @xmath199 .
however , in common with most literature in this area we will not make a notational difference between whether the phase is understood on the unit cell e.g. @xmath200 or on the lift , e.g. @xmath201 modulo @xmath202 .
consider with @xmath147 .
the _ asymptotic _ ( or _ latent _ ) _ phase _ of a point @xmath203 in the basin of attraction of the limit cycle @xmath182 of period @xmath94 is the value of @xmath204 such that @xmath205 where @xmath206 is a trajectory starting at @xmath203 .
thus if @xmath207 and @xmath206 are trajectories on and off the limit cycle respectively , they have the same asymptotic phase if the distance between @xmath207 and @xmath206 vanishes as @xmath208 .
the locus of all points with the same asymptotic phase is called an _
isochron_. thus an isochron extends the notion of phase off the cycle ( within its basin of attraction ) .
isochrons can also be interpreted as the leaves of the stable manifold of a hyperbolic limit cycle .
they fully specify the dynamics in the absence of perturbations @xcite .
there are very few instances where the isochrons can be computed in closed form ( though see the examples in @xcite for plane - polar models where the radial variable decouples from the angular one ) .
computing the isochron foliation of the basin of attraction of a limit cycle is a major challenge since it requires knowledge of the limit cycle and therefore can only be computed in special cases or numerically
. one computationally efficient method for numerically determining the isochrons is backward integration , however it is unstable and in particular for strongly attracting limit cycles the trajectories determined by backwards integration may quickly diverge to infinity .
see izhikevich @xcite for a matlab code which determines smooth curves approximating isochrons .
other methods include the continuation based algorithm introduced by osinga and moehlis @xcite , the geometric approach of guillamon and huguet to find high order approximations to isochrons in planar systems @xcite , quadratic and higher order approximations @xcite , and the forward integration method using the koopman operator and fourier averages as introduced by mauroy and mezi @xcite .
this latter method is particularly appealing and given its novelty we describe the technique below the koopman operator approach for constructing isochrons for a @xmath94-periodic orbit focuses on tracking observables ( or measures on a state space ) rather than the identification of invariant sets .
the koopman operator , @xmath209 , is defined by @xmath210 , where @xmath211 is some observable of the state space and @xmath212 denotes the flow evolved for a time @xmath169 , staring at a point @xmath213 .
the fourier average of an observable @xmath214 is defined as @xmath215 for a fixed @xmath213 , ( [ fourier ] ) is equivalent to a fourier transform of the ( time - varying ) observable computed along a trajectory .
hence , for a dynamics with a stable limit cycle ( of frequency @xmath216 ) , it is clear that the fourier average can be nonzero only for the frequencies @xmath217 , @xmath218 .
the fourier averages are the eigenfunctions of @xmath209 , so that @xmath219 perhaps rather remarkably the isochrons are level sets of @xmath220 for almost all observables .
the only restriction being that the first fourier coefficient of the fourier observable evaluated along the limit cycle is nonzero over one period .
an example of the use of this approach is shown in fig .
[ fig : isochron ] , where we plot the isochrons of a stuart - landau oscillator .
- z |z|^2 ( 1+i c)/2 $ ] , @xmath221 , with @xmath222 , @xmath223 and @xmath224 so that @xmath225 . the black curve represents the periodic orbit of the system ; in this case , we have @xmath226 .
the background colour represents the fourier average , whilst the coloured lines are the isochrons , given as level sets of the fourier average .
the white dots are the actual isochrons , computed analytically as @xmath227 , @xmath228 .
[ fig : isochron ] , width=336 ] an alternative ( non isochronal ) framework for studying oscillators with an attracting limit cycle is to make a transformation to a moving orthonormal coordinate system around the limit cycle where one coordinate gives the phase on the limit cycle while the other coordinates give a notion of distance from the limit cycle .
it has long been known in the dynamical systems community how to construct such a coordinate transformation , see @xcite for a discussion .
the importance of considering the effects of both the phase and amplitude interactions of neural oscillators has been highlighted by several authors including ermentrout and kopell @xcite and medvedev @xcite , and that this is especially pertinent when considering phenomenon such as oscillator death ( and see [ sec : weakandstrong ] ) .
phase - amplitude descriptions have already successfully been used to find equations for the evolution of the energies ( amplitudes ) and phases of weakly coupled weakly dissipative networks of nonlinear planar oscillators ( modelled by small dissipative perturbations of a hamiltonian oscillator ) @xcite .
et al_. @xcite use the notion of phase and amplitudes of large networks of globally coupled stuart - landau oscillators to investigate the effects of a spread in amplitude growth parameter ( units oscillating with different amplitudes and some not oscillating at all ) and the effect of a homogeneous shift in the nonlinear frequency parameter . we now discuss the phase - amplitude coordinate transformation detailed by hale @xcite , some of the situations where it has been employed and other areas in which a phase - amplitude description of oscillators is necessary to reveal the dynamics of the system .
consider again the system with @xmath147 which has an attracting hyperbolic periodic orbit .
make a transformation to a moving orthonormal coordinate system as follows .
choose one axis of the coordinate system to be in the direction of unit tangent vector along the periodic orbit , @xmath229 , given by @xmath230 the remaining coordinate axes can be expressed as the columns of an @xmath231 matrix @xmath232 .
we can then write an arbitrary point @xmath213 in terms of its phase @xmath197 and its amplitude @xmath233 : @xmath234 here , @xmath235 represents the euclidean distance from the limit cycle .
the vector of amplitudes @xmath236 allows us to consider points away from the limit cycle . upon projecting the dynamics onto the moving orthonormal system , we obtain the dynamics of the transformed system for @xmath197 and @xmath237 @xmath238 where @xmath239 , \nonumber\\ f_2(\vartheta,\rho)&=-\zeta^t{\frac{{{\rm d}}\zeta}{{{\rm d}}\vartheta}}\rho f_1+\zeta^t\left[f(u+\zeta\rho)-f(u)-{{\rm d}f}\zeta\rho\right ] , \nonumber\end{aligned}\ ] ] @xmath240^{-1 } \xi , \quad a(\vartheta)=\zeta^t\left[-{\frac{{{\rm d}}\zeta}{{{\rm d}}\vartheta}}+{{\rm d}f}\zeta\right ] \label{eq : h } , \end{aligned}\ ] ] and @xmath241 is the jacobian of the vector field @xmath121 , evaluated along the periodic orbit @xmath242 .
the technical details to specify the orthonormal coordinates forming @xmath232 can be found in the appendix of @xcite .
it is straight - forward to show that @xmath243 , @xmath244 and that @xmath245 . in the above
, @xmath246 describes the @xmath247-dependent rate of attraction or repulsion from cycle and @xmath248 captures the _ shear _ present in the system , that is , whether the speed of @xmath247 increases or decreases dependent on the distance from cycle .
a precise definition for shear is given in @xcite and will be discussed further in [ sec : shear ] .
some caution must be exercised when applying this transformation as it will break down when the determinant of the jacobian of the transformation vanishes .
this never occurs on cycle ( where @xmath249 ) but it may do so for some @xmath250 , setting an upper bound on how far from the limit cycle these phase - amplitude coordinates can be used to describe the system . in @xcite it is noted that for the planar ml model the value of @xmath251 can be relatively small for some values of @xmath247 , but that breakdown occurs where the orbit has high curvature . in higher dimensional systems
this issue would be less problematic .
similarly , the coordinate transformation can be applied to driven systems ( i.e. with @xmath193 ) where @xmath151 is not necessarily small . in this case
the dynamics in @xmath252 coordinates , where @xmath197 and @xmath253 , are @xmath254 with @xmath255 and @xmath256 is the @xmath257 identity matrix . here ,
@xmath28 and @xmath258 describe the effect in terms of @xmath247 and @xmath233 that the perturbations have . for planar models , @xmath259 .
medvedev @xcite has employed this phase - amplitude description to determine conditions for stability of the synchronised state in a network of identical oscillators with separable linear coupling .
medvedev @xcite has also used the framework to consider the effects of white noise on the synchronous state , identifying the types of linear coupling operators lead to synchrony in a network of oscillators provided that the strength of the interactions is sufficiently strong . since
phase - amplitude coordinates can capture dynamics a finite distance away from the limit cycle , ( and additionally have the advantage over isochronal coordinates of being defined outside of the basin of attraction of the limit cycle ) they can be used to model dynamical phenomena in driven systems where the perturbations necessarily push the dynamics away from the limit cycle .
there is no need to make any assumptions about the strength of the forcing @xmath151 .
the phase - amplitude description of a forced oscillator is able to detect the presence of other structures in the phase space .
for example if the system were multistable , phase - amplitude coordinates would track trajectories near these other structures and back again , should another perturbation return the dynamics to the basin of attraction of the limit cycle .
these coordinates would also detect the presence of other non - attracting invariant structures such as saddles in the unperturbed flow .
orbits passing near the saddle will remain there for some time and forcing may act to move trajectories near to this saddle before returning to the limit cycle
. it may also be the case that the forcing acts to create trapping regions if the forcing is strong compared to the attraction to the limit cycle .
another dynamical phenomenon which phase - amplitude coordinates are able to capture is the occurrence of shear - induced chaos . _ shear _ in a system near a limit cycle @xmath182 is the differential in components of the velocity tangent to the limit cycle as one moves further from @xmath182 in phase space .
shear forces within the system act to speed up ( or slow down ) trajectories further away from the limit cycle compared to those closer to it .
this phenomenon is illustrated in fig .
[ fig : shear ] . is kicked to the blue curve and is then allowed to relax back to cycle , passing through the red and green curves , which are images of the blue curve under the unperturbed flow .
[ fig : shear ] , width=451 ] as we show below , an impulsive force is applied ( the system is kicked ) a ` bump ' in @xmath182 is produced .
if there is sufficient shear in the system then the bump is folded and stretched as it is attracted back to the limit cycle .
such folding can potentially lead to the formation of horseshoes and strange attractors .
however , if the attraction to the limit cycle is large compared to the shear strength or size of the kick then the bumps will dissipate before any significant stretching occurs .
shear - induced chaos is most commonly discussed in the context of discrete time kicking of limit cycles .
wang and young @xcite prove rigorous results in the case of periodically kicked limit cycles with long relaxation times .
their results provide details of the geometric mechanism for producing chaos . here
we briefly review some of these results .
more detailed summaries can be found in @xcite and @xcite .
let @xmath260 be the flow of the unperturbed system , which has an attracting hyperbolic limit cycle @xmath182 .
we can think of a kick as a mapping @xmath261 .
the dynamics of the system with periodic kicking every @xmath262 units of time can be obtained by iterating the map @xmath263 .
provided that there is a neighbourhood @xmath264 of @xmath182 such that points in @xmath264 remain in the basin of attraction of @xmath182 under kicks and @xmath262 is long enough that the kicked points can return to @xmath264 , then @xmath265 is an attractor for the periodically kicked system .
if the kicks are weak then @xmath266 is generally expected to be a slightly perturbed version of @xmath182 and we should expect fairly regular behaviour . in this case @xmath266 is an invariant circle . to obtain more complicated dynamics
it is necessary to break the invariant circle .
this idea is illustrated by the following linear shear model considered in @xcite : @xmath267 where @xmath268 are coordinates in the phase space , @xmath269 are constants and @xmath270 is a nonconstant smooth function .
the unforced system ( @xmath271 ) has a limit cycle @xmath272 .
if the quantity @xmath273 is sufficiently large , then it is possible to show that there is a positive measure set @xmath274 such that for all @xmath275 , @xmath266 is a strange attractor of @xmath276 @xcite .
how large the quantity must be depends on the function @xmath277 .
since @xmath277 is nonconstant ( to create the bumps ) , the larger shear @xmath278 and the kick amplitude @xmath128 , the more folding will occur as the bump is attracted back to the limit cycle . also note that weaker limit cycles ( smaller @xmath279 ) create more favourable conditions for shear induced chaos as the shear has longer to act before the bump is attracted back to the limit cycle . for a more general system with periodic
forcing the shear may not appear explicitly as a parameter . to elucidate what kind of kicks
may cause shear induced chaos in this case we appeal to the isochrons of the system .
we think of the isochrons of @xmath182 as the strong stable manifold of @xmath182 .
suppose , as illustrated in fig .
[ fig : shear ] , that a section @xmath280 of the limit cycle is kicked upwards with the end points held fixed and assume @xmath281 for some @xmath282 .
since the isochrons are flow invariant , during relaxation the flow moves each point of the kicked curve @xmath283 back towards @xmath182 along the isochrons . in fig .
[ fig : shear ] we can clearly see the effect of the shear with a fold forming . from fig .
[ fig : shear ] one can see that kicks along isochrons or in directions roughly parallel to the isochrons will not produce strange attractors , nor will kicks that carry one isochron s manifold to another .
the cause of the stretching and folding is the variation in how far points @xmath284 are moved by @xmath261 in the direction transverse to the isochrons ( i.e. the ordering of points in terms of asymptotic phase is altered by the action of the kick ) .
lin and young @xcite emphasise that the occurrence of shear induced chaos depends on the interplay between the geometries of the kicks and the dynamical structures of the unforced system . in the case of the linear shear model above ,
the isochrons of the unforced system are simply the lines with slope @xmath285 in @xmath286 coordinates .
variation in kick distances in directions transverse to these isochrons is guaranteed with any nonconstant function @xmath287 , with greater variation given by larger values of @xmath288 and @xmath128 . beyond the rigorous results proved by wang and
young @xcite concerning periodically kicked limit cycles of the linear shear model and for supercritical hopf bifurcations , ott and stenlund @xcite prove that shear - induced chaos may exist near general limit cycles .
in addition , lin and young @xcite have carried out numerical studies including random kicks at times given by a poisson process and systems driven by white noise .
they also consider forcing of a pair of coupled oscillators . in all cases ,
shear - induced chaos occurs when the shearing and amplitude of the forcing are large enough to overcome the effects of damping .
lin _ et al_. @xcite demonstrate that the ml model can exhibit shear - induced chaos near the homoclinic bifurcation when periodically forced , by plotting images of the periodic orbit under successive applications of the kick map and calculating the maximum lyapunov exponent .
they also emphasise that the phenomenon of shear - induced chaos can not be detected by the perturbative techniques such as those outlined in
[ subsec : phase ] and [ subsec : phaseresponse ] below .
wedgwood _ et al_. @xcite go on to show that the phase - amplitude description is well - suited to understanding the behaviour of neural oscillators in response to external stimuli .
they consider a number of neural oscillator models in addition to a generic planar model and examine the response of the system to periodic pulsatile forcing by taking @xmath289 , @xmath290 and a variety of choices for @xmath291 in .
their numerical simulations indicate that for both the fhn and ml models the behaviour remains qualitatively similar when the relevant functions @xmath292 are replaced by @xmath293 , @xmath294 is dropped and @xmath246 is replaced with @xmath295 ( for a wide range of @xmath296 ) .
they then focus on the effect of the form of the forcing function in the phase - amplitude coordinates . evaluating the functions @xmath297 and @xmath298 of equations and respectively for the particular model and denoting the first component of @xmath28 as @xmath299 and the first component of @xmath229 as @xmath300 they find that forcing each system at the same ratio of the natural frequency of the underlying periodic orbit and implementing the choices above gives the following ode on @xmath197 , @xmath253 @xmath301 by developing a stroboscopic map for this system , wedgwood _
et al_. @xcite numerically evaluate the largest lyapunov exponent which is found to be positive for the morris - lecar model but negative for the fhn model .
the analysis of the behaviour of generic networks of oscillators within a phase - amplitude framework is a challenging open problem but such a description would allow for greater accuracy ( compared to the phase only methods traditionally used and described below ) in elucidating a richer variety of the complex dynamics of oscillator networks . to obtain a phase description
, one can consider the limit of strong attraction in equations ( [ drivetheta])-([driverho ] ) .
however , it is more appealing to consider a phase variable with a uniform rotation rate that assigns a phase coordinate @xmath197 to each point @xmath284 according to @xmath302 this reduction to a phase description gives a simple dynamical system , albeit one that can not describe evolution of trajectories in phase - space that are away from the limit cycle .
however , the phase reduction formalism is useful in quantifying how a system ( on or close to a cycle ) responds to weak forcing , via the construction of the infinitesimal phase response curve ( iprc ) .
this can be written , for a given ode model , as the solution to an _
adjoint equation_. for a given high dimensional conductance based model this can be solved for numerically , though for some normal form descriptions closed form solutions are also known @xcite .
the iprc at a point on cycle is equal to the gradient of the ( isochronal ) phase at that point . writing this vector quantity as @xmath122 means that weak forcing @xmath303 in the original high dimensional models transforms as @xmath304 where @xmath305 defines the standard inner product and @xmath151 is a small parameter . for periodic forcing such equations can be readily analysed , and questions relating to synchronisation , mode - locking and arnold tongues can be thoroughly explored @xcite .
moreover , this approach forms the basis for constructing models of weakly interacting oscillators , where the external forcing is pictured as a function of the phase of a firing neuron .
this has led to a great deal of work on phase - locking and central pattern generation in neural circuitry and see for example @xcite .
however , the assumption that phase alone is enough to capture the essentials of neural response is one made more for mathematical convenience than being physiologically motivated .
indeed for the popular type i ml firing model with standard parameters , direct numerical simulations with pulsatile forcing show responses that can not be explained solely with a phase model @xcite , as just highlighted in [ sec : shear ] ( since strong interactions will necessarily take one away from the neighbourhood of a cycle where a phase description is expected to hold ) .
it is common practice in neuroscience to characterise a neuronal oscillator in terms of its phase response to a perturbation @xcite .
this gives rise to the notion of a so - called phase response curve ( prc ) , which for a real neuron can be determined experimentally @xcite , and can also be related to the poststimulus time histogram @xcite .
consider a dynamical system @xmath173 , @xmath306 with a @xmath94-periodic solution @xmath307 and introduce an infinitesimal perturbation @xmath308 to the trajectory @xmath207 at time @xmath309 .
this perturbation evolves according to the linearised equation of motion:@xmath310 here @xmath311 denotes the jacobian of @xmath121 evaluated along @xmath242 . introducing a time - independent _ isochronal _ phase shift @xmath312 as @xmath313
, we have to first order in @xmath314 that@xmath315 where @xmath305 defines the standard inner product , and @xmath316 is the gradient of @xmath247 evaluated at @xmath207 . taking the time - derivative of ( [ prc1 ] )
gives @xmath317 since the above equation must hold for arbitrary perturbations , we see that the gradient @xmath316 satisfies the linear equation@xmath318 subject to the boundary conditions @xmath319 the first condition simply guarantees that @xmath320 ( at any point on the periodic orbit ) , and the second enforces periodicity .
note that the notion of phase response can also be extended to time - delayed systems @xcite . in general equation ( [ prc2 ] ) must be solved numerically to obtain the iprc , say , using the _ adjoint _ routine in xppaut @xcite or matcont @xcite . however , for the case of a nonlinear if model , defined by ( [ eq : one ] ) , the prc is given explicitly by @xmath321 where the period of oscillation is found by solving @xmath322 ( and the response function is valid for finite perturbations ) .
for example , for the quadratic if model , obtained from ( [ eq : one ] ) with the choice with @xmath323 , and taking the limit @xmath324 and @xmath325 we find @xmath326 with @xmath327 , recovering the iprc expected of an oscillator close to a snic bifurcation @xcite . for a comprehensive discussion of iprcs for various common oscillator models
see the excellent survey article by brown _
et al_. @xcite .
the iprc for planar piecewise linear if models can also be computed explicitly @xcite , although we do not discuss this further here . in fig .
[ fig : prc ] we show the numerically computed iprcs for the hodgkin - huxley model , the wilson cortical model and the morris - lecar model previously discussed in [ sec : neuronoscillators ] . .
orbits are in blue and iprcs in red .
[ fig : prc ] , width=432 ] for completeness it is well to note the iprc for a planar oscillator close to a supercritical hopf bifurcation has an adjoint with a shape given by @xmath328 for an orbit shape @xmath329 . having defined the phase in some neighbourhood of the limit cycle , we can write ( [ phasedot ] ) as @xmath330 for @xmath173 this gives @xmath331 .
we now consider the effect of a small external periodic force on the self sustained oscillations as in ( [ eq : driven ] ) , with @xmath332 , where in general @xmath333 is different to @xmath94 .
for weak perturbations ( @xmath334 ) the state point will leave @xmath182 , though will stay in some neighbourhood , which we denote by @xmath264 .
we extend the phase off cycle using isochronal coordinates so that @xmath331 holds for any point @xmath335 . for a perturbed oscillator , in these coordinates we have @xmath336 as a first approximation we evaluate the right hand side of the above on the limit - cycle to get an equation for the phase dynamics : @xmath337
the phase dynamics ( [ phase dynamics ] ) is still very hard to analyse , and as such it is common to resort to a further simplification , namely _
averaging_. first let us introduce a rotating phase @xmath338 , in which case ( [ phase dynamics ] ) becomes @xmath339 if both @xmath151 and @xmath118 are small then @xmath340 and @xmath341 evolves _
slowly_. thus we may set @xmath342 for @xmath343 $ ] .
averaging the above over one period gives @xmath344 where we have used the result that @xmath345 .
the function @xmath346 is @xmath94-periodic and can be written as a fourier series @xmath347 , with the simplest example of an averaged phase - dynamics being @xmath348 which is called the adler equation @xcite .
if we denote the maximum and minimum of @xmath287 by @xmath349 and @xmath350 respectively then for a _ phase - locked _ @xmath15:@xmath15 state defined by @xmath351
we require @xmath352 . in this case
there are two fixed points @xmath353 defined by @xmath354 .
one of these is unstable ( say @xmath355 , so that @xmath356 ) and the other stable ( @xmath357 , with @xmath358 ) .
this gives rise to a rotating solution with constant rotation frequency so that @xmath359 .
the two solutions coalesce in a saddle - node bifurcation when @xmath360 and @xmath361 ( or equivalently when @xmath362 ) . in the case of the adler model the parameter region for phase - locking
is given explicitly by a triangular wedge defined by @xmath363 a so - called arnold tongue .
outside of this tongue solutions _ drift _ ( they can not lock to the forcing period ) according to ( [ averaged ] ) , and the system evolves quasi - periodically .
we treat weakly coupled phase oscillators in [ sec : weakcoupling ] .
the theory of weakly coupled oscillators @xcite is now a standard tool of dynamical systems theory and has been used by many authors to study oscillatory neural networks , see for example @xcite .
the book by hoppensteadt and izhikevich provides a very comprehensive review of this framework @xcite , which can also be adapted to study networks of relaxation oscillators ( in some singular limit ) @xcite .
consider , for illustration , a system of interacting limit - cycle oscillators ( [ eq : pairwisecoupledcells ] ) . following the method in
[ subsec : phaseresponse ] , similar to ( [ averaged ] ) we obtain the network s phase dynamics in the form @xmath364 where the frequency @xmath365 allows for the fact that oscillators are not identical and , for this reason , we will assume that @xmath366 .
precisely this form of network model was originally suggested by winfree to describe populations of coupled oscillators .
the winfree model @xcite assumes a separation of time - scales so that an oscillator can be solely characterised by its phase on cycle ( fast attraction to cycle ) and is described by the network equations @xmath367 describing a globally coupled network with a biologically realistic prc @xmath368 and pulsatile interaction function @xmath369 . using a mixture of analysis and numerics winfree
found that with large @xmath140 there was a transition to macroscopic synchrony at a critical value of the _ homogeneity _ of the population .
following this kuramoto @xcite introduced a simpler model with interactions mediated by phase differences , and showed how the transition to collective synchronisation could be understood from a more mathematical perspective .
for an excellent review of the kuramoto model see @xcite and @xcite . the natural way to obtain a phase - difference model from ( [ phase dynamicsnetwork ] ) is , as in [ subsec : phaseresponse ] , to average over one period of oscillation . for simplicity
let us assume that all the oscillators are identical , and @xmath370 @xmath371 , in which case we find that @xmath372 where @xmath373 the @xmath202-periodic function @xmath287 is referred to as the _ phase interaction function_. if we write complex fourier series for @xmath122 and @xmath374 as @xmath375 respectively then @xmath376 with @xmath377 .
note that certain caution has to be exercised in applying averaging theory .
in general , one can only establish that a solution of the unaveraged equations is @xmath151-close to a corresponding solution of the averaged system for times of @xmath378 .
no such problem arises in the case of hyperbolic fixed points corresponding to phase - locked states . when describing a piece of cortex or a central pattern generator circuit with a set of oscillators , the biological realism of the model typically resides in the phase interaction function .
the simplest example is @xmath379 , which when combined with a choice of global coupling defines the well known kuramoto model @xcite .
however , to model realistic neural networks one should calculate ( [ h ] ) directly , using knowledge of the single neuron iprc and the form of interaction . as an example consider synaptic coupling , described in
[ sec : coupling ] , that can be written in the form @xmath380 , and a single neuron model for which the iprc in the voltage variable @xmath39 is given by @xmath368 ( say experimentally or from numerical investigation ) . in this case @xmath381
if instead we were interested in diffusive ( gap - junction ) coupling then we would have @xmath382 { { \rm d}}s .
\label{hgap } \nonumber\ ] ] for the hh model @xmath383 is known to have a shape like @xmath384 for a spike centred on the origin ( see fig . [
fig : prc ] ) . making the further choice that @xmath385 then ( [ hsynapse ] ) can be evaluated as @xmath386 \sin ( \psi)-2/\alpha \cos(\psi)}{2 \pi [ 1+(1/\alpha)^2]^2}. \label{halphahh}\ ] ] in the particular case of two oscillators with reciprocal coupling and @xmath224 then @xmath387 and we define @xmath388 .
a phase - locked solution satisfies constant phase difference @xmath389 that is a zero of the ( odd ) function @xmath390.\ ] ] a given phase - locked state is then stable provided that @xmath391 . note that by symmetry both the in - phase ( @xmath392 ) and anti - phase ( @xmath393 ) states are guaranteed to exist . for the form of phase interaction function given by @xmath394 , the stability of the synchronous solution is governed by the sign of @xmath395 : @xmath396 \right \ } .
\nonumber\ ] ] thus for inhibitory coupling ( @xmath397 ) synchronisation will occur if @xmath398 , namely when the synapse is _ slow _ ( @xmath399 ) .
it is also a simple matter to show that the anti - synchronous solution ( @xmath400 ) is stable for a sufficiently _ fast _ synapse ( @xmath401 ) .
it is also possible to develop a general theory for the existence and stability of phase - locked states in larger networks than just a pair .
now suppose we have a general population of @xmath402 coupled phase oscillators @xmath403 described by phases @xmath404 . for a particular continuous choice of phases @xmath405 for the trajectory one can define the _ frequency _ of the @xmath115th oscillator as @xmath406.\ ] ] this limit will converge under fairly weak assumptions on the dynamics @xcite , though it may vary for different attractors in the same system , for different oscillators @xmath115 and in some cases it may vary even for different trajectories within the same attractor .
we say two oscillators @xmath115 and @xmath251 are _ phase locked _ with ratio @xmath407 for @xmath408 with no common factors of @xmath29 and @xmath26 , if there is an @xmath409 such that @xmath410 for all @xmath411 .
the oscillators are _ frequency locked _ with ratio @xmath407 if @xmath412 if we say they are simply phase ( or frequency locked ) without explicit mention of the @xmath407 ratio , we are using the convention that they are @xmath413 phase ( or frequency ) locked .
the definition of @xmath414 means that if two oscillators are phase locked then they are frequency locked .
the converse is not necessarily the case : two oscillators may be frequency locked but not phase locked if the phase difference @xmath415 grows sublinearly with @xmath169 . for the special case globally coupled averaged networks ( @xmath416 for the system ( [ phasenetwork ] ) ) is @xmath417 equivariant . by topological arguments ,
maximally symmetric solutions describing synchronous , splay , and a variety of cluster states are expected to exist generically @xcite .
the system ( [ phasenetwork ] ) with global coupling is in itself an interesting subject of study in that it is of arbitrarily high dimension @xmath140 but is effectively determined by the single function @xmath418 that is computable from a single pair of oscillators .
the system ( and variants thereof ) have been productively studied by thousands of papers since the seminal work of kuramoto @xcite .
the collective dynamics of phase oscillators have been investigated for a range of regular network structures including linear arrays and rings with uni- or bi - directional coupling e.g. @xcite , and hierarchical networks @xcite . in some cases
the systems can be usefully investigated in terms of permutation symmetries of ( [ phasenetwork ] ) with global coupling , for example @xmath419 or @xmath420 for uni- or bi - directionally coupled rings .
in other cases a variety of approaches have been developed adapted to particular structures though these have not in all cases been specifically applied to oscillators ; some of these approaches are discussed in
[ sec : networks ] we recall that the form of the coupling in ( [ phasenetwork ] ) is special in the sense that it assumes the interactions between two oscillators are independent of any third - pairwise coupling @xcite .
if there are degeneracies such as @xmath421 which can appear when some of the fourier components of @xmath287 are zero , this can lead to degeneracies in the dynamics .
for example @xcite , while ( * ? ?
* theorem 7.1 ) shows that if @xmath287 satisfies ( [ eq : degenerate ] ) for some @xmath422 and @xmath140 is a multiple of @xmath26 then ( [ phasenetwork ] ) , with global coupling , will have @xmath26-dimensional invariant tori in phase space that are foliated by neutrally stable periodic orbits .
this degeneracy will disappear on including either non - pairwise coupling or introducing small but non - zero fourier components in the expansion of @xmath287 but as noted in @xcite this will typically be the case for the interaction of oscillators even if they are near a hopf bifurcation .
we examine in more detail some of the phase locked states that can arise in weakly coupled networks of identical phase oscillators described by ( [ phasenetwork ] ) .
we define a @xmath15:@xmath15 phase - locked solution to be of the form @xmath423 , where @xmath424 is a constant phase and @xmath425 is the collective frequency of the coupled oscillators .
substitution into the averaged system ( [ phasenetwork ] ) gives @xmath426 after choosing some reference oscillator , these @xmath140 equations determine the collective frequency @xmath425 and @xmath427 relative phases with the latter independent of @xmath151 .
it is interesting to compare the weak coupling theory for phase - locked states with the analysis of lif networks from
[ sec : ifnetworks ] .
equation ( [ ifphases ] ) has an identical structure to that of equation ( [ omega ] ) ( for @xmath428 for all @xmath114 ) , so that the classification of solutions using group theoretic methods is the same in both situations .
there are , however , a number of significant differences between phase - locking equations ( [ omega ] ) and ( [ ifphases ] ) .
first , equation ( [ ifphases ] ) is exact , whereas equation ( [ omega ] ) is valid only to @xmath429 since it is derived under the assumption of weak coupling .
second , the collective period of oscillations @xmath333 must be determined self - consistently in equation ( [ ifphases ] ) . in order to analyse the local stability of a phase - locked solution @xmath430
, we linearise the system by setting @xmath431 and expand to first - order in @xmath432 : @xmath433 where @xmath434 and @xmath435 .
one of the eigenvalues of the jacobian @xmath436 is always zero , and the corresponding eigenvector points in the direction of the flow , that is @xmath437 .
the phase - locked solution will be stable provided that all other eigenvalues have a negative real part .
we note that the jacobian has the form of a graph - laplacian mixing both anatomy and dynamics , namely it is the graph - laplacian of the matrix with components @xmath438 .
synchrony ( more precisely , exact phase synchrony ) is where @xmath439 for some fixed frequency @xmath425 is a classic example of a phase - locked state .
substitution into ( [ phasenetwork ] ) , describing a network of identical oscillators , shows that @xmath425 must satisfy the condition @xmath440 one way for this to be true for all @xmath114 is if @xmath441 , which is the case say for @xmath442 or for diffusive coupling , which is linear in the difference between two state variables so that @xmath441
. the existence of synchronous solutions is also guaranteed if @xmath443 is independent of @xmath114 .
this would be the case for global coupling where @xmath416 , so that the system has permutation symmetry .
if the synchronous solution exists then the jacobian is given by @xmath444 where @xmath445 is the graph - laplacian with components @xmath446 . .
[ cocomac],width=226 ] we note that @xmath445 has one zero eigenvalue , with eigenvector @xmath447 . hence if all the other eigenvalues of @xmath445 lie on one side of the imaginary axis then stability is solely determined by the sign of @xmath448 . in fig .
[ cocomac ] we show the eigenvalues of the graph - laplacian of the anatomical network structure of the macaque monkey brain , as determined from the cocomac database @xcite . here
all the eigenvalues lie to the left of the imaginary axis so that the stability of the synchronous solution ( should it exist ) is solely determined by the sign of @xmath448 . for global coupling
we have that @xmath449 , and the ( @xmath427 degenerate ) eigenvalue is @xmath450 .
hence the synchronous solution will be stable provided @xmath451 .
another example of a phase - locked state is the purely asynchronous solution whereby all phases are uniformly distributed around the unit circle .
this is sometimes referred to as a _ splay state _ or _ splay - phase state _ and can be written @xmath452 with @xmath453 . like the synchronous solution it will be present but not necessarily stable in networks with global coupling , with an emergent frequency that depends on @xmath287 : @xmath454 in this case the jacobian takes the form @xmath455 , \nonumber\ ] ] where @xmath456 and @xmath457 .
hence the eigenvalues are given by @xmath458/n$ ] , @xmath459 where @xmath460 are the eigenvalues of @xmath461 : @xmath462 , where @xmath463 denote the components of the @xmath464th eigenvector .
this has solutions of the form @xmath465 so that @xmath466 , giving @xmath467 , \nonumber\ ] ] and the splay state will be stable if @xmath468 . in the large @xmath140 limit @xmath469
we have the useful result that ( for global coupling ) network averages may be replaced by time averages : @xmath470 for some @xmath202-periodic function @xmath471 ( which can be established using a simple riemann sum argument ) , with a fourier series @xmath472 . hence in the large @xmath140 limit the collective frequency of a splay state ( global coupling ) is given by @xmath473 , with eigenvalues@xmath474 hence a splay state is stable if @xmath475 , where we have used the fact that since @xmath476 is real , then @xmath477 . as an example consider the case @xmath478 for @xmath200 and @xmath479 ( where @xmath287 is periodically extended outside @xmath480 ) .
it is straight forward to calculate the fourier coefficients ( [ eq : hfourier ] ) as @xmath481 , so that @xmath482 .
hence the asynchronous state is stable . if we flip any one of the coefficients @xmath483 then @xmath484 and the splay state will develop an instability to an eigenmode that will initially destabilise the system in favour of an @xmath26-cluster . for @xmath200 with complex fourier series coefficients given by @xmath481 .
the remaining panels show the effect of flipping the sign of just one of the co - efficients , namely @xmath485 .
( b ) @xmath486 , and the asynchronous solution will destabilise in favour of the synchronous solution since @xmath487 . ( c ) @xmath488 , and
the asynchronous solution will destabilise in favour of a @xmath489-cluster .
( d ) @xmath490 , and the asynchronous solution will destabilise in favour of a @xmath491-cluster .
note that only small changes in the shape of @xmath287 , as seen in panels ( c)-(d ) , can lead to a large change in the emergent network dynamics .
[ oddh],width=377 ] for reviews of the stability of cluster states ( in which subsets of the oscillator population synchronise , with oscillators belonging to different clusters behaving differently ) we refer the reader to @xcite ; here we use the notation of @xcite . if a group of @xmath140 oscillators is neither fully synchronised nor desynchronised it may be clustered .
we say @xmath492 is an @xmath409-cluster partition of @xmath493 if @xmath494 where @xmath495 are pairwise disjoint sets ( @xmath496 if @xmath497 ) .
nb if @xmath498 then @xmath499 one can refer to this as a cluster of type @xmath500 .
it is possible to show that any clustering can be realised as a stable periodic orbit of the globally coupled phase oscillator system @xcite for suitable choice of phase interaction function ; more precisely , there is a coupling function @xmath287 for the system ( [ phasenetwork ] ) , with global coupling , such that for any @xmath140 and any given @xmath409-cluster partition @xmath501 of @xmath502 there is a linearly stable periodic orbit realising that partition ( and all permutations of it ) .
see also @xcite where the authors consider clustering in this system where @xmath503 .
more generally , there are very many invariant subspaces corresponding to spatio - temporal clustering that we can characterise in the following theorem . *
( * ? ? ? * theorem 3.1 ) * [ thm : asiso ] the subsets of @xmath504 that are invariant for ( [ phasenetwork ] ) , with global coupling , because of symmetries of @xmath505 correspond to isotropy subgroups in the conjugacy class of @xmath506 where @xmath507 , @xmath508 and @xmath509 denotes the semidirect product .
the points with this isotropy have @xmath510 clusters that are split into @xmath511 groups of @xmath26 clusters of the size @xmath512 .
the clusters within these groups are cyclically permuted by a phase shift of @xmath513 .
the number of isotropy subgroups in this conjugacy class is @xmath514 $ ] .
it is a nontrivial problem to discover which of these subspaces contain periodic solutions .
note that in - phase corresponds to @xmath515 , @xmath516 while splay phase corresponds to @xmath517 , @xmath518 .
the stability of several classes of these solutions can be computed in terms of properties of @xmath418 ; see for example [ sec : synchrony ] and [ sec : asynchrony ] and for other classes of solution @xcite .
bifurcation properties of the globally coupled oscillator system ( [ phasenetwork ] ) on a varying parameter that affects the coupling @xmath418 are surprisingly complicated because of the symmetries present in the system ; see [ subsec : symmetry ] .
in particular , the high multiplicity of the eigenvalues for loss of stability of the synchronous solution means : * path following numerical bifurcation programmes such as auto or xppaut need to be done with great care when applying to problems with @xmath519 identical oscillators - these typically will not be able to find all solutions branching from one that loses stability . * a large number of branches with a range of symmetries
may generically be involved in the bifurcation ; indeed , all 2-cluster states @xmath520 .
* local bifurcations may have global bifurcation consequences owing to the presence of connections that are facilitated by the nontrivial topology of the torus @xcite .
* branches of degenerate attractors such as heteroclinic attractors may appear at such bifurcations for @xmath521 oscillators .
et al_. @xcite consider the system ( [ phasenetwork ] ) with global coupling and phase interaction function of the form @xmath522 for @xmath523 fixed parameters ; detailed bifurcation scenarios in the cases @xmath524 are shown in @xcite . as an example , figure [ fig : fouroscbifs ] shows regions on stability of synchrony , splay phase solutions and robust heteroclinic attractors as discussed later in [ sec : heteroclinic ] .
-oscillator system ( [ phasenetwork ] ) with phase interaction function ( [ eq : hmmcoupling ] ) and parameters in the region @xmath525
, @xmath526 @xcite .
the narrow stripes show the region of stability of synchrony , while the wide stripes show the region of stability of the splay phase solution .
the pink shaded area shows a region of existence of a robust heteroclinic network that is an attractor with in the checkerboard region ; the boundaries are described in @xcite .
[ fig : fouroscbifs ] , width=384 ] the phase reduction method has been applied to a number of important biological systems , including the study of travelling waves in chains of weakly coupled oscillators that model processes such as the generation and control of rhythmic activity in central pattern generators ( cpgs ) underlying locomotion @xcite and peristalsis in vascular and intestinal smooth muscle @xcite .
related phase models have been motivated by the observation that synchronisation and waves of excitation can occur during sensory processing in the cortex @xcite . in the former case
the focus has been on dynamics on a lattice and in the latter continuum models have been preferred .
we now present examples of both these types of model , focusing on _ phase wave _ solutions @xcite .
the lamprey is an eel - like vertebrate which swims by generating travelling waves of neural activity that pass down its spinal cord .
the spinal cord contains about @xmath110 segments , each of which is a simple half - center neural circuit capable of generating alternating contraction and relaxation of the body muscles on either side of body during swimming . in a seminal series of papers , ermentrout and kopell carried out a detailed study of the dynamics of a chain of weakly coupled limit cycle oscillators @xcite , motivated by the known physiology of the lamprey spinal cord .
they considered @xmath140 phase - oscillators arranged on a chain with nearest - neighbour anisotropic interactions and identified a travelling wave as a phase - locked state with a constant phase difference between adjacent segments .
the intersegmental phase differences are defined as @xmath527 .
if @xmath528 then the wave travels from head to tail whilst for @xmath529 the wave travels from the tail to the head .
phase oscillators @xmath530 with @xmath531 . ,
width=432 ] for a chain we set @xmath532 to obtain @xmath533 where @xmath534 . pairwise subtraction and substitution of @xmath527 leads to an @xmath427 dimensional system for the phase differences @xmath535 + w_- [ h ( -\varphi_i)-h ( -\varphi_{i-1 } ) ] , \nonumber\ ] ] for @xmath536 , with boundary conditions @xmath537 , where @xmath538 .
there are at least two different mechanisms that can generate travelling wave solutions .
the first is based on the presence of a gradient of frequencies along the chain , that is , @xmath539 has the same sign for all @xmath114 , with the wave propagating from the high frequency region to the low frequency region .
this can be established explicitly in the case of an isotropic , odd interaction function , @xmath540 and @xmath541 , where we have @xmath542 the fixed points @xmath543 satisfy the matrix equation @xmath544 , where @xmath545 , @xmath546 , and @xmath547 is a tridiagonal matrix with elements @xmath548 , @xmath549 .
for the sake of illustration suppose that @xmath550 . then a solution @xmath551 will exist if every component of @xmath552 lies between @xmath553 .
let @xmath554 . if @xmath555 then for each @xmath556 there are two distinct solutions @xmath557 in the interval @xmath558 $ ] with @xmath559 and @xmath560 . in other words , there are @xmath561 phase - locked solutions . linearising about a phase - locked solution and exploiting the structure of the matrix @xmath547 , it can be proven that only the solution @xmath562 is stable .
assuming that the frequency gradient is monotonic , this solution corresponds to a stable travelling wave . when the gradient becomes too steep to allow phase - locking ( i.e. @xmath563 )
, two or more clusters of oscillators ( frequency plateaus ) tend to form that oscillate at different frequencies .
waves produced by a frequency gradient do not have a constant speed or , equivalently , constant phase lags along the chain .
constant speed waves in a chain of identical oscillators can be generated by considering phase - locked solutions defined by @xmath564 for all @xmath114 , with a collective period of oscillation @xmath425 determined using @xmath565 to give @xmath566 .
the steady state equations are then @xmath567 , @xmath568 and @xmath569 , for @xmath570 . thus , a travelling wave solution requires that all frequencies must be the same except at the ends of the chain .
one travelling solution is given by @xmath571 with @xmath572 and @xmath573 .
for the choice @xmath574 we have that @xmath575 and @xmath576 .
if @xmath577 then @xmath578 and @xmath579 must be larger than @xmath580 and hence all the remaining @xmath365 for a forward travelling wave to exist .
backward swimming can be generated by setting @xmath581 and solving in a similar fashion .
there is solid experimental evidence for electrical waves in awake and aroused vertebrate preparations , as well as semi - intact and active invertebrate preparations , as nicely described by ermentrout and kleinfeld @xcite .
moreover , these authors argue convincingly for the use of phase models in understanding waves seen in cortex and speculate that they may serve to label simultaneously perceived features in the stimulus stream with a unique phase .
more recently it has been found that cortical oscillations can propagate as travelling waves across the surface of the motor cortex of monkeys ( macaca mulatta ) @xcite .
given that to a first approximation the cortex is often viewed as being built from a dense reciprocally interconnected network of corticocortical axonal pathways , of which there are typically @xmath582 in a human brain it is natural to develop a _ continuum _ phase model , along the lines described by crook _
et al_. @xcite .
these authors further incorporated axonal delays into their model to explain the seemingly contradictory result that synchrony is stable for short range excitatory coupling , but unstable for long range . to see how a delay induced
instability may arise we consider a continuum model of identical phase oscillators with space - dependent delays @xmath583 where @xmath584 and @xmath585 .
this model is naturally obtained as the continuum limit of ( [ phasenetwork ] ) for a network arranged along a line with communication delays @xmath586 set by the speed of an action potential @xmath39 that mediates the long range interaction over a distance between points in the tissue at @xmath213 and @xmath587 .
here we have used the result that for weak coupling then delays manifest themselves as phase - shifts .
the function @xmath588 sets the anatomical connectivity pattern .
it is convenient to assume that interactions are homogeneous and translation invariant , so that @xmath589 with @xmath590 , and either assume periodic boundary conditions or take @xmath591 . following @xcite we construct travelling wave solutions of equation ( [ phasecontinuum ] ) for @xmath591 of the form @xmath592 , with the frequency @xmath425 satisfying the dispersion relation @xmath593 when @xmath594 , the solution is synchronous . to explore the stability of the travelling wave we linearise ( [ phasecontinuum ] ) about @xmath595 and
consider perturbations of the form @xmath596 , to find that travelling phase waves solutions are stable if @xmath597 for all @xmath598 , where @xmath599 { { \rm d}}y .
\nonumber\ ] ] note that the neutrally stable mode @xmath600 represents constant phase - shifts .
for example , for the case that @xmath601 then we have that @xmath602 , \nonumber\ ] ] where @xmath603 and @xmath604 . a plot of the region of wave stability for the choice @xmath605 and @xmath479 in the @xmath606 plane is shown in fig .
[ fig : phasewave ] .
note that the synchronous solution @xmath594 is unstable for small values of @xmath39 . in the @xmath606 plane for the choice @xmath607 , @xmath605 and @xmath479 .
[ fig : phasewave ] , width=240 ] for a discussion of more realistic forms of phase interaction , describing synaptic interactions , see @xcite . for
appropriate choice of the anatomical kernel and the phase interaction function , continuum models of the form ( [ phasecontinuum ] ) can also support weak turbulent solutions reminiscent of those seen in the kuramoto - sivashkinsky ( ks ) equation .
the ks equation generically describes the dynamics near long wavelength primary instabilities in the presence of appropriate ( translational , parity and galilean ) symmetries , and is of the form @xmath608 where @xmath609 . for a further discussion of this model see @xcite . for the model ( [ phasecontinuum ] ) with decaying excitatory coupling excitation and purely sinusoidal phase coupling , simulations on a large domain show a marked tendency to generate phase - slips and spatiotemporal pattern shedding , resulting in a loss of spatial continuity of @xmath610 .
however , battogtokh @xcite has shown that a mixture of excitation and inhibition with higher harmonics in the phase interaction can counteract this tendency and allow the formation of turbulent states . to see how this can arise consider an extension of ( [ phasecontinuum ] ) to allow for a mixing of spatial scales and nonlinearities in the form @xmath611 where we drop the consideration of axonal delays . using the analysis of [ sec : phasewave ] the synchronous wave solution will be stable provided @xmath612 for all @xmath598 where @xmath613 , \qquad \widehat{w}_\mu(p ) = \int_{\rset } w_\mu(|y| ) { { \rm e}}^{i p y } { { \rm d}}y .
\nonumber\ ] ] after introducing the complex function @xmath614 and writing the phase interaction functions as fourier series of the form @xmath615 then we can re - write ( [ sumcontinuum ] ) as @xmath616 where @xmath617 the form above is useful for computational purposes , since @xmath618 can easily be computed using a fast fourier transform ( exploiting its convolution structure ) .
battogtokh @xcite considered the choice @xmath619 with @xmath620 , @xmath621 and @xmath622 with @xmath623 . in this case @xmath624 , so that @xmath625 by choosing a mixture of positive and negative coupling strengths the spectrum can easily show a band of unstable wave - numbers from zero up to some maximum as shown in fig .
[ fig : battogtokhspectrum ] . ) with a mixture of spatial scales and nonlinearities . here
@xmath620 , @xmath621 and @xmath622 with @xmath623 .
parameters are @xmath626 , @xmath627 , @xmath628 , @xmath629 , @xmath630 , and @xmath631 .
there is a band of unstable wave - numbers with @xmath632 , with @xmath633 .
[ fig : battogtokhspectrum ] , width=240 ] indeed this shape of spectrum is guaranteed when @xmath634 and @xmath635 .
similarly the ks equation ( [ ks ] ) has a band of unstable wave - numbers between zero and one ( with the most unstable wave - number at @xmath636 ) . for the case that all the spatial scales @xmath637 are small compared to the system size then we may develop a _ long wavelength _
argument to develop local models for @xmath618 .
to explain this we first construct the fourier transform @xmath638 , where @xmath639 is the fourier transform of @xmath640 and use the expansion @xmath641 .
after inverse fourier transforming we find @xmath642 z^n .
\nonumber\ ] ] noting that @xmath643 with all other fourier coefficients zero then ( [ zt ] ) yields @xmath644 where @xmath645 . expanding ( [ thetat ] ) to second order gives @xmath646 , where @xmath647 and @xmath648 . going to higher order yields fourth order terms and we recover an equation of ks type with the coefficient of @xmath649 given by @xmath650 . to generate phase turbulence we thus require @xmath651 , which is also a condition required to generate a band of unstable wave - numbers , and @xmath652 . a direct simulation of the model with the parameters for fig . [ fig : battogtokhspectrum ] , for which @xmath609 , shows the development of a phase turbulent state .
this is represented in fig .
[ fig : battogtokh ] where we plot the absolute value of the complex function @xmath653 .
with @xmath654 for which @xmath655 , @xmath656 and @xmath657 .
the physical domain size is @xmath658 and we have used a numerical mesh of @xmath659 points with matlab ode45 to evolve equation ( [ zt ] ) with convolutions computed using fast fourier transforms . as an order parameter describing the system we have chosen @xmath660 , where @xmath653 .
[ fig : battogtokh ] , width=384 ]
in addition to dynamically simple periodic attractors with varying degrees of clustering , the emergent dynamics of coupled systems such as ( [ phasenetwork ] ) can be remarkably complex even in the case of global coupling . in this case
, the dynamical complexity depends only on the phase interaction function @xmath287 and the number of oscillators @xmath140 .
chaotic dynamics @xcite can appear in four or more globally coupled phase oscillators for phase interaction functions of sufficient complexity .
we focus now on attractors that are robust and apparently prevalent in many such systems : robust heteroclinic attractors . in a neuroscience context
such attractors have been investigated under several related names , including _
slow switching _
@xcite where the system evolves towards an attractor that displays slow switching between cluster states where the switching is on a timescale determined by the noise , heteroclinic networks @xcite or _ winnerless competition _
@xcite where there are a number of possible patterns of activity that compete with each other but such that each pattern is unstable to some perturbations that take it to another pattern - this can be contrasted to _ winner - takes - all competition _ where there is attraction to a asymptotically stable pattern .
these attractors obtain their dynamical structure from the presence of invariant subspaces for the dynamics that allow the possibility of robust connections between saddles .
these subspaces may be symmetry - induced fixed point subspaces , spaces with a given clustering , subspaces forced by multiplicative coupling or subspaces forced by assumptions on the nature of the coupling . in all cases , there will be a number of dynamically simple nodes , usually equilibria or periodic orbits , say @xmath144 , @xmath661 each of which has saddle type .
these nodes have unstable manifolds that , within some invariant subspace , limit to other nodes within the network - usually because there is a robust ( transverse ) saddle - to - sink connection between the nodes within some invariant subspace ; see @xcite .
it can be verified that such heteroclinic networks can be attracting if , in some sense , the rate of expansion away from the nodes is not as strong as the rate of contraction towards the nodes - see @xcite for some precise results in this direction .
figure [ fig : shc ] illustrates some of the ingredients for a robust heteroclinic attractor .
( solid line ) approaching part of a robust heteroclinic network ( bold dashed lines ) .
the nodes @xmath144 represent equilibria or periodic orbits of saddle type and the invariant subspaces @xmath662 are forced to exist by model assumptions and there are connecting ( heteroclinic ) orbits @xmath663 that connect the nodes within the @xmath662 in a robust manner .
a neighbourhood of the connecting orbits @xmath663 is an absorbing _ stable heteroclinic channel _ that can be used to describe various aspects of neural system function in systems with this dynamics ; see for example @xcite .
[ fig : shc ] , width=336 ] hansel _
et al_. @xcite considered the dynamics of ( [ phasenetwork ] ) with global coupling and phase interaction function of the form ( [ eq : hmmcoupling ] ) for @xmath523 fixed parameters . for large @xmath140 , they find an open region in parameter space where typical attractors are heteroclinic cycles that show slow switching between states where the clustering is into two clusters of macroscopic size .
this dynamics is examined in more depth in @xcite where the simulations for typical initial conditions show a long and intermittent transient to a two - cluster state that , surprisingly , is unstable .
this is a paradox because only low dimensional subset of initial conditions ( the stable manifold ) should converge to a saddle .
the resolution of this paradox is a numerical effect : as the dynamics approaches the heteroclinic cycle where the connection is in a clustered subspace , there can be numerical rounding into the subspace . for weak perturbation of the system by additive noise , they find that the heteroclinic cycle is approximated by a roughly periodic transition around the cycle whose approximate period scales as the logarithm of the noise amplitude .
the bifurcations that give rise to heteroclinic attractors in this system on varying @xmath664 is quite complex even for small @xmath140 .
as discussed in @xcite one can only find attracting robust heteroclinic attractors in ( [ phasenetwork ] ) , ( [ eq : hmmcoupling ] ) for @xmath521 : in this case fig .
[ fig : fouroscbifs ] shows a region where robust heteroclinic attractors of the type illustrated in fig .
[ fig : fouroschet ] appear .
a trajectory approaching such a network will spend much of its time near a cluster state with two groups of two oscillators each . each time there is a connection between the states , one of the groups will break clustering for a short time , and over a longer period the clusters will alternate between breaking up and keeping together . qualitatively similar heteroclinic attractors can be found for example in coupled hodgkin - huxley type limit cycle oscillators with delayed synaptic coupling as detailed in @xcite and illustrated in fig . [ fig:22cycle ] .
that only equal one when the systems @xmath114 and @xmath115 are synchronised while the bottom panel shows the action potentials @xmath112 for the four oscillators ; see @xcite for more details .
observe that the mechanism of symmetry breaking and re - synchronisation of pairs is the same as in fig .
[ fig : fouroschet ] .
[ fig:22cycle ] , width=432 ] the heteroclinic attractors that appear for @xmath665 can have more complex structures . for @xmath666
this is investigated in @xcite for ( [ phasenetwork ] ) , ( [ eq : hmmcoupling ] ) and in @xcite for more general phase interaction function @xmath667 where @xmath103 , @xmath21 and @xmath668 are parameters .
figure [ fig : fiveoschetraster ] illustrates a trajectory of ( [ phasenetwork ] ) with global coupling and phase interaction function ( [ eq : ghmmcoupling ] ) as a raster plot for five phase oscillators with parameters @xmath669 , @xmath670 , @xmath671 and @xmath672 and with addition of noise of amplitude @xmath673 .
observe there is a dynamic change of clustering over the course of the time - series with a number of switches taking place between cluster states of the type @xmath674 where @xmath587 , @xmath117 and @xmath675 represent different relative phases that are coloured `` yellow '' , `` green '' or `` blue '' in fig .
[ fig : fiveoschetnetwork ] to other symmetrically related cluster states .
one can check that the group orbit of states with this clustering gives 30 symmetrically related cluster states for the system ; details of the connections are shown in fig .
[ fig : fiveoschetnetwork ] .
all cluster states connect together to form a single large heteroclinic network that is an attractor for typical initial conditions @xcite .
[ fig : fiveoschetraster ] illustrates the possible switchings between phase differences near one particular cluster state for the heteroclinic network in this case .
the dynamics of this system can be used for encoding a variety of inputs , as discussed in @xcite where it is shown that a constant heterogeneity of the natural frequencies between oscillators in this system leads to a spatio - temporal encoding of the heterogeneities .
the sequences produced by the system can be used by a similar system with adaptive dynamics to learn a spatio - temporal encoded state @xcite .
robust heteroclinic attractors also appear in a range of coupled phase oscillator models where the coupling is not global ( all - to - all ) but such that it still preserves enough invariant subspaces for the connections to remain robust .
for example , @xcite study the dynamics of a network `` motif '' of four coupled phase oscillators and find heteroclinic attractors that are `` ratchets '' , i.e. they are robust heteroclinic networks that wind preferentially around the phase space in one direction - this means that under the influence of small perturbations , phase slips in only one direction can appear . ) with phase interaction function ( [ eq : ghmmcoupling ] ) and a particular choice of parameters ( see text ) .
the vertical dark lines mark times ( @xmath169 in horizontal axis ) when the oscillator represented by @xmath676 ( @xmath251 in vertical axis ) passes through zero .
observe that most of the time there are three clusters .
occasionally the clustering splits and reforms different three clusters reforms , at times indicated by the grey bars , approximately every 70 time units .
[ adapted from @xcite ] [ fig : fiveoschetraster ] , width=384 ] ; see @xcite for more details of the dynamics .
each vertex represents a particular saddle periodic cluster state that is a permutation of the states shown in ( b - d ) .
note that there are precisely two incoming and two outgoing connections at each vertex .
( b - d ) show the relative phases of the five oscillators , indicated by the angles of the pendula " at the vertices of a regular pentagon , for a sequence of three consecutive three saddle cluster states visited on a longer trajectory that randomly explores graph ( a ) in the presence of noise .
inequivalent clusters are characterised by different coloured pendula " and numbers where yellow corresponds to 1 , green to 2 and blue to 3 .
the yellow cluster is stable to cluster - splitting perturbations while the blue cluster is unstable to such perturbations - observe that after the connection the yellow cluster becomes the blue cluster while the blue cluster splits up in one of two possible way .
[ fig : fiveoschetnetwork],width=364 ] heteroclinic connections play an key role in describing winnerless competition dynamics @xcite .
this dynamics is usually associated with firing rate model of coupled neurons of `` lotka - volterra '' form @xmath677 where @xmath166 is an instantaneous firing rate of a neuron or group of neurons , @xmath678 is a self - excitation term and @xmath156 represents coupling between the @xmath144 , though it is also possible to consider generalisations @xcite .
these models were originally developed for predator - prey interaction of species in an ecosystem .
having found wide use in population ecology , they have more recently be applied to evolutionary game dynamics @xcite , including applications in economics .
since the seminal paper of may @xcite it has been recognised that ( [ eq : lotkavolterra ] ) can have robust heteroclinic attractors for an open set of parameter choices @xmath678 and @xmath156 , as long as at least three species are involved . indeed
, the system can show a wide range of dynamics such as `` winner - takes - all '' equilibria where there is a stable equilibrium with @xmath679 and @xmath680 for @xmath681 ( @xmath114 is the `` winner '' ) , `` cooperative '' equilibria where several of the @xmath144 are non - zero as well as non - trivial periodic or chaotic dynamics .
the particular case of `` winnerless competition '' gives attractors consist of chains or networks of saddles joined by connections that are robust because the absence of a species ( or in our case , the lack of firing of one neuron or group of neurons ) @xmath682 is preserved by the dynamics of the model ( [ eq : lotkavolterra ] ) .
the simplest case of this appears in a ring of @xmath683 coupled neurons with dynamics @xmath684 for @xmath685 and @xmath686 @xcite , corresponding to cyclic inhibition of the neurons in one direction around the ring and cyclic excitation in the other direction .
this `` rock - paper - scissors '' type of dynamics leads to winnerless competition that has been applied in a variety of more complex models of neural systems .
the local behaviour near connections in such heteroclinic attractors has been called `` stable heteroclinic channels '' @xcite and used to model a variety of low - level and high - level neural behaviours including random sequence generation , information dynamics , encoding of odours and working memory .
we refer to the reviews @xcite .
analogous behaviour has been found in a range of other coupled systems , for example @xcite or delayed pulse - coupled oscillators @xcite .
recent work has also considered an explicit constructive approach to heteroclinic networks to realise arbitrary directed networks as a heteroclinic attractor of a coupled cell system @xcite .
more complex but related dynamical behaviour has been studied under the names of `` chaotic itinerancy '' ( see for example @xcite ) , `` cycling chaos '' @xcite , `` networks of milnor attractors '' @xcite and `` heteroclinic cycles for chaotic oscillators '' @xcite .
it has been is suggested that these and similar models are useful for modelling of the functional behaviour of neural systems @xcite .
because heteroclinic attractors are quite singular in their dynamical behaviour ( averages of observables need not converge , there is a great deal of sensitivity of the long - term dynamics to noise and system heterogeneity ) , it is important to consider the effect of noise and/or heterogeneities in the dynamics .
this leads to a finite average transition time between states determined by the level of noise and/or heterogeneity ( which may be due to inputs to the system ) and the local dynamics - see for example @xcite .
another useful feature of heteroclinic attractors is that they allow one to model `` input - output '' response of the system to a variety of inputs .
noise is well known to play a constructive role in the neural encoding of natural stimuli , leading to increased reliability or regularity of neuronal firing in single neurons @xcite and across populations @xcite . from a mathematical perspective
it is natural to consider how noise may affect the reduction to a phase - oscillator description .
naively one may simply consider the addition of noise to a deterministic phase - oscillator model to generate a stochastic differential equation . indeed
models of this type have been studied extensively at the network level to understand noise - induced first- and second - order phase transitions , and new phenomenon such as noise - induced synchrony @xcite or asynchrony @xcite , and noise - induced turbulence @xcite .
we refer the reader to the review by lindner @xcite for a comprehensive discussion .
more recently schwabedal and pikovsky have extended the foundations of deterministic phase descriptions to irregular , noisy oscillators ( based on the constancy of the mean first return times ) @xcite , ly and ermentrout @xcite and nakao _
et al_. @xcite have built analytical techniques for studying weak noise forcing , and moehlis has developed techniques to understand the effect of white noise on the period of an oscillator @xcite . at the network level ( global coupling ) a classic paper examining the role of external noise in if populations , using a phase description , is that of kuramoto @xcite , who analysed the onset of collective oscillations . without recourse to a phase - reduction
it is well to mention that medvedev has been pioneering a phase - amplitude approach to studying the effects of noise on the synchronisation of coupled stochastic limit cycle oscillators @xcite , and that newhall _
et al_. have developed a fokker - planck approach to understanding cascade - induced synchrony in stochastically driven if networks with pulsatile coupling and poisson spike - train external driven @xcite . more recent work on pairwise synchrony in network of heterogeneous coupled noisy phase oscillators receiving correlated and independent noise can be found in @xcite .
however , note that even in the absence of synaptic coupling , two or more neural oscillators may become synchronised by virtue of the statistical correlations in their noisy input streams @xcite . for clarity of exposition
let us consider the phase reduction of a planar system described by @xmath687 , where @xmath688 is white gaussian noise such that @xmath689 and @xmath690 , where @xmath691 denotes averaging over the realisation of @xmath229 , and @xmath692 scales the noise intensity .
we employ an stratonovich interpretation of the noise ( such that the chain rule of ordinary calculus holds ) . in the absence of noise
we shall assume that the system has a stable @xmath94-periodic limit - cycle solution , with a phase that satisfies @xmath693 .
for weak noise perturbations the state point will leave the cycle , though will stay in some neighbourhood , which we denote by @xmath34 . to extend the notion of a phase off the cycle we let @xmath247 be a smooth function of @xmath213 such that @xmath694 holds for any point @xmath335 .
we shall denote the other coordinate in @xmath264 by @xmath233 , and assume that there exists a smooth coordinate transformation @xmath695 . for a noise perturbed oscillator , in the new coordinates we have @xmath696 where we have introduced @xmath697 , @xmath698 and @xmath699 .
note that the full coordinate transformation @xmath700 is not prescribed here , though it is common to use one such that @xmath233 can be interpreted as some distance from cycle .
thus although results may be formally developed for the system ( [ thetarho ] ) they can not be directly interpreted in terms of a given model until the full coordinate transformation taking one from @xmath701 to @xmath702 is given
. one can transform ( [ thetarho ] ) into a stochastic phase - amplitude equation in the it sense , where it reads @xmath703 + h(\vartheta,\rho ) \xi(t ) , \nonumber \\
\frac{{{\rm d}}}{{{\rm d}}t } { \rho } & = f(\vartheta,\rho ) + d[g_\vartheta(\vartheta,\rho ) h(\vartheta,\rho)+ g_\rho(\vartheta,\rho ) g(\vartheta , r ) ] + g(\vartheta,\rho)\xi(t ) , \nonumber\end{aligned}\ ] ] where the subscripts @xmath247 and @xmath233 denote partial derivatives with respect to @xmath247 and @xmath233 respectively . using the it form
we may construct a fokker - planck equation for the time - dependent probability distribution @xmath704 as @xmath705 + d { \frac{\partial ^2 [ h^2 q]}{\partial \vartheta^2 } } \nonumber \\ & -{\frac{\partial } { \partial \rho } } \left [ \ { f + d(g_\vartheta h + g_\rho g ) \}q \right ] + 2 d { \frac{\partial ^2 [ h g q]}{\partial \vartheta \partial \rho } } + d { \frac{\partial ^2 [ g^2 q]}{\partial \rho^2 } } , \label{fp}\end{aligned}\ ] ] with periodic boundary condition @xmath706 . when @xmath707 the steady state distribution is given by @xmath708 .
for small noise amplitude @xmath709 we expect @xmath710 to still localise near @xmath249 @xcite , say over a region @xmath711 . in this case
it is natural to make the approximation that for large @xmath169 , @xmath712 at @xmath713 .
now introduce the marginal distribution @xmath714 following @xcite we can integrate ( [ fp ] ) over the interval @xmath715 $ ] and generate a fokker - planck equation for @xmath369 .
the last three terms in ( [ fp ] ) vanish after integration due to the boundary conditions , so that we are left with @xmath716 we now expand about @xmath249 to give @xmath717 and @xmath718 , where @xmath719 is identified as the infinitesimal phase response @xmath720 . in the limit of small @xmath709 where @xmath721 we note that , for an arbitrary function @xmath722 , @xmath723 making use of the above gives the fokker - planck equation as @xmath724 + d { \frac{\partial ^2[z^2p]}{\partial \vartheta^2 } } , \label{fpphase}\ ] ] where @xmath725 .
hence , the corresponding it equation is @xmath726 + z(\vartheta ) \xi(t ) , \label{itophase}\ ] ] while the stratonovich version is @xmath727 equations ( [ itophase ] ) and ( [ phasestrat ] ) are the stochastic phase oscillator descriptions for a limit cycle driven by weak white noise .
these make it clear that naively adding noise to the phase description misses not only a multiplication by the iprc but also the addition of a further term @xmath728 that contains information about the _ amplitude _ response of the underlying limit cycle oscillator .
we are now in a position to calculate the steady state probability distribution @xmath729 and use this to calculate the moments of the phase dynamics .
consider ( [ fpphase ] ) with the boundary condition @xmath730 , and set @xmath731 .
adopting a fourier representation for @xmath732 , @xmath733 and @xmath734 as @xmath735 , @xmath736 , @xmath737 , allows us to obtain a set of equations for the unknown amplitudes @xmath738 as @xmath739 for some constant @xmath209 .
for @xmath707 we have that @xmath740 , and after enforcing normalisation we may set @xmath741 . for small @xmath709 we may then substitute @xmath732 into ( [ amplitudes ] ) and work to next order in @xmath709 to obtain an approximation for the remaining amplitudes , @xmath742 , in the form @xmath743 using this we may reconstruct the distribution @xmath744 , for small d , from ( [ fourier ] ) as @xmath745 the mean frequency of the oscillator
is defined by @xmath746 .
this can be calculated by replacing the time average with the ensemble average . for an arbitrary @xmath747-periodic function
@xmath368 we set @xmath748 . using ( [ p ] ) and ( [ itophase ] ) we obtain @xmath749 where we have used the fact that @xmath750 .
we may also calculate the phase - diffusion @xmath751 as @xmath752\left [ \frac{{{\rm d}}}{{{\rm d}}t } { \vartheta}(t ) - \left \langle \frac{{{\rm d}}}{{{\rm d}}t } { \vartheta } \right \rangle \right ] \right \rangle { { \rm d}}\tau \nonumber \\ & = \frac{d}{\pi } \int_0^{2 \pi } z^2(\vartheta ) { { \rm d}}\vartheta + o(d^2 ) , \nonumber\end{aligned}\ ] ] where we use the fact that @xmath689 and @xmath690 .
this recovers a well known result of kuramoto @xcite . a recent paper by teramae _
et al_. @xcite shows that when one considers noise described by an ornstein - uhlenbeck ( ou ) process with a finite correlation time then this can interact with the attraction time - scale of the limit cycle and give fundamentally different results when compared to gaussian white noise ( which has a zero correlation time ) .
this observation has also been independently made in @xcite .
both approaches assume weak noise , though @xcite makes no assumptions about relative time - scales , and is thus a slightly more general approach than that of @xcite .
related work by goldobin _
et al_. @xcite for noise @xmath95 with zero - mean @xmath753 and prescribed auto - correlation function @xmath754 , yields the reduced stratonovich phase equation @xmath755 where @xmath756 where @xmath279 is the average rate of attraction to the limit cycle .
note that for @xmath757 , @xmath758 and ( [ phasegeneral ] ) reduces to ( [ phasestrat ] ) as expected .
to lowest order in the noise strength the steady state probability distribution will simply be @xmath759 .
therefore to lowest noise order the mean frequency is determined from an ensemble average as @xmath760 where the last term comes from using the it form of ( [ phasegeneral ] ) and the subscript @xmath14 notation is defined as in ( [ p ] ) .
the phase - diffusion coefficient at lowest noise order is given by @xmath761 let us now consider the example of ou noise so that @xmath762 . furthermore let us take the simultaneous limit @xmath763 ( zero correlation timescale ) and @xmath764 ( infinitely fast attraction ) , such that the ratio @xmath765 is constant .
in this case we have from ( [ ytilde ] ) that @xmath766 hence , when the correlation time of the noise is much smaller than the decay time constant @xmath767 and we recover the result for white gaussian noise . in the other extreme when @xmath401 , where the amplitude of the limit cycle decays much faster than the correlation time of the noise , then @xmath768 vanishes and the reduced phase equation is simply @xmath769 , as would be obtained using the standard phase reduction technique without paying attention to the stochastic nature of the perturbation .
the self - organisation of large networks of coupled neurons into macroscopic coherent states , such as observed in phase - locking , has inspired a search for equivalent low - dimensional dynamical descriptions .
however , the mathematical step from microscopic to macroscopic dynamics has proved elusive in all but a few special cases .
for example , neural mass models of the type described in [ subsec : neuralmass ] only track mean activity levels and not the higher order correlations of an underlying spiking model . only in the thermodynamic limit of a large number of neurons firing asynchronously ( producing null correlations )
are such rate models expected to provide a reduction of the microscopic dynamics .
moreover , even here the link from spike to rate is often phenomenological rather than rigorous .
unfortunately only in some rare instances has it been possible to analyse spiking networks directly ( usually under some restrictive assumption such as global coupling ) as in the spike - density approach @xcite , which makes heavy use of the numerical solution of coupled pdes .
recently however , exact results for globally pulse - coupled oscillators described by the winfree model @xcite have been obtained by paz and montbri @xcite .
this makes use of the ott - antonsen ( oa ) ansatz , which was originally used to find solutions on a reduced invariant manifold of the kuramoto model @xcite .
the major difference between the two phase - oscillator models being that the former has interactions described by a phase product structure and the latter a phase difference structure .
the winfree model is described in [ sec : weakcoupling ] as a model for weakly globally pulse - coupled biological oscillators , and can support incoherence , frequency locking , and oscillator death when @xmath770 and @xmath771 @xcite .
we note however that the same model is _ exact _ when describing nonlinear if models described by a single voltage equation , and that in this case we do not have to restrict attention to weak - coupling . indeed the oa ansatz has proved equally successful in describing both the winfree network with a sinusoidal prc @xcite and a network of qif neurons @xcite .
this is perhaps not surprising since the prc of a qif neutron can be computed using ( [ rif ] ) , and for the case described by ( [ eq : one ] ) with @xmath772 and @xmath323 and infinite threshold and reset then @xmath773 with @xmath774 for @xmath775 ( which is the shape expected for an oscillator near a snic bifurcation ) .
we shall now focus on this choice of prc and a pulsatile coupling that we write in the form @xmath776 where we have introduced a convenient fourier representation for the periodic function @xmath369 .
we now consider the large @xmath140 limit in ( [ winfree ] ) and let @xmath777 be the fraction of oscillators with phases between @xmath778 and @xmath779 and natural frequency @xmath780 at time @xmath169 .
the dynamics of the density @xmath233 is governed by the continuity equation ( expressing the conservation of oscillators ) @xmath781 , \label{continuity}\ ] ] where the mean - field drive is @xmath782 .
boundary conditions are periodic in the probability flux , namely @xmath783 .
a further reduction in dimensionality is obtained for the choice that the distribution of frequencies is described by a lorentzian @xmath784 with @xmath785 for fixed @xmath333 and @xmath786 ( controlling the width and mean of the distribution respectively ) , which has simple poles at @xmath787 .
a generalised set of order parameters is defined as @xmath788 the integration over @xmath780 can be done using a contour in the lower half complex plane so that @xmath789 , where we have introduced the inner product @xmath790 .
the oa ansatz assumes that the density @xmath233 can be written in a restricted fourier representation as @xmath791 where cc stands for complex conjugate .
substitution into the continuity equation ( [ continuity ] ) and balancing terms in @xmath792 shows that @xmath103 must obey @xmath793 moreover , using the inner product structure of @xmath794 we easily see that @xmath795^m$ ] .
thus the kuramoto order parameter @xmath796 is governed by ( [ alpha ] ) with @xmath797 yielding : @xmath798 .
\label{rpsi}\ ] ] to calculate the mean - field drive @xmath28 we note that it can be written as @xmath799 hence , we have explicitly that @xmath800 with @xmath801 the planar system of equations defined by ( [ rpsi ] ) and ( [ h ] ) can be readily analysed using numerical bifurcation analysis .
we note that the oa density ( [ oa ] ) can be written in the succinct form @xmath802 . \nonumber\ ] ] the form of this equation suggests recasting the density in terms of new variables @xmath803 and @xmath804 @xcite . in this case
we find that @xmath805 has a lorentzian shape given explicitly by @xmath806 with @xmath807 , where we have set @xmath808 and re - scaled @xmath1 by a factor of @xmath809 .
using the continuity equation for @xmath805 and then integrating over @xmath1 gives the evolution of @xmath588 as @xmath810 , \qquad v = \int_\infty^\infty { { \rm d}}v \rho(v|\omega , t ) v , \label{w}\ ] ] where we identify @xmath39 as the average of @xmath1 .
the firing probabilities for arbitrary cells at time @xmath169 are equal to the passage rate of the probability density through the spike phase , which gives the population firing rate as @xmath811 thus we may use ( [ w ] ) to determine the evolution of the coupled rate and average network activity by integrating over the distribution of parameters @xmath812 .
the evolution of @xmath813 is then found as @xmath814 where we have used the result that @xmath815 .
exploiting the fact that the order parameters in the ` @xmath778 ' and ` @xmath1 ' descriptions of macroscopic dynamics are related by a conformal mapping , namely @xmath816 , we may explore both the firing rate and degree of synchrony of the network .
thus ( [ mfoa ] ) is a mean field reduction of a population of spiking qif neurons , and unlike a phenomenological neural mass model it is able to track a measure of within population synchrony .
the system ( [ mfoa ] ) is capable of supporting bistable fixed point behaviour as shown in fig .
[ fig : roxin ] , where we also plot the basin boundary ( stable manifold of a saddle ) . '
dynamics showing the rate and average @xmath817 plane with three fixed points for @xmath818 and @xmath819 . for this parameter
set the system is bistable and the stable ( blue ) and unstable ( green ) manifolds of the saddle are plotted .
the remaining curves are nullclines .
[ fig : roxin ] , width=288 ] the oa ansatz has also proved remarkably useful in understanding non - trivial solutions such as chimera states ( where a sub - population of oscillators synchronises in an otherwise incoherent sea ) .
phase or cluster synchronised states in systems of identical coupled oscillators have distinct limitations as descriptions of neural systems where not just phase but also frequency clearly play a part in the processing , computation and output of information .
indeed , one might expect that for any coupled oscillator system that is homogeneous ( in the sense that any oscillators can be essentially replaced by any other by a suitable permutation of the oscillators ) , the only possible dynamical states are homogeneous in the sense that the oscillators behave in either a coherent or an incoherent way .
this expectation however is not justified - there can be many dynamical states that can not be easily classified as coherent or incoherent , but that seem to have a mixture of coherent and incoherent regions .
such states have been given the name `` chimera state '' by abrams and strogatz @xcite and have been the subject of intensive research over the past five years . for reviews of chimera state dynamics
we refer the reader to @xcite .
kuramoto and battogtokh @xcite investigated continuum systems of oscillators of the form @xmath820 where @xmath778 represent phases at locations @xmath584 , the kernel @xmath821 represents a non - local coupling and @xmath822 are constants .
interestingly this model is precisely in the form presented in [ sec : waves ] as equation ( [ phasecontinuum ] ) for an oscillatory model of cortex , although here there are no space - dependent delays and the interaction function is @xmath823 .
kuramoto and battogtokh found for a range of parameters near @xmath824 , and carefully selected initial conditions , that the oscillators can split into two regions in @xmath213 , one region which is frequency synchronised ( or coherent ) while the other region shows a nontrivial dependence of frequency on location @xmath213 .
an example of a chimera state is shown in fig .
[ fig : chimera ] . ) in a system of length @xmath825 using @xmath826 numerical grid points and periodic boundary conditions . here
@xmath654 , @xmath827 , @xmath828 and @xmath829 .
[ fig : chimera ] , width=302 ] note that a discretisation of ( [ eq : continuumnonlocal ] ) to a finite set of @xmath140 coupled oscillators is @xmath830 where @xmath366 represents the phase at location @xmath556 and the coupling matrix @xmath831 is the discretised interaction kernel ( assuming a domain of length 1 ) . using different kernels ,
@xmath832 and an approximation @xmath833 for small @xmath261 , abrams and strogatz @xcite identified similar behaviour and @xcite discussed a limiting case of parameters such that the continuum system provably has chimera solutions .
the oa reduction discussed in [ sec : ott - antonsen ] allows an exact reduction of oscillator networks of this form and in the continuum limit this can give a solvable low - order system whose solutions include a variety of chimera states @xcite .
it is useful to note that when @xmath824 , pure cosine coupling results in an integrable hamiltonian system @xcite , such that disordered initial states will remain disordered .
thus @xmath103 determines a balance between spontaneous order and permanent disorder .
however , it seems that chimera states are much more `` slippery '' in finite oscillator systems than in the continuum limit . in particular , wolfrum and omelchenko
@xcite note that for finite approximations of the ring ( [ eq : continuumnonlocal ] ) by @xmath140 oscillators , with a mixture of local and nearest @xmath834-neighbour coupling corresponding to ( [ eq : discretenonlocal ] ) with a particular choice of coupling matrix @xmath835 , chimera states apparently only exist as transients .
however , the lifetime of the typical transient apparently grows exponentially with @xmath140 .
thus , at least for some systems of the form ( [ eq : discretenonlocal ] ) , chimeras appear to be a type of chaotic saddle .
this corresponds to the fact that the boundaries between the regions of coherent and incoherent oscillation fluctuate apparently randomly over a long timescale .
these fluctuations lead to wandering of the incoherent region as well as change in size of the region .
eventually these fluctuations appear to result in typical collapse to either fully coherent or fully incoherent oscillation @xcite .
although this appears to be the case for chimeras for ( [ eq : discretenonlocal ] ) , there are networks such as coupled groups of oscillators ; @xcite or two dimensional lattices @xcite where chimera attractors can appear .
it is not clear what will cause a chimera to be transient or not , or indeed exactly what types of chimera - like states can appear in finite oscillator networks .
a suggestion of @xcite is that robust neutrally stable chimeras may be due to the special type of single - harmonic phase interaction function used in ( [ eq : continuumnonlocal],[eq : discretenonlocal ] ) .
more recent work includes investigations of chimeras ( or chimera - like states ) in chemical @xcite or mechanical oscillator networks @xcite ; chimeras in systems of coupled oscillators other than phase oscillators have been investigated in many papers ; for example in stuart - landau oscillators @xcite , winfree oscillators @xcite and models with inertia @xcite .
other recent work includes discussion of feedback control to stabilise chimeras @xcite , investigations of chimeras with multiple patches of incoherence @xcite and multicluster and traveling chimera states @xcite . in a neural context chimeras
have also been found in pulse - coupled lif networks @xcite , and hypothesised to underly coordinated oscillations in unihemispheric slow - wave sleep , whereby one brain hemisphere appears to be inactive while the other remains active @xcite .
we briefly review a few examples where mathematical frameworks are being applied to neural modelling questions .
these cover functional and structural connectivity in neuroimaging , central pattern generators ( cpgs ) and perceptual rivalry .
there are many other applications we do not review , for example to deep brain stimulation protocols @xcite or to modelling of epileptic seizures where network structures play a key role @xcite .
functional connectivity ( fc ) refers to the temporal synchronisation of neural activity in spatially remote areas .
it is widely believed to be significant for the integrative processes in brain function .
anatomical or structural connectivity ( sc ) , is widely believed to play an important role in determining the observed spatial patterns of fc .
however , there is clearly a role to be played by the dynamics of the neural tissue . even in a globally connected network
we would expect this to be the case , given our understanding of how synchronised solutions can lose stability for weak coupling .
thus it becomes useful to study models of brain like systems built from neural mass models ( such as the jansen - rit model of
[ subsec : neuralmass ] ) , and ascertain how the local behaviour of the oscillatory node dynamics can contribute to global patterns of activity . for simplicity , consider a network of @xmath140 globally coupled identical wilson - cowan @xcite neural mass models : @xmath836 for @xmath837 . here
@xmath838 represents the activity in each of a pair of coupled neuronal population models , @xmath121 is a sigmoidal firing rate given by @xmath839 and @xmath840 represent external drives .
the strength of connections within a local population is prescribed by the co - efficients @xmath841 , which we choose as @xmath842 and @xmath843 as in @xcite .
for @xmath147 it is straight forward to analyse the dynamics of a local node and find the bifurcation diagram in the @xmath840 plane as shown in fig .
[ fig : wc ] .
moreover , for @xmath334 we may invoke weak coupling theory to describe the dynamics of the full network within the oscillatory regime bounded by the two hopf curves shown in fig .
[ fig : wc ] left . from the theory of [ sec : synchrony ] we would expect the synchronous solution to be stable if @xmath844 .
taking @xmath845 we can consider @xmath846 as a proxy for the robustness of synchrony .
the numerical construction of this quantity , as in @xcite , predicts that there will be regions in the @xmath840 plane associated with a breakdown of fc ( where @xmath847 ) , as indicated by points a and b in fig .
[ fig : wc ] .
this highlights the role that local node dynamics can have on emergent network dynamics .
moreover , we see that simply by tuning the local dynamics to be deeper within the existence region for oscillatory solutions we can , at least for this model , enhance the degree of fc . plane . here
hb denotes hopf bifurcation and sn a saddle node of fixed - points bifurcation . at points
a and c we find @xmath848 and at point b @xmath849 .
a breakdown of fc ( loss of global synchrony ) within a globally coupled network is predicted at points a and c. [ fig : wc ] , height=240 ] it would be interesting to explore this simple model further for more realistic brain like connectivities , along the lines described in @xcite .
moreover , given that this would preclude the existence of the synchronous state by default ( since we would neither have that @xmath441 or @xmath850 would be independent of @xmath114 ) then it would be opportune to explore the use recent ideas in @xcite to understand how the system could organise into a regime of remote synchronisation whereby pairs of nodes with the same network symmetry could synchronise .
for related work on wilson - cowan networks with some small dynamic noise see @xcite , though here the authors construct a phase - oscillator network by linearising around an unstable fixed point , rather than use the notion of phase response .
cpgs are ( real or notional ) neural subsystems that are implicated in the generation of spatio - temporal patterns of activity @xcite , in particular for driving the relatively autonomous activities such as as locomotion @xcite or for driving involuntary activities such as heartbeat , respiration or digestion @xcite .
these systems are assumed to be behind the creation of the range of walking or running patterns ( gaits ) that appear in different animals @xcite .
the analysis of phase - locking provides a basis for understanding the behaviour of many cpgs , and for a nice overview see the review articles by marder and bucher @xcite and hooper @xcite . in some cases , such as the leech ( _ hirudo medicinalis _ ) heart or _
caenorhabditis elegans _ locomotion , the neural circuitry is well studied . for more complex neural systems and in more general cases cpgs
are still a powerful conceptual tool to construct notional minimal neural circuitry needed to undertake a simple task . in this notional sense
they have extensively been investigated to design control circuits for actuators for robots ; see for example the review @xcite .
recent work in this area includes robots that can reproduce salamander walking and swimming patterns @xcite .
since the control of motion of autonomous `` legged '' robots is still a very challenging problem in real - time control , one hope of this research is that nature s solutions ( for example , how to walk stably on two legs ) will help inspire robotic ways of doing this .
cpgs are called upon to produce one or more rhythmic patterns of actuation ; in the particular problem of locomotion , a likely cpg is one that will produce the range of observed rhythms of muscle actuation , and ideally the observed transitions between then . for an early discussion of design principles for modelling cpgs ,
see @xcite .
this is an area of modelling where consideration of symmetries as in [ subsec : symmetry ] has been usefully applied to constrain the models .
for example @xcite examine models for generating the gaits in a range of vertebrate animals , from those with two legs ( humans ) through those with four ( quadrupeds such as horses have a wide range of gaits - walk , trot , pace , canter , gallop - they may use ) or larger numbers of legs ( myriapods such as centipedes ) .
insects make use of six legs for locomotion while other invertebrates such as centipedes and millipedes have a large number of legs that are to some extend independently actuated . as an example , @xcite consider a schematic cpg of @xmath851 oscillators for animals with @xmath29 legs , as shown in fig .
[ fig : golcpg](a ) .
the authors use symmetry arguments and theorem [ thm : honk ] to draw a number of model - independent conclusions from the cpg structure .
coupled cells that is used to model gait patterns in animals with @xmath851 legs @xcite and ( b ) a three cell motif of bursters with varying coupling strengths , as considered in @xcite.,width=377 ] one can also view cpgs as a window into more fundamental problems of how small groups of neurons coordinate to produce a range of spatio - temporal patterns .
in particular , it is interesting to see how the observable structure of the connections influences the range and type of dynamical patterns that can be produced .
for example , @xcite consider a simple three - cell `` motif '' networks of bursters and classify a range of emergent spatio - temporal patterns in terms of the coupling parameters .
detailed studies @xcite investigate properties such multistability and bifurcation of different patterns and the influence of inhomogeneities in the system .
this is done by investigating return maps for the burst timings relative to each other .
the approach mentioned in [ subsec : groupoid ] and outlined in @xcite seems to be a promising way of providing a context in which view cpgs where there is a constrained connection structure suggested by either the neurobiology or from conceptual arguments , but this structure is not purely related to symmetries of the network .
for example , @xcite use that formalism to understand possible spatio - temporal patterns that arise in lattices or @xcite that relates synchrony properties of small motif networks to spectral properties of the adjacency matrix .
many neural systems process information - they need to produce outputs that depend on inputs .
if the system effectively has no internal degrees of freedom then this will give a functional relationship between output and input so that any temporal variation in the output corresponds to a temporal variation of the input .
however , this is not the case for all but the simplest systems and often outputs can vary temporally unrelated to the input .
a particularly important and well - studied system that is a model for autonomous temporal output is _ perceptual rivalry _
, where conflicting information input to a neural system results , not in a rejection or merging of the information , but in an apparently random `` flipping '' between possible `` rival '' states ( or percepts ) of perception .
this nontrivial temporal dynamics of the perception appears even in the absence of a temporally varying input .
the best studied example of this type is _ binocular rivalry _ , where conflicting inputs are simultaneously made to each eye .
it is widely reported by subjects that perception switches from one eye to the other , with a frequency that depends on a number of factors such as the contrast of the image @xcite .
more general perceptual rivalry , often used in `` optical illusions '' such as ambiguous figures - the rubin vase , the necker cube - show similar behaviour with percepts shifting temporally between possible interpretations .
various approaches @xcite have been made to construct nonlinear dynamical models of the generation of a temporal shifting between possible percepts such as competition models @xcite , bifurcation models , ones based on neural circuitry @xcite , or conceptual ones @xcite based on network structures@xcite or on heteroclinic attractors @xcite .
as with any review we have had to leave out many topics that will be of interest to the reader .
in particular we have confined ourselves to `` cell '' and `` system - level '' dynamics rather that `` sub - cellular '' behaviour of neurons .
we briefly mention some other active areas of mathematical research relevant to the science of rhythmic neural networks .
perhaps the most obvious area that we have not covered in any depth is that of single unit ( cell or population ) forcing , which itself is a rather natural starting point for gaining insights into network behaviour and how best to develop mathematical tools for understanding response @xcite . for a general perspective on mode - locked responses to periodic
forcing see @xcite and @xcite . for
a more recent discussion of the importance of mode - locking in auditory neuroscience see @xcite and in motor systems see @xcite .
however , it is well to note that not much is known about nonlinear systems with three or more interacting frequencies @xcite , as opposed to periodically forced systems where the notions of farey tree and the devil s staircase have proven especially useful .
we have also painted the notion of synchrony with a broad mathematical brush , and not discussed more subtle notions of envelope locking that may arise between coupled bursting neurons ( where the within burst patterns may desynchronise ) @xcite .
this is especially relevant to studies of synchronised bursting @xcite and the emergence of chaotic phenomena @xcite .
indeed , we have said very little about coupling between systems that are chaotic , such as described in @xcite , the emergence of chaos in networks @xcite or chaos in symmetric networks @xcite .
the issue of chaos is also relevant to notions of reliability , where one is interested in the stability of spike trains against fluctuations .
this has often been discussed in relation to stochastic oscillator forcing rather than those arising deterministically in a high dimensional setting @xcite .
of course , given the sparsity of firing in cortex means that it may not even be appropriate to treat neurons as oscillators .
however , some of the ideas developed for oscillators can be extended to excitable systems , as described in @xcite . as well as this
it is important to point out that neurons are not point processors , and have an extensive dendritic tree , which can also contribute significantly to emergent rhythms as described in @xcite , as well as couple strongly to glial cells .
although the latter do not fire spikes , they do show oscillations of membrane potential @xcite . at the macroscopic level it is also important to acknowledge that the amplitude of different brain waves can also be significantly affected by neuromodulation @xcite ,
say through release of norepinephrine , serotonin and acetylcholine , and the latter is also thought to be able to modulate the prc of a single neuron @xcite .
this review has focused mainly on the embedding of weakly coupled oscillator theory within a slightly wider framework .
this is useful in setting out some of the neuroscience driven challenges for the mathematical community in establishing inroads into a more general theory of coupled oscillators .
heterogeneity is one issue that we have mainly side - stepped , and remember that the weakly coupled oscillator approach requires frequencies of individual oscillators to be close .
this can have a strong effect on emergent network dynamics @xcite , and it is highly likely that even a theory with heterogeneous phase response curves @xcite will have little bearing on real networks . the _ equation - free _ coarse - graining approach may have merit in this regard , though is a numerically intensive technique @xcite .
we suggest a good project for the future is to develop a theory of strongly coupled heterogeneous networks based upon the phase - amplitude coordinate system described in [ subsec : phase - amplitude ] , with the challenge to develop a reduced network description in terms of a set of phase - amplitude interaction functions , and an emphasis on understanding the new and generic phenomena associated with nontrivial amplitude dynamics ( such as clustered phase - amplitude chaos and multiple attractors ) . to achieve this
one might further tap into recent ideas for classifying emergent dynamics based upon the group of structural symmetries of the network .
this can be computed as the group of automorphisms for the graph describing the network . for many real - world networks
, this can be decomposed into direct and wreath products of symmetric groups @xcite .
this would allow the use of tools from computational group theory @xcite to be used , and open up a way to classify the generic forms of behaviour that a given network may exhibit using the techniques of equivariant bifurcation theory .
the hodgkin - huxley description of nerve tissue is completed with : @xmath852 } , & \alpha_h(v ) & = 0.07 \exp [ -0.05(v+65 ) ] , \nonumber \\
\alpha_n(v ) & = \frac{0.01(v+55)}{1-\exp[-0.1(v+55 ) ] } , & \beta_m(v ) & = 4.0 \exp[-0.0556(v+65 ) ] , \nonumber \\
\beta_h(v ) & = \frac{1}{1 + \exp[-0.1(v+35 ) ] } , & \beta_n(v ) & = 0.125 \exp[-0.0125(v+65 ) ] , \nonumber\end{aligned}\ ] ] and @xmath853f @xmath108 , @xmath854mmho @xmath108 , @xmath855mmho @xmath108 , @xmath856mmho @xmath108 , @xmath857mv , @xmath858mv and @xmath859mv .
( all potentials are measured in mv , all times in ms and all currents in @xmath860a per @xmath861 ) .
we give a brief list of some of the abbreviations used within the review .
dde : : delay differential equation if : : integrate and fire ( model for neural oscillator ) iprc : : infinitesimal phase response curve isi : : inter - spike interval fhn : : fitzhugh - nagumo equation ( model for neural oscillator ) hh : : hodgkin - huxley equation ( model for neural oscillator ) lif : : leaky integrate and fire ( model for neural oscillator ) ml : : morris - lecar equation ( model for neural oscillator ) msf : : master stability function ode : : ordinary differential equation pde : : partial differential equation prc : : phase response curve qif : : quadratic integrate and fire ( model for neural oscillator ) sde : : stochastic differential equation shc : : stable heteroclinic channel snic : : saddle - node on an invariant circle ( bifurcation ) wlc : : winnerless competition
the authors declare that they have no competing interests .
pa , sc and rn contributed equally .
all authors read and approved the final manuscript .
we would like to thank kyle wedgwood and ine byrne for useful comments made on draft versions of this manuscript .
sc was supported by the european commission through the fp7 marie curie initial training network 289146 , nett : neural engineering transformative technologies .
l f abbott and t b kepler .
model neurons : from hodgkin huxley to hopfield . in luis garrido , editor , _ statistical mechanics of neural networks _ , number 368 in lecture notes in physics , pages 518 .
springer - verlag , berlin heidelberg , 1990 .
xerxes d. arsiwalla , riccardo zucca , alberto betella , enrique martinez , david dalmazzo , pedro omedas , gustavo deco , and paul f.m.j .
network dynamics with brainx3 : a large - scale simulation of the human brain network with real - time interaction . , 9(2 ) , 2015 .
o. benjamin , t.h.b .
fitzgerald , p. ashwin , k. tsaneva - atanasova , f. chowdhury , m.p .
richardson , and j.r .
a phenomenological model of seizure initiation suggests network structure may explain seizure frequency in idiopathic generalised epilepsy .
, 2:141 , 2012 .
m breakspear , ja roberts , john r terry , s rodrigues , n mahant , and pa robinson . a unifying explanation of primary generalized seizures through nonlinear brain modeling and bifurcation analysis .
, 16:12961313 , 2006 .
t g brown . on the nature of the fundamental activity of the nervous centres
; together with an analysis of the conditioning of rhythmic activity in progression and a theory of the evolution of function in the nervous system . , 48:1846 , 1914 .
fahmida a chowdhury , wessel woldman , thomas hb fitzgerald , robert dc elwes , lina nashef , john r terry , and mark p richardson .
revealing a brain network endophenotype in families with idiopathic generalised epilepsy .
, 9:e110136 , 2014 .
kiyoshi kotani , ikuhiro yamaguchi , yutaro ogawa , yasuhiko jimbo , hiroya nakao , and g. bard ermentrout .
adjoint method provides phase response functions for delay - induced oscillations .
, 109:044101 , 2012 .
y a kuznetsov , v v levitin , and a r skovoroda .
continuation of stationary solutions to evolution problems in content .
technical report report am - r9611 , centrum voor wiskunde en informatica , amsterdam , the netherlands , 1996 . | the tools of weakly coupled phase oscillator theory have had a profound impact on the neuroscience community , providing insight into a variety of network behaviours ranging from central pattern generation to synchronisation , as well as predicting novel network states such as chimeras .
however , there are many instances when this theory is expected to break down , say in the presence of strong coupling , or must be carefully interpreted , as in the presence of stochastic forcing .
there are also surprises in the dynamical complexity of the attractors that can robustly appear - for example , heteroclinic network attractors . in this review
we present a set of mathematical tools that are suitable for addressing the dynamics of oscillatory neural networks , broadening from a standard phase oscillator perspective to provide a practical framework for further successful applications of mathematics to understanding network dynamics in neuroscience . |
until recently the modelling of the atmospheres of stars ( e.g. * ? ? ? * ) and the modelling of the atmospheres of the earth and other solar - system planets ( e.g. * ? ? ?
* ) have developed largely independently .
models of stars applied to high temperature objects with effective temperatures @xmath0 3000k , with opacity dominated by the line and continuum absorption of atoms and atomic ions , whereas planetary atmosphere models applied to cool objects @xmath1 100 - 300k where the important processes were molecular absoprtion and scattering from molecules and cloud particles .
this situation changed with the discovery in the mid-1990s of the first unambiguous brown dwarf , gl229b . @xcite and the first hot jupiter planets beginning with 51 peg b @xcite .
many more such objects have now been discovered and reveal that planets and brown dwarfs populate an intermediate range of temperatures not explored previously .
this has led to the requirement to develop new methods to model these atmospheres that cover the effective temperature range from below 1000k to more than 2000k .
one widely used approach has been to adapt stellar atmosphere codes to handle the lower temperatures encountered in exoplanets and brown dwarfs .
models of this type are described for example by @xcite , @xcite , @xcite , and @xcite .
an alternative approach , and the one we follow in this paper , is to take models originally used for the atmospheres of the earth or other solar system planets and adapt them to handle the higher temperatures needed for exoplanets and brown dwarfs .
such an approach has been described by @xcite and @xcite who use a model based on one originally used to model the atmospheres of titan and uranus @xcite . one important difference between the techniques used in stellar atmosphere modelling and in earth atmosphere modelling is the approach to radiative transfer . in stellar atmospheres simplified treatments of scattering , such as the assumption of isotropic scattering , are commonly adopted .
this is justified by the fact that scattering is in most cases not an important source of opacity in stellar atmospheres , and where it does become significant , in the form of rayleigh scattering from molecules in cool stars , and scattering from electrons in hot stars , the phase functions are forward - backward symmetric .
scattering from clouds and aerosols in the earth atmosphere can , however , result in strongly forward peaked phase functions , and to properly model such cases radiative transfer methods that more rigorously handle multiple scattering with anisotropic phase functions are needed .
while these techniques build on the classic work of astronomers such as chandrasekhar and van de hulst , much recent development of such methods has been in the context of earth atmosphere research ( e.g. * ? ? ?
clouds are now known to be important not just in the atmospheres of all the solar system planets , but in many brown dwarfs and even in late m dwarfs .
thus for all these objects a more rigorous treatment of radiative transfer is desirable .
such an approach is particularly important for modelling the reflected light from exoplanets , because the angular dependence of scattering is an important factor in determining the phase variation around a planet s orbit ( e.g. * ? ? ?
* ; * ? ? ?
* ) . in this paper
we describe the methods used in the vstar ( versatile sofwtware for transfer of atmospheric radiaition ) atmospheric modelling software .
vstar was originally developed as a way of modelling the spectra of the solar system planets , and an early version of it is described by @xcite . in this paper
we describe the current version of vstar which can now handle a wide range of atmospheres ranging from those of the coolest solar system planets up to stars with temperatures of @xmath23000k , and thus including the brown dwarfs and hot jupiter type exoplanets .
the ionization and chemical equilibrium ( ice ) package of vstar is used to determine the equilibrium chemical composition of an atmospheric layer given its elemental abundances , pressure and temperature .
ice handles gas phase chemistry , ionized species and the formation of solid and liquid condensates .
full details of the methods employed in this package will be described elsewhere so only a brief description is given here .
the techniques used are similar to those described by @xcite , @xcite and @xcite . for each compound considered in the model
the equilibrium constant of formation @xmath3 from the elements is required .
these are used in a set of equations for the mass balance of each element , and for the charge balance to solve for the abundances of each molecular species .
@xmath3 is related to the gibbs free energy of formation @xmath4 through : @xmath5 where r is the gas constant .
this relationship is indicative of the link between the technique we use based on equilibrium coinstants , and the alternative technique for chemical models based on minimization of the total gibbs free energy of the system @xcite both @xmath4 and @xmath3 are functions of temperature ( t ) and are available in standdard compilations of thermochemical data .
our main source of thermochemical data was the fourth edition of the nist - janaf thermochemical tables @xcite .
we used data from @xcite and @xcite for several compounds for which the janaf data have been shown to contain errors .
additional sources of thermochemical data were @xcite and @xcite . some important species ,
however , do not have data in any of these tabulations , and for some others the available data do not extend to sufficiently high temperatures for stellar atmosphere models . for a number of gas - phase species we have therefore calculated our own thermochemical data . to do this we first calculate the partition function using the spectroscopic constants of the molecule .
we use a rigid rotator harmonic oscillator model ( with additonal corrections for anharmonicity and centrifugal distortion ) for the rotational and vibrational levels , and direct summation over the electronic levels @xcite .
the thermodynamic functions can then be derived from the partition function and the dissociation energy of the molecule .
the techniques are essentially the same as those used in constructing the nist - janaf thermochemical tables @xcite for gas phase species , and we have tested our methods by reproducing results published in these tables .
species we have calculated new thermochemical data for include feh , crh , cah , tih , rbcl and rbf .
our thermochemical model currently includes 143 gas - phase and condensed - phase compounds of 27 elements .
while this is not as extensive as some other models , we have been careful to include all species listed as important in previous studies such as @xcite .
our chemical model predictions show good agreement with previous results such as those of @xcite and @xcite .
figure [ fig_abund ] shows the results of calculations of the mixing ratios , or mole fractions , for a number of important species at 1 atmosphere pressure over a range of temperatures .
it can be directly compared with figure 17 of @xcite .
the model can easily be expanded by adding new species to our thermochemical database .
our current model assumes true equilibrium chemistry in that when condensates form , the condensed phase is assumed to remain in equilibrium with the gas phase .
this assumption is commonly made in such models .
however , it may not be true in real cases since condensed material can fall under gravity and rain out of the system .
models that take account of this `` rainout '' process are described by @xcite and @xcite .
absorption lines due to rovibrational and electronic transitions of molecules are the most important features of the spectra of planets , brown dwarfs and the coolest stars . for use with vstar we have collected a line database that currently contains more than 2.9 billion lines . the line lists used with vstar are listed in table [ tab_linelists ] .
the line parameters are in a variety of different formats .
the essential data needed for all lines are the line position ( as wavelength or wavenumber ) , the line intensity , and the lower state energy of the transition ( ususally given in @xmath6 or ev ) .
line intensity is usually given in one of three forms .
hitran , geisa ( see section [ sec_higeis ] ) and similar lists give a line intensity @xmath7 in units of cm molecule@xmath8 ( sometimes given as @xmath9 mol@xmath8 @xmath6 ) at a reference temperature @xmath10 . for hitran @xmath10
is 296 k , but some other lists use a different reference temperature .
the line intensity at temperature @xmath11 is then calculated using @xcite : @xmath12 } { [ 1 - \exp(-c_2\nu_0/t_0 ) ] } \label{eqn_hitran}\ ] ] where @xmath13 is the line frequency in @xmath6 , @xmath14 is the lower state energy in @xmath6 , @xmath15 is the second radiation constant ( @xmath16 ) and @xmath17 is the partition function ( or total internal partition sum ) .
line intensities may also be quoted in the form of einstein a coefficients ( @xmath18 ) .
the line intensity s ( in cm / mol ) at a temperature t can be calculated from @xmath18 using @xcite .
@xmath19 here @xmath20 is the statistical weight of the upper level of the transition . in most cases
@xmath20 can be written as @xmath21 where j is the rotational quantum number of the upper state , and @xmath22 is the nuclear spin degeneracy .
different formulae are needed if hyperfine structure is included in the line list ( see * ? ? ?
the g@xmath23 used in equation [ eqn_acoef ] needs to be consistent with that used in calculating the partition function ( see section [ sec_part_nuc ] ) . in astronomical line lists , such as those of @xcite , the oscillator strength @xmath24 , usually tabulated as @xmath25 or @xmath26 ( where @xmath20 is the statistical weight )
is the usual form for listing line strengths .
the line intensity at temperature t can be calculated from @xmath25 using : @xmath27 where @xmath28 is the electron charge , @xmath29 is the electron mass and other symbols are as above . [ cols="<,<,>,<",options="header " , ] the marcs stellar atmosphere code @xcite is a good choice for a comparison with vstar , since it uses modern abundances and opacities , similar to those used in vstar , and has been used to generate a large grid of model atmospheres with detailed information available on the model structure .
the comparison shown here is for a plane parallel model with @xmath30 = 2500 k , g = 1000 m s@xmath31 ( @xmath32 = 5 in c.g.s units ) , and solar metallicity . to make the comparison we take the temperature as a function of gas pressure from the marcs model , and use this as the input for calculating an equivalent vstar model .
molecular species included in the vstar model were : h@xmath33o ( bt2 ) , co , cah , mgh , feh , crh , tio and vo
. lines of alkali metals were taken from vald @xcite , and other atomic species were from the kurucz line lists .
figure [ fig_marcs ] shows some comparisons of data from marcs and vstar .
these include the height above the lowest level modelled ( determined from hydrostatic balance ) , and the mixing ratios of several important species . on these plots crosses
are the values from marcs and the lines are the values from vstar . in most cases
the agreement is excellence .
the behaviour for tio is however quite different in the cooler layers .
this is because the vstar chemical model includes condensate formation , whereas marcs only includes gas phase chemistry .
small differences in the co and h@xmath33o mixing ratios arise from the same cause .
the difference in tio does not have much effect on the strength of the tio bands in the spectrum as these are mostly formed at deeper levels .
the flux spectrum from vstar is compared with that from marcs in the bottom right panel of figure [ fig_marcs ] .
the fluxes provided with marcs are described as `` rough estimates of the surface fluxes '' and `` are not synthetic spectra '' .
nevertheless it is clear that the values of the fluxes are very similar and the same absorption fetaures are present in both spectra at similar depths .
a comparison of similar models with observations of an m dwarf spectrum are presented later .
in this section we present a brief comparison of vstar models with observed spectra for a range of objects .
full details on these and other results will be given elsewhere , but they are presented here to show the wide range of astronomical objects that can be successfully modelled using vstar .
the spectra of solar system objects are from our own observations with the anglo - australian 3.9 m telescope at siding spring observatory , and its iris2 instrument @xcite , and from the nifs instrument @xcite on the gemini north 8 m telescope at mauna kea hawaii .
the brown dwarf and stellar spectra are taken from the nasa infrared telescope facility ( irtf ) spectral library @xcite and are taken with the spex instrument on irtf at a resolving power of r@xmath22000 .
the spectra considered here are all in the near - infrared spectral region ( 1 2.5 @xmath34 m ) . for late type
dwarfs this is where the flux peaks , and for spectroscopy of planets , this is a region containing many interesting rovibrational molecular bands . while this is the wavelength region vstar
has normally been applied to , there is nothing inherent in vstar that restricts it to this region .
vstar should , in principle , be usable from uv to microwave wavelengths .
figure [ fig_ss ] shows a number of comparisons of observed spectra with vstar models for solar system atmospheres .
the earth atmosphere transmission comparison ( top left panel ) is described in more detail in @xcite and @xcite and uses the observed spectrum at a resolving power of r ( @xmath35 ) @xmath2 2400 of the g5v star bs 996 with the 3.9 m anglo - australian telescope ( aat ) and its iris2 instrument .
this spectrum is compared with a model of a solar spectrum as seen through a modelled earth atmosphere transmission spectrum calculated using vstar . of
the spectral features only that near 1.28 @xmath34 m is a stellar line .
other features are atmospheric absorptions of o@xmath33 , co@xmath33 and h@xmath33o , with the strongest feature being the o@xmath33 a - x band at 1.27 @xmath34 m and its associated collision induced absorption .
the data and model agree to better than 1% .
the bottom left panel shows the reflected light spectrum of jupiter in the near - infrared k band ( 2.04 2.38 @xmath34 m ) also observed with the aat and iris2 . the vstar model used here
is described in more detail in @xcite .
the absoprtion features present in this spectral region are due to methane , and to the collision induced absoprtion of h@xmath33-h@xmath33 ( see section [ sec_cia ] ) , with the latter producing the broad smooth feature centred on 2.11 @xmath34 m .
stratospheric and tropospheric clouds are included in the model with the details of the cloud properties and optical depths given in @xcite .
the right hand panels of figure [ fig_ss ] show spectra of titan and uranus in the spectral region covering the 1.55 @xmath34 m methane `` window '' that lies between strong methane absoprtion bands .
modelling of the spectra of these objects at this wavelength using line - by - line methods has only become possible very recently with the availability of the latest laboratory spectral line data for methane as described in section [ sec_ch4 ] , in particular the data of @xcite .
the titan spectrum was obtained with the near infrared integral field spectrometer ( nifs ) on the gemini north 8 m telescope , and the data and vstar model are described more fully by @xcite .
the absorption lines in this region are mostly weak lines of ch@xmath36 , but there are also lines of ch@xmath37d and co. a model for titan s aerosols based on that of @xcite is also used in this analysis . the models enabled titan s d / h ratio and co abundance to be determined as described in @xcite .
the uranus comparison in the lower right panel uses data from iris2 on the aat .
the vstar model spectrum is calculated from the new low temperature methane line data as described in section [ sec_ch4 ] .
the model includes clouds with a mean particle size of 1 @xmath34 m , and with the main cloud layers being at 2 3 bars and at 6 10 bars , which is similar to cloud distributions derived by @xcite and @xcite .
while not shown here vstar has also been used successfully to model the spectra of other solar system planets .
vstar models for the venus night side are used in the analyses described by @xcite and @xcite 1404489 with a model spectrum calculated with vstar for @xmath30 = 1100k and @xmath32 = 5 .
the observed spectrum is offset from the model upwards by 1 ( @xmath38 10@xmath39),width=340 ] .
the observed spectra are offset from the model upwards by 0.5 ( @xmath38 10@xmath39),width=340 ] figure [ fig_bd ] shows an observed spectrum of the t4.5 dwarf 2mass j05591915@xmath401404489 taken from the irtf spectral library @xcite compared with a vstar model spectrum .
the vstar spectra used for this comparison are based on brown dwarf model p - t structures taken from figures 8 and 9 of @xcite .
the vstar models included molecular absorption due to h@xmath33o ( bt2 ) , co , ch@xmath36 , cah , mgh , crh , feh , tih .
lines of alkali metals are included from vald @xcite .
the far wings of very strong sodium and potassium lines in the visible have a significant effect on the shape of brown dwarf spectra in the 1 @xmath34 m region .
we used a @xmath41 factor model for the far wing shapes of these lines with an exponentially decrease between 500 and 7500 @xmath6 from the line centre , with the parameters adjusted to provide a good match to the data .
more physically based models for the far wing shapes of alkali metal lines are described by @xcite and @xcite .
a number of models were tried with different temperatures and gravities , and best agreement with the data was found for a model with @xmath30 = 1100 k , g = 1000 m s@xmath31 ( @xmath32 = 5 in c.g.s .
units ) and solar metallicity .
this is in reasonable agreement with other results for this object.@xcite find @xmath30 = 1200 k , @xmath32 = 4.5 while @xcite find @xmath30 = 1002 k , @xmath32 = 4.9 using high resolution spectra .
the model spectrum agrees well with the observed spectrum except in the wavelength region from around 1.6 1.7 @xmath34 m .
as discused in section [ sec_ch4 ] , the methane line lists currently available do not include hot bands in this region and therefore underestimate the total absorption , and incorrectly model the observed structure .
this problem exists with all models for t - dwarf spectra .
there is also a discrepancy between the model and observations in the 2.1 @xmath34 m region where the model is slightly too low .
in other wavelength regions , where methane absorption is not significant , the agreement between model and observations is excellent .
what might , at first sight , appear to be noise on the spectra , is in fact spectral structure due to many absorption lines .
this is apparent from figure [ fig_bdh ] where two wavelength regions ( 1.1 1.27 @xmath34 m and 1.4 1.58 @xmath34 m ) are shown on an expanded scale .
the model used here did not include dust absorption and scattering .
this is acceptable for objects as late as t4.5 . for earlier type brown dwarfs , and
in particular the l - dwarf class , the effects of dust are important . while vstar is quite capable of modelling the radiative effects of dust in such systems , there are currently considerable diferences in the assumptions about dust properties used in different models .
some models use large ( @xmath2100 @xmath34 m ) dust particles @xcite while others use small ( sub-@xmath34 m ) particles @xcite .
for the purposes of this paper we have avoided the extra complexity of dusty brown dwarf models .
our models also assume equilibrium chemistry .
some brown dwarf models include a non - equlibrium treatment of the chemistry of some species such as ch@xmath36 , co and h@xmath33o @xcite which allows for the effects of vertical mixing in the atmosphere .
while this could be implemented as a modification to our chemical model we have not done so for the results presented here .
the effects on the near - ir spectrum appear to be small @xcite with more significant effects at 4
14 @xmath34 m wavelengths
. = 2700k and @xmath32 = 5 .
the observed spectrum is offset from the model upwards by 3 ( @xmath38 10@xmath42),width=340 ] .
the observed spectra are offset from the model upwards by 0.4 ( @xmath38 10@xmath42 ) .
the upper panel shows primarily absorption in the e - a band of feh .
the lower panel shows absorption due to co and h@xmath33o.,width=340 ] .
figure [ fig_vb10 ] shows an observed spectrum of the m8 dwarf vb10 ( gl 752b ) taken from the irtf spectral library @xcite compared with a model spectrum calculated with vstar .
the model structure was taken from the marcs grid of model atmospheres @xcite and is a model for @xmath30 = 2700 k , g = 1000 m s@xmath31 ( @xmath32 = 5 in c.g.s .
units ) and solar metallicity .
in other respects the model is similar to that described in section [ sec_marcs ] .
a number of different effective temperatures were tested with the 2700 k model providing the best match to the observed spectrum .
figure [ fig_vb10h ] shows expanded views of two regions of the spectrum .
the upper panel shows the 1.58 1.78 @xmath34 m region .
this is dominated by absorption lines in the e - a band of feh , which are included in our model using the line list of @xcite .
this band is essential to get a good model of the spectrum in this region .
however , including this line list results in a small step in the modelled spectrum at 1.582 @xmath34 m ( visible in figure [ fig_vb10 ] ) which is the shortest wavelength included in this list .
this step is not seen in the observed spectrum , and suggests that there is more feh absorption at shorter wavelengths that should be included .
the lower panel of figure [ fig_vb10h ] is 2.25 2.5 @xmath34 m region showing that the model does a good job of representing the detailed structure of the co and h@xmath33o bands in this region . as for the brown dwarf case
we have used a dust free model .
it is quite possible that dust is present in such a late type m dwarf , and its inclusion might change our conclusions about the best fitting effective temperature .
the upper limit on the layer temperature for modelling stellar atmospheres using vstar is currently 6000 k set by the use of data from the janaf tables in our chemical model . since a model usually requires a range of layer temperatures extending to about a factor of two above and below the effective temperature of the star being modelled , the maximum effective temperature is around 3000 k.
we have desribed the techniques used in the vstar code to calculate model spectra for objects including solar - system planets , brown dwarfs and m dwarfs . while there are other codes used for modelling the spectra of these objects , vstar has a number of unique features . *
it is capable of being used for a very wide range of objects , ranging from the coolest solar system planets ( uranus with layer temepratures down to 60 k ) to stars with layer temperatures up to 6000 k. * it has been tested by comparison with both the earth atmosphere radiative transfer code ( rfm ) and a stellar atmosphere code ( marcs ) . * not only is a rigorous approach to radiative transfer used , but this has been tested against radiative transfer benchmarks and shown to be accurate to levels of 10@xmath43 to 10@xmath44 .
an obvious application of vstar is to the modelling of exoplanet spectra .
vstar has the proven ability to model both cool planets in our own solar system , and hotter objects such as brown dwarfs and m dwarfs , and thus covers the full temperature range expected to be encountered in exoplanets .
the full treatment of the angular dependence of scattering incorporated in vstar means that it can be used to predict the phase dependence of an exoplanet spectrum around its orbital cycle .
work is currently in progress on extending vstar to include polarized radiative transfer .
the phase variation of polarization can potentially provide important information about an exoplanet s atmosphere or surface @xcite based on observations obtained at the gemini observatory which is operated by the association of universities for research in astronomy , inc . under a cooperative agreement with the nsf on behalf of the gemini partnership : the national science foundation ( united states ) , the science and technology facilities council ( united kingdom ) , the national research council ( canada ) , conicyt ( chile ) , the australian research council ( australia ) , ministerio da cienca e tecnologia ( brazil ) and ministerio de ciencia , tecnologia e innovacion productiva ( argentina ) .
the titan observations were obtained as part of the system verification of the nifs instrument under program gn-2006a - sv-128 .
we thank the staff of the australian astronomical observatory for assistance in obtaining observations with the anglo - australian telescope .
albert , s. , bauerecker , s. , boudon , v. , brown , l.r . , champion , j .-
p . , loete , m. , nikitin , a. & quack , m. , 2009 , chem .
phys . , 356 , 131 .
allard , f. , hauschildt , p.h .
, schwenke , d. , 2000 , apj , 540 , 1005 .
allard , f. , hauschildt , p.h . ,
alexander , d.r . ,
tamanai , a. , schweitzer , a. , 2001 , apj , 556 , 357 .
allard , n.f . , allard , f. , hauschildt , p.h .
, kielkopf , j.f . ,
machin , l. , 2003 , a&a , 411 , 473 .
asplund , m. , grevesse , n. , sauval , a.j . , scott , p. , 2009
, ara&a , 47 , 481 .
barber , r.j . , tennyson , j. , harris , g.j . ,
tolchenov , r.n .
, 2006 , mnras , 368 , 1087 .
bailey , j. , 2006 , in forget , f. et al .
, ( eds ) proceedings mars atmosphere modelling and observations workshop , granada , 148 , laboratoire meteorologie dynamique , paris .
bailey , j. , 2007 , astrobiology , 7 , 320 .
bailey , j. , simpson , a. , crisp , d. , 2007 , pasp , 119 , 228 .
bailey , j. , ulanowski , z. , lucas , p.w . ,
hough , j.h .
, hirst , e. , tamura , m. , 2008a , mnras , 386 , 1016 .
bailey , j. , meadows , v.s . , chamberlain , s. , crisp , d. , 2008b , icarus , 197 , 247 .
bailey , j. , 2009 , icarus , 201 , 444 .
bailey , j. , ahlsved , l. , meadows , v.s . , 2011 ,
icarus , 213 , 218 .
barin , i. , 1995 , thermochemical data of pure substances ( 3rd ed ) ; vch - verlag , weinheim barman , t.s . ,
hauschildt , p.h . ,
allard , f. , 2001 , apj , 556 , 885 .
bell , k.l .
, 1980 , j. phys .
b , 13 , 1859 .
bell , k.l .
& berrington , k.a . , 1987 , j. phys .
b , 20 , 801 .
borysow , a. , 2002 , a&a , 390 , 779 .
borysow , a. & frommhold , l. , 1989 , apj , 341 , 549 .
borysow , a. , frommhold , l. , massimo , m. , 1989 , apj , 336 , 495 .
borysow , a. , jrgensen , u.g . ,
zheng , c. , 1997 , a&a , 324 , 185 .
borysow , a. , jrgensen , u.g . , fu , y. , 2001 , j. quant .
trans . , 2001 , 68 , 235 .
borysov , a. , champion , j.p . ,
jrgensen , u.g . , wenger , c. , 2002 , mol .
, 100 , 3585 .
boudon , v. , rey , m. , loete , m. , 2006 , j. quant .
trans . , 98 , 394 .
brown , l.r .
, 2005 , j. quant .
, 96 , 251 .
brown , l.r . & peterson , d.b . , 1994 ,
spectrsoc . , 168 , 593 .
burch , d.e . ,
singleton , e.b . ,
williams , d. , 1962 , appl . opt .
, 1 , 359 .
burch , d.e .
, gryvnak , d.a . ,
patty , r.r . ,
bartky , c.e . , 1969 ,
, 59 , 267 . burrows , a. & sharp , c.m . , 1999 ,
apj , 512 , 843 .
burrows , a. , volobuyev , m. , 2003 , apj , 583 , 985 .
burrows , a. , ram , r.s . , bernath , p. , sharp , c.m .
, milsom , j.a . , 2002 , apj , 577 , 986 .
burrows , a. , sudarsky , d. , hubeny , i. , 2003 , apj , 588 , 1121 .
burrows , a. , dulick , m. , bauschlicher , c.w . , bernath , p.f . , ram , r.s . , sharp , c.m . , milson , j.a .
, 2005 , apj , 624 , 988 .
burrows , a. , sudarsky , d. , hubeny , i. , 2006 , apj , 640 , 1063 .
cahoy , k.l . ,
marley , m.s .
, fortney , j.j . , 2010 , apj , 724 , 189 .
campargue , a. , wang , l. kassi , s. , mast , m. , ondrej . ,
v. , 2010 , j. quant .
spectrosc . radiat .
trans . , 111 , 1141 .
chandrasekhar , s. , 1960 , radiative transfer , dover , new york .
chase , m.w .
, 1998 , nist - janaf thermochemical tables , j. phys .
ref . data monograph no .
9 . cotton , d.v . ,
bailey , j. , crisp , d. , meadows , v.s .
, 2011 , icarus , in press .
cox , a.n . , 2000 , allen s astrophysical quantities , 4th edition , springer - verlag , new york .
cushing , m.c . ,
rayner , j.t .
, vacca , w.d . , 2005 , apj , 623 , 1115 .
cushing , m.c .
et al . , 2006 ,
apj , 648 , 614 .
dalgarno , a. & williams , d.a . , 1962 ,
apj , 136 , 690 . de bergh , c. et al . , 2011 , planet .
space sci .
, in press doi:10.1016/j.pss.2011.05.003 .
delaye , c. , hartmann , j .-
m . , taine , j. , 1989 , appl .
, 28 , 5080 .
del burgo , c. , martin , e.l . , zapatero osorio , m.r . ,
hauschildt , p.h . , 2009 , a&a , 501 , 1059 .
dulick , m. , bauschlicher , c.w . , burrows , a. , sharp , c.m . , ram , r.s . ,
bernath , p. , 2003 ,
apj , 594 , 651 .
edwards , d.p . , 1992 , ncar technical note ncar / tn-367+str , national centre for atmospheric research , boulder , colorado .
fischer , j. , gamache , r.r . ,
goldman , a. , rothman , l.s .
, perrin , a. , 2003 , j. quant . spectrosc . radiat .
trans . , 82 , 401 .
fortney , j.j . ,
marley , m.s . , lodders , k. , saumon , d. , freedman , r. , 2005 , apj , 627 , l69 .
freedman , r.s . ,
marley , m.s . ,
lodders , k. , 2008 , apjs , 174 , 504 .
gamache , r.r . ,
lynch , r. , brown , l.r .
, 1996 , j. quant .
trans . , 56 , 471 .
garcia , r.d.m .
& siewert , c.e . , 1985 ,
transport theory and statistical physics , 14 , 437 .
gao , b. , kassi , s. , campargue , a. , 2009 , j. mol .
spectrosc . , 253 , 55 .
goody , r. , west , r. , chen , l. , crisp , d. , 1989 , j. quant .
, 42 , 539 .
goorvitch , d. , 1994 , apjs , 95 , 535 .
gray , d.f .
, 2005 , the observation and analysis of stellar photospheres ( 3rd ed ) ; cambridge univ . press , cambridge grevesse , n. , asplund , m. , sauval , a.j . , 2007 , space sci .
, 130 , 115 .
gustaffson , b. , edvadrsson , b. , eriksson , k. , jrgensen , u.g . , nordlund , a. , plez , b. , 2008 , a&a , 486 , 951 .
hansen , j.e . & travis , l.d .
, 1974 , space sci .
, 16 , 527 . hargreaves , r.j . ,
hinkle , k.h .
, bauschlicher , c.w . ,
wende , s. , seifahrt , a. , bernath , p.f .
, 2010 , aj , 140 , 919 .
harris , g.j . ,
tennyson , j. , kaminsky , b.m . , pavlenko , ya . v. , jones , h.r.a .
, 2006 , mnras , 367 , 400 .
hartmann , j.m .
, rosenmann , l. , perrin , m.y . , taine , j. , 1988 , appl .
, 27 , 3063 .
henyey , l.g . & greenstein , j.l . , 1941 , apj , 93 , 70 .
homeier , d. , hauschildt , p.h . ,
allard , f. , 2003 , in hubeny , i. , mihalas , d. , werner , k. ( eds ) stellar atmosphere modelling , asp conf .
, 288 , p357 , astronomical society of pacific , san francisco .
hubeny , i. , burrows , a. , 2007 , apj , 669 , 1248 .
humlicek , j. , 1982 , j. quant .
trans . , 27 , 437 .
irwin , a.w . , 1981 , apjs , 45 , 621 .
irwin , a.w .
, 1988 , a&as , 74 , 145 .
irwin , p.g.j . ,
sromovsky , l.a . , strong ,
sihra , k. , teanby , n.a . , bowles , n. , calcutt , s.b . ,
remedios , j.j . , 2006 ,
icarus , 181 , 309 .
irwin , p.g.j . ,
teanby , n.a . ,
davis , g.r .
, 2007 , apj , 665 , l71 .
jacquinet - husson , n. et al .
, 2008 , j. quant .
trans . , 109 , 1043 .
jones , h.r.a . ,
pavlenko , y. , viti , s. , tennyson , j. , 2003 , in brown , a. , harper , g.m . & ayres , t.r .
( eds ) the future of cool - star astrophysics : 12th cambridge workshop on cool stars , stellar systems , and the sun , p899 , univ . of colorado .
jrgensen , u.g . , 1994 ,
a&a , 284 , 179 .
jrgensen , u.g . ,
larsson , m. , 1990 , a&a , 238 , 424 .
jrgensen , u.g . ,
larsson , m. , iwamae , a. , yu , b. , 1996 , a&a , 315 , 204 .
jrgensen , u.g . , jensen , p. , srensen , g.o .
, aringer , b. , 2001 , a&a , 372 , 249 .
karkoschka , e. & tomasko , m.g . , 2010 ,
icarus , 205 , 674 .
kedziora - chudczer , l. , and bailey , j. , 2011 , mnras , 414 , 1483 .
kupka , f. , piskunov , n. , ryabchiokva , t.a . , stempels , h.c . , weiss , w.w .
, 1999 , a&as , 138 , 119 .
kurucz , r. , 1970 , atlas : a computer program for calculating model stellar atmospheres , sao special report 309 , smithsonian astrophysical observatory , cambridge , mass .
kurucz , r. , 2005 , mem s. a. it .
suppl . 8 , 86 .
kurucz , r. & avrett , e.h . , 1981 , solar spectrum synthesis i. a sample atlas from 224 to 300 nm , sao special report 391 , smithsonian astrophysical observatory , cambridge , mass .
laraia , a.l . ,
gamache , r.r .
, lamouroux , j. , gordon , i.e. , rothman , l.s . , 2011 , icarus , 215 , 391 .
lenoble , j. ( ed ) , standard procedures to compute atmopsheric radiative transfer , national centre for atmospheric research , boulder , colorado .
liou , k.n .
, 2002 , an introduction to atmospheric radiation , academic press , san diego lodders , k. , 1999 , j. phys .
chem . ref .
data , 28 , 1705 .
lodders , k. , 2004 , j. phys .
chem . ref .
data , 33 . 357 .
lodders , k. & fegley , b. , 2002 , icarus , 155 , 393 .
lodders , k. , palme , h. , gail , h .-
, 2009 , arxiv:1001.4635 .
marcy , g.w . ,
butler , r.p . ,
williams , e. , bildsten , l. , graham , j.r . ,
ghez , a.m. , jernigan , j.g .
, 1997 , apj , 481 , 926 .
margolis , j.s .
, 1990 , appl .
29 , 2295 .
marley , m.s .
& mckay , 1999 , icarus , 138 , 268 .
marley , m.s .
, seager , s. , saumon , d. , lodders , k. , ackerman , a.s . ,
freedman , r.s . ,
fan , x. , 2002 , apj , 568 , 335 mayer , j.e .
& mayer , m.g . , 1940 statistical mechanics , wiley , new york .
mayor , m. & queloz , d. , 1995 , nature , 378 , 355 .
mckay , c.p . ,
pollack , j.b . ,
courtin , r. , 1989 , icarus , 80 , 23 .
mckellar , a.r.w . , 1989 , icarus , 80 , 361 .
mcgregor , p.j .
et al . , 2003 , spie proc . , 4841 , 1551 .
meadows , v.s . , crisp , d. , 1996 , j. geophys .
res . , 101 , 4595 .
mishchenko , m.i . ,
1991 , j. opt .
, 8 , 871 .
mishchenko , m.i . ,
travis , l.d . ,
lacis , a.a . , 2002 , scattering , absorption and emission of light by small particles , cambridge university press , cambridge mondelain , d. , kassi , s. , wang , l. , campargue , a. , 2011 , phys .
phys . , 13 , 7985 .
nakajima , t. , oppenheimer , b.r . ,
kulkarni , s.r .
, golimowski , d.a . ,
matthews , k. , durrance , s.t .
, 1995 , nature , 378 , 463 .
nikitin , a.v .
et al . , 2010 , j. quant .
trans . , 111 , 2211 .
nikitin , a.v .
et al . , 2011 , j. quant .
trans . , 112 , 28 .
nouri , s. , orphal , j. , aroui , h. , hartmann , j .-
m . , 2004 , j. mol .
spectrosc . , 227 , 60 .
old , j.g .
, gentili , k.l . , peck , e.r .
, 1971 , j. opt .
a. , 61 , 89 .
oppenheimer , b.r . ,
kulkarni , s.r . ,
matthews , k. , nakajima , t. , 1995 , science , 270 , 1478 .
partridge , h. , & schwenke , d.w .
, 1997 , j. chem .
phys . , 106 , 4618 .
pavlenko , ya .
v. , jones , h.r.a . ,
lyubchik , yu . ,
tennyson , j. , pinfield , d.j .
, 2006 , a&a , 447 , 709 .
peck , e.r . & reeder , k. , 1972 , j. opt .
, 62 , 950 .
perrin , m.y .
& hartmann , j.m . , 1989 , j. quant .
trans . , 42 , 311 .
pine , a.s .
, 1992 , j. chem .
phys . , 97 , 773 .
pine , a.s .
& gabard , t. , 2003 , j. mol .
spectrosc . , 217 , 105 .
piskunov , n.e . ,
kukpa , f. , ryabchikova , t.a . ,
weiss , w.w . , jeffery , c.s . , 1995 , a&as , 112 , 525 .
plez , b. , 1998 , a&a , 337 , 495 .
pollack , j.b . ,
dalton , j.b . ,
grinspoon , d. , wattson , r.b . ,
freedman , r. , crisp , d. , allen , d.a .
, bezard , b. , de bergh , c. , giver , l.p . , ma , q. , tipping , r. , 1993 , icarus,103 , 1 .
querci , f. , querci , m. , kunde , v.g . , 1971 , a&a , 15 , 256 .
querci , f. , querci , m. , tsuji , t. , 1974 , a&a , 31 , 265 .
rayner , j.t . ,
cushing , m.c . ,
vacca , w.d . , 200 , apjs , 185 , 289 .
rgalia - jarlot , l. , thomas , x. , von der heyden , p. , barbe , a. , 2005 , j. quant .
, 91 , 121 .
robert , o. , hilico , j .- c . ,
loete , m. , champion , j .-
p . , brown , l.r .
, 2001 , j. mol .
spectrosc . , 209 , 14 .
robie , r.a .
& hemingway , b.s . , 1995 , us geological survey bull .
rothman , l.s .
, wattson , r.b . ,
gamache , r. , schroeder , j.w . ,
mcann , a. , 1995 , in daintey , j.c .
( ed ) , atmospheric propagation and remote sensing iv .
spie proc .
2471 , 105 , spie , bellingham , washington .
rothman , l.s .
et al . , 1998 ,
trans . , 60 , 665 .
rothman , l.s .
et al . , 2005 , j. quant .
trans . , 96 , 139 .
rothman , l.s .
et al . , 2009 , j. quant .
trans . , 110 , 533 .
rothman , l.s .
et al . , 2010 , j. quant .
trans . , 111 , 2139 .
saumon , d. , marley , m.s . , lodders , k. , freedman , r.s . , 2003 , in martin , e.l .
( ed ) , `` brown dwarfs '' , iau symp , 211 , astronomical society of the pacific , san francisco , p. 345 .
saunders , r. et al .
, 2007 , j. geophys .
res . , 112 , d01s90 .
sauval , a.j . & tatum , j.b . , 1984 ,
apjs , 56 , 193 .
schreier , f. , 1992 , j. quant .
spectrosc . radiat .
trans . , 27 , 437 .
schwenke , d. , 1998 , faraday discussions , 109 , 321 .
schwenke , d. , partridge , h. , 2000 , j. chem .
113 , 129 .
seager , s. , whitney , b.a . ,
sasselov , d.d . , 2000 , apj , 540 , 504 .
seiff , a. et al . , 1985 , adv .
space res . , 5 , 3 .
sharp , c.m .
& huebner , w.f . , 1990 ,
apj , 72 , 417 .
sharp , c.m . &
burrows , a. , 2007 , apjs , 168 , 140 .
simeckov , m. , jacquemart , d. , rothman , l.s . ,
gamache , r.r . ,
goldman , a. , 2006 , j. quant .
trans . , 98 , 130 .
skory , s. , weck , p.f . ,
stancil , p.c . ,
kirby , k. , 2003 , apjs , 148 , 599 .
smith , k.m .
& newnham , d.a . , 1999 ,
lett . , 308 , 1 .
smith , k.m .
& newnham , d.a . , 2000 , j. geophys .
res . , 105 , d6 , 7383 .
sneep , m. & ubachs , w. , 2005 , j. quant .
trans . , 92 , 293 .
sromovsky , l.a . , irwin , p.g.j . , fry , p.m. , 2006 , icarus , 182 , 577 .
stam , d.m .
, 2008 , a&a , 482 , 989 .
stamnes , k. , tsay , s.c . , wiscombe , w. , jayaweera , k. , 1988 , appl .
, 27 , 2502 .
stephens , d.c .
et al . , 2009 , apj , 702 , 154 .
strong , k. , taylor , f.w . ,
calcutt , s.b . ,
remedios , j.j . , ballard , j. , 1993 , j. quant .
trans . , 50 , 363 .
swain , m.r . ,
vasisht , g. , tinnetti , g. , 2008 , nature , 452 , 329 .
swain , m.r . , vasisht , g. , tinetti , g. , bouwman , j. , chen , p. , yung , y. , deming , d. , deroo , p. , 2009a
, apjl , 690 , l114 .
swain , m.r .
et al . , 2009b , apj , 704 , 1616 .
takano , y. & liou , k .-
n . , 1989 , j. atmos .
tashkun , s.a . ,
perevalov , v.i . ,
teffo , j - l . ,
bykov , a.d . ,
lavrentieva , n.n . , 2003 ,
trans . , 82 , 165 .
tinney , c. , et al . , 2004 , spie proc . , 5492 , 998 .
tjemkes , s.a .
et al . , 2003 , j. quant .
trans . , 77 , 433 .
tomasko , m.g . ,
doose , l. , engel , s. , dafoe , l.e . , west , r. , lemmon , m. , karkoschka , e. , see , c. , 2008 , plan .
space sci .
, 56 , 669 .
tonkov , m.v .
et al . , 1996 ,
, 35 , 4863 .
tsuji , t. , 1973 , a&a , 23 , 411 .
tsuji , t. , ohnaka , k. , aoki , w. , nakajima , t. , 1996 , a&a , 308 , l29 .
van vleck , j.h . & weisskopf , v.f . , 1945 , rev .
phys . , 17 , 227 .
vidler , m. & tennyson , j. , 2000 , j. phys .
, 113 , 9766 .
voronin , b.a . ,
tennyson , j. , tolchenov , r.n . ,
lugovskoy , a.a . , yurchenko , s.n .
, 2010 , mnras . 402 . 492 .
wang , l. , kassi , s. , campargue , a. , 2010 , j. quant .
, 111 , 1130 .
wang , l. , kassi , s. , liu , a.w . , hu , s.m . , campargue , a. , 2011 , j. quant .
trans . , 112 , 937 .
waterman , p.c .
, 1971 , phys .
d. , 3 , 825 .
weck , p.f . , schweitzer , a. , stancil , p.c . ,
hauschildt , p.h . ,
kirby , k. , 2003 , apj , 582 , 1059 .
wenger , c. & champion , j.p . , 1998 , j. quant .
trans . , 59 , 471 .
wenger , c. , champion , j.p .
, boudon , v. , 2008 , j. quant .
trans . , 109 , 2697 .
wishart , a.w . , 1979 ,
mnras . 187
young , a.t . , 1981 , appl . opt .
, 20 , 533 .
yamamura , i. , tsuji , t. , tanabe , t. , 2010 , apj , 722 , 682 .
yurchenko , s.n . ,
barber , r.j . , yachmenev , a. , thiel , w. , jensen , p. , tennyson , j. , 2009 , j. phys .
, 113 , 11845 .
yurchenko , s.n . ,
barber , r.j . ,
tennyson , j. , 2011 , mnras , 413 , 1828 .
zugger , m. , kasting , j.f . , williams , d.m . , kane , t.j .
, philbrick , c.r . , 2010 ,
apj , 723 , 1168 . | we describe a new software package capable of predicting the spectra of solar - system planets , exoplanets , brown dwarfs and cool stars .
the versatile software for transfer of atmospheric radiation ( vstar ) code combines a line - by - line approach to molecular and atomic absorption with a full multiple scattering treatment of radiative transfer .
vstar is a modular system incorporating an ionization and chemical equilibrium model , a comprehensive treatment of spectral line absorption using a database of more than 2.9 billion spectral lines , a scattering package and a radiative transfer module .
we test the methods by comparison with other models and benchmark calculations .
we present examples of the use of vstar to model the spectra of terrestrial and giant planet in our own solar system , brown dwarfs and cool stars .
[ firstpage ] radiative transfer techniques : spectroscopic planets and satellites : atmospheres stars : atmospheres brown dwarfs . |
the hadron colliders , due to the biggest center - of - mass energy and their relatively compact sizes , still remain a main tool for discoveries of very heavy particles .
thus , in 1983 the two dedicated experiments ua1 @xcite and ua2 @xcite discovered the intermediate vector bosons at the cern sps collider .
one faces , however , a very large background from the strong interactions . in any case , besides the simple manifestation of the existence of the weak bosons , one needs a precise study of their properties following from the standard model ( sm ) .
this task has been excellently fulfilled by the large electron - positron ( lep ) storage ring at cern and the stanford linear collider ( slc ) at slac . however , the masses of the @xmath0 quark and the undiscovered yet higgs boson happened to be too high to be discovered at these colliders .
i remember the words by samuel ting at one of the lep meetings in defence of continuation of the lep running : `` each collision at the lepton colliders is an event , while it is a background at the hadron colliders '' .
so , the precision of the electroweak measurements at the lepton colliders was so high , that the predicted from the radiative loop corrections mass of the top - quark @xmath1 gev @xcite has been found in agreement and with a comparable accuracy of its first direct measurements at the fermilab tevatron by the cdf @xcite @xmath2 gev and the d0 @xcite @xmath3 gev collaborations .
nevertheless , in spite of the overwhelming background for the top - quark pair production by the strong interactions at the hadron collider , the uncertainty of the top - quark mass @xmath4 gev @xcite is considerably reduced at present . moreover ,
recently , the evidence for a single top - quark production @xcite through the weak interactions and the direct measurement of @xmath5 at the fermilab tevatron hadron collider became possible .
another achievement in precise measurements at the hadron collider is the @xmath6-mass measurement @xmath7 gev @xcite by the cdf collaboration at a comparable with the lep experiments accuracy , which represents the single most precise measurement to date .
all these measurements will allow to constrain further the mass of the higgs particle , which discovery is a priority task of the running tevatron and the large hadron collider ( lhc ) .
the discovery of the theoretically predicted heavy particles and the establishment of the sm without any surprises are characteristic for the experimental high energy physics during the last thirty years .
therefore , the lhc construction is connected not only with the higgs discovery , but with the hope to find the physics beyond the sm . up to
now it is not clear what kind of physics it will be . therefore , any inputs like constraints on the new physics from low - energy precise experiments or from the presently most powerful tevatron collider at fnal are badly needed when discussing the properties of future colliders , in particular , the international linear collider ( ilc ) .
this talk is dedicated to the energy requirements for the future lepton colliders , which follow from the constraints on the new boson production at the tevatron . in order to investigate the properties of the new bosons and eventually to distinguish among different models of the new physics
, the energy of the future icl should be enough to produce them .
although it is still possible to investigate some properties of the new bosons at low - energy , we will consider the case of their resonance or threshold production , as an optimal possibility .
in the second part of the talk we will consider one of the possible scenarios of new physics in the boson sector , for which some confirmation from the tevatron data already exists .
a quantitative model of such a new physics will be very valuable in interpreting the data from the hadron colliders , tevatron and lhc , that presents concrete requirements for the ilc energy design .
let us start with the case of new neutral massive bosons , @xmath8 , which can be produced at the lepton colliders as resonances .
such a type of bosons is very difficult to detect in the low - energy experiments due to the huge background from the electromagnetic interactions .
some guiding principle is necessary to distinguish them from the known interactions .
for example , the neutral weak currents were detected in the deep - inelastic electron scattering through the measurements of @xmath9-odd quantities .
therefore , we expect that direct constraints from the high - energy hadron colliders should be more restrictive .
moreover , up to now , the drell - yan process with high - energy invariant mass of the lepton pairs remains the most clear indication of the heavy boson production at the hadron colliders .
therefore , the constraints from these investigations can be directly applied to the resonance boson production at the lepton colliders .
so , using only a modest integrated luminosity of 200 pb@xmath10 collected during run ii , the d0 collaboration puts tight restrictions on the @xmath8 masses for the different models from the di - electron events @xcite : @xmath11 gev , @xmath12 gev , @xmath13 gev , @xmath14 gev and @xmath15 gev .
a comparable statistics in the di - muon channel leads approximately to the same constraint @xmath16 gev @xcite .
the cdf constraints from the di - electron channel are based on more data , 1.3 fb@xmath10 , which lead to tighter restrictions @xcite : @xmath17 gev , @xmath18 gev , @xmath19 gev , @xmath20 gev and @xmath21 gev .
another possible channel , which can indicate the production of the neutral heavy bosons , is their hadronic decay into @xmath22 pairs . while the light quark decay channels are swamped by multijet background ,
the @xmath22 pairs can be detected , for example , through their decays into two energetic @xmath23-jets and two @xmath6 s , where one @xmath6 boson decays hadronically and one leptonically .
although the constraints from this channel can not be applied directly to the energy requirements for the lepton collider due to the possible leptophobic character of the bosons , it is interesting to detect the eventual peaks in the tevatron data .
so the latest results both of the d0 @xcite and of the cdf @xcite collaborations show some excess in the invariant mass distributions around 700 gev ( fig .
[ fig : mttbar ] ) . a possible explanation of this excess will be discussed in the next section . in comparison with the predicted leptophobic topcolor @xmath8 cross section ( left panel from @xcite
d0 data , right panel from @xcite cdf data of 680 pb@xmath10).,title="fig : " ] in comparison with the predicted leptophobic topcolor @xmath8 cross section ( left panel from @xcite
d0 data , right panel from @xcite cdf data of 680 pb@xmath10).,title="fig : " ] let us consider the case of the new heavy charged bosons , generically noted by @xmath24
. they could be produced at the lepton colliders only in pairs or in association with other charged boson , like @xmath6 .
therefore , restrictions on their masses lead to the following energy requirements for their threshold production @xmath25 at the lepton colliders . here again we will consider leptonic and hadronic channels of their decays . the leptonic decay of the new heavy charged boson into high - energy pair of a lepton and a corresponding antineutrino is the most clear signature of its production at the hadron colliders .
so , already from 205 pb@xmath10 of run ii data , the cdf collaboration obtained a tight constraint on possible @xmath24 mass @xmath26 gev @xcite .
the most rigid constraint comes from the d0 collaboration @xcite @xmath27 gev , based on bigger statistics , 900 pb@xmath10 , and better calorimetry than the cdf detector .
the hadronic decay of the new heavy charged boson into a @xmath28 pair of a heavy @xmath23 quark and a short living @xmath0 quark with its subsequent decay to @xmath29 pair allows to make jet @xmath23-tagging , where one of the jets must have a displaced secondary vertex .
a search for the intermediate heavy bosons in this channel has been fulfilled by both the d0 and cdf collaborations , and for this purpose the part of the same data sets of the single top production analysis has been used . owing to boson high masses
this analysis is even simpler than the single top production searches , because at such energies the background is considerably reduced .
so the d0 collaboration , based on 230 pb@xmath10 of integrated luminosity , puts the following constraints on the @xmath24 mass depending on the model : @xmath30 gev , @xmath31 gev , and @xmath32 gev @xcite .
the cdf constraints are tighter ( fig .
[ fig:23j_obs ] ) : @xmath33 gev for @xmath34 and @xmath35 gev for @xmath36 , since they are r0.5 based on 955 pb@xmath10 @xcite .
taking into account the most stringent constraints from the tevatron data , we can conclude , that in order to produce the heavy charged boson in association with the @xmath6 boson or the heavy neutral boson a lepton collider with energy above 1 tev is necessary .
also it is interesting to note the presence of some excesses in the observed data in figs .
[ fig : mttbar ] and [ fig:23j_obs ] , which we will discuss in the next section .
additional chiral bosons , which have anomalous interactions with fermions , were proposed in @xcite .
an exchange through these bosons leads to effective tensor interactions with the coupling constant by two orders of magnitude smaller than @xmath37 .
this follows from the precise low - energy experiments of the radiative pion decay @xcite . assuming the universality of these interactions we can explain the long standing discrepancy between the two pion production in the @xmath38 annihilation and the @xmath39 decay @xcite , which now reaches 4.5@xmath40 @xcite .
the universality of the interactions of the new bosons and the hypothesis about a dynamical generation of their kinetic terms allow to predict their masses @xcite .
due to the mixing between two charged chiral bosons the lightest state corresponds to @xmath41-boson with a mass @xmath42 gev and the heaviest one is @xmath43-boson with a mass @xmath44 gev .
the neutral physical states come as @xmath45-even @xmath46 and @xmath45-odd @xmath47 bosons with approximately r0.5 the same masses @xmath48 gev , which couple only to the up fermions , and analogous but heavier bosons @xmath49 and @xmath50 with a common mass @xmath51 gev , coupling to the down fermions . due to the anomalous interactions the angular distribution of the chiral boson decays differs drastically from the analogous distribution of the gauge bosons .
this leads to a specific transverse momentum distribution @xcite , which has a broad smooth bump with a maximum below the kinematical endpoint @xmath52 , instead of a sharp jacobian peak ( fig .
[ fig:3 ] ) .
the form of the decay distribution for the chiral bosons resembles the bump anomalies in the inclusive jet @xmath53 distribution ( fig .
[ fig:4 ] ) , reported by the cdf collaboration @xcite many years ago . analysing the bumps in the jet transverse energy distribution in fig .
[ fig:4 ] , we can find the endpoint of the first bump at 250 gev and guess about the second bump endpoint from the minimum around 350 gev . if we assign the first bump to the hadron decay products of the lightest charged bosons , which exactly corresponds to the estimated mass , the second endpoint hints to a mass around 700 gev , which is also in a quantitative agreement with our estimations for the mass of the lightest neutral boson .
however , taking into account the large systematic uncertainties in jet production , these conclusions may be premature , unless they are confirmed in other channels .
r0.5 indeed , an excess about 2@xmath54 in the lepton channel has been pointed out recently by the cdf collaboraion @xcite in the region @xmath55 gev @xmath56 gev .
at the same time the same collaboration , however , denies the peak in the quark channel in the same region ( fig .
[ fig:23j_obs ] ) , claiming that `` since the predictions in _ the neighboring bins _ agree with the observation , and since _ the three jet bin _ does not show a similar excess , we anticipate that the excess in this region is a statistical fluctuation '' .
but this signature means just a resonance and this excess is in some sense a confirmation of the excess in the leptonic channel !
therefore , the independent result from the d0 collaboration is very important .
their published result @xcite is based on 230 pb@xmath10 of integrated luminosity and does not show any excess in the histogram with the bin s width of 50 gev .
however , it should always be taken into account that the narrow peak could be missed due to the smearing effect of the detector resolution or an insufficient statistics . indeed , the right histogram in the fig . 3 of the conference paper @xcite of the same collaboration with the bin s width of the 45 gev reveals , nevertheless , the weak peak in the same region of the 500 gev .
all these not statistically significant results for the separated analyses may give a more conclusive answer after their combination and an additional investigation of the angular distributions of the events in this region . the small excess in the @xmath22 channel around 700 gev ( fig .
[ fig : mttbar ] ) can be explained in the framework of our model by the production and the decay of the lightest neutral chiral boson .
the latter shows ` leptophobic ' property , since it decays to ` invisible ' @xmath57 leptonic channel , and can be detected only through its decay into a pair of up quarks .
the d0 collaboration even superimposed its plot of the @xmath22 invariant mass distribution with the expected signal for a topcolor - assisted technicolor @xmath8 with @xmath58 gev , which perfectly agrees with the data .
there are some hints for the existence of a lightest charged chiral boson with a mass around 500 gev and a lightest neutral ` leptophobic ' chiral boson with a mass around 700 gev in the tevatron data . in the positive case
the lhc would be able to discover all predicted charged and neutral chiral bosons spanning in mass up to around 1 tev ( see their leptonic decay distributions in the fig .
[ fig : lhc ] ) .
the ilc with such energy would be an ideal place to produce and to study these particles .
99 g. arnison _ et al . _
( ua1 collaboration ) , phys .
b122 * 103 ( 1983 ) ; phys . lett . *
b126 * 398 ( 1983 ) .
m. banner _
( ua2 collaboration ) , phys .
b122 * 476 ( 1983 ) ; p. bagnaia _ et al . _
( ua2 collaboration ) , phys .
* b129 * 130 ( 1983 ) .
p. antilogus _ et al .
_ ( lep collaborations and lep electroweak working group ) , cern - ppe/95 - 172 ( 1995 ) .
f. abe _ et al . _
( cdf collaboration ) , phys .
lett . * 74 * 2626 ( 1995 ) .
s. abachi _ et al . _
( d0 collaboration ) , phys .
rev . lett . * 74 * 2632 ( 1995 ) .
tevatron electroweak working group , arxiv : hep - ex/0703034 ( 2007 ) .
abazov _ et al . _
( d0 collaboration ) , phys .
lett . * 98 * 181802 ( 2007 ) .
t. aaltonen _ et al . _
( cdf collaboration ) , arxiv:0708.3642 [ hep - ex ] ( 2007 ) .
d0 collaboration , d0 note 4375-conf ( 2004 ) .
d0 collaboration , d0 note 4375-conf ( 2004 ) .
t. aaltonen _ et al . _
( cdf collaboration ) , arxiv:0707.2524[hep - ex ] ( 2007 ) .
d0 collaboration , d0 note 5443-conf ( 2007 ) .
t. aaltonen _ et al . _
( cdf collaboration ) , arxiv:0709.0705[hep - ex ] ( 2007 ) .
a. abulencia _ et al .
_ ( cdf collaboration ) , phys . rev . * d75 * 091101 ( 2007 ) .
d0 collaboration , d0 note 5191-conf ( 2006 ) .
v. m. abazov _
( d0 collaboration ) , phys .
* b641 * 423 ( 2006 ) .
cdf collaboration , cdf note 8747 ( 2007 ) .
m. v. chizhov , mod .
a8 * 2753 ( 1993 ) [ arxiv : hep - ph/0401217 ( 2004 ) ] .
bolotov _ et al .
_ , phys . lett . *
b243 * 308 ( 1990 ) ; e. frle _ et al._. phys .
* 93 * 181804 ( 2004 ) .
chizhov , arxiv : hep - ph/0311360 ( 2003 ) .
m. davier , nucl .
* 169 * , 288 ( 2007 ) .
chizhov , arxiv : hep - ph/0609141 ( 2006 ) .
chizhov , arxiv:0705.3944 ( 2007 ) .
f. abe _ et al .
lett . * 77 * 438 ( 1996 ) .
d0 collaboration , d0 note 5024-conf ( 2006 ) . | direct constraints on the masses of new heavy bosons by the tevatron data are discussed . some excesses in the experimental data
are interpreted as a resonance production of new charged and ` leptophobic ' neutral chiral bosons with masses around 500 gev and 700 gev , respectively .
the interpretation was provided on the basis of the theoretical model , proposed by the author about 15 years ago .
new tevatron data and the lhc results will definitely confirm or reject this interpretation .
the ilc with an energy above 1 tev would be an ideal place to produce and to study the properties of these particles . |
the transition between quantum hall plateaux in a two - dimensional electron gas is characterized by a divergent localization length @xmath1 with a critical exponent @xmath0 and a non - zero longitudinal conductivity @xmath4 .
@xmath1 is finite and @xmath4 is zero inside the hall plateaux whereas @xmath5 is a constant . a direct measurement of the localization length exponent in a algaas / gaas sample by koch et al .
@xcite gave a value for @xmath0 very close to @xmath6 .
recent high precision measurements on si mosfet @xcite and algaas / gaas samples @xcite , however , indicate that @xmath1 diverges with the electron density @xmath7 like @xmath8 or with the magnetic field @xmath9 like @xmath10 , where the quantum hall transition ( qht ) is at @xmath11 or @xmath12 , respectively .
the exponent @xmath13 appears to be almost independent of the material or the hall plateaux .
on the other hand , the coefficients @xmath14 , @xmath15 are sensitive to disorder : they _ increase _ with increasing disorder @xcite .
this is a remarkable observation because in the scaling theory of anderson localization @xcite the coefficient is related to the mean free path .
that means it would _ decrease _ with increasing disorder .
the observation of @xmath13 is in sharp contrast to the experiment by koch et al .
the disagreement was explained in @xcite by insufficient sample size in the earlier experiment .
the localization length scale near the qht was also studied intensively in a number of numerical simulations using the network model @xcite and the lowest landau level approximation @xcite .
these calculations agree on the result that the critical exponent is @xmath16 .
concerning the exponent @xmath0 there is a calculation by affleck @xcite based on the @xmath17 nonlinear sigma model with topological term in the replica limit @xmath18 @xcite .
he obtains @xmath2 .
the same value was found for dirac fermions with random vector potential if the average density is finite @xcite .
a reason for the impressive agreement of the numerical calculations , on the one hand , and the disagreement between experiments , numerical and analytic calculations on the other hand could be the sensitivity of the qht to the size of the system and to the type of disorder . in particular , it may be related to the existence of a large characteristic length scale depending on disorder .
the existence of such a typical scale is also indicated by the numerical results due to the fact that there is a cross - over from the pure network model ( @xmath3 ) to the random network model ( @xmath19 ) .
the purpose of this article is to investigate the role of disorder induced length scales in a tight binding model with strong magnetic field .
the work is based on an effective supersymmetric field theory for dirac fermions with a random mass which enables us to study large scale properties .
the main results are : + 1 .
spontaneous breaking of the supersymmetry .
i.e. , the effective field theory for the qht is _ not _ a nonlinear sigma model .
the scaling behavior of the localization length depends on the characteristic scale @xmath20 , where @xmath21 is the strength of disorder : if @xmath22 the effective exponent is @xmath3 whereas for @xmath23 the exponent is @xmath2 .
there is a universal value for the conductivity @xmath24 .
+ the article is organized as follows : after the definition of the model in sect . ii an effective field theory is constructed for the averaged green s functions ( sect .
this field theory includes the description of the conductivity according to kubo s formula .
then a collective field representation is introduced in order to cover symmetry breaking effects ( sect .
the latter are discussed using a saddle point approximation for the collective field ( sect .
gaussian fluctuations around the saddle points and corrections to gaussian fluctuations are studied in sects .
iii.c and iii.d , respectively . finally , the localization length ( sect .
iv ) and the conductivity ( sect .
v ) are evaluated .
a lattice model is considered in this article , stressing the universality in terms of the electron density and the magnetic field observed in the experiment , to study the asymptotic behavior of the localization length near the qht .
starting point is a microscopic model for non - interacting electrons on a regular lattice in a homogeneous magnetic field .
disorder enters only through a random potential on the lattice .
this choice guarantees that the disorder does not affect the homogeneity of the magnetic field .
the model is defined by the tight - binding hamiltonian on a square lattice with magnetic flux @xmath25 , where @xmath26 is the lattice constant and @xmath27 the homogeneous external magnetic field .
there is nearest neighbor hopping with rate @xmath28 and next nearest neighbor hopping with rate @xmath29 .
the hamiltonian reads in landau gauge @xmath30+\sum_rv(r)c(r ) c^\dagger(r ) .
\label{hamiltonian}\end{aligned}\ ] ] @xmath31 are lattice unit vectors , and @xmath32 and @xmath33 are fermion creation and annihilation operators , respectively .
@xmath34 is a random potential representing disorder on the lattice . without disorder , i.e. , for @xmath35
, this model was discussed extensively in the literature @xcite .
a central result is the occurence of electron bands with a quantized hall conductivity in each band .
gaps can be created in the model , for instance , by choosing a staggered potential @xmath36 in the hamiltonian ( [ hamiltonian ] ) @xcite . by varying the staggered chemical potential
one varies the concentration of electrons in the system .
there are other methods to create gaps in a tigh - binding model .
e.g. , one could vary the magnetic field .
this , however , would lead to a more complicated situation because the corresponding vector potential depends on space .
in general , the relevant parameter for the quantum hall transition is the filling factor @xmath37 .
this is essentially determined by the ratio of the concentration of electrons @xmath7 and the magnetic field @xmath27 .
therefore , the variation of the concentration of electrons ( i.e. , the chemical potential ) is equivalent with variation of the magnetic field in the quantum hall system . in general ,
the creation of new bands ( `` gap opening '' ) can be described by dirac fermions @xcite .
starting from the tight - binding hamiltonian the dirac fermions can be derived in a large scale approximation .
the simplest case is that with half a flux quantum per lattice plaquette ( @xmath38 ) @xcite .
( such a strong magnetic flux is unrealistic in real crystals but typical for arrays of quantum dots in moderate magnetic fields @xcite . ) for half a flux quantum per plaquette it is easy to derive the dirac theory from a sublattice representation which takes into account the phase factor @xmath39 of the tight - binding hamiltonian and the staggered potential .
the fourier components of the non random part @xmath40 read @xmath41 with @xmath42 .
all elements of the matrix are measured in units of the nearest neighbor hopping rate @xmath28 .
after expansion of @xmath43 for small @xmath44 vectors around the four nodes and a global orthogonal transformation @xmath45 with @xmath46 the hamiltonian becomes @xmath47 the corresponding green s function @xmath48 decays asymptotically into two diagonal blocks @xmath49 the lattice constant @xmath26 is implicitly scaled out in @xmath29 , @xmath50 and @xmath51 .
@xmath52 means asymptotics with respect to @xmath53 and @xmath54 .
thus the approximation breaks up the hamiltonian ( [ hamiltonian ] ) into two independent dirac hamiltonians @xmath55 with pauli matrices @xmath56 .
the two dirac theories describe particles with different masses @xmath57 , respectively .
the next nearest neighbor hopping term lifts the degeneracy of the two dirac particles .
therefore , it plays an important role in this model and must be taken into account .
variation of the chemical potential implies a variation of the dirac mass .
in particular , dirac fermions undergo a hall transition if the mass vanishes @xcite .
this is a consequence of the fact that the mass breaks the time - reversal symmetry : depending on the sign of @xmath58 there is a clockwise or counterclockwise hall current . if the light dirac particle undergoes a hall transition at @xmath59 its contribution to the hall conductivity changes from @xmath60 for @xmath61 to @xmath62 for @xmath63 .
( the conductivity is in units of @xmath64 . )
the heavy dirac particle contributes @xmath62 because its mass is positive .
thus the combined effect is a hall step from @xmath65 to @xmath66 .
this picture is particularly simple for @xmath67 but should also hold for other values of the flux as long as the low energy excitations are linear and can be described by dirac fermions . for large scale properties like for the critical behavior near the hall transition it is sufficient to consider only the light particle with @xmath68 @xmath69 where @xmath70 is the lattice gradient operator .
disorder , originally introduced in @xmath9 by the random potential @xmath71 , appears in @xmath72 as a diagonal matrix @xmath73 with independent random elements @xmath74 , @xmath75 .
the appearence of two random variables per site is a consequence of the sublattice representation required by the phase factor @xmath39 of the tight - binding hamiltonian .
the random matrix @xmath73 is equivalent to a random mass @xmath76 and a random energy @xmath77 . for technical reasons
the random energy term will be neglected in the following .
it should be noticed that the random mass is marginally irrelevant on a perturbative level @xcite .
however , going beyond perturbation theory , it turns out that the random mass leads to spontaneous symmetry breaking which changes the properties significantly @xcite .
this effect has not been included in previous studies of the localization properties near the hall transition .
it will be important for the considerations in this article .
dirac fermions can also be derived as the large scale approximation of the network model @xcite .
it was recently pointed out by ho and chalker @xcite that the network implies a random dirac mass ( due to fluctuations in the tunneling rates ) , a random energy ( due to fluctuations in the flux per plaquette ) and a random vector potential ( due to fluctuations in the phase of the hopping elements ) .
i.e. , in terms of the network model the random dirac mass requires a fixed flux per plaquette and a fixed phase for the current between the vertices of the network .
this is probably the simplest situation for the realization of a qht . after averaging with respect to the random mass
the localization length @xmath1 , measured in lattice units @xmath78 for electrons in a magnetic field @xmath27 , is defined as the decay length of the function @xmath79 .
the relation @xmath80 means that @xmath1 is given by the product of two green s function at frequencies with _ opposite sign _ ( retarded and advanced green s functions ) . due to the @xmath81 block structure of @xmath82 there exists a relation between green s functions at @xmath83 and green s functions at @xmath84 : @xmath85 this identity reflects the lorentz - covariance of the dirac theory .
it implies @xmath86 and @xmath87 this means that only the green s functions with one frequency is required for the evaluation of localization properties in the relativistic model .
the averaged quantity @xmath88 is translational invariant .
therefore , it can be expressed by its fourier components @xmath89 .
this can be used to calculate the localization length @xmath1 .
( the following discussion holds for any choice of @xmath90 , @xmath91 . therefore
, these labels are not written explicitly . )
the correlation function @xmath92 for large @xmath93 is proportional to @xmath94 with some exponent @xmath95 for which we assume that it is fixed for the model and does not depend on the parameters .
this implies that @xmath96 dropping the constant term from the ratio of sums on the right hand side of ( [ ll ] ) , the localization length can be defined in terms of the fourier components as @xmath97 the localization length is finite for @xmath98 but diverges in the regime of delocalized states with @xmath99 .
the localization length for massless dirac fermions without disorder diverges like @xmath100 if @xmath101 is approached .
this behavior is probably unstable against arbitrarily weak randomness , as it will be shown in this article .
however , it has been shown in a previous paper @xcite that the localization length of the averaged correlation function has a lower bound which is the energy energy correlation length of the 2d random bond ising model .
since the latter diverges at the critical points , this implies a divergent localization length at the qht .
the calculation for the network model of chalker and coddington @xcite , indicates that the critical exponent @xmath3 of the pure model may change to @xmath102 due to disorder .
this was discussed as a possible appearence of a new random fixed point of the random model @xcite . however , a new fixed point with such a behavior has not been found so far in terms of renormalization group calculations .
it is convenient to introduce a functional integral representation for @xmath92 , because this provides a basis to apply an approximation using a saddle point integration . the product of green s functions on the right hand side of ( [ id1 ] ) and ( [ id2 ] )
can formally be written as @xmath103 with the quadratic form of the superfield @xmath104
@xmath105 @xmath106 is a complex field and @xmath107 a grassmann field , respectively .
it is important to notice that @xmath9 appears in the quadratic form for the complex field whereas @xmath108 appears for the grassmann field .
this difference will turn out to be crucial for the localization properties of the dirac fermions .
in particular , it will give all the delocalized states expected near the hall transition .
in contrast , the quadratic form where @xmath9 is used in the grassmann sector instead of @xmath108 does not give these critical properties @xcite .
averaging with respect to disorder leads to @xmath109 a gaussian distribution of the random dirac mass is assumed in the following with mean @xmath58 and variance @xmath21 . then the average can be performed exactly giving an additional quartic interaction term in @xmath110 with coupling constant @xmath21 . for weak disorder ,
i.e. , small @xmath21 , one could apply perturbation theory .
unfortunately , this does not lead to interesting results because it can not catch spontaneous symmetry breaking . in order to deal with the latter one must construct a representation which describes the field which is the conjugate to the symmetry breaking terms of the dirac theory , the mass @xmath58 and the frequency @xmath111 .
the appearence of the symmetry breaking terms in @xmath110 dictates to choose @xmath112 and @xmath113 as the collective fields . in general ,
products the fields @xmath114 in @xmath110 can be replaced by the collective fields as @xmath115 , @xmath116 , @xmath117 and @xmath118 .
( some care is necessary to choose the paths of integration correctly @xcite . )
one obtains for ( [ 8 ] ) in collective field representation ( for details see @xcite ) @xmath119 with the supersymmetric effective action @xmath120 \nonumber\\ & & + \ln\det\big\lbrack{\bf 1}-4\tau{\bar \theta}\tau(h_0^t+i\omega\sigma_0 + 2i \tau p\tau)^{-1 } \tau\theta\tau ( h_0+i\omega\sigma_0 - 2\tau q\tau)^{-1}\big\rbrack.\end{aligned}\ ] ] @xmath121 is the average dirac hamiltonian and @xmath122 the diagonal matrix @xmath123 . the introduction of the collective fields is important to discover the finite length scale @xmath124 , created by disorder , which is crucial for the properties of the random dirac mass model @xcite .
a saddle point ( sp ) approximation is a crude approach for a two dimensional system because it usually gives wrong results for low dimensional systems due to strong effects of fluctuations . in the model under consideration
it will be used as a starting point to study also the fluctuations . an argument in favor of a sp approach is the fact that some features of the model can be described which are not available from pertubation theory .
an example is the creation of states in the massive dirac theory due to randomness @xcite .
the hope is that the fluctuations around the sp are controlled by gaussian fluctuations , at least if randomness is weak .
this will be supported by the discussion presented below .
another argument for the sp approximation is its equivalence with the @xmath125limit , where @xmath126 is a formal extension of the model to one for electrons with @xmath126 states per lattice site @xcite .
the sp of the functional integral is given by the equations @xmath127 .
the two sp equations are identical if one substitutes @xmath128 : @xmath129 an ansatz for a uniform sp solution reads @xmath130 .
the sp equations imply a shift of the frequency @xmath131 with @xmath132 and a shift of the average mass @xmath133 with @xmath134 @xmath135 with @xmath136 . in the pure limit @xmath137 the sp equations lead to @xmath138 and @xmath139 . for a given @xmath140
the sp depends on two parameters , @xmath58 and @xmath111 . for @xmath101 and large @xmath141
there is only a trivial solution of ( [ spea ] ) with @xmath142 because @xmath143 .
as one varies @xmath141 there is a critical point @xmath144 where @xmath143 approaches @xmath145 . as a consequence , the sp solution of ( [ spea ] ) bifurcates from @xmath146 to @xmath147 at @xmath148 , and @xmath146 becomes unstable @xcite . in the following only the region with @xmath147
is considered , where @xmath149 .
this has a non
zero density of states @xmath150 which describes a semicircular behavior . of course
, this must be normalized with an energy cut
off dependent constant . as an ansatz for the sp approximation with @xmath151 one can write @xmath152 where the last equation follows from the sp equation for @xmath101
this implies @xmath153 and for eq .
( [ spea ] ) @xmath154 the expansion of the logarithm for @xmath155 yields a cubic equation for @xmath156 @xmath157 although this equation could be solved directly it is simpler to distinguish two different asymptotic regimes : \(i ) @xmath158 : @xmath159 and \(ii ) @xmath160 : @xmath161 in order to evaluate the localization length @xmath1 the gaussian fluctuations around the sp must be calculated ( semi - classical approximation ) . since @xmath162 , @xmath163 and @xmath164
are @xmath165 matrices , the fluctuations can also be written as 4-component vector fields : @xmath166 , @xmath167 , @xmath168 , @xmath169 with analogous definitions for @xmath170 and for the grassmann field @xmath171 .
the action of the gaussian fluctuations reads in fourier representation @xmath172d^2k\ ] ] with the fluctuation matrices @xmath173 and @xmath174 . for the large scale properties one needs only the asymptotic behavior for small wave vectors @xmath175 .
in particular , for a vanishing wave vector there is @xmath176 @xmath177 with @xmath178 and @xmath179 these quantities become quite large for small @xmath180 indicating a short range behavior of the related modes .
however , the second diagonal element of @xmath181 vanishes for vanishing @xmath111 because of @xmath182 .
this is a direct consequence of the sp condition ( [ spea ] ) and indicates a critical mode @xmath183 for all @xmath184 .
moreover , it implies a divergent behavior of the localization length if @xmath99
. the corresponding correlation function of the critical mode can be calculated in the large scale limit by expanding @xmath185 in powers of @xmath175 as @xmath186 where @xmath187 .
\label{3.21}\ ] ] @xmath188 is like a diffusion coefficient .
it is real and it never becomes zero .
the critical behavior describes a phase with a divergent sum @xmath189 ; i.e. , the correlation function decays non exponentially .
it implies that @xmath190 .
this holds not only on the sp level but in general due to the identity @xmath191 .
@xmath192 in ( [ cricor ] ) is an approximation based on @xmath193 .
since @xmath194 diverges like @xmath195 it is not possible to use this expression for the pure limit .
this also reflects the non - perturbative character of the sp approximation .
apart from the critical ( delocalized ) fermion mode there is a boson mode which becomes critical at @xmath196 .
this is due to a vanishing eigenvalue of @xmath197 because @xmath146 at @xmath196 .
thus there are delocalized states for @xmath198 due to massless fermions whereas a combination of critical fermions _ and _ critical bosons controls the qht .
the band of delocalized fermions simplifies the study of transport properties of the model away from the critical points @xmath196 .
an analogous calculation for the gaussian fluctuations can be performed for anderson localization ( i.e. , for @xmath199 in eq .
( [ hamiltonian ] ) ) . in that case
the green s function of eq .
( [ green ] ) must be replaced by @xmath200 now the matrix @xmath201 is symmetric in contrast to @xmath202 . as a consequence
the corresponding fluctuation matrix is degenerated for the fermion and the boson sector ( i.e. , @xmath203 ) with @xmath204 where @xmath205 and @xmath206 there are two vanishing eigenvalues of @xmath197 for vanishing @xmath111 in the grassmann as well as in the complex sector , each of them is @xmath207 .
therefore , the large scale behavior of this model is very different , and the critical properties are described by a nonlinear sigma model including fermion as well as boson degrees of freedom @xcite .
this is a consequence of the fact that @xmath199 preserves the supersymmetry whereas the model with @xmath38 breaks the supersymmetry implying @xmath208 .
the latter implies that there is only one massless ( delocalized ) fermion field , and all other fields are massive ( localized ) .
gaussian fluctuations are usually not sufficient to describe the properties of a critical system , especially at low dimensionality , because interaction between the fluctuations are a relevant perturbation . for anderson localization
the interaction is marginal in @xmath209 as one finds from power counting of the scaling behavior .
this method can also be applied to the effective field theory of random dirac fermions .
it provides a first check for the effect of the interaction among the fluctuations . in the following
we will see that the perturbation term for random dirac fermions has dimensionality @xmath210 .
therefore , the interaction of the fluctuations is irrelevant .
( it seems that the dimensionality is reduced by 2 in comparison to anderson localization .
this is similar to the dimensional reduction by 2 in supersymmetric theories , applied to the average density of states for a particle in a strong magnetic field and a random potential @xcite . )
away from the critical points @xmath211 it is sufficient to study the grassmann fluctuations because the complex fluctuations are massive .
their action reads for terms up to fourth order in the fluctuations @xmath212,\ ] ] where @xmath213 .
@xmath214 is the only critical mode of the collective grassmann field @xmath215 .
thus the interaction term can also be written as @xmath216 { \bar\psi}_{r_1,2}\psi_{r_2,2}{\bar\psi}_{r_3,2}\psi_{r_4,2}. \label{int}\ ] ] the trace term can be evaluated and yields after some straightforward calculation together with the approximation @xmath217 $ ] the following expression @xmath218\ ] ] where the index 2 of the grassmann field has been dropped and @xmath219 .
this result reflects the fact that terms with an odd number of @xmath70operators cancels in ( [ int ] ) and terms quadratic in @xmath70 cancel each other because of the anticommutation rule of the grassmann field : @xmath220=0 .
higher order terms with at most second order gradients disappear individually because they contain grassmann fields at the same site .
simple power counting indicates that this interaction term has dimensionality @xmath210 .
therefore , the interaction is irrelevant in comparison with gaussian fluctuations considered in the previous section , and it scales quickly to zero under renormalization group transformations . the situation is different if we approach @xmath196 because the complex field also becomes critical . as a consequence ,
the corrections to gaussian fluctuations are marginal rather than irrelevant then .
it is possible that the localization length becomes finite at @xmath196 due to renormalization effects .
this phenomenon near @xmath221 requires a separate treatment which will not be considered in this article .
the irrelevance of the interaction terms is very special for the model under consideration . in similar two - dimensional models , like the gross - neveu model or the tight binding model without magnetic field ( orthogonal nonlinear sigma model ) or with weak magnetic field ( unitary nonlinear sigma model ) ,
the interaction is always marginal ( i.e. , dimensionality of the interaction term is zero ) .
according to the discussion of the gaussian fluctuations in sect . iiic the leading large scale behavior is given by the correlation function of ( [ cricor ] ) . from the latter the localization length
can be calculated , using eq .
( [ local ] ) , as @xmath222 .
together with the sp results ( [ 3.11 ] ) and ( [ 3.21 ] ) @xmath1 reads @xmath223 the localization length diverges like @xmath224 because @xmath194 and @xmath225 remain non - zero for @xmath198 according to the results of the sp approximation . to compare @xmath1 with numerical or experimental results
it is important that @xmath194 can be large ( @xmath226 ) and @xmath225 can be small .
therefore , several quasi
scaling regimes exist as indicated by the graph in fig .
1 which is simply the plot of ( [ fig1 ] ) together with ( [ spe ] ) . only in the asymptotic regime @xmath227 ,
i.e. , for @xmath160 , the localization length diverges like @xmath224 @xmath228 surprisingly , the exponent @xmath2 agrees with that of a completely different approach to the qht by affleck @xcite .
moreover , @xmath2 was also found by ludwig et al . for dirac fermions with a random vector potential at that special point on their fixed point line where the average density of states is finite @xcite .
a quasi scaling regime occurs for @xmath158 where we have @xmath229 i.e. , the effective exponent is @xmath230 .
going back to the sp equation ( [ spe ] ) one could also assume @xmath231 .
it gives @xmath232 this result implies for the localization length @xmath233 i.e. , the effective exponent is @xmath3 . the coefficient in ( [ coeff1 ] ) _ decays _ for small @xmath21 like @xmath234 whereas the corresponding coefficient in ( [ coeff2 ] ) _ grows _ like @xmath235 with the strength of disorder .
the coefficients of the power law for @xmath230 ( @xmath236 ) , for @xmath2 ( @xmath237 ) and for @xmath3 ( @xmath238 ) are plotted in fig .
2 . @xmath3 and the behavior of the corresponding coefficient @xmath238 agree with the observation in experiments @xcite .
the longitudinal conductivity can be calculated via kubo s formula @xmath239 the correlation function is again the expression we have considered in the effective field theory .
since for small @xmath111 only the large scale part of the correlation contributes significantly , we can use @xmath240 of ( [ cricor ] ) to write @xmath241 for weak disorder we use the approximation @xmath242 .
furthermore , for @xmath227 we get @xmath243 .
therefore , the conductivity reads in the dc
limit @xmath244 for @xmath245 the dc
conductivity is independent of disorder @xmath246 this result is in agreement with a calculation for dirac fermions with a random vector potential@xcite .
thus there is a universal conductivity in @xmath209 at the center of the band of extended states @xmath247 .
the hall conductivity can be calculated using a simpler field theory which does not break the supersymmetry @xcite . in units of @xmath64
it was found @xmath248.\ ] ] for @xmath245 the hall conductivity is always @xmath249 whereas for @xmath250 it depends on disorder via @xmath251 .
the resistivity , calculated from the averaged conductivities ( which is an additional approximation because one should actually evaluate the averaged resistivity ) yields @xmath252 this value agrees with an error of about @xmath253 with various experimental results @xcite .
for a given strength of randomness @xmath21 there are three different regimes of the behavior of the localization length , depending on the value of @xmath156 .
@xmath156 is directly related to the frequency @xmath111 according to the equations ( [ spe1 ] ) , ( [ spe2 ] ) and ( [ spe3 ] ) , respectively .
the corresponding power laws @xmath254 appear with effective exponents @xmath255 and @xmath256 and with the asymptotic exponent @xmath257 .
the exponents and the corresponding coefficients @xmath258 are shown in table i , and the @xmath21 dependence of the coefficients is plotted in fig .
2 . ludwig et al . evaluated the exponent @xmath0 for two - dimensional dirac fermions with random vector potential with variance @xmath259 . in that case
the exponent of the asymptotic localization length depends smoothly on randomness as @xmath260 @xcite .
it should be noticed that the inequality @xmath261 of chayes et al .
@xcite for two dimensional random systems does not apply to the problem under consideration ( cf also the discussion in ref.@xcite ) .
it is remarkable though that there is agreement of the asymptotic result @xmath2 between the present calculation for the random dirac mass , the random vector potential with finite average density of states ( where @xmath262 ) @xcite and the nonlinear sigma model with topological term @xcite .
the result @xmath230 is not reliable because it appears close to the critical points @xmath211 , where the bosonic degrees of freedom become critical .
moreover , the asymptotic regime with @xmath2 is not realistic : the typical width of the fluctuations @xmath263 in a sample is about @xmath264 of the hopping rate .
i.e. , @xmath265 is immeasurably small .
therefore , only the regime @xmath266 would be accessible , and only the effective exponent @xmath3 , together with the coefficient @xmath267 , is of practical relevance .
consequently , only the coefficient of the localization length is affected by randomness .
this is in agreement with the observations in refs.@xcite .
however , the divergency of the localization length is controversal among the various experiments . in this context
it would be interesting if the cross - over length , evaluated in the article as @xmath268 , can be observed experimentally .
the density of states is non
zero near the qht .
this result is non perturbative because @xmath269 , and it reflects spontaneous symmetry breaking @xcite .
it agrees with a monte carlo simulation for the network model by lee and wang @xcite , with an exact diagonalization of a finite system @xcite , with a rigorous estimation @xcite and with an exact calculation for lorentzian disorder @xcite . the longitudinal conductivity @xmath270 is non zero between @xmath271 and @xmath272 , the transition region between two hall plateaux . in an experiment it may not be possible to resolve the width of this band of delocalized states since @xmath251 is too small .
therefore , the width of the transition between the hall plateaux will always be dominated by thermal broadening .
thus a power law for the width @xmath273 at temperature @xmath274 like @xmath275 @xcite is a realistic ansatz .
the conductivity at the qht should agree with the value of @xmath4 at @xmath245 , i.e. , @xmath276 .
converting this value into the corresponding value of @xmath277 gives @xmath278 which is in good agreement with experiments .
the universal value @xmath276 agrees with the value found for random vector potential @xcite but not with the result found from the numerical simulation of the network model , where @xmath279 was found @xcite .
this value was also obtained for the lowest landau level projection @xcite .
it seems that our results for the 2d dirac fermions with random mass are in good agreement with the corresponding results for 2d dirac fermions with random vector potential and with recent experiments .
but there is disagreement with the results of the numerical simulation of the network model .
this may be related to different types of randomness ( e.g. , strong randomness in the magnetic field ) or to different geometries .
it is also possible that there is a strong renormalization of the localization length near @xmath196 due to extended boson fields leading to an exponent @xmath19 .
this , however , would raise the question about the origin of the experimental value @xmath13 .
moreover , the disagreement of the values of @xmath280 for the network model on the one side and for dirac fermions with @xmath245 on the other side can not be explained by renormalization effects . | a two - dimensional lattice system of non - interacting electrons in a homogeneous magnetic field with half a flux quantum per plaquette and a random potential is considered . for the large scale behavior a supersymmetric theory with collective fields
is constructed and studied within saddle point approximation and fluctuations .
the model is characterized by a broken supersymmetry indicating that only the fermion collective field becomes delocalized whereas the boson field is exponentially localized .
power counting for the fluctuation terms suggests that the interactions between delocalized fluctuations are irrelevant . several quasi
scaling regimes , separated by large cross over lengths , are found with effective exponents @xmath0 for the localization length @xmath1 . in the asymptotic regime
there is @xmath2 in agreement with an earlier calculation of affleck and one by ludwig et al . for finite density of states .
the effective exponent , relevant for physical system , is @xmath3 where the coefficient of @xmath1 is growing with randomness .
this is in agreement with recent high precision measurements on si mosfet and algaas / gaas samples . |
@xmath0 : : - the entry angle from workpiece to the roller , also known as the attack angle [ @xmath1 ( fig
. 4 ) @xmath2 : : - overall contact area [ mm@xmath3 ( section [ sec : application ] ) @xmath4 : : - @xmath5 planar projection of the contact area [ mm@xmath3 ( section [ sec : application ] ) @xmath6 : : - @xmath7 planar projection of the contact area [ mm@xmath3 ( section [ sec : application ] ) @xmath8 : : - @xmath9 planar projection of the contact area [ mm@xmath3 ( section [ sec : application ] ) @xmath10 : : - the trailing angle from the workpiece to the roller , also known as exit angle or planishing angle [ @xmath1 ( fig .
4 ) @xmath11 : : - distance between analytical surface and experimental surface nearest neighbour points ( section [ sec : validation ] ) @xmath12 : : - distance between analytical surface and experimental surface interpolant ( section [ sec : validation ] ) @xmath13 : : - the distance between the center of the mandrel and the center of the roller [ mm](fig . 4 , @xmath14 ) @xmath15
: : - axial feed rate of the roller down the face of the cylinder , along the @xmath16 direction [ mm / min ] mse : : - mean square error ( eq . [ eq : mse ] ) @xmath17 : : - the mandrel rate of rotation [ revolutions / min ] @xmath18 : : - the roller path pitch or distance traveled axially by the roller in one revolution [ mm ] ( @xmath19 ) @xmath20 : : - roller nose radius [ mm ] ( fig .
4 ) @xmath21 : : - numeric resolution of the solution ( section [ sec : surdef ] ) @xmath22 : : - initial workpiece radius ( @xmath23 ) @xmath24 : : - the mandrel radius
[ mm ] ( fig . 4 ) @xmath25 : : - the roller radius excluding the radius of the nose [ mm ] ( fig .
4 ) @xmath26 : : - intermediate set of radial quantities used to find @xmath27 and @xmath28 ( eq . [ eq : s ] ) @xmath29 : : - the angle of contact between the roller and workpiece [ rad](fig . 2 , eq . [ eq : tf ] ) @xmath30 : : - intermediate value of @xmath31 used for the iterative solution of the contact area [ rad ] ( eq .
[ eq : tf ] ) @xmath32 : : - maximum angular limit of the solution ( eq . [ eq : thetamax ] ) @xmath33 : : - angular coordinates used to define boundary surfaces ( eq . [ eq : tcoords ] ) @xmath34 : : - starting material thickness [ mm ] ( fig .
4 ) @xmath35 : : - the final material thickness [ mm ] ( fig .
4 ) @xmath36 : : - @xmath37 coordinates lying on the instantaneous roller position ( eq . [ eq : x_i ] ) @xmath38 : : - @xmath37 coordinate used for intersection conditioning ( eq . [ eq : xi ] ) @xmath39 : : - @xmath37 coordinate used for intersection conditioning ( eq . [ eq : xlower ] ) @xmath40 : : - @xmath37 coordinate lying on the cylinder defined by @xmath22 ( eq . [ eq : x_m ] ) @xmath41 : : - maximum limit in the @xmath37 direction of the solution ( eq . [ eq : xmax ] ) @xmath27 : : - @xmath37 coordinates lying on the previous roller path ( eq . [ eq : xp ] ) @xmath42 : : - @xmath37 coordinates within the roller / workpiece contact area ( eq . [ eq : xs ] ) @xmath43 : : - @xmath37 coordinate used for intersection conditioning ( eq .
[ eq : xupper ] ) @xmath44 : : - @xmath45 coordinates lying on the instantaneous roller position ( eq . [ eq : yi ] ) @xmath46 : : - @xmath45 coordinates lying on the cylinder defined by @xmath22 ( eq .
[ eq : y_m ] ) @xmath47 : : - maximum limit in the @xmath45 direction of the solution ( eq . [ eq : ymax ] ) @xmath28 : : - @xmath45 coordinates lying on the previous roller path ( eq . [ eq : yp ] ) @xmath48 : : - @xmath45 coordinates used to define boundary surfaces ( eq . [ eq : ycoords ] ) @xmath49 : : - @xmath45 coordinates within the roller / workpiece contact area ( eq . [ eq : ys ] ) @xmath50 : : - axial limits of the workpiece / roller contact area , @xmath16 coordinate of the endpoint of contour 1 and starting point of contour 2 ( eq . [ eq : lowerlimflat1 ] and [ eq : lowerlimitflat2 ] ) @xmath51 : : - axial limits of the workpiece / roller contact area , @xmath16 coordinate of the endpoint of contour 1 and starting point of contour 3 ( eq . [ eq : za ] to [ eq : zd ] ) @xmath52 : : - @xmath16 coordinates lying on the instantaneous roller position ( eq . [ eq : zi ] ) @xmath53 : : - @xmath16 coordinate used for intersection conditioning ( eq . [ eq : zi ] ) @xmath54 : : - @xmath16 coordinate used for intersection conditioning ( eq . [ eq : zlower ] ) @xmath55 : : - @xmath16 coordinates lying on the cylinder defined by @xmath22 ( eq . [ eq : zm ] ) @xmath56 : : - @xmath16 coordinates lying on the previous roller path ( eq . [ eq : zp ] ) @xmath57 : : - @xmath16 coordinates used to define boundary surfaces ( eq . [ eq : zcoords ] ) @xmath58 : : - @xmath16 coordinates within the roller / workpiece contact area ( eq . [ eq : zs ] ) @xmath59 : : - @xmath16 coordinate used for intersection conditioning ( eq . [ eq : zupper ] )
to determine the energy required to form a component , the size and orientation of the tooling interface on the workpiece is necessary . while purely analytical models describing this contact are preferable , they are usually difficult to attain for complex metal forming processes . in this study ,
an analytical approach is presented to model the tooling / workpiece contact area in an application of rotary forming . while the present work focuses on an implementation for flow forming ,
the applied technique can be applied to other variants of rotary forming operations such as metal spinning , shear forming , thread rolling and crankshaft fillet rolling .
flow forming , a variant of metal spinning , is a process used to fabricate rotationally symmetrical parts from ductile materials , after @xcite . during flow forming ,
the workpiece is clamped to a rotating mandrel and pressed into contact with the mandrel by rollers .
the rollers induce high levels of plasticity in the workpiece causing it to undergo both reduction in thickness and axial lengthening .
since the rollers press on only a very small area of the overall workpiece at any given time , the deformation is highly localized between the roller and workpiece . to properly understand the distribution of this intense local plastic deformation it is essential to be able to calculate the roller / workpiece contact area from the geometric parameters that govern the flow forming process .
in addition , the roller / workpiece contact area is critical to coupling other experimental findings , such as power consumption , frictional effects , force , stress and strain distributions through the workpiece back to geometric process parameters . in flow forming , the combined mandrel rotation and linear movement of the rollers induce contact on the workpiece along a helical path .
this helical tool path , coupled with the curved profile of the rollers leads to a very complicated roller / workpiece contact area . in terms of related tool contact studies , an important analytical derivation of the workpiece contact in shear spinning
was completed by @xcite . however , in a comprehensive review of metal spinning processes , @xcite highlighted that the mechanics of flow forming are quite different than shear spinning .
this is also true for the contact area formulation as there is little roller penetration into the workpiece and deformation proceeds according to the sine rule . in terms of flow forming specific research , investigations made by gur and tirosh ( 1982 ) , singhal et al .
( 1995 ) , ma ( 1993 ) and jahazi and ebrahimi ( 2000 ) have proposed analytical models of this contact .
gur and tirosh ( 1982 ) developed the formulation of a planar contact area in each of the primary rolling and extrusion deformation directions in backwards flow forming .
singhal et al .
( 1995 ) derived the contact area imposed by tooling in the flow forming of small diameter tubes where the assumption made was that material is assumed to be perfectly plastic , and the tools were assumed rigid .
ma ( 1993 ) extended the work of gur and tirosh ( 1982 ) to derive a critical angle of attack and jahazi and ebrahimi ( 2000 ) extended the contact formulation made by gur and tirosh ( 1982 ) to investigate the mechanics in a specific application of flow forming .
more recently , kemin et al .
( 1996 ) , xu et al .
( 2001 ) and hua et al .
( 2005 ) have developed finite element ( fe ) models of single roller flow forming . in each of these studies
, contact was modeled explicitly within each respective fe model . furthermore , with the exception of the work by @xcite and @xcite , all previous works have made assumptions concerning the roller / workpiece contact geometry that do not necessarily reflect the actual contact during flow forming .
these assumptions include : 1 . idealized roller geometry ( i.e. no blending radii ) ( fig .
1 ) ; 2 .
the use of two - dimensional treatments that do not account for the three - dimensional aspects of the workpiece contact ; 3 . not considering the influence of prior forming steps ( i.e. roller path overlap ) on the instantaneous roller / workpiece contact area . the most successful technique for modelling the roller / workpiece contact area , and other facets of the flow forming process , has been through fe analyses .
@xcite addressed items 1 and 2 listed above in their work to numerically calculate the roller / workpiece of a single roller flow forming .
however , @xcite did not give the details of their calculation of the contact area , nor did they specifically address item 3 .
@xcite has developed a thorough 3-d fe model that addresses all three items , but an fe approach is still limited to case - by - case application involving extensive pre - processing and explicit geometric modelling .
an analytical solution provides a solution with significantly lower effort . in the present work
, a generalized solution is developed for the roller / workpiece contact area during a single roller flow forming operation that accommodates items 1 to 3 . to accomplish this ,
the following assumptions are made : 1 . the single roller flow forming process proceeds under steady state conditions . the final and starting thickness ,
mandrel rotation and feed rate are constant .
the deformation response of the workpiece is perfectly plastic .
elastic effects are not considered .
volume of the flow formed workpiece is conserved outside the tool interface .
no material build - up occurs in front of the roller as the workpiece conforms completely with the rigid roller .
during flow forming , the roller contacts the workpiece along a path having a constant pitch ( fig .
1 ) . the profile of the roller can be divided into three regions ; the entry region , the nose region and the exit region .
these regions dictate the size and shape of the roller / workpiece contact area .
the contact area is bounded by three contours : the tangential exit contour , the axial entry contour and the axial exit contour , labeled 1 - 3 respectively in fig .
the contact area extends angularly from the tangential exit contour ( @xmath60 ) through to @xmath61 ( fig .
2(b ) ) . [ fig:1 ]
if the roller has an archetypal flow forming profile similar to that shown in fig .
1 with distinct flat entry and exit regions and a blending radius between the two that creates a nosed roller , the final contact area is dependent on six surfaces ( fig .
3 ) . contour 1 , and the starting points of contours 2 and 3 ( fig .
2 ) can be calculated directly as they lie exclusively on the @xmath7-plane .
the _ a priori _ @xmath16-coordinates of the extents of contour 1 define the axial limits of the roller / workpiece contact area .
once the a solution has been found for the starting and ending points of contour 1 ( by definition the starting points of contours 2 and 3 ) , the common end point of contours 2 and 3 is then solved using an iterative technique . [ fig:3 ] it is first necessary to calculate the axial limits of contact by determining the endpoints of contour 1 .
contour 2 is a function of the instantaneous roller contact with the workpiece at pitch @xmath62 .
contour 3 is a function of the instantaneous roller contact on the material as well as the tool contact on the workpiece one revolution of the mandrel beforehand , at @xmath63 .
contour 1 exists solely on the @xmath7 plane and is bound by the points of intersection with contours 2 and 3 .
contour 1 is both dependent on roller geometry and the roller path pitch , @xmath18 .
there are four possible conditions describing the intersection of the current roller position with that of its position on the previous mandrel revolution ( fig .
[ fig:4 ] corresponds to the instantaneous roller position and position @xmath63 corresponds to the roller at one mandrel revolution beforehand .
the upper endpoint of contour 1 can occur within the nosed region of the roller on both the instantaneous position and the position on the previous mandrel revolution ( condition a ) .
it can also occur at the intersection of the exit / entry profiles ( condition b ) , the nosed / entry profiles ( condition c ) or the exit / nosed profiles ( condition d ) of the instantaneous and the previous roller positions.,title="fig:",width=4 ] calculation of the location of the upper end point of contour 1 for the four conditions shown in fig .
4 is accomplished through comparison of the endpoints of the roller nose profile on the @xmath7 plane . for comparison purposes ,
the local coordinate system is moved on the @xmath37 axis from the global origin by @xmath64 ( fig .
the @xmath37 and @xmath16 coordinates of the upper end point of the nosed region of contour 1 , @xmath43 and @xmath59 ( fig .
5 ) : [ fig:5 ] @xmath65 @xmath66 for the lower @xmath37 and @xmath16 coordinates of the end point of the nose region of contour 1 , @xmath39 and @xmath54 ( fig .
5 ) : @xmath67 @xmath68 the entry profile of the previous roller path and the instantaneous exit profile of the roller will occur at @xmath38 and @xmath53 . these are expressed as : @xmath69 @xmath70 the values of @xmath43 , @xmath39 , @xmath38 , @xmath59 , @xmath54 and @xmath53 can be compared to identify which contact condition shown in fig .
4 applies .
the conditions and the relationships that must be simultaneously satisfied are shown in table [ table : conditions ] .
.upper axial limits of contact [ cols="^,^,^ " , ] the ofat analysis technique is limited as , by definition , it does not allow for simultaneous changes in multiple variables . for the given geometry , however , this analysis does display the following important observations : * in terms of the largest effect on the overall contact area @xmath2 , changing the material starting and final thicknesses and the pitch had the largest effect .
this is also true for all of the area components , @xmath4 , @xmath6 and @xmath8 . in order of precedence ,
the variables that had largest sensitivity on the overall area other than thicknesses and pitch were the radius of the mandrel , the attack angle , radius of the roller , with the nose radius having the least effect overall .
* varying the the roller nose radius had the least effect on the contact area as well as the @xmath4 and @xmath8 components . *
the rolling component , @xmath8 , followed the same trends as the overall area for changes in thicknesses , pitch , mandrel / roller radii and attack angle .
this component decreased while the extrusion / drawing component and the overall area increased for larger roller nose radii .
furthermore , @xmath8 , is more sensitive to the attack angle than the mandrel radius . * the largest effect on the @xmath4 component , or the drawing / extrusion part of the deformation saw the same precedence of variables as for the overall contact area .
this component showed the same response to variable changes as the overall area . *
the tangential deformation component , @xmath6 , is marginally more sensitive to the radius of the roller than the radius of the mandrel , and the roller nose radius has approximately the same sensitivity as the attack angle .
this component also increased for larger values of pitch and mandrel radius , but decreased for larger roller radii and nose radii .
@xmath6 remained unaffected by changes in attack angle .
* the overall contact area increased with increased variable values in all cases except for the final thickness value and the attack angle .
this decrease was a linear for the former and non - linear for the latter . *
the effect of changing the starting thickness , final thickness and roller nose radius is a linear change for all area components while all others are non - linear .
these findings are of practical importance to flow forming .
if a worn roller is to be re - used after resurfacing , it may be necessary to modify the pitch in order to maintain the same forming geometry when the process was first commissioned .
if a single set of rollers are to be used with different mandrels , it is also important from a process design standpoint so that the same forming zone geometry can be maintained . furthermore , knowing the sensitivity of each of the variables on the overall contact and therefore deformation mode also permits easier troubleshooting of existing processes .
an analytical model of the roller / workpiece interface in flow forming has been developed such that it may predict the contact area .
this model is applicable to all tooling geometries for both forward and backward flow forming processes . due to the general nature of the description of the geometry
, the approach taken can be used for other rotary forming operations where a die or a roller is used to deform a cylindrical workpiece locally .
this model has been compared to experimental data generated from physical modelling and shows excellent correspondence . specifically , the analytical model was found to describe the experimental surface within 0.4 mm based on mean square error .
an example of the application of the model has been demonstrated in the form of a ofat sensitivity analysis applied to independent geometric variables determining tooling interaction .
the independent geometric variables examined were starting and final thicknesses ( @xmath34 , @xmath35 ) , forming pitch ( @xmath18 ) , mandrel radius ( @xmath24 ) , attack angle ( @xmath0 ) as well as roller and roller nose radii ( @xmath25 , @xmath20 ) . these variables were modified over a range of @xmath71 for the starting thickness and 150% for all others .
specific findings showed that on the basis of a unit change in the respective variables : * @xmath34 had four times the effect on the the change in overall area and components * @xmath35 had 33% more of an effect * @xmath18 had 50% less of an effect , with the exception of the tangential deformation component , @xmath6 which had 50% more of an effect * @xmath24 , @xmath0 , and @xmath25 have less than a @xmath72 effect * @xmath20 caused the least change : less than 7% change in area the present work could be extended to study the multi - variant effects on the contact area to fully account for the geometric changes during complicated forming processes .
however , geometric factors are not the only process parameters which govern the process .
the main direction of future work is to link the geometric factors to other process factors such as workpiece material properties and tribological considerations to gain deeper insight into the overall process mechanics . | flow forming involves complicated tooling / workpiece interactions .
purely analytical models of the tool contact area are difficult to formulate , resulting in numerical approaches that are case - specific .
provided are the details of an analytical model that describes the steady - state tooling / workpiece contact area allowing for easy modification of the dominant geometric variables .
the assumptions made in formulating this analytical model are validated with experimental results attained from physical modelling .
the analysis procedure can be extended to other rotary forming operations such as metal spinning , shear forming , thread rolling and crankshaft fillet rolling .
flow forming , metal forming , physical modelling , contact interface , analytical model |
clusters are cosmological probes for the formation and evolution of the universe through measurement of their mass - to - light ratios or clustering properties . in addition , they are very useful astrophysical laboratories for the study of the properties of their galactic population . a number of methods to detect clusters have been developed based on the clusters x - ray emission @xcite , weak lensing @xcite and sunyaev - zeldovich effect @xcite .
furthermore , a number of cluster detection methods based on optical data have provided a large dataset of clusters .
these methods are based on modeling different properties of the clusters : geometric distribution of the galaxies @xcite ; luminosity and density profiles @xcite and cluster red sequence , galaxy colors and brightest cluster galaxy ( bcg ) magnitudes @xcite .
all of these different kind of methods have just put in evidence the need to have an unbiased cluster sample that covers both a wide range in mass and redshift in order to constrain the selection functions and the cosmological parameters of the universe .
a large number of studies of individual clusters at moderate and high redshift @xcite have showed statistical trends even with wide dispersion for a number of cluster properties such as the color - magnitude relation ( cmr ) , the blue fraction or the luminosity function .
however , these clusters have often been selected to be at the high end of the mass function .
the ideal goals are to determine the least unbiased sample of clusters , the highest completeness rates and the lowest cluster mass limit as possible .
the motivation of this work is to take advantage of all the characteristics of the present methods by modeling each cluster property .
however , we still want to detect a cluster if one of these properties is not present , like for example the cmr in high redshift clusters . to accomplish this ,
we have designed a bayesian cluster finder where each galaxy is assigned a bayesian probability that the galaxy belongs to a cluster at a certain redshift .
this probability can be decomposed into a likelihood which is based on a variation of the matched filter algorithm @xcite including photo - z information , and a bayesian prior , where we include previously known cluster properties ( ascaso et al . 2010b , apj , in prep .
) the likelihood models the probability that a galaxy with its position , photometric redshift , magnitude and morphological type belongs to a cluster at that position , with a given redshift and richness .
it is the product of the model probability for a cluster spatial profile ( we use a plummer profile ; @xcite ) , a luminosity function ( we use a schechter function , @xcite ) and a redshift probability distribution ( either from a photometric redshift software or from a gaussian approximation ) .
we show in table [ tab : all ] the parameters used in this work .
@xmath3 [ tab : all ] the prior enhances the probability that a cluster exists at a given position by including any a priori information about clusters .
we consider two main cluster properties to be included as prior information : the relation between the cluster cmr and its redshift and the bcg magnitude - redshift relation obtained from the maxbcg sample of 13823 bcgs @xcite .
we created redshift slices from 0.1 @xmath41.2 in steps of 0.1 .
then , for each redshift slice , we assigned each galaxy a probability .
we fit a gaussian to the overall probability for each redshift slice and use the peak and width of the gaussian as the background and the dispersion respectively .
galaxies 3@xmath5 above the background probability form the basis of cluster candidates .
then , we select the maximum probability as the initial center of the detection .
we make radial density profiles and consider the limits of the cluster when the profile becomes constant . after that , we center iteratively the structure in the brightest galaxy within 1.5 mpc .
the output consists of a richness @xmath6 ( ie , the effective number of l@xmath7 galaxies in the cluster , @xcite ) , the position of the cluster and the redshift slice which maximizes the probability .
we also obtained a different estimation of the redshift by fitting a gaussian to the photometric redshift distribution of the galaxy population .
finally , we filtered those candidates by requiring that the difference between the two redshift estimates is smaller than @xmath8 , i.e : the width of the bin multiplied by a factor depending on the redshift .
we also merged two or more detections if they were closer than 1.5 mpc and with estimated redshift difference smaller than 0.3 .
we have performed simulations to test the reliability of the results .
we simulated sets of clusters ranging 10 @xmath9 200 and 0.1 @xmath10 1.2 .
the magnitudes were distributed by following a schechter luminosity function with fixed parameters @xmath11 and @xmath12 .
the positions of the galaxies in the cluster were distributed according to a plummer profile and the redshifts of the member galaxies were spread following a gaussian function to mimic photometric redshift errors . finally , we simulated the galaxy cluster colors by using the empirical cluster color distribution by @xcite , combined with a shift to the expected color , obtained from synthetic spectral templates .
these simulated clusters were embedded into a field galaxy distribution .
the magnitudes , colors and photo - z distribution were taken from the original data and we redistributed the positions by following a rayleigh - levy galaxy pair distribution function .
the results of the simulations show a completeness fraction always higher than 80% for clusters up to redshift @xmath0 1.2 and for every richness , while the purity is over @xmath13 80% for clusters with @xmath14 up to redshift 1.2 .
we find that the purity decreases about 20% for detections at [email protected] and @xmath15 if we do not include the prior information in our probability .
however , the completeness stays approximately the same .
hence , the prior helps to discriminate the spurious detections that might be detected as clusters at low redshift and to increase purity . in view of these results , to keep the completeness and purity rates over 80% , we will consider real cluster detections those with @xmath14 up to redshift 1.2 with the prior turned on .
we applied this algorithm to two different surveys , obtaining the following results .
the cfhtls - archive - research survey ( cars ; @xcite ) consists of catalogues obtained from the public archive images from the cfhtls - wide ( 37 square degrees ) .
the survey has five optical bands ( ugriz ) , it is complete up to 24 in r band and it has photometric redshift available obtained from bpz @xcite .
we detected galaxy clusters in this survey and compared the results with other surveys which use the cfhtls data , such as @xcite in cfhtls - deep ( which overlaps one square degree with cars ) and @xcite which used both wide and deep fields .
we detected @xmath16 90% of the detections found in works @xcite and @xcite , whereas @xcite and @xcite detect approx 70% and 80% of our detections .
in particular , we find a number of extra detections at z @xmath16 0.9 - 1 that look probably real ( ascaso et al .
we plan to observe a sample of these candidates to confirm its existence .
the deep lens survey ( dls ; @xcite ) is a very deep @xmath17 imaging survey of five @xmath18 degree fields .
the depth is 27 mag arcsec@xmath19 in r and 26 mag arcsec@xmath19 in b , v and z. we detected more than 780 galaxy clusters in the whole dls up to redshift 1.2 , detecting 100% of other optical detections from shallower galaxy cluster samples such as maxbcg , @xcite up to [email protected] .
we also detect 100% of all the spectroscopically confirmed clusters @xcite detected from weak lensing ( wl ) .
we are also able to map large scale structure and compare the optical detections with the already available wl detections ( ascaso et al . 2010c , apj , in prep . ) . in figure
[ fig : wl ] , we show the wl mass contours in color , with the the optical cluster detection weighted by the luminosity and the wl kernel in white .
large scale structures may be seen in both mass and weighted light .
dls field 2 ( 4 square degrees ) .
the color map shows the weak lensing mass map , while the white contours refer to the optical galaxy cluster density map weighted by absolute luminosity and the weak lensing kernel .
the most prominent structures are abell 781 ( upper left quadrant ) at z=0.3 , and a string of clusters at z=0.53 at lower right.,width=313 ]
the bayesian cluster finder is the first optical cluster detection technique that combines the matched filter algorithm with a probability enhancement provided from cluster cmr and bcg magnitude if any .
we tested the method with simulations , which show 100% completeness and @xmath20 80% purity up to z @xmath01.2 for clusters richer than @xmath21 .
we have successfully detected clusters in the dls and cars , providing good agreement with previous work and some additional detections at high redshift .
this algorithm is ready to be applied to any survey .
we are very grateful to the loc and soc for the excellent organization .
ba and dw acknowledge the support of nasa grant nng05gd32 g .
adami , c. , durret , f. , benoist , c. , et al .
2010 , , 509 , a81 ascaso , b. , moles , m. , aguerri , j. a. l. , et al . , & varela , j. 2008 , , 487 , 453 ascaso , b. , aguerri , j. a. l. , moles , m. , et al .. 2009 , , 506 , 1071 baldry , i. k. , glazebrook , k. , brinkmann , j. , et al .
2004 , , 600 , 681 bentez , n. 2000 , , 536 , 571 botzler , c. s. , snigula , j. , bender , r. , et al .
2004 , , 349 , 425 00 carlstrom , j. e. , holder , g. p. , & reese , e. d. 2002 , , 40 , 643 de propris , r. , colless , m. , peacock , j. a. , et al .
2004 , , 351 , 125 eisenhardt , p. r. m. , brodwin , m , gonzalez , a. h. , et al .
2008 , , 684 , 905 erben , t. , hildebrandt , h. , lerchster , m. , et al .
2009 , , 493 , 1197 gal , r. r. , de carvalho , r. r. , lopes , p. a. a. , et al .
2003 , , 125 , 2064 gladders , m. d. , & yee , h. k. c. 2000 , , 120 , 2148 gladders , m. d. , & yee , h. k. c. 2005 , , 157 , 1 harsono , d. , & depropris , r. 2009 , , 137 , 3091 kim , r. s. j. , kepner , j. v. , postman , m. , et al .
2002 , , 123 , 20 koester , b. p. ; mckay , t. a. ; annis , j. , et al .
2007 , , 660 , 221 lopes , p. a. a. , de carvalho , r. r. , gal , r. r. , et al .
2004 , , 128 , 1017 lpez - cruz , o. , barkhouse , w. a. , & yee , h. k. c. 2004 , , 614 , 679 milkeraitis , m. , van waerbeke , l. , heymans , c. , et al . 2010 , , 701 olsen , l. f. , benoist , c. , cappi , a. , et al .
2007 , , 461 , 81 postman , m. , lubin , l. m. , gunn , j. e et al .
1996 , , 111 , 615 postman , m. , lauer , t. r. , oegerle , w. , et al .
2002 , , 579 , 93 ramella , m. , boschin , w. , fadda , d. et al .
2001 , , 368 , 776 ramella , m. , geller , m. j. , pisani , a. et al .
2002 , , 123 , 2976 rosati , p. , borgani , s. , & norman , c. 2002 , , 40 , 539 schechter , p. 1976
, , 203 , 297 tyson , j. a. , wenk , r. a. , & valdes , f. 1990 , , 349 , l1 wilson , g. , muzzin , a. , lacy , m. , et al . 2008 , infrared diagnostics of galaxy evolution , 381 , 210 wittman , d. , tyson , j. a. , margoniner , v. e. et al .
2001 , , 557 , l89 wittman , d. m. , et al .
2002 , , 4836 , 73 wittman , d. , margoniner , v. e. , tyson , j. a. et al .
2003 , , 597 , 218 wittman , d. , dellantonio , i. p. , hughes , j. p. , et al . 2006 , , | the detection of galaxy clusters in present and future surveys enables measuring mass - to - light ratios , clustering properties or galaxy cluster abundances and therefore , constraining cosmological parameters .
we present a new technique for detecting galaxy clusters , which is based on the matched filter algorithm from a bayesian point of view .
the method is able to determine the position , redshift and richness of the cluster through the maximization of a filter depending on galaxy luminosity , density and photometric redshift combined with a galaxy cluster prior .
we tested the algorithm through realistic mock galaxy catalogs , revealing that the detections are 100% complete and 80% pure for clusters up to z @xmath01.2 and richer than @xmath1 25 ( abell richness @xmath2 0 ) .
we applied the algorithm to the cfhtls archive research survey ( cars ) data , recovering similar detections as previously published using the same data plus additional clusters that are very probably real .
we also applied this algorithm to the deep lens survey ( dls ) , obtaining the first sample of optical - selected galaxy in this survey .
the sample is complete up to redshift 0.7 and we detect more than 780 cluster candidates up to redshift 1.2 .
we conclude by discussing the differences between previous weak lensing detections in this survey and optical detections in both samples . |
entanglement entropy , a popular measure to quantify quantum entanglement , has become a subject of intensive theoretical investigation , especially for systems with many degrees of freedom @xcite and is being used to characterize properties of a wide spectrum of systems like quantum information processing @xcite , quantum phase transition @xcite and entropy of black holes @xcite .
however , the entanglement entropy of free quantized fields is found to be divergent @xcite and some form of regularization has to be used in order to extract useful information from it ( also see @xcite in this context ) .
furthermore , in general , it is difficult to get an analytic handle on entanglement entropy except for a few special cases like @xmath0 dimensional cfts @xcite . in this work , with the aim to gain a better understanding of the divergence of entanglement entropy as well as to have better analytic control , we consider in detail the entanglement entropy of free scalar field regularized on a spatial lattice in @xmath0-dimensional space - time . with the insights gained in @xmath2 dimensions
, we extend the results to higher dimensions .
specifically , we obtain analytical expression for the entanglement entropy by tracing over a single oscillator of the lattice regularized scalar field .
the analytical expression provides two interesting features which , to our knowledge , have not been noted in the literature : ( i ) entanglement entropy is invariant under a scaling transformation even when the hamiltonian is not , and ( ii ) the divergence in entanglement entropy in @xmath3 dimensions in the continuum limit is due to the presence of a large number of near zero modes ( and is not of uv origin as commonly believed ) . in the case of higher dimensions ,
accumulation of zero modes occur , however , the entropy remains finite ( non - divergent ) . the work is organized as follows . in the next section
we start with the discretized version of free , massive scalar field in @xmath0 dimensions and define the covariance matrix for the corresponding hamiltonian and show its relation to the entanglement entropy . in section ( 3 ) , after performing a canonical transformation on the phase space of the scalar field theory , we show that the divergence of entanglement entropy in the continuum is due to the presence of near zero frequency modes .
it is further shown that the entropy is divergent even when a single oscillator is traced over .
section ( 4 ) is devoted to regulating this divergence using a suitable infra - red cut - off . in section ( 5 )
we consider the effect of spatial dimensions on entanglement entropy of free fields .
we conclude in section ( 6 ) .
we set @xmath4 .
the system of interest is the @xmath0 dimensional massive , free scalar field theory described by the lagrangian @xmath5 where @xmath6 is the mass of the scalar field .
the corresponding hamiltonian is @xmath7 as mentioned earlier , the ground state entanglement entropy for such a system is divergent . in order to gain better understanding of the divergence
, we place the system on a spatial lattice with lattice spacing @xmath8 .
using the notation @xmath9 @xmath10 , where @xmath11 denotes the position of the lattice points , the discretized hamiltonian is @xmath12 where @xmath13 is the canonically conjugate momentum of @xmath14 .
it is important to note that we have an infinite lattice in mind so that @xmath15 ; the continuum limit corresponds to @xmath16 . in the following
we assume periodic boundary conditions @xmath17 ( the final results are , however , independent of the specific choice of boundary conditions ) .
our aim is to analytically calculate the entanglement entropy of the ground state wave - function of the above hamiltonian obtained by tracing over first @xmath18 oscillators .
this can be done by finding the ground state wave - function and then performing the partial trace . however , for greater generality , instead of taking this route , we calculate the entanglement entropy by finding the covariance matrix , which for the hamiltonian is the following @xmath19 matrix @xcite @xmath20.\end{aligned}\ ] ] in the above equation @xmath21 are the normal mode frequencies @xmath22 and @xmath23 is the potential matrix of the hamiltonian .
the matrix elements of the ` position ' correlation @xmath24 and ` momentum ' correlation @xmath25 depend only on the separation @xmath26 between the @xmath27th and the @xmath28th oscillators . the reduced state obtained after tracing over @xmath29 oscillators can be characterized from its covariance matrix @xmath30 by picking appropriate elements from the total matrix .
the entanglement entropy is given by @xcite @xmath31 where @xmath32 are the symplectic eigen values of the reduced covariance matrix .
to have better analytic control in order to identify the scaling symmetry , we consider the simplest case of the single oscillator reduced system @xmath33 , for which the covariance matrix is @xmath34.\ ] ] here we would like to note a couple of things regarding the covariance matrix .
the @xmath35 ( @xmath36 ) element of the covariance matrix , often referred to in the literature as the ` position ' ( momentum ) covariance @xcite , is @xmath37 in the continuum limit ( @xmath16 ) , the position covariance diverges while the momentum covariance is finite ( non - zero value ) .
hence , the product @xmath38 , which is the determinant of the covariance matrix , diverges .
this determinant @xmath39 is the eigenvalue @xmath40 , from which the entropy can be calculated using .
the following points are worth noting regarding the above result : ( i ) the entropy is invariant under the scaling transformations @xmath41 and ( ii ) in the continuum limit , @xmath16 , @xmath42 and , with the summations going over to integrals , the entanglement entropy @xmath43 since the numerator in diverges for @xmath44 .
this is the familiar uv - divergence of the entanglement entropy in the continuum . in the next section ,
we show that the canonical transformation of the variables @xmath45 which accounts for the scaling symmetry will lead to a hamiltonian that can be separated into a scale invariant and a scale dependent part .
the entanglement entropy of the resultant hamiltonian leads to a new way of identifying the cause of the divergence .
to take into account the scaling symmetry , we introduce the following canonical rescaling of the hamiltonian @xmath46 in terms of these variables the hamiltonian can be written as @xmath47 here we have defined @xmath48 it is interesting to note that ( i ) under the scaling transformation , @xmath49 while @xmath50 is scale invariant .
thus , in writing the hamiltonian , we have separated the scale invariant part of the hamiltonian from the scale dependent part ( which appears solely in the factor @xmath51 ) .
( ii ) the canonical transformations ( [ rescaling the canonical variables ] ) are well - defined for all values of @xmath8 ( including the continuum limit whereby @xmath52 and @xmath53 ) .
( the above canonical transformations have been discussed by botero and reznik @xcite in a different context . ) since the determinant of the covariance matrix and , hence , the entanglement entropy are invariant under the canonical transformations , we can express the determinant of the covariance matrix ( [ determinant reduced state ] ) in terms of @xmath50 by pulling out a factor of @xmath54 from both the numerator and the denominator leading to @xmath55 this points to the fact that , for the purpose of evaluation of the entanglement entropy , instead of working with the full hamiltonian ( [ discretized hamiltonian in terms of rescaled variables ] ) , it is sufficient to work with the following ` effective ' hamiltonian @xmath56 from which we identify the following normal mode frequencies @xmath57 note that @xmath58 is related to the normal mode frequency @xmath59 defined in by @xmath60 thus , the resultant entanglement entropy depends _ only _ on the scale - invariant parameter @xmath50 . as pointed out earlier , in the continuum limit @xmath50 is finite while @xmath61 diverges .
( [ covariance matrix for discretized scalar field in terms of beta ] ) gives insight into the real cause of the divergence . to understand this ,
let us first consider the case where @xmath52 . for @xmath62 mode ,
the effective normal mode frequency is @xmath63 .
this implies that in eq . the discrete sum over @xmath27 would diverge because of the appearance of ( at least ) one term with zero in the denominator ( the zero mode term )
this shows that the divergence is due to the zero mode of @xmath58 .
in fact , in this limit , large number of near - zero modes _ accumulate _ leading to the vanishing of the effective normal mode frequencies and hence divergence of the entanglement entropy .
another ( heuristic ) way to understand how the innocuous canonical transformation ( [ rescaling the canonical variables ] ) identifies the divergence of the entanglement entropy due to the near - zero modes is to look at the effective hamiltonian ( [ hamiltonian without e0 ] ) . in the continuum limit ,
the last term on the right becomes @xmath64 and this term cancels the second term in the hamiltonian ( [ hamiltonian without e0 ] ) .
thus , in the continuum limit , there are large number of modes which effectively behave as free particles . in the context of quantum field theory ,
zero modes are excluded on the grounds that they are not normalizable and that they do not have particle interpretation .
although the zero modes have no physical effects , they carry @xcite an undetermined , non - zero energy @xmath65 .
this undetermined non - zero energy indicates the divergence of the entanglement entropy in @xmath0dimensions .
having identified the divergence of the entanglement entropy in the continuum due to the presence of zero frequency modes , we still need to isolate the divergent term from the finite terms .
we start by taking @xmath67 limit in eq . , and replace the summations by integrals ( with @xmath68 ) @xmath69 in the continuum limit @xmath70 , and as noted earlier , the first integral diverges . to isolate this divergence we introduce a cut - off function @xmath71 which has the desired property @xmath72 from the first integral in it is clear that in the continuum ( @xmath73 ) the divergent contribution comes only from the region @xmath74 where @xmath75 and we get @xmath76 rhs of the above expression , in the limit when the cut - off parameter @xmath77 , can be approximated by @xmath78 where @xmath79 is the euler - mascheroni constant .
the remaining terms in the series expansion of @xmath80 in the first integral in give a finite contribution of @xmath81 while the contribution from the second integral is @xmath82 ( in evaluating these finite contributions we have put @xmath83 ) . finally , using for @xmath84
, we find the entropy to be the sum of a finite contribution and a divergent contribution @xmath85 it is interesting to note that the divergence in entropy is very slow going only as a double logarithm .
using the boltzmann definition of entropy as logarithm of number of states , eq .
( [ entanglement entropy in 1 + 1 continuum ] ) gives @xmath86 the above expression explicitly shows that large number of near - zero modes ( @xmath87 ) leads to the divergence of the entanglement entropy .
until now , we have focused on the entanglement in @xmath88 partition as entanglement entropy can be evaluated analytically .
it is straight forward to extend to a general partition @xmath89 using eq . .
however , for a general partition it is no longer possible to obtain analytic results and the determinant of the covariance matrix , for instance , has to be computed numerically .
figure [ entropy in 1 + 1 dim as a function of m ] shows the plot of entropy as a function of @xmath90 ( the number of oscillators traced over ) for different values of @xmath50 and it is seen to be an increasing function of @xmath90 . since entanglement entropy depends on the correlations across boundary @xcite , one would expect that only oscillators near the boundary of the traced out region would contribute which would imply that the entropy should be a constant in @xmath0 dimensions even as more and more oscillators away from the boundary get traced out . in one dimensional chains ,
although this holds for weakly coupled chains as shown by plenio et.al
@xcite , it breaks down in the strongly coupled systems @xcite and the entropy grows as @xmath91 .
, the area law is violated , width=302 ] + for the critical case of @xmath73 , introducing the cut off function @xmath92 defined earlier , the entropy @xmath93 for @xmath90 oscillator reduced state gives a numerical fitting ( ignoring the terms vanishing in @xmath77 limit ) @xmath94 where @xmath95 is a constant .
to investigate the effects of dimensionality on the behavior of entanglement entropy , we first consider the scalar field in @xmath96 dimensions ( discretized on a square lattice ) .
the discretized version of the hamiltonian for a scalar field in @xmath96 dimensions is @xmath97 ^ 2 + \frac{1}{2}\left[\eta(i , j+1)-\eta(i , j)\right]^2 .\nonumber\end{aligned}\ ] ] here @xmath98 denotes the oscillator at position labeled by @xmath99 on the two dimensional lattice and @xmath100 is the corresponding conjugate momentum
. the hamiltonian can be diagonalized by going to normal coordinates and the corresponding normal mode frequencies are ( compare with eq . for @xmath0 dimensions ) @xmath101 where @xmath102 . to explicitly see the finiteness of the determinant ,
let us replace the summations in ( [ position covariance 2 + 1 dim ] , [ momentum covariance 2 + 1 dim ] ) with integrals ( with @xmath106 ) @xmath107 notice that , unlike @xmath0 dimensions , the result is finite as can easily be seen by replacing @xmath108 ( regions away from @xmath109 not contributing to the divergence ) .
the entanglement entropy is given by and as in @xmath2dimensional case , it is invariant under the scaling transformations : @xmath110 and in the continuum limit , the entropy is divergent due to the uv modes .
taking into account the scaling symmetry , introduce a canonical rescaling ( similar to that in @xmath111dimensions ) @xmath112 in terms of the transformed variables the hamiltonian becomes @xmath113 where @xmath114 it is important to highlight the differences between @xmath2dimensions and higher dimensions : first , unlike eqs .
( [ scaling transformation ] ) , canonical transformations ( [ scaling transformation 2 + 1 dim ] ) in @xmath105 dimensions have an explicit @xmath8 dependence .
this is due to the fact that the field and the canonically conjugate momentum @xmath115 have the dimensions @xmath116^{-1/2 } , [ l]^{1/2})$ ] . in the case of @xmath2 dimensions
@xmath117 ( @xmath118 ) are dimensionless .
hence , the strict @xmath119 is ill - defined in the case of @xmath105 ( or higher ) dimensions , second , like in @xmath2dimensions , the hamiltonian ( [ transformed hamiltonian 2 + 1 dim ] ) is separated into a scale invariant part and a scale dependent part . in the small @xmath8 limit , @xmath120 .
third , the position and the momentum covariance corresponding to the rescaled hamiltonian ( [ transformed hamiltonian 2 + 1 dim ] ) are @xmath121 + 2 \beta \sin^2 \left[\frac{\pi m}{n}\right]}\end{aligned}\ ] ] are finite .
this should be contrasted with the @xmath0 dimensional case where the canonical rescaling does not change the behavior of the position and momentum covariance .
finally , like in @xmath0 dimensions , the entropy depends _ only _ on @xmath50 and , hence , it is sufficient to work with the following effective hamiltonian @xmath122 whose normal mode frequencies are @xmath123}.\ ] ] although the canonical transformations are not well - defined in @xmath124 limit , @xmath50 and normal mode frequencies are well - defined in this limit . the above calculation can be extended to any higher space dimensions .
the only difference is that the critical value of @xmath125 where @xmath126 is the number of space dimensions .
heuristically , as in the previous case , in @xmath124 limit , the last three terms in the rhs of the hamiltonian cancel giving rise to the free particle hamiltonian .
this shows that there is an accumulation of near - zero modes , however , unlike the previous case the contribution is non - divergent .
it is important to note that in both the cases [ @xmath2 and @xmath127dimensional space - time ] , in the continuum limit , the entanglement entropy behaves differently .
one possible explanation for this difference is the following : entanglement entropy arises due to the correlations across the boundary @xcite . in @xmath0 dimensions the boundary separating the two regions is
the zero dimensional point and a single oscillator , occupying a point in space , is dense in the boundary ( in the set - theoretic sense ) leading to a divergent entropy . on the other hand ,
the boundary in @xmath96 dimensions is one dimensional and the single traced over oscillator , occupying a point is not dense in the 1-dimensional spatial boundary and , hence , the entropy is finite .
entanglement entropy for free fields is divergent . in this work we revisited the question of the divergence in the light of zero modes . to obtain better analytic control we focused on the case where only a single oscillator ( in the discretized version of free scalar field )
was traced over .
we found that the entanglement entropy has a scaling symmetry which the hamiltonian does not possess . in terms of the canonically
transformed variables , the hamiltonian can be separated into a scale dependent and scale invariant part .
we have shown that the origin of the divergence of the entanglement entropy in @xmath111dimensions in the continuum limit is due the presence of accumulation of large number of ( near-)zero frequency modes . in the context of quantum field theory ,
zero modes are excluded on the grounds that they are not normalizable and that they do not have particle interpretation .
although the zero modes have no physical effects , they carry @xcite an undetermined , non - zero energy @xmath65 . using an ir cutoff
, we have shown that the undetermined non - zero energy leads to the divergence of the entanglement entropy in @xmath111dimensions . in higher dimensions ,
although , there is an accumulation of near - zero modes , their contribution to the entanglement entropy is non - divergent .
one possible explanation for this difference between @xmath0 and higher dimensions is the following : entanglement entropy arises due to the correlations across the boundary @xcite . in @xmath0 dimensions the boundary separating the two regions is the zero dimensional point and a single oscillator , occupying a point in space ,
is dense in the boundary ( in the set - theoretic sense ) leading to a divergent entropy . on the other hand ,
the boundary in @xmath96 dimensions is one dimensional and the single traced over oscillator , occupying a point is not dense in the 1-dimensional spatial boundary and , hence , the entropy is finite .
we would like to thank a.p .
balachandran , samuel braunstein , sourendu gupta , namit mahajan and v.p .
nair for useful discussions .
km is supported by dst , government of india through kvpy fellowship .
the work of ss and rt is supported by max planck partner group in india .
ss is partly supported by ramanujan fellowship of dst , india .
work of tp is partly supported by j. c. bose fellowship of dst , india . | we investigate the cause of the divergence of the entanglement entropy for the free scalar fields in @xmath0 and @xmath1 dimensional space - times . in a canonically equivalent
set of variables , we show explicitly that the divergence in the entanglement entropy of the continuum field in @xmath2 dimensions is due to the accumulation of large number of near - zero frequency modes as opposed to the commonly held view of divergence having uv origin .
the feature revealing the divergence in zero modes is related to the observation that the entropy is invariant under a hidden scaling transformation even when the hamiltonian is not .
we discuss the role of dispersion relations and the dimensionality of the space - time on the behavior of entanglement entropy . |
the key to all matter - wave interferometry are coherent beamsplitters which divide each incident wave into separated wavelets with well - defined phase relations @xcite .
while modern atom interferometry often utilizes the momentum recoil of resonant laser light to split the atomic wave @xcite , mechanical masks @xcite as well as optical phase @xcite or absorption @xcite gratings can also be used to divide a matter - wave front .
since the universality of mechanical gratings is compromised by the van der waals ( vdw ) potential it is important to ask to what extent it is possible to minimize this interaction by reducing the grating thickness .
this technological feat - to actually create free - standing nanogratings in even a single layer of atoms @xcite - is accompanied by the fundamental question whether the path of a massive particle through the multi - slit array will become significantly entangled with the mechanical motion of the recoiling ultra - thin structure .
if this were the case we would expect to observe loss of the interference contrast .
using a focused ion beam we were able to mill a periodic diffraction grating into a single layer of graphene that was suspended over a silicon nitride ( sin@xmath1 ) membrane ( fig .
[ fig : gratings]a , see appendix sec.[app : gratings ] for details ) .
the nanoribbons ( @xmath2 nm wide ) were written with a period of @xmath3 nm ( fig .
[ fig : gratings]c ) .
they spontaneously transform into carbon nanoscrolls @xcite , here with a diameter down to @xmath4 nm ( fig .
1b and appendix sec.[app : scrolls ] ) . for shorter grating bars scanning transmission electron microscopy ( stem ) reveals , however , stable flat single - layer graphene structures ( fig .
[ fig : gratings]e ) .
we were also able to write gratings into bilayer graphene suspended across a lacey carbon mesh ( fig .
[ fig : gratings]f ) .
the second layer of carbon atoms suppresses the formation of nanoscrolls entirely on the observed scale .
bilayer graphene may also form bonds at open cuts and thus expose closed edges and a thicker wall than expected based on the number of layers alone @xcite .
we compare these structures finally to an insulating structure of almost identical thickness , the carbonaceous biphenyl membrane @xcite ( also on lacey carbon , fig .
[ fig : gratings]g , ) and reference all images to the diffraction of molecules at @xmath5 nm thick silicon nitride ( fig .
[ fig : gratings]h ) @xcite .
figure [ fig : gratings ] demonstrates that stable structures can be written even into atomically thin membranes . the diffraction pattern behind a purely absorptive periodic mask - _ i.e. _ without any phase modulation can be described as the convolution of two contributions : the diffraction at each single slit of width @xmath6 and the diffraction at an array of infinitely thin slits of period @xmath7 .
the fringe positions are well described by wave theory .
the fringe amplitudes are modulated by the single - slit diffraction pattern . comparing fig .
[ fig : overview](a ) - ( e ) , we see immediately vast differences in the diffraction at geometrically similar gratings . for @xmath5 nm thick silicon nitride ( fig .
[ fig : overview]a ) we observe interference up to the @xmath8th diffraction order , which can only be understood if we complement the quantum wave model by the assumption that the position - dependent phases accumulated in the presence of vdw forces alter the effective transmission function . approximating this interaction by reducing
the effective slit width @xcite ( see appendix sec.[app : kirchhoff ] ) allows us to estimate the strength of the vdw interactions . for silicon nitride the analysis yields an effective slit width of @xmath9 nm - a reduction of the open width by @xmath10 .
this influence is strongly reduced for the single - layer graphene grating , the probably thinnest conceivable grating ( fig .
[ fig : overview]e , @xmath11 ) .
as expected this leads to a strong suppression of all diffraction orders beyond the first one .
this holds true - and even more so - for gratings made from nanoscrolls ( @xmath12 ) , which maximize the opening fraction , _
i.e. _ the ratio of slit width to period .
in contrast to that , the diffraction pattern both behind the carbonaceous biphenyl membrane ( @xmath13 ) and the bilayer graphene ( @xmath14 ) show substantial populations up to the fifth diffraction order .
this indicates a stronger influence of the attractive van der waals force .
while the carbonaceous biphenyl membrane should be an insulator , the bilayer sample is conducting .
however , both imprint similar phase shifts onto the transmitted molecules .
we attribute the remaining van der waals forces in particular also to the lacey carbon support structure which is thicker ( up to @xmath15 nm ) and spatially less well controlled than the sin@xmath1 support of the other ultra - thin gratings .
the central result of our diffraction experiments is the observation of high - contrast interference for all gratings , where the contrast reaches its maximum when the interference minima approach the zero ( background ) level .
it is the absence of signal in these minima which contradicts any classical expectation the most .
our result relates to a thought experiment between bohr and einstein who elucidated the role of decoherence in double slit diffraction @xcite .
einstein predicted that the observer should be capable of extracting which - path information from the recoil that the double slit receives upon diffraction of a particle ( fig .
[ fig : born ] ) . invoking heisenberg s uncertainty relation ,
bohr was able to demonstrate , however , that the recoil imparted by the diffracted particles remains within the momentum uncertainty of the grating if this is well positioned .
different recent studies have realized the probably smallest double slits and analogs to this einstein - bohr debate using diatomic molecules as the diffraction elements in a photoionization process @xcite .
they were destroyed in the diffraction process and therefore were capable of carrying away path information .
our carbon nanostructures are probably the thinnest durable realization of the bohr - einstein idea .
these atomically thin membranes are flexible but they last for months and survive the diffraction of millions of molecules .
the observed full contrast shows that coherence prevails , which we interpret as an indication that the momentum exchange with the grating remains within the intrinsic momentum uncertainty of the grating bars ( see appendix [ app : heisenberg ] ) @xcite .
hence , diffraction does not lead to a loss of fringe visibility even though it can cause a sizable matter - wave ( vdw ) phase shift .
graphene gratings are ten times thinner than any other beamsplitter for atoms , molecules or clusters before and they are four orders of magnitude thinner than the width of a typical laser grating @xcite .
our results on single - layer graphene are the best approach with regard to the diffraction of high - mass molecules at nanomechanical gratings , since they show that the effect of molecule - surface interactions can be substantially reduced , even though they do not eliminate them entirely .
this property makes graphene nanogratings also appealing for use with other kinds of matter - waves .
the reduced phase components may enable new coherence experiments with slow ions or anti - matter @xcite .
it will also be interesting to explore atomically thin but insulating 2d sheets like boron nitride or molybdenum disulfide , in the future . even though van der waals forces in thick gratings can be prohibitive for high - mass diffraction
, they can also serve a purpose in metrology : our experiments show that a thick material grating ( sin@xmath16 , fig .
[ fig : overview]a ) can act as a large momentum transfer beamsplitter . if we were to calibrate the momentum exchange between the extreme diffraction orders ( @xmath17 order for @xmath18 nm ) in units of the photon momentum ( @xmath19 ) that is usually imparted in state - of - the - art beamsplitters @xcite for ( rubidium ) atoms , we see that the diffraction at our nanogratings amounts to @xmath20 .
photolithography can reliably provide thick gratings with a periodicity of better than @xmath21 nm over large areas @xcite .
stable thick membranes with narrow slits are predicted to provide a momentum transfer even beyond @xmath22 @xmath19 .
finally , our work shows that path - decoherence close to material gratings is still negligible in all settings discussed here .
this applies in particular for the bohr - einstein gedankenexperiment .
even the conceivably thinnest durable mechanical grating is still sufficiently massive and sufficiently localized not to encode sizable recoil - information .
the intrinsic momentum uncertainty of each grating bar is larger than the recoil imparted on each diffracted molecule ( see appendix [ app : heisenberg ] ) .
this holds for all particles , independent of their mass , as long as we can keep their de broglie wave - length , _
i.e. _ their momentum , the same .
for instance , quantum diffraction of insulin molecules seems conceivable once we can provide a directed molecular beam of about @xmath23 m/s velocity .
finally , it will also be intriguing to explore the role of quantum friction and path - decoherence @xcite which may become particularly relevant for polar molecules in the future .
we acknowledge support by the european commission ( 304886 ) , the european research council ( 320694 ) and the austrian science funds ( dk coqus w1210 - 3 ) .
cb acknowledges the financial support of the alexander von humboldt foundation through a feodor lynen fellowship .
jk acknowledges the austrian science fund fwf for funding through project m 1481-n20 .
jm and cm acknowledge support from the austrian science funds fwf project p 25721-n20 .
aw and at acknowledge support of the dfg ( spp `` graphene '' tu149/2 - 2 , heisenberg program tu149/3 - 1 ) .
tj acknowledges support by the gordon and betty moore foundation .
we thank the group of prof .
schattschneider , ustem tu vienna for assistance in recording image [ fig : gratings](h ) .
we thank stefan scheel and johannes fielder ( univ .
rostock ) for fruitful discussions .
41ifxundefined [ 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 @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) in @noop _
_ , ( , , ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop _ _ ( , )
coherent diffraction of a matter - wave is possible as long as the momentum transferred by the particle to the grating is within the momentum uncertainty of the grating itself . in diffraction experiments at a single slit , the relationship @xmath24 holds @xcite , when @xmath25 designates the width of the single slit and @xmath26 the full width at half maximum ( fwhm ) of the diffraction curve . the momentum uncertainty @xmath27 is the fwhm of the envelope to all diffraction orders of the grating .
this momentum is transferred to the grating , which itself is described by the uncertainty relationship @xmath28 . to estimate @xmath29 within the plane of the grating we need an independent measure of its position and movement at room temperature .
for a freely suspended single - layer graphene ribbon ( length @xmath30 width = @xmath31 @xmath32 m ) clamped on both sides capacity measurements by garcia - sanchez _ et al . _
@xcite allowed them to estimate an out - of - plane vibration amplitude of @xmath33 nm . the vibration amplitude of shorter ribbons and in particular of in - plane modes are expected to be even smaller - due to their large width - to - thickness ratio .
the value of @xmath21 nm therefore constitutes an upper boundary for the in - plane vibration amplitude for this system .
the largest amplitudes for in - plane vibrations are expected for the nanoscrolls grating because of the ( nearly ) rotational symmetry of the scrolls .
their vibrational properties might be inferred from the results for carbon nanotubes . here , the maximum displacement of a doubly clamped single - walled carbon nanotube due to thermal fluctuations is given as @xcite : @xmath34 where @xmath35 is the length of the nanotube , @xmath36 tpa is the young s modulus , here taken to be the value of a single wall carbon nanotubes , and @xmath37 is the area moment of inertia , with @xmath7 the scroll diameter . for the geometry of our nanoscroll bars
@xmath38 @xmath32 m , @xmath39 nm , @xmath40 k ) this yields the thermal oscillation amplitude @xmath41 nm . inserting this into heisenberg s uncertainty relation , we can derive a minimal slit width that is still compatible with the observation of coherent matter - wave diffraction , for which @xmath42 holds : @xmath43 this amounts to @xmath44 nm for our nanoscrolls .
such closely spaced nanoscrolls can not be grown from cutting extended nanoribbons , since these ribbons define their minimal separation . in the future experiment it may become possible to tailor an array of single - wall carbon nanotubes by advanced nanomechanical manipulation techniques .
but although it is conceivable to place them as close as @xmath45 nm , the inter - tube forces will attract the tubes to each other and prevent the formation of a stable grating .
it thus seems that for foreseeable future practical nanogratings the momentum recoil will remain within the quantum uncertainty .
all diffraction gratings were milled with a focused ion beam ( raith ionline fib ) of @xmath46 kev ga@xmath47 ions @xcite into the freestanding membranes .
spots with homogeneous membranes were first identified using scanning electron microscopy ( sem , jeol @xmath48 , or raith @xmath49 ebeam writer , used as sem ) .
after the writing process all conducting gratings ( single and bilayer graphene and the nanoscroll grating ) were investigated with transmission electron microscopy ( tem , delong instruments lvem5 ) or scanning transmission electron microscopy ( stem , nion ultrastem 100 ) .
the stem was operated at @xmath50 kv with a medium angle annular dark field ( maadf ) detector .
this inspection is needed to ensure the presence of a clean binary diffraction grating , rather than a patterned heterostructure which may form in single - layer graphene at low ion doses @xcite .
the gratings fabricated from insulating membranes ( carbonaceous biphenyl membrane and silicon nitride membrane ) were investigated either with environmental scanning electron microscopy ( esem , fei quanta @xmath51 ) or electron energy loss spectrometry ( eels , fei tecnai f@xmath52 ) to determine the parameter of the gratings .
all gratings maintained their structure and diffraction properties over time .
virtually identical interferograms can be recorded after storing the gratings for several months in air . some ambiguity in the local thickness of all gratings remains since even though we can confirm the predicted structure and thickness . as for instance shown for single - layer graphene in fig .
[ fig : gratings]d , residual atomic and molecular contaminations on the membranes can not be excluded , as shown in the right lower corner of the same figure . in the following the beam parameters for milling of the ultra - thin gratings
are given .
the dose of the fib is given in pc / cm instead of pc/@xmath53 .
since we were writing single lines the dose is given along the line direction .
also the step size refers to this direction .
the width is defined by the thickness of the beam , which is about @xmath54 nm .
the carbonaceous biphenyl membrane @xcite is deposited on a copper tem mesh ( plano @xmath55 ) with an additional layer of lacey carbon for stabilization . to mill the grating we used a dose of @xmath56
nc / cm with a dwell time of @xmath57 ms , a current of @xmath58 pa and a step size of @xmath59 nm .
the overall size of the grating in the diffraction experiments is @xmath60 @xmath32m@xmath61 .
the grating has a periodicity of @xmath62 nm , and the grating bars are @xmath63 nm wide and 9@xmath64 nm long . from the mean width of the slits of @xmath65
nm we determine the opening fraction to be @xmath66% .
the error bar on the slit width , here and for all other gratings , is the @xmath67 standard deviation which was derived from about @xmath15 individual measurements at different locations of each membrane .
the diffraction gratings were written into single - layer graphene ( ted pella inc . , redding / ca @xmath68 pelco ) which was freely suspended over circular holes in a silicon nitride ( sin@xmath1 ) membrane .
each slit was manufactured by milling two adjacent lines separated by @xmath52 nm into the membrane . for each line
the gallium ion beam was set to a current of @xmath69 pa with a dose of @xmath70 nc / cm , a step size of @xmath0 nm and a dwell time of @xmath71 ms .
the grating region has an extension of @xmath72 holes of the support structure , resulting in a size of @xmath73 @xmath32m@xmath61 .
the grating has a periodicity of @xmath74 nm .
the bars have a mean width of @xmath75 nm and are separated by @xmath76 nm wide slits , resulting in an opening fraction of @xmath77% .
the length of the grating bars is @xmath78 nm . as several lines of grating are written on top of each other we have another grating perpendicular to the first one .
the periodicity of this second grating is @xmath79 nm . diffraction at this grating parallel to gravity can in principle lead to a stronger mixing of the velocities .
however , the periodicity of @xmath80 nm is too large to result in a visible effect at the detector . for this grating we used the same material as for the single layer graphene grating .
the dwell time for fib milling was @xmath81 ms , the current @xmath82 pa at a step size of @xmath59 nm and a dose of @xmath83 nc / cm .
the grating region used for the diffraction experiments extends over @xmath84 holes of the silicon nitride support structure which corresponds to an overall size of @xmath85 @xmath32m@xmath61 . from parts of the grating
which are not rolled up we determined the parameter of the original grating .
the bars are @xmath2 nm wide , have a length of @xmath86 @xmath32 m and are separated by @xmath87 nm wide slits .
the periodicity determined is @xmath88 nm and the opening fraction is @xmath89% . as expected , the curling does not alter the periodicity in the middle of the grating ( @xmath3 nm ) , but increases the opening fraction by almost a factor of @xmath70 to @xmath90% .
the minimum diameter observed in this region is around @xmath4 nm .
foldings in the membrane as well as contaminations ( see fig . [
fig : scrolls ] ) can disturb this process and lead to a mean width of the grating bars of @xmath91 nm .
the sin@xmath1 grating has an overall size of @xmath92 @xmath32m@xmath61 .
the grating bars are @xmath93 nm wide and @xmath94 nm long . from the periodicity of @xmath95
nm the mean width of the slits results to be @xmath96 nm , and the resulting opening fracture is @xmath97% . for this sample we used a bilayer graphene sample deposited on a lacey carbon film suspended across a copper tem mesh ( pelco@xmath98 @xmath59-layer graphene tem support films on lacey carbon , @xmath99 mesh copper grids ) . the parameters during the milling process were as follows : current @xmath100 pa , step size @xmath59 nm , dwell time @xmath101 ms and a dose of @xmath102 nc / cm .
the resulting grating has a size of @xmath103 @xmath32m@xmath61 and consists of grating bars which are @xmath104 nm long and @xmath66 nm wide . from the width of the slits ( @xmath105 nm ) and the periodicity of the grating of @xmath106 nm , the opening fraction was determined to be @xmath107% . for comparison we state the masses of the molecule and a single bar of the diffraction gratings in table [ tab
: masses ] . | matter - wave interferometry has become an essential tool in studies on the foundations of quantum physics @xcite and for precision measurements @xcite .
mechanical gratings have played an important role as coherent beamsplitters for atoms @xcite , molecules and clusters @xcite since the basic diffraction mechanism is the same for all particles . however , polarizable objects may experience van der waals shifts when they pass the grating walls @xcite and the undesired dephasing may prevent interferometry with massive objects @xcite . here
we explore how to minimize this perturbation by reducing the thickness of the diffraction mask to its ultimate physical limit , i.e. the thickness of a single atom .
we have fabricated diffraction masks in single - layer and bilayer graphene as well as in @xmath0 nm thin carbonaceous biphenyl membrane .
we identify conditions to transform an array of single layer graphene nanoribbons into a grating of carbon nanoscrolls .
we show that all these ultra - thin nanomasks can be used for high - contrast quantum diffraction of massive molecules .
they can be seen as a nanomechanical answer to the question debated by bohr and einstein @xcite whether a softly suspended double slit would destroy quantum interference . in agreement with bohr
s reasoning we show that quantum coherence prevails even in the limit of atomically thin gratings . |
in a supernova ( sn ) explosion , the shock wave must ultimately emerge from the body of the star , and begin to propagate down the exponential density gradient of the stellar atmosphere .
in newtonian fluid dynamics , the propagation of such a shock is described by a self similar solution ( raizer 1964 ; grover and hardy 1966 ; hayes 1968 ) . this self
similar solution is unavoidable ( raizer 1964 ) .
in fact , it turns out that , as the shock accelerates , a sonic point is formed , separating matter located immediately behind the shock from the flow s initial conditions ; this prevents pressure waves ( _ i.e. _ , causal information ) to reach post shock material from the the area where the flow s initial conditions are set .
thus all flows will converge to the same solution , irrespective of their initial conditions .
the lack of dependence upon initial conditions severely constrains the ensuing flow , by restricting the number of parameters upon which the shock evolution may depend ; in fact , this constraint is so strong that only a single ( self similar ) solution exists . in the newtonian solution
, it is found that the shock velocity increases exponentially , in such a way that the shock reaches spatial infinity in a finite time .
this is clearly impossible when account is taken of special relativity , and therefore the newtonian solution must break down at high shock velocities , as expected .
the question of precisely how the shock evolves in relativistic conditions has not been investigated so far .
this problem has acquired a new urgency within the hypernova model ( paczyski 1998 ; macfadyen & woosley 1999 ) for gamma ray bursts ( grbs ) .
in fact , meszaros and rees ( 2001 ) have shown that , despite the large assumed energy release of the central engine , the outwardly moving shock in the large star hypothesized to give rise to the grb can only reach a lorentz factor of @xmath0 at the end of the h envelope , and they had to invoke ( without explicit computations ) shock acceleration down the star exponential atmosphere to reach the required lorentz factors of @xmath1 .
the existence of a self - similar temporal structure in grbs is also suggested by an analysis of their power spectra ( beloborodov , stern , & svensson 2000 ; see also sivron 1998 ) .
it is the purpose of this paper to derive a fully ( special ) relativistic self similar solution for the problem of shock propagation in an exponential atmosphere ( section 2 ) , and to discuss its use in the hypernova model for grbs ( section 3 ) .
we shall assume that a cold ( @xmath2 ) material is stratified with density distribution @xmath3 ; the shock is supposed to move toward @xmath4 , thus with positive velocity @xmath5 .
the symmetry is assumed to be planar , since both the length scale of stellar atmospheres and the total extension of the atmosphere are much smaller than the stellar radius . in the newtonian analog ,
the shock speed is set by a purely dimensional argument ( e.g. chevalier 1990 ) : @xmath6 , with @xmath7 an adimensional constant to be determined . in the special relativistic problem ,
dimensional arguments alone fail because the presence of the light speed @xmath8 allows the construction of a new adimensional quantity ( @xmath9 ) , thusly spoiling the above argument .
we can however recover from this impasse by appealing to both dimensional and covariance arguments .
the time like part of the shock four speed @xmath10 is of course determined by the identity @xmath11 , while the space like part must be built in covariant fashion from the available quantities . defining a four vector @xmath12 , we can then write for the spatial part of @xmath13 @xmath14 where @xmath15 .
this is the only sensible solution to our dimensional / covariant problem .
we see from this that , as long as we use proper time , the structure of the problem is identical to that of the newtonian analog .
in particular , the shock reaches spatial infinity within a finite proper time .
we may choose @xmath16 for the moment when this occurs , so that the flow is restricted to @xmath17 .
incidentally , note that this implies @xmath18 .
physically , this makes perfect sense : as the shock accelerates , its proper time is contracted by its lorentz factor ( @xmath19 ) with respect to the fluid time .
the shock speed in terms of the fluid time can be obtained by remembering that @xmath20 , and substituing in the above equation one finds ( @xmath21 from now on ) @xmath22 taking the time derivative of both sides we obtain @xmath23 which can be immediately integrated to yield @xmath24 the low speed limit @xmath25 is @xmath26 , in agreement with the newtonian solution .
the hyperrelativistic limit ( @xmath27 ) is @xmath28 where we have introduced an initial shock lorentz factor @xmath29 at time @xmath30 , for ease of use in the future .
since the shock moves essentially at speed @xmath31 , we can rewrite the lhs of the above equation as @xmath32 , from which we see that the initial and the final ( i.e. , when the shock leaves the exponential atmosphere ) shock lorentz factors are related by @xmath33 the above clearly shows that all we have left to do is to determine the value of the parameter @xmath7 . to this purpose
, we must consider the fluid equations for the post shock material , which we have found it convenient to take in the form given by blandford and mckee ( 1976 , their eqs .
14 - 15 ) @xmath34 @xmath35 @xmath36 where @xmath37 is the convective derivative , and we have dropped curvature terms ( which are contained in the blandford and mckee equations ) , to keep with our planar symmetry approach . here
@xmath38 is the local energy density , @xmath39 is the local fluid lorentz factor , as seen from the reference frame of the unshocked material ( to be distinguished from @xmath19 , the shock lorentz factor in the same reference frame ) , and @xmath40 is the baryon number density always in the reference frame of the unshocked fluid .
the above equations assume an hyperrelativistic equation of state for the post shock material of the form @xmath41 , which is correct in the limit @xmath42 .
indeed , for the large shock lorentz factors appropriate to this problem ( @xmath43 ) heavens & drury ( 1988 ; see also iwamoto 1989 ) showed that the adiabatic index of the flow is within @xmath44 of its asymptotic relativistic value of @xmath45 , irrespective of the upstream temperature and plasma composition ( see especially their fig.2 , and eqs .
25 - 26 ) .
it should be remarked here that eq .
( [ important ] ) remains valid whatever we assume for the equation of state , but the precise value of @xmath7 depends instead on the particular choice of the equation of state .
the boundary conditions for this problem are provided by taub s jump conditions ( taub 1948 ) , which again we take in the hyperrelativistic limit given by blandford and mckee ( 1976 ) : @xmath46 here the subscripts @xmath47 refer to pre and post shock quantities , respectively ; @xmath40 and @xmath39 are always defined in the unshocked fluid frame , and @xmath38 in the comoving frame .
the enthalpy of the unshocked material , @xmath48 , equals @xmath49 since this material is assumed to be cold . in our problem , @xmath49 is not a constant , because the atmosphere is stratified .
we have @xmath50 so that , ultimately , @xmath51 which is the form we shall use in the following .
an identical argument shows that @xmath52 where @xmath53 . in order to search for a self similar solution
, we need to assume a form for the similarity variable . in the newtonian analog , this is clearly @xmath54 , where @xmath55 is the instantaneous shock location . in this problem
, we take @xmath56 the rationale for this is that , since post shock material has a lorentz factor which is @xmath57 times smaller than the shock s , the post shock material falls behind the shock by an amount @xmath58 as the shock covers an exponential length .
please notice that , with our notation , @xmath59 for the shocked material .
the post shock quantities @xmath60 , @xmath38 and @xmath40 can then be taken of the form @xmath61 with boundary conditions given by eqs .
( [ taub ] ) ( the second one ) , ( [ taube ] ) , ( [ taubn ] ) as @xmath62 substitution of eqs . ( [ scalingvariable ] ) , ( [ scaledquantities ] ) into eqs .
( [ bmck1 ] ) , ( [ bmck2 ] ) and ( [ bmck3 ] ) yields , after some algebra ( the dot indicates derivation with respect to @xmath63 ) @xmath64 @xmath65 @xmath66 from these , we see what fixes @xmath7 : the denominator of the rhs of eq .
( [ grand1 ] ) goes to zero at a critical point @xmath67 and , unless the numerator simultaneously does the same ( which will only occur for a special value of @xmath7 ) , a non
integrable singularity will ensue .
the same phenomenon occurs in the newtonian analog , where it has been shown that this critical point is a sonic point .
it is exactly because there is a sonic point that a self similar solution can develop : in fact , material between the shock and the sonic point is not in causal contact with post sonic point material , and thus its properties can not be determined by the problem s initial conditions . to keep the problem non singular ,
we impose that the numerator and denominator of eq .
( [ grand1 ] ) vanish simultaneously , obtaining @xmath68 from the numerator , and @xmath69 for the denominator .
the conditions above lead to two solutions for @xmath7 , one positive and another negative .
the positive solution must be discarded because otherwise the shock would not reach spatial infinity for @xmath70 ( eq . [ propertime ] ) .
we are therefore left with only one sensible solution , @xmath71 the location of the sonic point , @xmath67 , where both the numerator and the denominator vanish , is determined through numerical integration of eqs .
( [ grand1 ] ) , ( [ grand2 ] ) and ( [ grand3 ] ) .
this yields @xmath72 .
also note that @xmath73 , and therefore the denominator of eq .
( [ grand3 ] ) is always well - behaved . the full numerical solution for @xmath74 , @xmath75 and @xmath76
is illustrated in fig .
it can be seen from eq .
( [ important ] ) that the necessary asymptotic lorentz factor required for a proper modelling of the properties of grbs , i.e. @xmath77 ( remember that the lorentz factor of the matter is only @xmath78 that of the shock ) can be reached by crossing a modest factor of @xmath79 density decrease in the exponential atmosphere ( assuming @xmath80 ) .
yet , the total energy of material moving at these large lorentz factors is only modest .
in fact , let us compute the distribution of kinetic energy with baryon number .
consider a cylindrical fluid element with surface area @xmath81 ( with the normal along the direction of fluid motion @xmath82 ) , and length @xmath83 along the direction @xmath82 .
this element will have a bulk kinetic energy of @xmath84 .
what we want to know is the energy distribution at the moment in which the shock gets out of the exponential atmosphere .
let us indicate with @xmath85 the shock lorentz factor at that moment .
then @xmath86 which can be coupled to the solution @xmath87 to yield a parametric representation of @xmath88 , the distribution of kinetic energy with respect to the final lorentz factor . the adimensional part of the function in the previous equation is plotted in fig .
the numerical factor can be estimated using eq .
( [ important ] ) : @xmath89 in this expression , the values for @xmath90 ( the edge of the h
shell ) , @xmath91 ( the matter density at the end of the h shell , and thus presumably at the beginning of the exponential atmosphere ) , and @xmath29 ( the shock lorentz factor at the end of the h shell @xmath92 beginning of the exponential atmosphere ) are taken from meszaros and rees ( 2001 ) .
taking the adimensional factor from fig.2 , we see that the kinetic energy falls short of the average isotropic burst energy ( schmidt 1999 ) by about three orders of magnitude ( note that here what we are computing is the isotropic energy ) . before showing how to come out of this impasse ,
we need to consider which part of fig .
2 can actually be obtained in a realistic model .
the reason for this limitation comes from the fact that the idealized model presented here has been propagating down an exponential atmosphere forever , while in a realistic model only a finite amount of matter can have converged onto this self similar solution , given the finite dimension of the star . for this reason ,
we have plotted in fig .
2 three ticks , which correspond to the present position of matter which was located , before being reached by the shock , at @xmath93 exponential scale
lengths from the shock s present location .
if we think that the initial atmosphere extends for @xmath93 exponential scale lengths , then we can believe the part of fig . 2 located to the right of their respective tickmarks .
leftward of them , the true , physical solution will depart from the one shown here , and the total kinetic energy at the corresponding values of lorentz factor will be much smaller than the values plotted ( obviously , since the physical solution has finite mass and energy , which is not true for the idealized one ) .
let us now suppose that the putative burst progenitor possesses a wide exponential atmosphere , spanning some @xmath94 orders of magnitude in density .
then , for an initial lorentz factor @xmath95 , the shock lorentz factor at the end of the atmosphere is @xmath96 ( eq . [ important ] ) ; once again , the overall energy of matter moving at this speed would be modest ( @xmath97 erg ) , but fig .
2 shows that most of the energy ( before the cutoff implied by the finite extent of the atmosphere ) will come out at a lorentz factor @xmath98 of the shock s factor , _
i.e. _ at @xmath99 . with the values
above , the total energy then amounts to @xmath100 erg , in reasonable agreement with schmidt s ( 1999 ) isotropic estimates . it should be noticed that inspection of the numerical solution for large values of the distance parameter @xmath63 shows the material to be cold ( @xmath101 ) , so that there will be negligible further acceleration of these slower shells by @xmath102 work , and @xmath103 remains a good estimate of the coasting lorentz factor . to summarize
, the application of our solution to shock acceleration in the atmosphere of a massive star has shown that , in atmospheres with small density range , the amount of energy carried by material accelerated to the typical grb lorentz factors falls short of the grb required energetics . however , for stars with a wide density range in their atmospheres ( @xmath104 orders of magnitude ) , a sufficient quantity of energy is carried by later shells of material moving at the typical grb lorentz factors . in this model , the early emission would be dominated by an ultra hard component , due to the very large@xmath19 shells slowing down in the ism .
this is at least qualitatively consistent with virtually all bursts studied in some detail by bepposax ( frontera _ et al .
_ , 2000 , see especially their fig .
2 ) , which exhibit , in their first few seconds only , spectra peaking beyond the instrument s ( grbm ) observing limit of @xmath105 kev .
beloborodov , a. m. , stern , b. e. , & svensson 2000 , apj , 535 , 58 blandford , r.d . ,
mckee , c.f .
, 1976 , phys . of fluids , 19 , 1130 chevalier , r.a .
, 1990 , apj , 359 , 463 frontera , f. , et al .
, 2000 , apjs , 127 , 59 grover , r. , hardy , j.w . , 1966 , apj , 143 , 48 hayes , w.d . , 1968 , j. fluid .
mech . , 32 , 305 heavens , a. f. & drury , l. oc .
1988 , mnras , 235 , 997 iwamoto , n. 1989 , phys .
a , 39 , 4076 kallaman , t. r. , meszaros , p. , & rees , m. j. 2001 , astro - ph/0110654 macfadyen , a. , & woosley , s. e. 1999 , apj , 524 , 262 meszaros , p. , rees , m.j . , 2001 ,
apjl , 556 , 37 paczyski , b. 1998 , apj , 494 , l45 raizer , yu.p .
, 1964 , zh.prikl.mat.tekh.fiz.,4,49 .
sivron , r. 1998 , apj , 503 , l57 schmidt , m. , 1999 , apjl , 523 , l117 taub , h. 1948 , phys . | we derive a fully relativistic , self similar solution to describe the propagation of a shock along an exponentially decreasing atmosphere , in the limit of very large lorentz factor .
we solve the problem in planar symmetry and compute the acceleration of the shock in terms of the density gradient crossed during its evolution .
we apply our solution to the acceleration of shocks within the atmosphere of a hypernova , and show that velocities consistent with the requirements of grb models can be achieved with exponential atmospheres spanning a wide density range . |
the prospect of observing parity violation from the strong interaction in relativistic heavy
ion collisions has recently gained great attention @xcite . in a highly excited dense state ,
quantum fluctuations of the gluonic field can produce configurations with different local parities .
the interaction of these local bubbles of chiral symmetry restoration and non - vanishing topological charges , with the strong and short lived magnetic field produced in non central heavy
ion collisions @xcite gives rise to the _ chiral magnetic effect_. the effect is argued to be reflected in the asymmetry in the emission of particles relative to the reaction plane : opposite charged particles will be emitted preferentially in different directions across the reaction plane . in this article , we report the results for the two and three particle charge dependent azimuthal correlations with respect to the reaction plane for pb pb collisions at @xmath0 tev recorded in 2010 with alice @xcite at the lhc .
the tools used for this study are the 2 and 3particle azimuthal correlators represented by @xmath1 and @xmath2 @xcite , respectively .
the dependence of the 3particle correlator on the : ( a ) @xmath3 difference of the pair ; ( b ) average @xmath3 of the pair ; ( c ) @xmath4 of the pair .
the legend indicates the connection between the charge combinations and the different marker styles used.,width=604,height=268 ]
data from around 50 m pb pb events at @xmath0 tev were analyzed .
only events with a proper online trigger were considered in the analysis .
an offline event selection was also applied to reduce the contamination from background events .
events having a reconstructed vertex were used in this study .
the centrality of the collision was estimated by using the distribution of the signal from the vzero scintillator detectors , placed around the beam pipe on either side of the interaction region and covering the pseudorapidity ranges@xmath5 and @xmath6 @xcite , and fitting it with a glauber model .
the tracks were reconstructed by the main tracking device of alice , the tpc @xcite , that provides a uniform acceptance with minimal corrections .
adequate selection criteria were applied to minimize the contribution from background tracks .
finally , the phase space analyzed was given by @xmath7 and @xmath8 gev / c .
figure [ fig : aliceresults ] shows the dependence of the 3particle correlator on the @xmath3 difference , the mean @xmath3 and the pseudo rapidity difference of the pair , in the left , middle and right plots , respectively .
the black circles correspond to pairs having the same charge , while the red squares representing the correlations between opposite charges . in all plots , the 3040@xmath9 centrality percentile is shown .
the opposite charged pairs do nt show any dependence on either of the three variables used and their values are consistent with 0 . on the other hand ,
the same charged pairs demonstrate a different pattern : the data points do nt indicate a significant contribution from short range correlations ( left plot ) , the magnitude of the correlations seems to increase with increasing mean @xmath3 of the pair ( middle plot ) , whereas the @xmath4 dependence shows a width of one unit in @xmath4 .
the comparison of the integrated 2 ( left plot ) and 3particle correlator ( right plot ) between alice and star . the legend indicates the connection between the charge combinations and the different color and marker styles used.,title="fig:",width=302,height=268 ] the comparison of the integrated 2 ( left plot ) and 3particle correlator ( right plot ) between alice and star .
the legend indicates the connection between the charge combinations and the different color and marker styles used.,title="fig:",width=302,height=268 ] figure [ fig : alicevsstar ] shows the centrality dependence of the integrated 2 ( left plot ) and 3particle ( right plot ) correlators measured by alice in pb
pb collisions at @xmath0 tev ( red circles ) compared to the values reported by star in au
au collisions at @xmath10 tev ( green stars ) @xcite . in both plots ,
the error bars correspond to the statistical uncertainties whereas the systematic ones are represented by the shaded areas .
the full markers correspond to the correlations between oppositely charged pairs , with the open ones showing the same charged pairs .
the 2particle correlation analysis indicates that there is a change of sign in the correlations of same charged pairs between the lhc and rhic energies , indicating in turn a change of the correlation pattern in and out of plane in the two experiments .
this manifests itself by the fact that the results for the correlations of same charged pairs at rhic energies have the same magnitude when measured with the 2 and 3particle correlation technique : at rhic energies particles exhibit a stronger magnitude of their correlations in the direction along than across the reaction plane .
the results obtained from the 3particle correlation analysis show a clear charge separation when moving to more peripheral collisions .
there is a remarkable agreement between the measurements at two different energies . combining both correlation techniques for alice ,
we conclude that the relevant correlations have stronger magnitude out of than in plane .
( a ) comparison of the centrality dependence of the integrated 3particle correlators @xmath11 between data and hijing @xcite .
( b ) comparison of the centrality dependence of the measured 3particle correlator with theory predictions.,width=529,height=268 ] figure [ fig : alicecomparison]left presents the comparison of the centrality dependence of the integrated correlator @xmath2 @xcite , between the experimental points and hijing @xcite .
the blue squares represent the analysis of hijing events without any after burner . in addition , an after burner was also used to modify the relevant distributions of generated particles with the realistic values of the differential elliptic flow for each centrality bin @xcite .
the corresponding points are represented by the purple triangles .
the correlations between opposite charged pairs are described well by the hijing points , the latter exhibiting no major differences related to the usage of the after burner .
figure [ fig : alicecomparison]right shows the comparison of the experimental data ( red points ) with predictions from theory @xcite for lhc energies .
the blue line shows the prediction for the centrality dependence of the correlations between same charged pairs , assuming the existence of the _ chiral magnetic effect _
, with a certain value for the starting time of the evolution of the magnetic field .
this particular model clearly underestimates the experimental result .
the experimental observation of no apparent change in the magnitude of the effect between rhic and lhc contradicts the quantitative predictions . on the other hand , a hint that the signal might be similar at different energies
was already given in @xcite .
we presented the first measurement of the 2 and 3particle correlators at lhc energies with the alice experiment .
the results indicate that there is a change in the correlation pattern ( i.e. larger out
of than in plane correlations ) between lhc and rhic energies .
the results of the integrated 3particle correlator measured by alice show a remarkable agreement with the published data from star . | separation of charges along the extreme magnetic field created in non - central relativistic heavy
ion collisions is predicted to be a signature of local parity violation in strong interactions .
we report on results for charge dependent two particle azimuthal correlations with respect to the reaction plane for pb
pb collisions at @xmath0 tev recorded in 2010 with alice at the lhc .
the results are compared with measurements at rhic energies and against currently available model predictions for lhc .
systematic studies of possible background effects including comparison with conventional ( parity - even ) correlations simulated with monte carlo event generators of heavy
ion collisions are also presented . |
scattering resonances play a central role in cold - atom physics @xcite .
in particular , @xmath0-wave feshbach resonances induced by a magnetic field have been utilized to control the atom - atom interaction and have led to experimental realization of a rich variety of physics , such as the bcs - bec crossover in fermi gases @xcite .
although a @xmath1-wave analog of the bcs - bec crossover has been predicted theoretically @xcite and @xmath1-wave and higher partial - wave resonances have been observed experimentally @xcite , @xmath1-wave superfluids have not been realized in cold - atom experiments so far .
this is because fermi gases in the vicinity of the @xmath1-wave feshbach resonances are unstable owing to inelastic collisions decaying into deeply bound dimers and do not reach their equilibrium within their short lifetime @xcite .
in contrast , fermi gases in the vicinity of the @xmath0-wave feshbach resonances are long lived because such inelastic collisions are strongly suppressed by the pauli exclusion principle @xcite . in this paper
, we propose a way to induce @xmath1-wave and higher partial - wave resonances from the @xmath0-wave feshbach resonance only .
this can be achieved by using a mixture of two atomic species @xmath2 and @xmath3 in mixed dimensions , where @xmath2 atoms are confined in lower dimensions [ quasi - zero dimension ( 0d ) , one dimension ( 1d ) , or two dimensions ( 2d ) ] and interact with @xmath3 atoms living in three dimensions ( 3d ) @xcite ( see fig .
[ fig : molecule ] ) . indeed , the authors of ref .
@xcite mention the possibility of @xmath1-wave resonances in the 0d-3d mixture .
higher partial - wave resonances induced from a purely @xmath0-wave interaction seem counterintuitive but can be understood by generalizing the argument for @xmath0-wave resonances given in ref .
@xcite .
suppose an @xmath0-wave interspecies interaction between @xmath2 and @xmath3 atoms supports a bound state with its binding energy @xmath4 .
we confine only the @xmath2 atom by a 3d harmonic potential with the oscillator frequency @xmath5 . because the @xmath6 molecule also feels the confinement potential , its center - of - mass motion is quantized . in the limit @xmath7 or @xmath8 ,
the energy of the @xmath6 molecule @xmath9 is given by a sum of its center - of - mass energy and binding energy : @xmath10 where @xmath11 is the mass of the @xmath12 atom , @xmath13 is an orbital angular momentum quantum number , and @xmath14 is an integer .
we can view these quantized energy levels as a tower of @xmath15th partial - wave `` feshbach molecules . '' when one of them coincides with the @xmath2-@xmath3 scattering threshold at @xmath16 , the coupling between the center - of - mass and relative motions leads to the resonance occurring in the @xmath15th partial - wave channel .
furthermore , there exists a series of resonances in each partial - wave channel corresponding to different values of @xmath17 .
-wave molecules in mixed dimensions . shown are the asymptotic angular wave functions of @xmath3 atoms in 3d relative to @xmath2 atoms confined in 0d ( origin ) , 1d ( @xmath18 axis ) , and 2d ( @xmath19 plane ) , respectively .
see the text for details .
[ fig : molecule ] ] similarly , in the 1d-3d mixture , the @xmath2 atom is confined by a 2d harmonic potential and the energy of the @xmath6 molecule is quantized as @xmath20 where @xmath21 is a magnetic quantum number .
when one of them coincides with the @xmath2-@xmath3 scattering threshold at @xmath22 , the resonance occurs in the @xmath23th partial - wave channel .
there exists a series of resonances in each partial - wave channel corresponding to different values of @xmath17 . in the 2d-3d mixture ,
the @xmath2 atom is confined by a 1d harmonic potential and the energy of the @xmath6 molecule is quantized as @xmath24 when one of them coincides with the @xmath2-@xmath3 scattering threshold at @xmath25 , the resonance occurs . because the wave function of the @xmath6 molecule has the even ( odd ) parity for an even ( odd ) @xmath17 , the associated resonance is in the even ( odd)-parity channel , which is also phrased as an @xmath0-wave ( @xmath1-wave ) resonance in the 2d-3d mixture .
there exists a series of resonances in each parity channel corresponding to different values of @xmath17 . in order to achieve these higher partial - wave resonances in mixed dimensions
, one needs to apply a strong optical lattice only on @xmath2 atoms without affecting @xmath3 atoms @xcite .
such a species - selective optical lattice has been successfully implemented in a bose - bose mixture of @xmath26k and @xmath27rb @xcite and the 2d-3d mixed dimensions have been realized in the florence group experiment @xcite .
the above simple arguments illustrate how the higher partial - wave resonances are induced in mixed dimensions from the purely @xmath0-wave interaction in a free space . in this paper , we develop two - body scattering theories in mixed dimensions , assuming the harmonic confinement potential and the short - range interspecies interaction characterized by an @xmath0-wave scattering length ( @xmath28 ) and effective range ( @xmath29 ) , which are extensions of analyses in refs .
@xcite aimed for @xmath0-wave scatterings only .
the two - body scattering of a 0d , 1d , or 2d atom with a 3d atom is three - dimensional in the sense that it is described by three relative coordinates , and thus such scattering theories can be developed in parallel with the ordinary scattering theory in 3d . in particular , we determine the positions of all partial - wave resonances and will see that they are well understood from the above simple arguments in a wide range of the mass ratio @xmath30 .
we also compute the low - energy scattering parameters in the @xmath1-wave channel ( scattering volume and effective momentum ) that are necessary for the low - energy effective theory of the @xmath1-wave resonance in mixed dimensions .
our analyses and results for the 0d-3d , 1d-3d , and 2d-3d mixtures are presented in secs .
[ sec:0d-3d ] , [ sec:1d-3d ] , and [ sec:2d-3d ] , respectively , while some details are shown in appendixes [ app:0d-3d ] , [ app:1d-3d ] , and [ app:2d-3d ] , respectively . also in sec .
[ sec:2d-3d ] , we point out that some of the resonances observed in the florence group experiment @xcite can be interpreted as the @xmath1-wave resonances in the 2d-3d mixed dimensions .
finally , sec .
[ sec : summary ] is devoted to a summary of this paper and the stability of confinement - induced molecules and their analogy with kaluza - klein modes in extra - dimension models are discussed . for later use , we define the reduced mass @xmath31 , the total mass @xmath32 , and the harmonic oscillator length @xmath33 . below we set @xmath34 and the range of integrations
is assumed to be from @xmath35 to @xmath36 unless otherwise specified .
the scattering of a quasi-0d @xmath2 atom with a @xmath3 atom in 3d is described by a schrdinger equation , @xmath37 for @xmath38 .
the short - range interspecies interaction is implemented by the generalized bethe - peierls boundary condition @xcite : @xmath39\chi({{\bm{r}}}).\ ] ] here @xmath40 is the energy - dependent `` scattering length '' defined by @xmath41 and the collision energy operator @xmath42 in the present case is given by @xmath43 the solution to the schrdinger equation ( [ eq : schrodinger_0d ] ) can be written as @xmath44 where @xmath45 is a solution in the noninteracting limit and @xmath46 is the retarded green s function for the noninteracting hamiltonian : @xmath47 here @xmath48 with quantum numbers @xmath49 is the normalized wave function of an @xmath2 atom in the 3d harmonic potential .
we now consider the low - energy scattering in which @xmath50 is satisfied , and then @xmath45 becomes @xmath51 which represents the @xmath2 atom in the ground state of the 3d harmonic potential and the plane wave of @xmath3 atom with the wave vector @xmath52 .
the asymptotic form of the wave function at a large distance @xmath53 is given by @xmath54\phi_{{\bm{0}}}(r_a),\ ] ] where @xmath55 with @xmath56 defines the two - body scattering amplitude in the 0d-3d mixed dimensions : @xmath57 we note that @xmath58 has an implicit @xmath52 dependence .
the unknown function @xmath58 can be determined by substituting the solution ( [ eq : solution_0d ] ) into the bethe - peierls boundary condition ( [ eq : short - range_0d ] ) . defining the regular part of the green s function @xmath59 by @xmath60 we obtain @xmath61 this integral equation determines @xmath62 , which in turn provides @xmath63 from eq .
( [ eq : f_0d ] ) . because of the 3d rotational symmetries of the system , @xmath63 , @xmath58 , and @xmath59 can be decomposed into their partial - wave components : @xmath64 equations ( [ eq : f_0d ] ) and ( [ eq : chi_0d ] ) lead to the @xmath15th partial - wave scattering amplitude given by @xmath65 with @xmath66 from the general argument based on the 3d rotational symmetries and unitarity @xcite , or from the explicit calculation that uses the green s function in eq .
( [ eq : g_0d ] ) , we can show that @xmath67 has the usual low - energy expansion : @xmath68 where @xmath69 and @xmath70 are effective scattering `` length '' and `` range '' parameters in the @xmath15th partial - wave channel .
note that @xmath69 has the dimension of @xmath71 and @xmath70 has @xmath72 . substituting the expansion of @xmath67 into eq .
( [ eq : partial - wave_0d ] ) , we can determine the low - energy scattering parameters .
in particular , the effective scattering length @xmath69 is given by @xmath73 with @xmath74 the @xmath15th partial - wave resonance in the 0d-3d mixed dimensions is defined by the divergence of @xmath75 , which occurs when @xmath76 is satisfied .
+ we now solve the integral equation ( [ eq : resonance_0d ] ) numerically to determine the positions of @xmath15th partial - wave resonances .
for the purpose of illustrating qualitative results , we shall set @xmath77 .
for quantitative predictions in a specific atomic mixture , it is necessary but straightforward to include the effective range correction @xcite .
some details of our method to solve the integral equation are shown in appendix [ app:0d-3d ] .
figure [ fig : resonance_0d ] shows the positions of the lowest five resonances in terms of @xmath78 for @xmath79 partial - wave channels as functions of the mass ratio @xmath80 . for completeness , we have included the result for the @xmath0-wave ( @xmath81 ) resonance , which has been reported in ref .
@xcite . as we have discussed in sec .
[ sec : introduction ] , there exists a series of resonances in each partial - wave channel induced from the purely @xmath0-wave interaction in a free space .
indeed , the resonance positions are well described by the approximate formula @xmath82 by using eq .
( [ eq : approx_0d ] ) with @xmath83 in a wide range of the mass ratio @xmath30 .
for such a mass ratio , as is evident from eq .
( [ eq : approx_0d ] ) , @xmath84-wave ( @xmath63-wave ) resonances are nearly degenerate with @xmath0-wave ( @xmath1-wave ) resonances , and thus it could be difficult to distinguish them practically .
we also note that the @xmath15th partial - wave resonance is ( @xmath85)-fold degenerate in a spherically symmetric potential .
this degeneracy is lifted when the confinement potential is deformed , and therefore the one resonance at a given @xmath78 splits into more resonances . on the @xmath86 side of every resonance , a shallow @xmath6 molecule is formed . in the vicinity of the resonance @xmath87 , its binding energy @xmath88
is determined by the pole of the scattering amplitude [ eq .
( [ eq : expansion_0d ] ) ] with keeping the two dominant terms at @xmath89 : @xmath90 away from the resonance , these universal formulas are no longer valid .
the binding energy @xmath91 has to be determined by solving the integral equation @xmath92 where @xmath93 is a component of the spherical harmonics expansion of @xmath58 : @xmath94 we note that the solution is independent of @xmath95 , and therefore there are @xmath85 degenerate molecules for a spherically symmetric potential .
this degeneracy is lifted when the confinement potential is deformed .
we now derive the asymptotic form of the molecular wave function in the vicinity of the @xmath15th partial - wave resonance where @xmath96 is satisfied . from eqs .
( [ eq : solution_0d ] ) and ( [ eq : g_0d ] ) with the replacement @xmath97 , we find @xmath98 the angular parts of asymptotic wave functions of three degenerate shallow @xmath1-wave molecules @xmath99 are illustrated in fig .
[ fig : molecule ] .
the effective scattering length in the @xmath0-wave channel @xmath100 has been computed in ref . @xcite . here
we focus on the @xmath1-wave ( @xmath101 ) channel and determine its two low - energy scattering parameters , namely , the effective scattering volume @xmath102 and the effective momentum @xmath103 . the effective scattering volume can be computed by eliminating @xmath104 from eqs .
( [ eq : a_eff_0d-1 ] ) and ( [ eq : a_eff_0d-2 ] ) and solving the resulting integral equation numerically ( see appendix [ app:0d-3d ] for details ) . in fig .
[ fig : parameter_0d ] , @xmath105 for @xmath77 is plotted as a function of @xmath78 for three mass ratios @xmath106 , @xmath107 , and @xmath108 .
we confirm the existence of a series of @xmath1-wave resonances ( @xmath109 ) induced from the purely @xmath0-wave interaction in a free space , while they become narrower for larger @xmath78 .
we also find that the resonance is wider when a lighter atom is confined in lower dimensions .
similarly , the effective momentum can be computed from eq .
( [ eq : partial - wave_0d ] ) , and @xmath110 for @xmath77 is plotted in fig .
[ fig : parameter_0d ] as a function of @xmath78 for the same three mass ratios . in the vicinity of the @xmath1-wave resonance @xmath111 , @xmath112 and
@xmath113 determine the binding energy of three degenerate shallow @xmath1-wave molecules via the universal formula ( [ eq : binding_0d ] ) .
both @xmath112 and @xmath113 are important to the low - energy effective theory of the @xmath1-wave resonance discussed below .
the low - energy effective theory of the @xmath1-wave resonance in the 0d-3d mixed dimensions is provided by the action @xmath114 , \end{split}\ ] ] where the summation over @xmath115 is implicitly understood . @xmath116 and @xmath117 fields represent the @xmath2 and @xmath3 atoms in 0d and 3d , respectively .
the interaction between @xmath2 and @xmath3 atoms is described through their coupling with three @xmath1-wave molecular fields @xmath118 .
@xmath119 is their coupling strength and @xmath120 is the detuning from the resonance .
these cutoff ( @xmath121)-dependent bare parameters can be related to the effective scattering volume @xmath112 and effective momentum @xmath113 by matching the two - body scattering amplitude from the action ( [ eq : action_0d ] ) with that shown in eqs .
( [ eq : decomposition_0d ] ) and ( [ eq : expansion_0d ] ) .
the standard diagrammatic calculation leads to the following scattering amplitude with collision energy @xmath122 : @xmath123 by defining @xmath124 and @xmath125 we reproduce the scattering amplitude ( [ eq : expansion_0d ] ) in the @xmath1-wave ( @xmath101 ) channel up to a kinematical factor : @xmath126 this low - energy effective theory can be generalized easily to the case with more than one lattice site where @xmath3 atoms are confined and could be used to investigate the many - body physics across the @xmath1-wave resonance as in the 3d case @xcite .
we note that an effective field theory of the @xmath1-wave resonance in 3d has been developed in connection with the @xmath127-@xmath17 scattering @xcite . when @xmath128 and @xmath129 , the confinement - induced resonances can be understood in a different way , as is discussed in refs .
@xcite . to the leading order in the weak - coupling expansion @xmath130 , an @xmath2 atom occupies the ground state in a 3d harmonic potential and creates a mean - field attractive potential felt by a @xmath3 atom .
therefore , the scattering of the @xmath3 atom by the confined @xmath2 atom is described by @xmath131\psi({{\bm{r}}}_b ) = \frac{k^2}{2m_b}\psi({{\bm{r}}}_b),\ ] ] where @xmath132 .
this schrdinger equation , which is valid in the weak - coupling limit , is equivalent to the integral equation ( [ eq : chi_0d ] ) , where only the @xmath133 term is kept in the green s function ( [ eq : g_0d ] ) and @xmath134 is identified as @xmath135 . by matching the solution of eq .
( [ eq : weak - coupling_0d ] ) with the asymptotic form ( [ eq : asymptotic_0d ] ) , we can determine the scattering amplitude and low - energy scattering parameters in the weak - coupling limit @xmath129 .
in particular , the resonance occurs when a new bound state appears .
this is possible even in the weak - coupling limit because the attractive potential becomes strong compared to the kinetic term by decreasing the mass ratio down to @xmath136 .
we find that the resonances are achieved at the critical values of @xmath137 in the @xmath81 channel @xcite , @xmath138 in the @xmath101 channel , @xmath139 in the @xmath140 channel , and @xmath141 in the @xmath142 channel . here
we present some details of our method to solve integral equations derived in sec .
[ sec:0d-3d ] .
the central issue is the evaluation of the regular part of the green s function defined in eq .
( [ eq : regular_0d ] ) . when @xmath298 , it is useful to represent @xmath59 as @xmath299 \\ & \quad + \int_0^\infty\!d\tau \left(\frac{m_{ab}}{2\pi\tau}\right)^{3/2}\delta({{\bm{r}}}-{{\bm{r } } } ' ) .
\end{split}\ ] ] its partial - wave projection gives @xmath300 \\
& \quad \times \exp\!\left[-\left(\frac{m_a\omega\cosh\omega\tau}{\sinh\omega\tau } + \frac{m_b}\tau\right)\frac{r^2+r'^2}2\right ] + \int_0^\infty\!d\tau\left(\frac{m_{ab}}{2\pi\tau}\right)^{3/2 } \frac{\delta(r - r')}{4\pi r^2}. \end{split}\ ] ] we now evaluate the matrix elements of @xmath301 with respect to the eigenfunctions of 3d harmonic oscillator with orbital angular momentum @xmath15 : @xmath302 which form an orthonormal basis : @xmath303 a lengthy but straightforward calculation leads to @xmath304 + \delta_{ij}\sqrt{\frac{m_{ab}}{2\pi m_a } } \int_0^\infty\!\frac{\omega d\tau}{(\omega\tau)^{3/2 } } , \end{split}\ ] ] where @xmath305 is the hypergeometric function and @xmath306 expanding @xmath307 in terms of @xmath308 , the integral equations in sec .
[ sec:0d-3d ] reduce to linear algebraic equations that have been solved numerically .
the scattering of a quasi-1d @xmath2 atom with a @xmath3 atom in 3d is described by a schrdinger equation , @xmath143 for @xmath144 , where @xmath145 , @xmath146 , and the center - of - mass motion in the @xmath18 direction is eliminated . the short - range interspecies interaction is implemented by the generalized bethe - peierls boundary condition @xcite : @xmath147\chi(\bm\rho ) .
\end{split}\ ] ] the collision energy operator @xmath42 in eq .
( [ eq : generalized_a ] ) in the present case is given by @xmath148 the solution to the schrdinger equation ( [ eq : schrodinger_1d ] ) can be written as @xmath149 where @xmath45 is a solution in the noninteracting limit and @xmath46 is the retarded green s function for the noninteracting hamiltonian : @xmath150 here @xmath48 with quantum numbers @xmath151 is the normalized wave function of an @xmath2 atom in the 2d harmonic potential , and @xmath152 are coordinates of the @xmath3 atom relative to the confined @xmath2 atom .
the anisotropic factor is such because separations in the @xmath153 and @xmath154 directions are associated with @xmath155 while a separation in the @xmath18 direction is associated with @xmath156 [ see the last two terms in eq .
( [ eq : schrodinger_1d ] ) ] .
we now consider the low - energy scattering in which @xmath157 is satisfied , and then @xmath45 becomes @xmath158 which represents the @xmath2 atom in the ground state of the 2d harmonic potential and the plane wave of @xmath3 atom with the wave vector @xmath52 .
the asymptotic form of the wave function at a large distance @xmath159 is given by @xmath160\phi_{{\bm{0}}}(\rho_a),\ ] ] where @xmath161 with @xmath162 defines the two - body scattering amplitude in the 1d-3d mixed dimensions : @xmath163 we note that @xmath58 has an implicit @xmath164 dependence and both @xmath58 and @xmath63 depend on @xmath165 through the green s function ( [ eq : g_1d ] ) .
the unknown function @xmath58 can be determined by substituting the solution ( [ eq : solution_1d ] ) into the bethe - peierls boundary condition ( [ eq : short - range_1d ] ) .
defining the regular part of the green s function @xmath59 by @xmath166 we obtain @xmath167 this integral equation determines @xmath62 , which in turn provides @xmath63 from eq .
( [ eq : f_1d ] ) .
because the system has a rotational symmetry in the @xmath19 plane , @xmath63 , @xmath58 , and @xmath59 can be decomposed into their partial - wave components : @xmath168 because @xmath95 and @xmath169 are degenerate , we assume @xmath170 in this section without loss of generality
. equations ( [ eq : f_1d ] ) and ( [ eq : chi_1d ] ) lead to the @xmath95th partial - wave scattering amplitude given by @xmath171 with @xmath172 from the explicit calculation that uses the green s function in eq .
( [ eq : g_1d ] ) , we can show that @xmath173 has the following low - energy expansion : @xmath174 } \\ &
\quad + o(k^{m+2}_\perp k'^m_\perp , k^m_\perp k'^{m+2}_\perp ) , \end{split}\ ] ] where @xmath175 and @xmath176 are effective scattering `` length '' and `` range '' parameters in the @xmath95th partial - wave channel .
note that @xmath175 has the dimension of @xmath177 and @xmath176 has @xmath178 .
this unusual form of the low - energy expansion is owing to the lack of full 3d rotational symmetries . + substituting the expansion of @xmath173 into eq .
( [ eq : partial - wave_1d ] ) , we can determine the low - energy scattering parameters .
in particular , the effective scattering length @xmath175 is given by @xmath179 with @xmath180 the @xmath95th partial - wave resonance in the 1d-3d mixed dimensions is defined by the divergence of @xmath181 , which occurs when @xmath182 is satisfied .
we now solve the integral equation ( [ eq : resonance_1d ] ) numerically to determine the positions of @xmath23th partial - wave resonances .
for the purpose of illustrating qualitative results , we shall set @xmath77 .
for quantitative predictions in a specific atomic mixture , it is necessary but straightforward to include the effective range correction @xcite .
some details of our method to solve the integral equation are shown in appendix [ app:1d-3d ] .
figure [ fig : resonance_1d ] shows the positions of the lowest five resonances in terms of @xmath78 for @xmath183 partial - wave channels as functions of the mass ratio @xmath80 . for completeness
, we have included the result for the @xmath0-wave ( @xmath184 ) resonance , which has been reported in ref .
@xcite . as we have discussed in sec .
[ sec : introduction ] , there exists a series of resonances in each partial - wave channel induced from the purely @xmath0-wave interaction in a free space .
indeed , the resonance positions are well described by the approximate formula @xmath185 by using eq .
( [ eq : approx_1d ] ) with @xmath83 in a wide range of the mass ratio @xmath30 .
for such a mass ratio , as is evident from eq .
( [ eq : approx_1d ] ) , @xmath84-wave ( @xmath63-wave ) resonances are nearly degenerate with @xmath0-wave ( @xmath1-wave ) resonances , and thus it could be difficult to distinguish them practically .
we also note that the @xmath23th partial - wave resonance for @xmath186 is twofold degenerate in a circularly symmetric potential .
this degeneracy is lifted when the confinement potential is deformed , and therefore the one resonance at a given @xmath78 splits into two resonances . on the @xmath187 side of every resonance , a shallow @xmath6 molecule is formed . in the vicinity of the resonance @xmath188 , its binding energy @xmath189
is determined by the pole of the scattering amplitude [ eq .
( [ eq : expansion_1d ] ) ] with keeping the two dominant terms at @xmath89 : @xmath190 away from the resonance , these universal formulas are no longer valid .
the binding energy @xmath91 has to be determined by solving the integral equation @xmath191 we note that the solution for @xmath186 is independent of the sign of @xmath95 , and therefore there are two degenerate molecules for a circularly symmetric potential .
this degeneracy is lifted when the confinement potential is deformed .
we now derive the asymptotic form of the molecular wave function in the vicinity of the @xmath23th partial - wave resonance where @xmath192 is satisfied . from eqs .
( [ eq : solution_1d ] ) and ( [ eq : g_1d ] ) with the replacement @xmath97 , we find @xmath193 one can see that different spherical harmonics @xmath194 with the same magnetic quantum number @xmath95 contribute owing to the lack of full 3d rotational symmetries .
the @xmath0-wave nature of the free - space interaction ensures that only @xmath195 contribute so that the wave function is an even function of @xmath196 .
the asymptotic behavior at a large separation @xmath197 is dominated by the @xmath198 component .
the angular parts of asymptotic wave functions of two degenerate shallow @xmath1-wave molecules @xmath199 are illustrated in fig .
[ fig : molecule ] .
the effective scattering length in the @xmath0-wave channel @xmath100 has been computed in ref .
@xcite . here
we focus on the @xmath1-wave ( @xmath200 ) channel and determine its two low - energy scattering parameters , namely , the effective scattering volume @xmath102 and the effective momentum @xmath103 .
the effective scattering volume can be computed by eliminating @xmath104 from eqs .
( [ eq : a_eff_1d-1 ] ) and ( [ eq : a_eff_1d-2 ] ) and solving the resulting integral equation numerically ( see appendix [ app:1d-3d ] for details ) . in fig .
[ fig : parameter_1d ] , @xmath105 for @xmath77 is plotted as a function of @xmath78 for three mass ratios @xmath106 , @xmath107 , and @xmath108 .
we confirm the existence of a series of @xmath1-wave resonances ( @xmath109 ) induced from the purely @xmath0-wave interaction in a free space , while they become narrower for larger @xmath78 .
we also find that the resonance is wider when a lighter atom is confined in lower dimensions .
similarly , the effective momentum can be computed from eq .
( [ eq : partial - wave_1d ] ) , and @xmath110 for @xmath77 is plotted in fig .
[ fig : parameter_1d ] as a function of @xmath78 for the same three mass ratios . in the vicinity of the @xmath1-wave resonance @xmath111 , @xmath112 and
@xmath113 determine the binding energy of two degenerate shallow @xmath1-wave molecules via the universal formula ( [ eq : binding_1d ] ) .
both @xmath112 and @xmath113 are important to the low - energy effective theory of the @xmath1-wave resonance discussed below .
the low - energy effective theory of the @xmath1-wave resonance in the 1d-3d mixed dimensions is provided by the action @xmath201 , \end{split}\ ] ] where the summation over @xmath202 is implicitly understood . @xmath116 and @xmath117 fields represent the @xmath2 and @xmath3 atoms in 1d and 3d , respectively .
the interaction between @xmath2 and @xmath3 atoms is described through their coupling with two @xmath1-wave molecular fields @xmath118 .
@xmath119 is their coupling strength and @xmath120 is the detuning from the resonance .
these cutoff ( @xmath121)-dependent bare parameters can be related to the effective scattering volume @xmath112 and effective momentum @xmath113 by matching the two - body scattering amplitude from the action ( [ eq : action_1d ] ) with that shown in eqs .
( [ eq : decomposition_1d ] ) and ( [ eq : expansion_1d ] ) .
the standard diagrammatic calculation leads to the following scattering amplitude with collision energy @xmath122 : @xmath203 by defining @xmath204 and @xmath205 we reproduce the scattering amplitude ( [ eq : expansion_1d ] ) in the @xmath1-wave ( @xmath200 ) channel up to a kinematical factor : @xmath206 this low - energy effective theory can be generalized easily to the case with more than one tube where @xmath3 atoms are confined and could be used to investigate the many - body physics across the @xmath1-wave resonance as in the 3d case @xcite .
we note that the low - energy effective theory of the @xmath0-wave resonance in the 1d-3d mixed dimensions has been derived and used to study three - body problems in ref .
@xcite .
when @xmath128 and @xmath129 , the confinement - induced resonances can be understood in a different way . to the leading order in the weak - coupling expansion @xmath130 , an @xmath2 atom occupies the ground state in a 2d harmonic potential and creates a mean - field attractive potential felt by a @xmath3 atom .
therefore , the scattering of the @xmath3 atom by the confined @xmath2 atom is described by @xmath207 \\ & \times \psi(\bm\rho_b , z_{ab } ) = \frac{k^2}{2m_b}\psi(\bm\rho_b , z_{ab } ) , \end{split}\ ] ] where @xmath208 .
this schrdinger equation , which is valid in the weak - coupling limit , is equivalent to the integral equation ( [ eq : chi_1d ] ) , where only the @xmath133 term is kept in the green s function ( [ eq : g_1d ] ) and @xmath209 is identified as @xmath210 . by matching the solution of eq .
( [ eq : weak - coupling_1d ] ) with the asymptotic form ( [ eq : asymptotic_1d ] ) , we can determine the scattering amplitude and low - energy scattering parameters in the weak - coupling limit @xmath129 .
in particular , the resonance occurs when a new bound state appears .
this is possible even in the weak - coupling limit because the attractive potential becomes strong compared to the kinetic term by decreasing the mass ratio down to @xmath136 .
we find that the resonances are achieved at the critical values of @xmath211 in the @xmath184 channel , @xmath212 in the @xmath213 channel , @xmath214 in the @xmath215 channel , and @xmath216 in the @xmath217 channel . here
we present some details of our method to solve integral equations derived in sec .
[ sec:1d-3d ] .
the central issue is the evaluation of the regular part of the green s function defined in eq .
( [ eq : regular_1d ] ) .
when @xmath309 , it is useful to represent @xmath59 as @xmath310 \\ & \quad + \int_0^\infty\!d\tau \left(\frac{m_{ab}}{2\pi\tau}\right)^{3/2}\delta(\bm\rho-\bm\rho ' ) .
\end{split}\ ] ] its partial - wave projection gives @xmath311 \\
& \quad \times \exp\!\left[-\left(\frac{m_a\omega\cosh\omega\tau}{\sinh\omega\tau } + \frac{m_b}\tau\right)\frac{\rho^2+\rho'^2}2\right ] + \int_0^\infty\!d\tau\left(\frac{m_{ab}}{2\pi\tau}\right)^{3/2 } \frac{\delta(\rho-\rho')}{2\pi\rho}. \end{split}\ ] ] we now evaluate the matrix elements of @xmath312 with respect to the eigenfunctions of 2d harmonic oscillator with magnetic quantum number @xmath95 : @xmath313 which form an orthonormal basis : @xmath314 a lengthy but straightforward calculation leads to @xmath315 + \delta_{ij}\sqrt{\frac{m_{ab}}{2\pi m_a } } \int_0^\infty\!\frac{\omega d\tau}{(\omega\tau)^{3/2}}. \end{split}\ ] ] expanding @xmath316 in terms of @xmath317 , the integral equations in sec .
[ sec:1d-3d ] reduce to linear algebraic equations that have been solved numerically .
the scattering of a quasi-2d @xmath2 atom with a @xmath3 atom in 3d is described by a schrdinger equation , @xmath218 for @xmath219 , where @xmath220 and the center - of - mass motions in the @xmath153 and @xmath154 directions are eliminated . the short - range interspecies interaction is implemented by the generalized bethe - peierls boundary condition @xcite : @xmath221\chi(z ) .
\end{split}\ ] ] the collision energy operator @xmath42 in eq .
( [ eq : generalized_a ] ) in the present case is given by @xmath222 the solution to the schrdinger equation ( [ eq : schrodinger_2d ] ) can be written as @xmath223 where @xmath45 is a solution in the noninteracting limit and @xmath46 is the retarded green s function for the noninteracting hamiltonian : @xmath224 here @xmath225 is the normalized wave function of an @xmath2 atom in the 1d harmonic potential , and @xmath226 are coordinates of the @xmath3 atom relative to the confined @xmath2 atom .
the anisotropic factor is such because a separation in the @xmath18 direction is associated with @xmath155 while separations in the @xmath153 and @xmath154 directions are associated with @xmath156 [ see the last two terms in eq .
( [ eq : schrodinger_2d ] ) ] .
we now consider the low - energy scattering in which @xmath227 is satisfied , and then @xmath45 becomes @xmath228 which represents the @xmath2 atom in the ground state of the 1d harmonic potential and the plane wave of @xmath3 atom with the wave vector @xmath52 .
the asymptotic form of the wave function at a large distance @xmath159 is given by @xmath229\phi_0(z_a),\ ] ] where @xmath230 with @xmath231 defines the two - body scattering amplitude in the 2d-3d mixed dimensions : @xmath232 we note that @xmath58 has an implicit @xmath233 dependence and both @xmath58 and @xmath63 depend on @xmath165 through the green s function ( [ eq : g_2d ] ) .
the unknown function @xmath58 can be determined by substituting the solution ( [ eq : solution_2d ] ) into the bethe - peierls boundary condition ( [ eq : short - range_2d ] ) . defining the regular part of the green s function @xmath59 by @xmath234 we obtain @xmath235 this integral equation determines @xmath62 , which in turn provides @xmath63 from eq .
( [ eq : f_2d ] ) . because the system has a reflection symmetry about the @xmath18 axis , @xmath63 , @xmath58 , and @xmath59 can be decomposed into their even- and odd - parity components : @xmath236 equations ( [ eq : f_2d ] ) and ( [ eq : chi_2d ] ) lead to the even- and odd - parity scattering amplitudes given by @xmath237 with @xmath238 from the explicit calculation that uses the green s function in eq .
( [ eq : g_2d ] ) , we can show that @xmath239 has the following low - energy expansion : @xmath240 } \\ &
\quad + o(k^2_z , k'^2_z ) \end{split}\ ] ] and @xmath241 } \\ & \quad + o(k^3_zk'_z , k_zk'^3_z ) , \end{split}\ ] ] where @xmath242 and @xmath243 are effective scattering `` length '' and `` range '' parameters in the even- and odd - parity channels .
note that @xmath242 has the dimension of @xmath244 and @xmath243 has @xmath245 .
this unusual form of the low - energy expansion is owing to the lack of full 3d rotational symmetries . substituting the expansion of @xmath239 into eq .
( [ eq : partial - wave_2d ] ) , we can determine the low - energy scattering parameters .
in particular , the effective scattering length @xmath242 is given by @xmath246 and @xmath247 where we have eliminated @xmath104 .
the even- and odd - parity resonances in the 2d-3d mixed dimensions are defined by the divergence of @xmath248 , which occurs when @xmath249 is satisfied .
we now solve the integral equation ( [ eq : resonance_2d ] ) numerically to determine the positions of even- and odd - parity resonances .
for the purpose of illustrating qualitative results , we shall set @xmath77 .
for quantitative predictions in a specific atomic mixture , it is necessary but straightforward to include the effective range correction @xcite .
some details of our method to solve the integral equation are shown in appendix [ app:2d-3d ] .
figure [ fig : resonance_2d ] shows the positions of the lowest five resonances in terms of @xmath78 for even- and odd - parity channels as functions of the mass ratio @xmath80 . for completeness , we have included the result for the even - parity resonance , which has been reported in ref .
@xcite . as we have discussed in sec .
[ sec : introduction ] , there exists a series of resonances in each parity channel induced from the purely @xmath0-wave interaction in a free space .
indeed , the resonance positions are well described by the approximate formula @xmath250 by using eq .
( [ eq : approx_2d ] ) with @xmath83 in a wide range of the mass ratio @xmath30 . on the @xmath251 side of every resonance , a shallow @xmath6 molecule is formed . in the vicinity of the resonance @xmath252 , its binding energy @xmath253
is determined by the pole of the scattering amplitude [ eqs .
( [ eq : expansion_2d_even ] ) and ( [ eq : expansion_2d_odd ] ) ] with keeping the two dominant terms at @xmath89 : @xmath254 away from the resonance , these universal formulas are no longer valid .
the binding energy @xmath91 has to be determined by solving the integral equation @xmath255 we now derive the asymptotic form of the molecular wave function in the vicinity of the even- or odd - parity resonance where @xmath192 is satisfied . from eqs .
( [ eq : solution_2d ] ) and ( [ eq : g_2d ] ) with the replacement @xmath97 , we find @xmath256 one can see that different spherical harmonics @xmath257 contribute owing to the lack of full 3d rotational symmetries .
the @xmath0-wave nature of the free - space interaction ensures that only the @xmath184 component contributes so that the wave function is independent of @xmath258 .
the asymptotic behavior at a large separation @xmath197 is dominated by the @xmath81 or @xmath107 component in the even- or odd - parity channel , respectively .
therefore , we can phrase the even ( odd)-parity resonance as an @xmath0-wave ( @xmath1-wave ) resonance in the 2d-3d mixture .
the angular part of the asymptotic wave function of a shallow @xmath1-wave molecule @xmath259 is illustrated in fig .
[ fig : molecule ] . the effective scattering length in the @xmath0-wave ( even - parity ) channel @xmath260 has been computed in ref .
@xcite . here
we focus on the @xmath1-wave ( odd - parity ) channel and determine its two low - energy scattering parameters , namely , the effective scattering volume @xmath261 and the effective momentum @xmath262 . the effective scattering volume can be computed by solving the integral equation ( [ eq : v_eff_2d ] ) numerically ( see appendix [ app:2d-3d ] for details ) . in fig .
[ fig : parameter_2d ] , @xmath105 for @xmath77 is plotted as a function of @xmath78 for three mass ratios @xmath106 , @xmath107 , and @xmath108 .
we confirm the existence of a series of @xmath1-wave resonances ( @xmath109 ) induced from the purely @xmath0-wave interaction in a free space , while they become narrower for larger @xmath78 .
we also find that the resonance is wider when a lighter atom is confined in lower dimensions .
similarly , the effective momentum can be computed from eq .
( [ eq : partial - wave_2d ] ) , and @xmath110 for @xmath77 is plotted in fig .
[ fig : parameter_2d ] as a function of @xmath78 for the same three mass ratios . in the vicinity of the @xmath1-wave resonance @xmath111 , @xmath112 and
@xmath113 determine the binding energy of a shallow @xmath1-wave molecule via the universal formula ( [ eq : binding_2d ] ) .
both @xmath112 and @xmath113 are important to the low - energy effective theory of the @xmath1-wave resonance discussed below .
the low - energy effective theory of the @xmath1-wave resonance in the 2d-3d mixed dimensions is provided by the action @xmath263 .
\end{split}\ ] ] @xmath116 and @xmath117 fields represent the @xmath2 and @xmath3 atoms in 2d and 3d , respectively .
the interaction between @xmath2 and @xmath3 atoms is described through their coupling with a @xmath1-wave molecular field @xmath264 .
@xmath119 is their coupling strength and @xmath120 is the detuning from the resonance .
these cutoff ( @xmath121)-dependent bare parameters can be related to the effective scattering volume @xmath112 and effective momentum @xmath113 by matching the two - body scattering amplitude from the action ( [ eq : action_2d ] ) with that shown in eqs .
( [ eq : decomposition_2d ] ) and ( [ eq : expansion_2d_odd ] ) .
the standard diagrammatic calculation leads to the following scattering amplitude with collision energy @xmath122 : @xmath265 by defining @xmath266 and @xmath267 we reproduce the scattering amplitude ( [ eq : expansion_2d_odd ] ) in the @xmath1-wave ( odd - parity ) channel up to a kinematical factor : @xmath268 this low - energy effective theory can be generalized easily to the case with more than one layer where @xmath3 atoms are confined and could be used to investigate the many - body physics across the @xmath1-wave resonance as in the 3d case @xcite .
we note that the low - energy effective theory of the @xmath0-wave resonance in the 2d-3d mixed dimensions has been derived and used to study many - body problems in refs .
@xcite .
finally , we compare our predictions to the experimental measurement of resonance positions reported in ref .
the florence group has realized the 2d-3d mixed dimensions by using a bose - bose mixture of @xmath269k and @xmath270rb with a species - selective 1d optical lattice : @xmath271 here @xmath0 is the lattice depth parameter and @xmath272 with @xmath273 nm is the wave vector of the laser light . by monitoring three - body inelastic losses ,
a series of resonances has been observed as a function of the magnetic field .
the measured resonance positions are shown in fig .
[ fig : experiment ] for five values of the lattice depth parameter , @xmath274 , and @xmath275 . as for theoretical predictions , we assume that the lattice potential ( [ eq : lattice ] )
can be approximated by a harmonic potential : @xmath276 the harmonic oscillator length is given from the laser wavelength by @xmath277 . with the use of the free - space value of the effective range @xmath278 for the @xmath26k-@xmath27rb mixture
, we solve the integral equation ( [ eq : resonance_2d ] ) numerically to determine the resonance positions in terms of @xmath78 .
the free - space scattering length @xmath279 thus obtained is converted into the magnetic field @xmath3(g ) by the following empirical formula @xcite : @xmath280 where @xmath281 is the bohr radius and the resonance in a free space occurs at @xmath282 g .
these critical magnetic field values for the @xmath0-wave ( @xmath283 ) and @xmath1-wave ( @xmath109 ) resonances are plotted in fig .
[ fig : experiment ] as functions of the lattice depth parameter @xmath0 . as functions of the depth parameter @xmath0 of an optical lattice @xcite . also shown
are the predicted positions of @xmath0-wave ( @xmath283 ; solid curves ) and @xmath1-wave ( @xmath109 ; dashed curves ) resonances in a harmonic potential as well as the resonance position in a free space ( @xmath284 ; dotted line ) .
[ fig : experiment ] ] in fig .
[ fig : experiment ] , the resonances are labeled by an integer @xmath285 in descending order , corresponding to @xmath17 in eq . ( [ eq : approx_2d ] ) .
one can see the reasonable agreement between the experimental measurements and the theoretical predictions for @xmath286 ( @xmath0-wave ) and @xmath287 ( @xmath1-wave ) resonances .
they begin to deviate for larger @xmath17 simply because the harmonic potential approximation ( [ eq : harmonic ] ) becomes worse for higher excited states in an optical lattice .
although the lattice potential ( [ eq : lattice ] ) has to be taken into account to improve the quantitative agreement @xcite , for a sufficiently strong optical lattice , the resonances corresponding to odd @xmath17 should be the @xmath1-wave resonances in the 2d-3d mixed dimensions , as we have discussed in this section .
further experimental investigations to confirm the @xmath1-wave nature of these resonances would be worthwhile . the @xmath1-wave nature of shallow molecules as illustrated in fig .
[ fig : molecule ] could be seen as a suppression of the density of @xmath3 atoms , as opposed to an enhancement for @xmath0-wave molecules , at the center of the confinement potential after a sudden change of @xmath288 to its negative side . here
we present some details of our method to solve integral equations derived in sec .
[ sec:2d-3d ] .
the central issue is the evaluation of the regular part of the green s function defined in eq .
( [ eq : regular_2d ] ) . when @xmath318 , it is useful to represent @xmath59 as @xmath319 \\ & \quad + \int_0^\infty\!d\tau \left(\frac{m_{ab}}{2\pi\tau}\right)^{3/2}\delta(z - z ' ) .
\end{split}\ ] ] its parity projection gives @xmath320 \\
& \quad \times \exp\!\left[-\left(\frac{m_a\omega\cosh\omega\tau}{\sinh\omega\tau } + \frac{m_b}\tau\right)\frac{z^2+z'^2}2\right ] + \int_0^\infty\!d\tau\left(\frac{m_{ab}}{2\pi\tau}\right)^{3/2 } \frac{\delta(z - z')\pm\delta(z+z')}2 .
\end{split}\ ] ] we now evaluate the matrix elements of @xmath321 with respect to the eigenfunctions of 1d harmonic oscillator : @xmath322 which form an orthonormal basis : @xmath323 a lengthy but straightforward calculation leads to @xmath324 + \delta_{ij}\sqrt{\frac{m_{ab}}{2\pi m_a } } \int_0^\infty\!\frac{\omega d\tau}{(\omega\tau)^{3/2 } } \end{split}\ ] ] for @xmath325 and @xmath326 + \delta_{ij}\sqrt{\frac{m_{ab}}{2\pi m_a } } \int_0^\infty\!\frac{\omega d\tau}{(\omega\tau)^{3/2 } } \end{split}\ ] ] for @xmath327 . expanding @xmath328 in terms of @xmath329 , the integral equations in sec .
[ sec:2d-3d ] reduce to linear algebraic equations that have been solved numerically . c. chin , r. grimm , p. julienne , and e. tiesinga , rev .
* 82 * , 12251286 ( 2010 ) .
w. ketterle and m. w. zwierlein , arxiv:0801.2500 , in _ proceedings of the international school of physics `` enrico fermi , '' _ edited by m. inguscio , w. ketterle , and c. salomon ( ios press , amsterdam , 2008 ) , and references therein .
i. bloch , j. dalibard , and w. zwerger , rev .
phys . * 80 * , 885 ( 2008 ) .
s. giorgini , l. p. pitaevskii , and s. stringari , rev .
* 80 * , 1215 ( 2008 ) .
l . ho and r. b. diener , phys .
lett . * 94 * , 090402 ( 2005 ) .
s. s. botelho and c. a. r. s de melo , j. of low temp .
phys . * 140 * , 409 ( 2005 ) .
y. ohashi , phys .
94 * , 050403 ( 2005 ) .
v. gurarie , l. radzihovsky , and a. v. andreev , phys .
lett . * 94 * , 230403 ( 2005 ) .
cheng and s .- k . yip , phys .
* 95 * , 070404 ( 2005 ) ; phys .
b * 73 * , 064517 ( 2006 ) . c. a. regal , c. ticknor , j. l. bohn , and d. s. jin , phys .
* 90 * , 053201 ( 2003 ) .
j. zhang , e. g. m. van kempen , t. bourdel , l. khaykovich , j. cubizolles , f. chevy , m. teichmann , l. tarruell , s. j. j. m. f. kokkelmans , and c. salomon , phys .
a * 70 * , 030702(r ) ( 2004 ) . c. h. schunck , m. w. zwierlein ,
c. a. stan , s. m. f. raupach , w. ketterle , a. simoni , e. tiesinga , c. j. williams , and p. s. julienne , phys .
a * 71 * , 045601 ( 2005 ) .
f. chevy , e. g. m. van kempen , t. bourdel , j. zhang , l. khaykovich , m. teichmann , l. tarruell , s. j. j. m. f. kokkelmans , and c. salomon , phys .
a * 71 * , 062710 ( 2005 ) .
k. gnter , t. stferle , h. moritz , m. khl , and t. esslinger , phys .
lett . * 95 * , 230401 ( 2005 ) .
j. p. gaebler , j. t. stewart , j. l. bohn , and d. s. jin , phys .
lett . * 98 * , 200403 ( 2007 ) .
d. s. jin , j. p. gaebler , and j. t. stewart , in _ proceedings of the international conference on laser spectroscopy , _ edited by l. hollberg , j. bergquist , and m. kasevich ( world scientific , singapore , 2008 ) .
j. fuchs , c. ticknor , p. dyke , g. veeravalli , e. kuhnle , w. rowlands , p. hannaford , and c. j. vale , phys .
a * 77 * , 053616 ( 2008 ) .
y. inada , m. horikoshi , s. nakajima , m. kuwata - gonokami , m. ueda , and t. mukaiyama , phys .
* 101 * , 100401 ( 2008 ) .
r. a. w. maier , c. marzok , c. zimmermann , and ph . w. courteille , phys .
a * 81 * , 064701 ( 2010 ) .
j. levinsen , n. r. cooper , and v. gurarie , phys .
lett . * 99 * , 210402 ( 2007 ) ; phys .
a * 78 * , 063616 ( 2008 ) .
m. jona - lasinio , l. pricoupenko , and y. castin , phys .
a * 77 * , 043611 ( 2008 ) .
d. s. petrov , c. salomon , and g. v. shlyapnikov , phys .
lett . * 93 * , 090404 ( 2004 ) ; phys .
a * 71 * , 012708 ( 2005 ) ; j. phys .
b * 38 * , s645 ( 2005 ) .
p. massignan and y. castin , phys .
a * 74 * , 013616 ( 2006 ) .
y. nishida and s. tan , phys .
* 101 * , 170401 ( 2008 ) .
l. j. leblanc and j. h. thywissen , phys .
a * 75 * , 053612 ( 2007 ) .
j. catani , g. barontini , g. lamporesi , f. rabatti , g. thalhammer , f. minardi , s. stringari , and m. inguscio , phys .
lett . * 103 * , 140401 ( 2009 ) .
g. lamporesi , j. catani , g. barontini , y. nishida , m. inguscio , and f. minardi , phys .
* 104 * , 153202 ( 2010 ) .
d. s. petrov , phys .
lett . * 93 * , 143201 ( 2004 ) .
j. levinsen , t. g. tiecke , j. t. m. walraven , and d. s. petrov , phys .
lett . * 103 * , 153202 ( 2009 ) .
see , for example , j. j. sakurai , _ modern quantum mechanics _
( addison - wesley , reading , ma , 1994 ) .
v. gurarie and l. radzihovsky , ann .
* 322 * , 2 ( 2007 ) . c. a. bertulani , h. w. hammer , and u. van kolck , nucl .
a * 712 * , 37 ( 2002 ) .
f. bedaque , h. w. hammer , and u. van kolck , phys .
b * 569 * , 159 ( 2003 ) .
e. vernier , d. pekker , m. w. zwierlein , and e. demler , arxiv:1010.6085 [ cond-mat.quant-gas ] .
y. nishida and s. tan , phys .
a * 79 * , 060701(r ) ( 2009 ) .
y. nishida , ann .
324 * , 897 ( 2009 ) .
y. nishida , phys .
a * 82 * , 011605 ( 2010 ) .
u. gavish and y. castin , phys .
lett . * 95 * , 020401 ( 2005 ) .
m. antezza , y. castin , and d. a. w. hutchinson , phys .
a * 82 * , 043602 ( 2010 ) .
m. taglieber , a .- c . voigt , t. aoki , t. w. hnsch , and k. dieckmann , phys .
lett . * 100 * , 010401 ( 2008 ) .
e. wille , f. m. spiegelhalder , g. kerner , d. naik , a. trenkwalder , g. hendl , f. schreck , r. grimm , t. g. tiecke , j. t. m. walraven , s. j. j. m. f. kokkelmans , e. tiesinga , and p. s. julienne , phys .
* 100 * , 053201 ( 2008 ) .
voigt , m. taglieber , l. costa , t. aoki , w. wieser , t. w. hnsch , and k. dieckmann , phys .
lett . * 102 * , 020405 ( 2009 ) .
t. g. tiecke , m. r. goosen , a. ludewig , s. d. gensemer , s. kraft , s. j. j. m. f. kokkelmans , and j. t. m. walraven , phys .
lett . * 104 * , 053202 ( 2010 ) .
d. naik , a. trenkwalder , c. kohstall , f. m. spiegelhalder , m. zaccanti , g. hendl , f. schreck , r. grimm , t. m. hanna , and p. s. julienne , arxiv:1010.3662 [ cond-mat.quant-gas ] .
in this paper , we showed that a purely @xmath0-wave interaction in a free space can induce higher partial - wave resonances in mixed dimensions .
we developed two - body scattering theories in all three cases of 0d-3d , 1d-3d , and 2d-3d mixtures , and determined the positions of higher partial - wave resonances in terms of the free - space scattering length , assuming a harmonic confinement potential .
we also computed the low - energy scattering parameters in the @xmath1-wave channel ( effective scattering volume and momentum ) that are necessary for the low - energy effective theory of the @xmath1-wave resonance .
potentially our study paves the way for a variety of physics , such as anderson localization of matter waves under @xmath1-wave resonant scatterers in the 0d-3d mixed dimensions @xcite , shiba bound states in a superfluid medium @xcite , and bose - einstein condensation of @xmath1-wave molecules in the 2d-3d mixed dimensions , provided that such molecules are long lived . when both @xmath2 and @xmath3 atoms are fermionic ( such as for the @xmath289li-@xmath290k mixture @xcite ) , inelastic three - body , atom - molecule , and molecule - molecule collisions decaying into deeply bound dimers whose size is set by the range of interatomic potential @xmath291 are suppressed .
this is because at a short distance @xmath292 , the confinement potential is irrelevant , and in the @xmath0-wave interspecies interaction the pauli exclusion principle is effective to suppress the inelastic collisions decaying into the deeply bound dimers @xcite .
an order - of - magnitude estimate of such a three - body recombination rate can be found in refs .
@xcite and @xcite for wide feshbach resonances and in ref .
@xcite for narrow feshbach resonances
. therefore , for a fermi - fermi mixture in mixed dimensions , the relaxation of molecules is dominated by the decay into deeper molecular states of size @xmath293 if they exist . in the vicinity of the @xmath0-wave resonance at the smallest value of @xmath78 , indicated by thick curves in figs .
[ fig : resonance_0d ] , [ fig : resonance_1d ] , and [ fig : resonance_2d ] , there is no such molecular state and thus associated @xmath0-wave molecules are long lived .
however , in the vicinity of the higher partial - wave resonances , there is always at least one deeper molecular state .
the inelastic atom - molecule and molecule - molecule collisions decaying into such molecular states are not generally suppressed , and thus @xmath1-wave or higher partial - wave molecules in mixed dimensions would be short lived .
it is therefore an important future problem to investigate as to whether the inelastic collisions of @xmath1-wave molecules decaying into the lowest @xmath0-wave state of size @xmath293 could be suppressed , for example , by controlling the system parameters .
we conclude this paper by discussing a simple analogy between a tower of our confinement - induced molecules and kaluza - klein modes in extra - dimension models @xcite .
suppose our 3d world has one compact extra dimension with an extent of @xmath294 .
then the momentum of a 4d particle in the extra fourth direction is quantized as @xmath295 .
because we live in 3d , such a particle can be viewed as a tower of `` new particles '' with the quantized masses @xmath296 .
these new particles are called kaluza - klein modes and could be observed by colliding high - energy particles @xmath297 at the large hadron collider . now in our case , for example , in the 2d-3d mixed dimensions , the motion of a 3d molecule in the `` extra '' @xmath18 direction is quantized as in eq .
( [ eq : approx_2d ] ) because of the confinement potential .
we can view such quantized energy levels as a tower of new `` feshbach molecules '' analogous to the kaluza - klein modes . unlike in extra - dimension models , our particle is a composite molecule and we can control its binding energy to shift the energies of the tower of new molecules .
when each of them crosses the scattering threshold , the resonance occurs . a series of such resonances caused by the `` kaluza - klein '' tower of confinement - induced molecules has been investigated in this paper and observed in the cold - atom experiment @xcite .
the authors thank yvan castin , eugene demler , david pekker , andrey turlapov , fa wang , zhenhua yu , and martin zwierlein for valuable discussions and jacopo catani , giacomo lamporesi , and francesco minardi for providing the experimental data .
y.n . was supported by mit pappalardo fellowship in physics and the doe office of nuclear physics under grant
de - fg02 - 94er40818 .
thanks the institute for advanced study in tsinghua university for hospitality when this work was near completion . | we show that a purely @xmath0-wave interaction in three dimensions ( 3d ) can induce higher partial - wave resonances in mixed dimensions .
we develop two - body scattering theories in all three cases of 0d-3d , 1d-3d , and 2d-3d mixtures and determine the positions of higher partial - wave resonances in terms of the 3d @xmath0-wave scattering length assuming a harmonic confinement potential .
we also compute the low - energy scattering parameters in the @xmath1-wave channel ( scattering volume and effective momentum ) that are necessary for the low - energy effective theory of the @xmath1-wave resonance .
we point out that some of the resonances observed in the florence group experiment [ phys .
rev .
lett . * 104 * , 153202 ( 2010 ) ] can be interpreted as the @xmath1-wave resonances in the 2d-3d mixed dimensions .
our study paves the way for a variety of physics , such as anderson localization of matter waves under @xmath1-wave resonant scatterers . |
in hybrid systems of ferromagnetic and normal metals , interesting phenomena can appear due the interplay between charge and spin .
the discovery of the giant magnetoresistance ( gmr ) effect in metallic magnetic multilayers,@xcite has motivated a large number of studies on the transport properties of such systems.@xcite the gmr is caused by spin dependent scattering in the system .
most studies concentrated on collinear configurations ( parallel and antiparallel configurations ) .
there are several papers which cover non - collinear magnetizations , both theoretical @xcite and experimental.@xcite magnetoelectronic multiterminal devices reveal new physics,@xcite but may also lead to novel applications , _ e.g.
_ non - volatile electronics .
johnson and silsbee investigated spin dependent effects in a 3-terminal device.@xcite they found transistor effects that depend on the relative orientation of the magnetization of the ferromagnets .
@xcite more recently , a ferromagnetic single - electron transistor in a three terminal configuration has been realized@xcite and studied theoretically.@xcite in this case the source - drain current also depends on the relative orientation of the magnetizations .
brataas _ et al._@xcite give a unified semiclassical picture for electron and spin transport in such systems .
their formalism is inspired by the circuit theory of the andreev reflection,@xcite and is applicable to systems with non - collinear magnetization directions and an arbitrary number and variety of contacts between the ferromagnetic and the normal metals .
however , the simple circuit theory of ref . only holds when the resistances of the contacts between the ferromagnetic and the normal metals are much higher than the resistance of the normal metal itself , thus fails when the size @xmath2 of the system in the transport direction becomes too large .
moreover , when the size of the system is larger than the spin diffusion length ( @xmath3 ) , the presence of spin - diffusion in the normal metal requires a more complicated description with spatially dependent spin distribution functions . in the present paper ,
we present a study of the transport properties of simple f - n - f systems ( see fig .
1 ) , taking into account different magnetizations of the ferromagnetic reservoirs and spin - diffusion in the normal metal . at low temperatures
, spin - flip can be due to spin - orbit interactions and scattering by defects or impurities .
exchange scattering by paramagnetic impurities also flips the spin ( see _ e.g. _ appendix a in ref . ) .
the length of the normal metal @xmath2 is assumed to be much larger than the mean - free path @xmath4 , so electronic transport may be described by the diffusion equation . on the other hand
, we allow the spin diffusion length @xmath5 , which is the length scale on which an electron looses its spin in diffusive transport , to be much smaller , of the same order , or much larger than the size of the system @xmath2 . under an applied bias ,
ferromagnetic reservoirs inject a spin - current , causing a non - equilibrium magnetization or `` spin accumulation '' in the normal metal .
we are interested in the different mechanisms that reduce and also rotate this spin accumulation . for non - collinear configurations
the physics of spin injection is more subtle than in the simple collinear case , since it requires generalized boundary conditions for transport through a single ferromagnetic - normal metal ( f - n ) contact.@xcite in general , such a contact is charaterized not only by the conventional spin dependent conductances @xmath6 , which describe the transport of spins collinear to the magnetization of the ferromagnetic reservoir , but also by the @xmath7 _ mixing _ conductance @xmath8 ( see ref . ) , that contains information about the transport of spins oriented perpendicular to the magnetization of the ferromagnetic reservoir .
we are also interested in the effect of a magnetic field applied to the diffusive normal metal in arbitrary directions . in this case
we assume that the magnetic field only couples to the spin degrees of freedom .
our approach is similar to the treatment of a precessing magnetic field applied to a diffusive metal in ref .. in section ii we introduce and solve the basic equations for the diffusive spin transport , showing the general expression for the non - equilibrium distribution function in the normal metal . in section
iii we discuss the boundary conditions of the problem . in section iv we obtain analytical expressions for the total conductance of the system in collinear configurations and in the absence of applied magnetic field .
we also obtain analytical expressions for the total conductance in the case of non - collinear magnetization directions , zero magnetic field and no spin - flip scattering . in section
v we calculate numerically the conductance in the general case . in section
vi we summarize and discuss our results .
when a bias is applied to our f - n - f device , a spin current is injected from the ferromagnetic reservoirs into the normal metal , causing a non - equilibrium magnetization or spin accumulation . for an arbitrary magnetic configuration of the system ,
the spin accumulated in the normal metal can be oriented in differents directions . if we take the spin quantization axis parallel to the magnetization of one of the ferromagnetic reservoirs , we need to take into account spins oriented perpendicular to this quantization axis , which can be described as a superposition of up ( @xmath9 ) and down ( @xmath10 ) spin states .
we study a geometry invariant to translations in the lateral direction , so all quantities depend only on one spatial coordinate @xmath11 .
the spin - polarized electron distribution is characterized by a @xmath12 matrix in spin space of the form : @xmath13 when the size of the system @xmath2 is larger than the spin diffusion length @xmath14 , @xmath15 depends on the position . here
we are interested in transport under the condition @xmath16 , where @xmath17 is the _
mean free path _
, @xmath18 is the fermi velocity , @xmath19 the spin - conserving scattering time and @xmath20 the spin - flip scattering time .
both @xmath19 and @xmath21 are considered isotropic in momentum space@xmath22 the spin diffusion length @xmath5 is defined as @xmath23 , where @xmath24 , is the spin - independent diffusion coefficient of the normal metal ( @xmath25 is the dimension of the normal metal ) .
so under the condition @xmath16 , we obtain for diffusive spin transport in the steady state the following @xmath26 matrix equations for @xmath15 @xmath27 where @xmath28 is the unit matrix and where the electron charge @xmath29 is assumed to be equal to one .
( [ eqspinrelax ] ) describes the relaxation of the spin accumulation due to spin - flip scattering , and ( [ currentdens . ] ) relates the current density matrix @xmath30 and @xmath15 . in the case of collinear transport
, our matrix equations simply reduce to @xmath31 where @xmath32 .
( [ valetfert1 ] ) and ( [ valetfert2 ] ) have been extensively used for collinear transport in f - n multilayers in which the current flows perpendicular to the planes of the interfaces ( cpp geometry ) .
@xcite we are also interested in the effect of an external magnetic field applied to the normal metal in an arbitrary direction . we know that the magnetic zeeman energy associated with the coupling between the magnetic field and the spin of the electrons is given by @xmath33 , where @xmath34 is the bohr magneton , @xmath35 is the gyromagnetic ratio , @xmath36 is the vector of pauli matrices and @xmath37 is the external magnetic field .
semiclassically , we can write for the spin dynamics ( see _ e.g. _ ref . ) @xmath38 _ { -}. \label{comutf}\ ] ] then , in the steady state : @xmath39 _ { -}. \label{eqcompmagnet}\ ] ] using the properties of the pauli matrices we can express the non - equilibrium distribution matrix @xmath15 as : @xmath40 where @xmath41 is a scalar and @xmath42 is a three component vector .
@xmath41 is the particle or _ spin - independent _ distribution function . on the other hand
, @xmath43 describes the `` spin polarization '' on the system , and @xmath44 and @xmath45 contain information about the spins oriented perpendicular to the quantization axis .
we call the three component vector @xmath46 the _ spin - dependent _ distribution
_ _ function . using ( [ transpaulimat ] )
, we separate ( [ eqcompmagnet ] ) into two contributions , one for the _ spin - independent _ part and another for the _ spin - dependent _ part : @xmath47 [ finaleqs ] the spin - independent part ( eq.([finaleqs1 ] ) ) , is the conventional result for diffusive particle transport .
similar to eq .
( [ valetfert2 ] ) , the _ particle _ current density @xmath48 reads @xmath49 } { \partial x}=-2d\frac{\partial f_{0}(x)}{\partial x}. \label{partcurrentdens.}\ ] ] eqs .
( [ finaleqs1 ] ) and ( [ partcurrentdens . ] ) express the particle current conservation @xmath50 the general solution of eq .
( [ finaleqs1 ] ) is @xmath51 eq .
( [ finaleqs3 ] ) describes how the spin accumulation relaxes by spin - flip scattering and by the spin _ precession _ around the magnetic field .
this equation can be written in a general matrix form as : @xmath52 the eigengenvalues associated with the matrix @xmath53 are : @xmath54 where we have introduced the vector @xmath55 , which describes the `` effectiveness '' of the magnetic field in a diffusive metal .
the eigenvector associated with @xmath56 is @xmath57 on the other hand , @xmath58 and @xmath59 have associated two complex conjugated eigenvectors @xmath60 and @xmath61 , where @xmath62 and @xmath63 the general solution of eq .
( [ finaleqs3 ] ) , can then be written in terms of the eigenvalues @xmath64
@xmath65 @xmath59 and vectors @xmath66 @xmath67 @xmath68 as : @xmath69 \\
-{\cal d}\text { } \left [ \vec{v}_{1}\sinh ( x)\sin ( y)+\vec{v}_{2}\cosh ( x)\cos ( y)\right ] \\ + { \cal e}\text { \ } \left [ \vec{v}_{1}\sinh ( x)\cos ( y)-\vec{v}_{2}\cosh ( x)\sin ( y)\right ] \\
-{\cal f}\text { \ } \left [ \vec{v}_{1}\cosh ( x)\sin ( y)+\vec{v}_{2}\sinh ( x)\cos ( y)\right ] .
\end{array } \right\}\ ] ] where @xmath70 and where the dimensionless constant @xmath71 @xmath72 is the ratio between spin - flip and precession relaxation mechanisms .
@xmath73 is the ( larmor ) frequency for the spin precession around the magnetic field.@xcite the solution associated with @xmath56 describes the relaxation of the spin accumulation due to spin - flip scattering , and the two complex conjugated solutions associated with @xmath74 and @xmath59 describe the relaxation and precession of the spins due to the coupling with the magnetic field .
the eight * * real constants ( @xmath75 ) must be determined by the boundary conditions .
we consider two ferromagnetic reservoirs attached to a diffusive normal metal through some arbitrary contacts , as shown in fig .
1 . the ferromagnetic reservoirs are supposed to be large and in local equilibrium at chemical potentials @xmath76 ( @xmath77 , @xmath78 denotes left and right reservoir respectively ) , and with energy - dependent diagonal distribution matrices in spin space @xmath79 .
the components of @xmath80 are given by the fermi - dirac distribution function @xmath81 , and the direction of the magnetization in each ferromagnetic reservoir is denoted by the unit vector @xmath82 .
the current through the system and the non - equilibrium distribution function in the normal metal are completely determined by the relative orientation of the magnetization directions in the ferromagnetic reservoirs , the contact conductances , the normal metal conductance , the spin diffusion length and the magnetic field .
the current through an f - n contact is given in ref . in terms of the microscopic scattering matrices of the landauer - bttiker formalism.@xcite according to eq .
( 3 ) of ref .
, the particle current through a single contact directed into the normal metal can be written as @xmath83 where @xmath15 and @xmath84 are isotropic distribution functions , @xmath85 and @xmath86 are the conventional spin - dependent conductances , which describe the transport of spins oriented in the direction of the magnetization of the adjacent _ _ ferromagnetic reservoir , and @xmath87 is the _ mixing conductance , _ which contains information about the transport of spins oriented in perpendicular direction to the magnetization of the ferromagnetic reservoir . the matrices @xmath88 and @xmath89 define the basis in which the spin - quantization axis is parallel to the magnetization of the ferromagnet ( for details see ref . ) .
( [ contactcurrent ] ) relates the spin current through the contact @xmath90 and the non - equilibrium distribution matrix @xmath15 in the normal metal . due to current conservation ,
( [ contactcurrent ] ) is equal , at each contact , to the particle current per energy interval in the normal metal ( see fig .
the particle current per energy interval is related with the current density @xmath30 as , @xmath91 @xmath92 @xmath30 , where @xmath93 is the surface perpendicular to the transport direction and @xmath94 is the density of states of the normal metal . using eq .
( [ currentdens . ] ) @xmath95 is @xmath96 so we have @xmath97 for the left contact ( @xmath98 ) and @xmath99 for the right contact ( @xmath100 ) . by substituting ( [ contactcurrent ] ) and ( [ currentnormal ] ) into ( [ boundary1 ] ) and ( [ boundary2 ] )
, we obtain the boundary conditions for the left contact ( @xmath98 ) : @xmath101 and for the right contact ( @xmath100 ) : @xmath102 the set of parameters @xmath103 is in general different for each contact , but we have omitted the indices @xmath104 and @xmath78 in ( [ system1 ] ) and ( [ system2 ] ) for brevity . eqs .
( [ system1 ] ) and ( [ system2 ] ) are two @xmath12 matrix equations , that provide us a system of linear equations that determinate the eight unknown constants ( @xmath105 ) .
from ( [ partcurrentdens . ] ) we can see that the total particle current @xmath106 can be written in terms of one of these constants as : @xmath107 where @xmath108 is the normal metal conductance and @xmath109 is the volume of the normal metal . by solving the system of equations ( [ system1 ] ) and ( [ system2 ] )
, we can calculate this total particle current .
@xmath110 is proportional to the difference between the distribution functions of the ferromagnets @xmath111 , times a quantity which does not depends on energy . from this quantity
is possible to obtain the total conductance @xmath112 : @xmath113 where @xmath112 is in principle a function of the relative orientation of the magnetization directions in the ferromagnetic reservoirs , the contacts and normal metal conductances , the spin diffussion length and also of the magnetic field:@xcite @xmath114 by studying @xmath112 for different values of these parameters , we obtain information about the physics of the spin accumulation in diffusive systems .
the properties of the contacts are parametrized by the spin - dependent conductances @xmath115 . for collinear configurations of the ferromagnetic reservoirs ( parallel and antiparallel ) , it is easy to obtain simple expressions for the conductance , which can be interpreted by simple equivalent circuits .
when @xmath116 , there is no mixing between spin - up ( @xmath117 ) and spin - down ( @xmath10 ) channels , and we obtain the conductance for the parallel configuration @xmath118 and for the antiparallel configuration : @xmath119 on the other hand , when @xmath120 , spin - up ( @xmath117 ) and spin - down ( @xmath121 ) channels are completely mixed due to spin - flip scattering and the spin accumulation vanishes . in this limit
we have : @xmath122 these expressions correspond to the simple equivalent circuits displayed in fig .
( [ gpgeneral ] ) corresponds to a circuit in which the two spin channels are independent in the parallel configuration ( fig .
3a ) . eq .
( [ gapgeneral ] ) corresponds to the anti - parallel configuration ( fig .
3b ) . eq .
( [ gogeneral ] ) is equivalent to a circuit with a complete mixing between spin up ( @xmath117 ) and spin - down ( @xmath10 ) channels ( fig .
3c ) , in which there is no difference between parallel and anti - parallel configurations . for symmetric contacts @xmath123 , we find analytical expressions for the conductance of the system for any value of @xmath124 , in the parallel configuration : @xmath125 and in the antiparallel configuration : @xmath126 where @xmath127 .
in the limit @xmath116 , these equations reduce to , for parallel configuration : @xmath128 for anti - parallel configuration : @xmath129 and in the limit @xmath130 for parallel and anti - parallel configurations : @xmath131 for non - collinear configurations there is no simple circuit analogy , but we can still find an analytical expression for the total conductance of the system as a function of the angle between the magnetizations of the different ferromagnets @xmath132 , when @xmath116 , at zero magnetic field ( @xmath133 ) and for symmetric contacts : @xmath134 in the limit of @xmath135 and @xmath136 , eq .
( [ g(alpha ) ] ) simplifies to eqs .
( [ gspinp ] ) and ( [ gspinap ] ) respectively .
when the resistance of the normal metal is negligible compared to the contacts resistance ( @xmath137 ) , this reduces to @xmath138 where @xmath139 is the polarization and @xmath140 is the ( complex ) relative mixing conductance .
( [ g_arne ] ) can also be obtained by means of the circuit theory.@xcite
the total conductance depends on the spin - dependent conductances of the contacts .
we mostly set the polarization @xmath141 ( for real metallic ferromagnets like fe or co , @xmath142 is 0.4 and 0.35 respectively@xcite ) , which corresponds to a ratio @xmath143 . on the other hand ,
the real part of the mixing conductance obeys @xmath144 .
@xcite the conductances of the contacts and the diffusive normal metal are considered to be of the same order @xmath145 the total conductance depends on the magnetic configuration .
we plot in fig . 4a and fig .
4b , @xmath146 as function of the relative angle between magnetizations @xmath132 , for symmetric contacts , zero magnetic field ( @xmath147 ) and in the absence of spin - relaxation in the normal metal @xmath148 , as given by eq .
( [ g(alpha ) ] ) for different values of @xmath149 and @xmath150 respectively . for @xmath151{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion } ,
$ ] @xmath152{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion } $ ] and @xmath153{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion } $ ] the total conductance does not depend on the mixing conductance and the values of @xmath146 at @xmath151{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion } , 360^{% % tcimacro{\unicode[m]{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion } $ ] and @xmath154 @xmath155{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion } $ ] are given by eq .
( [ gspinp ] ) and eq .
( [ gspinap ] ) respectively . on the other hand , for non - collinear configurations ,
the total conductance increases with increasing mixing conductance ( the dip become more sharp ) .
this enhancement is due to the contributions of non - collinear spins to the transport , in which electrons with spins oriented in different directions than the magnetization of the adjacent ferromagnet are transmitted or reflected at the contact .
these processes are described by the real and the imaginary part of the mixing conductance .
when spin - flip scattering is caused by spin - orbit interaction in the normal metal , the spin diffusion length @xmath5 can be estimated to be equal to @xmath156 , where @xmath157 is the relativistic fine structure constant , @xmath158 is the atomic number and @xmath4 is the mean free path ( see _ e.g. _ ref . ) . in co
/ cu multilayers , the spin diffusion length @xmath5 is of the order of a few hundred angstrom ( see appendix a in ref . ) . for al ,
@xmath5 can be estimated to be of the order of a few micrometers for polycrystalline al ( see ref . ) , or even between @xmath159 for al - single crystals .
@xcite in the case of very pure na , @xmath160.@xcite in this case , @xmath5 limited by spin - orbit interactions can be estimated to be of the order of @xmath161 @xmath162 . in fig .
5a we plot , at zero magnetic field and for symmetric contacts , the conductance of the system @xmath163 , normalized to the conductance @xmath164 given by eq .
( [ goap ] ) , as a function of @xmath124 .
the length of the normal metal section @xmath2 is set to be constant , in order to keep a constant value of @xmath165 .
when @xmath166 the conductance of the system depends on the magnetic configuration . by decreasing @xmath5 ,
all configurations converge to the same value of conductance @xmath167 .
all configurations reach the same value of the conductance long before @xmath167 , since for @xmath168 both contacts become independent and as the relative magnetic configuration is irrelevant . in fig .
5b we plot @xmath169 in the case of antiparallel configuration for differents values of the relative polarization @xmath170 @xmath171 and for @xmath172 constant .
when @xmath116 the configuration with large relative polarization @xmath173 gives a small conductance and _ vice versa_. the spin accumulation increases with increasing polarization of the ferromagnet and causes a reduction of the total conductance of the system .
for @xmath120 we also see that in each case the conductance approaches @xmath164 asymptotically in different ways , depending on the magnitude of the spin accumulation . in a diffusive system
the presence of an external magnetic field relaxes the spin accumulation , in addition to the usual precession of the spin .
semiclassically , the spin accumulation at a certain position @xmath174 is the average contribution of the spin of all electrons . in a diffusive metal
each electron diffuses along a random trajectory , while its spin precesses with frequency @xmath175 around the magnetic field .
since each trajectory has a different length , the spins of the electrons at a certain point @xmath174 are oriented in different directions , which in average relaxes the local spin polarization .
the length scale of both relaxation and precession processes is the _ precession _
length @xmath176 @xmath177 , where @xmath178 is the diffusion coefficient , @xmath179 is the magnetic field , @xmath34 is the bohr magneton and @xmath35 is the spin gyromagnetic ratio .
the external magnetic field may also influence the transport processes described by the mixing conductance @xmath0 at the contacts .
let us consider for simplicity that @xmath180 , and @xmath181 in this limit the distribution function of the normal metal does not depend on position .
from current conservation we have : @xmath182 where eq .
( [ invariant1 ] ) corresponds to the particle current , eq .
( [ invariant2 ] ) corresponds to the spin current and @xmath183 is the volume of the normal metal .
the current is defined to be positive when injected into the normal metal by the ferromagnetic reservoirs .
we can re - write ( [ invariant2 ] ) as @xmath184 where @xmath185 is the larmor frequency vector
. from this expression follows that the time scale relevant for @xmath175 is the escape time @xmath186 @xmath187 , where @xmath188 is the average contact conductance .
@xmath189 is the time in which an electron escapes from the normal metal into the ferromagnetic reservoirs .
it is also the time scale relevant for the precession of the electrons around the magnetic field . on the other hand ,
if @xmath190 , @xmath189 is of the order of the thouless time @xmath191 , which is the average time in which an electron passes through the diffusive normal metal .
when @xmath165 @xmath192 diffusion , precession and transmission or reflection at the contacts , happen on the same time scale . from these estimates
we see that the ballistic or diffusive nature of the normal metal is not going to change the effect of the magnetic field on the physics at the contacts .
the results obtained for @xmath193 , should therefore be valid when @xmath165 @xmath194 .
we now make a perturbation expansion in small magnetic fields ( see appendix a ) . to first order ,
the current depends on the expansion parameter @xmath179 as : @xmath195 where @xmath196 and @xmath197 are vectors associated with the spin current injected into the normal metal ( see appendix a ) and where the matrix @xmath198 describes the contacts and @xmath199 the magnetic field contribution .
as detailed in appendix a , @xmath200 has a symmetric part @xmath201 , which only includes three of the four contact conductances , _
i.e. , _ only _ _ the conductances _
_ @xmath202 of each contact ( @xmath77 and @xmath78 ) respectively . on the other hand
, @xmath200 has also an antisymmetric part which only depends on the imaginary part of the mixing conductance of each contact @xmath203 the matrix @xmath204 which describes the precession of spins due to the magnetic field , is also antisymmetric . using the symmetry properties of the matrices @xmath205 and @xmath199 we can determine from eq .
( [ finalexpans ] ) the symmetry properties of the total conductance of the system @xmath206 @xmath207 with respect to the magnetic field @xmath179 . when @xmath200 is a symmetric matrix , @xmath208 for small values of magnetic field .
the conductance of the system is then symmetric with respect to a change of sign of the magnetic field ( @xmath209 ) , _ i.e. _ , with respect to time - reversal . on the other hand ,
if @xmath200 is antisymmetric , @xmath210 and we can expect asymmetric behavior of the conductance with respect to change of sign of magnetic field . in the following ,
we discuss the dependence of the conductance on the magnetic field .
we obtain @xmath169 as a function of @xmath211 for different magnetic configurations , @xmath212 and @xmath213 constant . in fig .
6a we plot @xmath169 in the case of symmetric contacts , where the magnetic field is perpendicular to both magnetization directions @xmath214 . in this case
all injected spins precess around the magnetic field .
when @xmath215 , the spins injected from one ferromagnet are not strongly affected by the magnetic field , so they travel through the normal metal and reach the other ferromagnet without relaxation . as a result ,
the total conductance depends on the relative magnetic configuration . by decreasing @xmath216 , the spin accumulation precesses and relaxes on the scale of @xmath216 . due to the precession of spins
, the conductance displays in general a non - monotonic behavior with @xmath217 .
this modulation of the conductance can be understood in terms of the `` matching '' of the spins at the contacts after precession . according to the values of the contact conductances for the different magnetic configurations the spins
are reflected or transmitted at the contacts depending on its orientation . concerning the relaxation , the configuration with more spin accumulation ( in this case , the antiparallel configuration )
is the most sensitive to the magnetic field ( increases faster than the other ones ) , since there are more spins to be rotated by the magnetic field in this configuration than in others . in fig .
6a , the conductance of @xmath153{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion } $ ] ( antiparallel ) and @xmath218{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion } $ ] configurations cross the conductance for @xmath151{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion } $ ] ( parallel ) around @xmath219 .
that means that at this point the spins accumulated in these two configurations have been reduced to the value of spin accumulation of @xmath151{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion } $ ] configuration .
after the point @xmath219 , the parallel configuration ( @xmath220{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion } $ ] ) gives a smaller conductance and the antiparallel configuration ( @xmath221{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion } $ ] ) gives the highest conductance . as a result , for @xmath222 , the parallel configuration is more sensitive to the magnetic field than the antiparallel configuration ( now the one which increases faster ) .
the relaxation of spins via the precession around the magnetic field depends on the amount of spin accumulation in the system .
this non - mononotic behavior of the conductance is specially relevant between parallel and antiparallel configurations , because the difference between the conductance of both configurations @xmath223 can be modulated from positive to negative values by the external magnetic field . when @xmath214 , according to eq.([buvw ] ) in appendix a , @xmath37 has only one component @xmath224
@xmath225 @xmath226 ( see ref . ) .
moreover , the spins are injected with directions along @xmath227 and @xmath228 , so the precession due to the magnetic field only switches the spin directions between @xmath227 and @xmath228 . as a result , the distribution function given by ( [ expansionuvw ] ) , has only two components @xmath229 in this particular case , @xmath200 reduces to @xmath201 , which is a symmetric matrix .
the same holds for the matrix @xmath199 * , * which reduces to its @xmath230 upper box , which only includes @xmath231 ( see appendix a ) . as @xmath198 reduces to @xmath232
we expect @xmath233 .
6b shows the dependence of @xmath169 on @xmath234 for different magnetic configurations , where @xmath235 is the scale of magnetic fields relevant for precession in a diffusive medium .
as expected , all configurations are symmetric with respect to a change of sign in magnetic field ( @xmath236 ) .
now we want to investigate the role of @xmath149 . to this end , the magnetic field
is assumed to be oriented perpendicular to both magnetizations when the system is in collinear configurations , and parallel to the direction of one of the magnetizations when the system is in the @xmath237{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion $ ] configuration .
according to eq.([buvw ] ) in appendix a , for @xmath237{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion $ ] , the magnetic field is along @xmath238 and as a result from the injection and precession , there are spins in the three directions @xmath239 _ i.e. _ , the precession of spins around the magnetic field induces spins along the perpendicular direction @xmath240 to the injection orientations @xmath227 and @xmath241 .
@xmath200 is then an antisymmetric matrix , due to the contributions of the terms which includes @xmath242 so for @xmath243 @xmath244 which means asymmetric behavior of the conductance with respect to time reversal . on the other hand ,
if we put @xmath245 , @xmath200 is symmetric and @xmath246 in fig .
7a we obtain @xmath169 _ vs _ @xmath217 , for the same set of parameters as in fig .
6a . for parallel and antiparallel configurations ,
the results are not modified compared to fig .
however , for @xmath237{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion $ ] the relative conductance @xmath169
does not approach unity asymptotically . in this configuration
there are some injected spins , which are parallel to the magnetic field and which do not precess at all .
so this part of the spin accumulation remains in the system and does not relax irrespective of the values of the magnetic field .
more interesting is the appearence of a dip in the conductance for small values of the magnetic field .
if we repeat the calculation of @xmath169 vs @xmath217 for the same set of parameters except for @xmath247 we see that the dip disappears ( fig .
7b ) , so according to our discussion , it is related with asymmetric properties of the conductance . in fig .
7c , we plot @xmath169 vs @xmath248 .
as we expect , @xmath237{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion $ ] configuration presents asymmetric behavior respect time reversal , whereas both parallel and antiparallel configurations remain symmetric ( fig.7c ) . in particular the @xmath237{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion $ ] conductance is _ antisymmetric _ with respect to time reversal , for small values of magnetic field . from this discussion
, we understand that the real part of the mixing conductance describes processes at the contacts in which spins perpendicular to the magnetization direction , are transmitted or reflected obeying time - reversal symmetry . on the other hand ,
the imaginary part of the mixing conductance describes processes in which the spins precess around the magnetization vector of the ferromagnet . as a result of the precession , the orientation of the spin changes .
the latter processes are _ antisymmetric _ with respect to time - reversal .
spin - flip scattering causes relaxation of the spin accumulation in the normal metal and as a result , supression of the spin - dependent properties on the system .
now we want to investigate how the spin - flip affect the magnetic field effects show above .
the existence of spin - flip scattering reduces @xmath14 .
if @xmath249 there is no strong spin - flip scattering in the system and it is possible to observe spin - dependent effects . on the other hand , if @xmath120 , the injected spins relax very fast due to spin - flip processes and no spin - dependent effects can be observed . in particular in fig .
8 we show how the dip of @xmath237{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion $ ] configuration from figs . 7a and 7c , is suppressed by spin - flip scattering on the system . also by decreasing @xmath250 @xmath169 increases for constant magnetic field , to the value @xmath251 .
that simply means , that the spin accumulation relaxes due to spin - flip scattering , as is expected .
in this paper , the normal metal in our f - n - f device is considered three dimensional , but can also be two dimensional ( 2d ) , _
e.g. _ , a two dimensional electron gas ( 2deg ) attached to ferromagnetic reservoirs , or even one dimensional ( 1d ) , if the normal metal is a quantum wire or a carbon nanotube.@xcite in this case electron - electron interaction should be taken into account.@xcite the non - magnetic material can also be a semiconductor , as shown in recent spin - injection experiments.@xcite in the case of a 2deg attached to metallic ferromagnets , the large difference between the conductivities of the 2deg and the ferromagnetic reservoirs suppresses the spin - injection via metallic contacts . for a significant spin - injection into the 2deg , tunnel contacts , a semiconductor ferromagnet or a half - metallic ferromagnet are required.@xcite in this paper
we have shown how the spin - dependent transport through a f - n - f double heterojunction can be described in terms of the spin dependent conductances of the contacts @xmath252 the magnetization direction @xmath253 of the ferromagnetic reservoirs , and the normal metal conductance @xmath254 the dependence of the conductance on the relative angle between the magnetizations of the different ferromagnets is affected by the mixing conductance @xmath0 . for non - collinear transport between the ferromagnetic reservoirs , @xmath87 describes transport of spins perpendicular to the magnetization direction of the ferromagnets .
these processes enhaces the conductance for non - collinear configurations , which may be used in multiterminal devices for modulation of the transport properties .
@xcite this modulation could be useful for future applications as spin - dependent transistors .
we find that spin injection can be symmetric and antisymmetric with respect to time - reversal .
the symmetric processes are described by @xmath150 and the antisymmetric ones are described by @xmath149 .
it is interesting to observe that the antisymmetric processes described by @xmath149 correspond to spin precession around the magnetization vector of the ferromagnet which couples to an external magnetic field . in a diffusive system ,
an applied magnetic field produces both precession and relaxation of the spin accumulation .
the conductance displays a non - monotonic behavior on the scale of the _ precession length _
@xmath216 , which is the distance for the precession of the spin around the magnetic field in the normal metal . due to this modulation ,
the difference between the conductances of the parallel and antiparallel configurations @xmath255 can be positive and negative as a function of the magnetic field . a possible candidate to observe this effect is al , which has a large @xmath5 and which can be coupled to ferromagnetic reservoirs ( _ e.g. _ , fe , cofe , nife , co , .. ) via metallic junctions or also al@xmath256o@xmath257 tunnel junctions@xmath22 let us estimate the values of magnetic fields for al - single crystal associated with the points @xmath258 and @xmath259 of fig .
6a , where @xmath223 is positive and negative respectively .
if the length of the system is @xmath260 which is comparable with the spin diffusion length @xmath261@xcite we obtain for @xmath262 @xmath263 and for @xmath264 .
if @xmath265 @xmath266 we obtain for @xmath267 and for @xmath268 . in both cases
we see that the values of magnetic field in the case of al - single crystal are quite reasonable and also that the change from positive to negative @xmath223 can be achieved by an increase of the magnetic field by one order of magnitude .
the same estimate for polycrystalline al gives us for @xmath269@xcite the values of @xmath270 and @xmath271 respectively .
these fields are much higher than the typical switching field for a ferromagnet , so polycrystalline al does not appear to be a good candidate . for very pure na , if @xmath272 the corresponding values of magnetic fields are @xmath273 and @xmath274 respectively .
this modulation of @xmath223 by a magnetic field can also be explored in semiconductors ( sc ) 2deg , as _
e.g. _ gaas and inas . when @xmath275 the following expression holds for the magnetic field corresponding to @xmath276 which depends on the spin - flip time @xmath21 and on the gyromagnetic ratio @xmath35 of the semiconductor material . for sc
, @xmath35 depends strongly on the material @xmath277 so depending on the values of @xmath20 , one can obtain the corresponding values of magnetic field .
kikkawa and awschalom@xcite report @xmath278 in n - type gaas system , but this values corresponds to spin lifetimes of optically pumped carriers , and not to the usual carriers relevant for transport .
the corresponding value for the magnetic field for this case is @xmath279 @xmath280 .
on the other hand , we are not aware of reliable values of @xmath20 for transport in these systems . in conclusion , from our estimates of the relevant values of magnetic fields , al - single crystals with ferromagnetic contacts are good candidates to test our predictions and possibly lead to the discovery of other new physical phenomena of spin - transport .
* acknowledgments * this work is part of the research program for the `` stichting voor fundamenteel onderzoek der materie '' ( fom).we acknowledge support from the nedo joint research program ( ntdp-98 ) . it is a pleasure to acknowledge useful discussions with w. belzig , d. pfannkuche and y. tokura .
eqs . ( [ invariant1 ] ) and ( [ invariant2 ] ) can be written as follows : @xmath281 where @xmath288 and where @xmath289 .
we can also express the magnetic field in this basis as @xmath290 in terms of this expansion , we can combine ( [ eqinv1 ] ) and ( [ eqinv2 ] ) into a compact matrix form as @xmath293 \begin{array}{c } \sqrt{\frac{1+m}{2}}\frac{p_{{\cal l}}+p_{{\cal r}}}{2 } \end{array } & \frac{(g_{{\cal l}}+g_{{\cal r}})\left ( 1+m\right ) } { 4}+\frac{\left ( % tcimacro{\func{re}}% % beginexpansion \mathop{\rm re}% % endexpansion g_{{\cal l}}^{\uparrow \downarrow } + % % tcimacro{\func{re}}% % beginexpansion \mathop{\rm re}% % endexpansion g_{{\cal r}}^{\uparrow \downarrow } \right ) ( 1-m)}{2}\text { } & \begin{array}{c } \text { } \left ( \frac{g_{{\cal l}}-g_{{\cal r}}}{2}-% % tcimacro{\func{re}}% % beginexpansion \mathop{\rm re}% % endexpansion g_{{\cal l}}^{\uparrow \downarrow } + % % tcimacro{\func{re}}% % beginexpansion \mathop{\rm re}% % endexpansion g_{{\cal r}}^{\uparrow \downarrow } \right ) \left ( \frac{\sqrt{1-m^{2}}}{2}% \right ) \end{array } \\[3ex ] \begin{array}{c } \sqrt{\frac{1-m}{2}}\frac{p_{{\cal l}}-p_{{\cal r}}}{2 } \end{array } & \left ( \frac{g_{{\cal l}}-g_{{\cal r}}}{2}-% % tcimacro{\func{re}}% % beginexpansion \mathop{\rm re}% % endexpansion g_{{\cal l}}^{\uparrow \downarrow } + % % tcimacro{\func{re}}% % beginexpansion \mathop{\rm re}% % endexpansion g_{{\cal r}}^{\uparrow \downarrow } \right ) \left ( \frac{\sqrt{1-m^{2}}}{2}% \right ) & \frac{(g_{{\cal l}}+g_{{\cal r}})\left ( 1-m\right ) } { 4}+\frac{% \left ( % tcimacro{\func{re}}% % beginexpansion \mathop{\rm re}% % endexpansion g_{{\cal l}}^{\uparrow \downarrow } + % % tcimacro{\func{re}}% % beginexpansion \mathop{\rm re}% % endexpansion g_{{\cal r}}^{\uparrow \downarrow } \right ) ( 1+m)}{2 } \end{array } \right ) \label{smo}\ ] ] by a perturbation expansion in small magnetic fields , we may study how the magnetic field is coupled with the physics at the contacts . to zeroth order in magnetic field we simply have @xmath297 to first order the total particle current in the system
is given by @xmath299 where @xmath300 the dependence of the current on the expansion parameter @xmath179 is given by @xmath301 which to first order reduces to @xmath302 m. n. baibich , j. m. broto , a. fert , f.nguyen van dau , f.petroff , p. etienne , g. creuzet , a. friederich , and j. chazelas , phys .
* 61 , * 2472 ( 1988 ) ; g. binach , p. grunberg , f. saurenbach , and w.zinn , phys .
rev b * 39 , * 4828 * ( * 1989 * ) . * r.e .
camley and j. barnas , phys .
lett . * 63 * , 664 ( 1989 ) ; m.levy , s. zhang , and a.fert , phys . rev . lett . * 65 * , 1643 ( 1990 ) ; s. f. lee , w. p. pratt , q. yang , p. holody , r. loloee , p.a .
schroeder , and j. bass , j. magn .
magn . matter . *
118 * , 1 ( 1993 ) ; p.m. levy , sol .
. phys . * 47 * , 367 ( 1994 ) ; m. a. m. gijs and g.e.w .
bauer , adv .
phys . * 46 * , 285 ( 1997 ) .
j. c. slonczewski , phys .
b * 39 * , 6995 ( 1989 ) ; v. v. ustinov and e. a. kravtsov , j. phys . : cond . matter * 7 * , 3471 ( 1995 ) ; h. e. camblong , p.m. levy and s. zang , phys . rev .
b * 51 * , 16052 ( 1995 ) , a. vedyayev , n. ryzhanova , b. dieny , p. dauguet , p. gandit and j. chaussy , phys . rev .
b * 55 * , 3728 ( 1997 ) .
j. barnas and a. fert , phys .
lett . * 80 * , 1058 ( 1998 ) ; s. takahashi and s. maekawa , phys .
lett . * 80 * , 1758 ( 1998 ) ; a. brataas , yu . v. nazarov , j. inoue , and g.e.w .
bauer , phys .
b * 59 * , 93 ( 1999 ) ; x. h. wang and a. brataas , phys . rev . lett .
* 83 * , 5138 ( 1999 ) ; j. barnas and a. fert , j. magn .
magn . matter .
* 192 , * 391 ( 1999 ) ; a. brataas , yu . v. nazarov , j. inoue , and g.e.w .
bauer , eur .
j. b * 9 * , 421 ( 1999 ) .
the energy dependence of @xmath112 is only relevant for energy scales of the order of the fermi energy e@xmath303 which is much higher than the revelant energy scale for transport , so we can consider @xmath112 as energy independent .
m.oestreich , j. hubner , d. hagele , p. j. klar , w. heimbrodt , w. w. ruhle , d.e .
ashenford and b. lunn , appl .
* 74 , * 1251 ( 1999 ) ; r. fiederling , m. keim , g. reuscher , w. ossau , g. schmidt , a. waag and l. w. molenkamp , nature * 402 * , 787 ( 1999 ) ; y. ohno , d. k. young , b. beschoten , f. matsukura , h. ohno and d.d .
awschalom , nature * 402 * , 790 ( 1999 ) . * fig .
1 : * ferromagnetic reservoirs attached to diffusive normal metal through arbitrary contacts . arbitrary but fixed magnetization directions of the ferromagnetic reservoirs and orientations of a magnetic field applied to the normal metal are taken into account .
the contacts are described by the spin - dependent conductances @xmath307 and the normal metal is characterized by the normal metal conductance @xmath165 .
the ferromagnetic reservoirs are supposed to be large enough and in local equilibrium . *
fig . 2 : * current conservation imposes boundary conditions on the system .
the current through the contacts @xmath90 is equal to the current into the normal metal @xmath95 at each contact .
@xmath308 depends on the contact conductances and on the direction of the magnetization of the adjacent ferromagnet reservoir and @xmath309 is the current for the normal metal . * fig . 3 : * equivalent circuits for parallel and antiparallel configurations in the limits @xmath116 and @xmath120 .
( a ) and ( b ) correspond to the parallel and anti - parallel configuration respectively , when @xmath116 . in this limit ,
the two spin channels are independent and there is no mixing between them .
( c ) parallel and anti - parallel configurations when @xmath120 .
in this case , there is complete mixing between spin - up ( @xmath117 ) and spin - down ( @xmath10 ) channels and the spin accumulation vanishes . *
4 : mixing conductance * : dependence of @xmath146 on the relative angle @xmath132 between the magnetizations of the ferromagnetic reservoirs , for symmetric contacts , zero magnetic field and in the absence of spin - flip scattering . ( a ) the following set of parameters is chosen : @xmath310 and @xmath311 takes values 0.0 , 0.5 , 1.0 , 2.0 and 10.0 corresponding to the different plotted lines .
( b ) in this case , @xmath312 and @xmath313 changes with values 0.7 , 1.0 , 2.0 and 10.0 . according to the condition
@xmath314 @xmath313 can not be , smaller than 0.65 .
* fig . 5 : * * effect of spin - flip scattering on the system * : for symmetric contacts and zero magnetic field .
( a ) @xmath112 normalized to @xmath164 as a function of @xmath124 , for the following set of parameters : @xmath315 @xmath316 @xmath317 @xmath318 .
( b ) @xmath169 versus @xmath124 in the case of antiparallel configuration , for different values of the relative polarization @xmath319
@xmath320 @xmath321 and for @xmath322 constant .
6 : magnetic field dependence in the absence of spin - flip scattering * : we consider symmetric contacts and the following set of parameters : @xmath315 @xmath316 @xmath317 @xmath323
. moreover the magnetic field is always perpendicular to both magnetizations directions @xmath324 .
( a ) @xmath169 as a function of @xmath217 .
( b ) @xmath325 versus @xmath248 . * fig . 7 : magnetic field dependence in the absence of spin - flip scattering * : we consider @xmath326 , @xmath327{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion , 180% % tcimacro{\unicode[m]{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion } = 0 $ ] and @xmath37 @xmath328 @xmath329{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion } $ ] ( or @xmath330 , @xmath331{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion , 180% % tcimacro{\unicode[m]{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion } = 0 $ ] and @xmath332 @xmath328 @xmath333{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion } $ ] ) .
( a ) @xmath169 vs @xmath334 for symmetric contacts and the following set of parameters : @xmath335 @xmath336 @xmath317 @xmath323.(b ) same as ( a ) but for @xmath337 ( c ) @xmath169 versus @xmath248 , for @xmath335 @xmath316 @xmath317 @xmath338 * fig . 8 : magnetic field dependence and spin - flip scattering * : conductance for @xmath218{0xb0}}% % beginexpansion { { } ^\circ}% % endexpansion } $ ] configuration of fig
. 7a , for different ratios @xmath339 0.1 , 0.3 , 1 , 3 , 10 , 100 . | we study diffusive transport through ferromagnet - normal metal - ferromagnet ( f - n - f ) systems , with arbitrary but fixed magnetization directions of the ferromagnetic reservoirs and orientations of a magnetic field applied to the normal metal . for non - collinear configurations , the complex mixing conductance @xmath0 describes the transport of spins non - collinear to the magnetizations of the ferromagnetic reservoirs .
when @xmath1 , the total conductance of the system in the presence of a magnetic field can be asymmetric with respect to time reversal .
the total conductance changes non - monotonically with the magnetic field strength for different magnetic configurations .
this modulation of the conductance is due to the precession of the spin accumulation in the normal metal .
the difference between the conductance of the parallel and antiparallel configurations can be either positive or negative as a function of the applied magnetic field .
this effect should be best observable on al single crystals attached to ferromagnetic electrodes by means of tunnel junctions or metallic contacts . |
phase ordering @xcite is one of the important topics in non - equilibrium statistical mechanics . for systems with two coexisting phases the situation is generally well understood from the analytic and numeric points of view @xcite . for the potts model @xcite with a @xmath3-degenerated ground state ,
the situation is not so clear in general .
also in this case , an allen - cahn power - law regime @xcite and dynamical scaling relations for the structure factor and for the distribution of domain sizes @xcite have been predicted @xcite and found numerically @xcite . on the other hand ,
a singular behaviour of the potts model has been recently observed by de oliveira et .
@xcite when the degeneracy ( @xmath0 ) is large enough : in the thermodynamic limit the model was shown to relax towards a ` glassy ' , disordered , phase with a non negligible density of defects when it is quenched at zero temperature .
this paper is devoted to discuss some ideas for understanding this glassy phase . in the next section we recall the phase and
report new results on the ordering of the potts model after a quench at low but finite temperature .
the system equilibrates locally nucleating domains that eventually become of the size of system . in section 3
we study the @xmath4 case , and we interpret the impossibility to equilibrate in the thermodynamic limit as an impossibility of breaking the local zero magnetisation .
section 4 is to conclude and summarise the main results .
de oliveira et .
@xcite have recently observed an interesting slow relaxation in the dynamics of the potts model after a quench from a disordered state to zero temperature , and the impossibility for the system to achieve the ground state in the @xmath5 limit , being @xmath6 the linear size of the system .
let us briefly describe this effect . given a lattice @xmath7 , in which each site
@xmath8 can take @xmath0 equivalent values , or _
colours _ , the potts model @xcite is defined by the hamiltonian : @xmath9 where @xmath10 means @xmath11 , @xmath12 is a neighbour of @xmath13 , and where @xmath14 is the colour of site @xmath13 .
when quenching at zero temperature a two - dimensional system on a square lattice using glauber single spin - flip dynamics , systems with @xmath2 obey the allen - cahn law @xcite @xmath15 ( @xmath16 being the energy per site ) up to a time @xmath17 , increasing with @xmath6 , when they get trapped in a blocked configuration , invariant with respect to the single spin - flip dynamics at zero temperature @xcite and it is negligible in the thermodynamic limit . ] .
inverting the usual order of the @xmath18 and @xmath5 limits , i.e. , assuming that the thermodynamic limit is taken before the infinite - time limit , and extrapolating to @xmath19 the @xmath20 data with the allen - cahn power - law @xmath15 @xcite , one finds a positive energy @xmath21 for @xmath2 and , in particular the data for @xmath22 is very well fitted by the expression @xmath23 , @xmath24 being real constants .
this would imply that an infinite - size system with @xmath25 relaxes after an infinite time towards a phase with stationary observables and positive energy , different from the ground state with zero energy , and this is the reason for which this out - of - equilibrium ` phase ' is called ` glassy phase ' in @xcite .
a noticeable point is that the onset of this glass - like phase occurs in the two - dimensional potts model for @xmath2 , and this @xmath0-range coincides with the one in which the system presents a first - order phase transition @xcite . a justification of this fact is in progress @xcite .
we now give a general characterisation of the equilibration of the system at finite temperature , and an interpretation of the non - equilibrium phase in terms of the symmetries of the problem in section 3 .
we have studied the dynamics after a quench at finite temperature , @xmath26 , of the @xmath27 colours potts model with periodic boundary conditions , being @xmath28 the critical temperature of the model @xcite . in fig .
1 we present the energy per site ( on top ) and the magnetisation ( centre ) with respect to @xmath29 of a 2-d square lattice system with nearest - neighbours interaction and periodic boundary conditions .
the magnetisation , @xmath30 , is @xmath31 where @xmath32 is the fraction of colour @xmath33 ,
@xmath34 , @xmath35 being the number of sites in the system with colour @xmath33 and @xmath36 is the total number of sites .
the magnetisation is zero when @xmath37 , and one when all spins have the same colour .
initially the system is in an uncorrelated configuration at infinite temperature , with @xmath38 and @xmath39 . for @xmath40 , systems with @xmath2 coarsen with equal proportion of all colours and with the expected power - law dependence of the energy on time @xcite while , for @xmath41 , finite - size systems equilibrate , breaking the @xmath39 symmetry and approaching gradually the ground state with @xmath42 , @xmath43 .
the above description of the problem suggests that , for @xmath2 and for times larger than @xmath44 , coalescence effects , not considered in the derivation of the allen - cahn law , become relevant in the dynamics , and allow the mean domain size , @xmath45 @xcite , to grow up to the system size , @xmath6 .
the numerical results reported in fig .
1 suggest that , for @xmath46 , @xmath44 is not divergent in @xmath6 , but constant above a certain @xmath6 , i.e. , that the `` nucleation '' needed to equilibrate the system is a local process no longer dependent on @xmath6 .
in fact , the @xmath47 and @xmath48 curves of fig . 1
coincide within @xmath49 , the variance of the distribution of energy values corresponding to different realisations of the quench ( shown in error bars ) .
we observe from fig .
1 that the @xmath39 symmetry is broken later in larger systems . as in @xcite for the continuous ising model , the magnetisation is supposed to be zero for all times in the absence of finite - size effects , and in fact we find @xcite a similar scaling relation @xmath50 for @xmath51 , and even a slower dependence of @xmath52 on @xmath6 for @xmath2 . one could ask why @xmath44 seems to approach a limit value with @xmath53 , while the magnetisation at fixed times tends to zero in this limit . to answer this question we propose the following argument : systems with @xmath2 equilibrate leaving at @xmath54 the power - law dependence of the energy on time and forming domains that will eventually become system - sized , and @xmath44 seems to be a characteristictime of the model .
it is this local nucleation the origin of the ordering , rather than the changes on the global colour fraction , which is a finite - size effect @xcite . in order to supply a quantitative support to this argument
we define an order parameter , @xmath55 , accounting for the spatial ordering , which seems to characterise the dynamical ordering more than the global magnetisation , @xmath30 @xcite .
let us define @xmath55 as a distance from the @xmath56 , @xmath57 , situation , where @xmath58 is the magnetisation of a cell centred in @xmath59 and of size @xmath60 , independent of @xmath6 , in such a way that @xmath61 in the thermodynamic limit . in particular , @xmath62 where @xmath63 is the volume of the general system and @xmath64 $ ] is the proportion of colour @xmath33 in the cell centered in @xmath59 .
the expression at the right of the arrow in ( 3 ) is the definition of @xmath55 in the lattice @xmath65 and @xmath66 is a cell centred in the position of site @xmath13 . , we have taken the nearest 24 neighbours of each site for the definition of the mentioned cell . ] for @xmath5 ,
@xmath55 so defined is zero in the completely uncorrelated configuration , and one in the ordered configuration , when all the sites have the same colour and @xmath42 .
since @xmath67 , @xmath55 is a distance , functional of @xmath58 , and can be used as an alternative order parameter accounting for the spatial ordering of the system .
we see in fig .
1 ( bottom ) that also @xmath55 presents a power - law dependence on time .
as expected by the argument exposed above , @xmath55 seems not to depend on @xmath6 for large @xmath6 : curves for @xmath48 and @xmath68 coincide within their standard deviation limits .
moreover , @xmath55 seems to leave the power - law regime and converge to 1 at @xmath44 .
at zero temperature , the situation is different : the above defined time , @xmath44 , needed to leave the allen - cahn law is divergent with @xmath6 , as described in @xcite , and an infinite - size system is always supposed to follow the allen - cahn power - law . in larger and larger systems , it is less and less likely to find a blocked configuration or a path in phase space to the ground state .
a measure of this fact can be seen in the value of the energy variance , @xmath49 , which decreases with @xmath6 ( see inset of fig .
this means that in a large system it is less likely to find a deviation from the power - law regime , in which blocked dynamics @xcite or breaking of the @xmath39 symmetry @xcite are not present .
this explains at a qualitative level the fact that in large systems the magnetisation is broken later ( and hence the system equilibrates later ) , but it does not address the fact that the energy of an infinite size system with @xmath2 converges to a positive value for @xmath19 . in other words , it does not explain why the term @xmath22 of the generalised allen - cahn law @xmath69 is different from zero for @xmath2 .
one could ask why systems with @xmath2 can not converge to a zero energy phase in the limit @xmath5 , even respecting the @xmath39 symmetry , as the @xmath70 , 3 cases do .
we discuss this point in the following .
numerical simulations show that in the latest stage of the coarsening at zero temperature , systems with @xmath70 , 3 and with large enough @xmath6 ( to respect the @xmath39 symmetry ) present final configurations formed by domains of the size of system ( fig .
2 , left ) and @xmath71 , the mean domain size , is proportional to @xmath6 when @xmath19 .
the energy , or the perimeter of the interface , is proportional to @xmath6 , and thus , the energy per site is zero in the thermodynamic limit .
in fact we have @xmath72 for @xmath70 , 3 , as said before . on the other hand , for @xmath2
, we observe that the system presents not only @xmath39 , but also a local symmetry that is a local equal fraction of all colours .
we define this local symmetry by introducing a certain scale , @xmath60 , with @xmath73 , such that the magnetisation is also zero in every cell of the system of size greater than @xmath60 and _ being @xmath60 such that @xmath61 for @xmath74_. we propose , and we argue below , that the zero temperature dynamics can not form domains of the size of the system for @xmath2 , hence the perimeter of the interface ( as well as the final number of domains ) grows with @xmath75 and the energy per site of the glassy phase is not zero in the thermodynamic limit , while , in this limit , systems with @xmath76 do not break the @xmath39 symmetry when forming domains of the size of the system but they can break the local @xmath39 symmetry .
a numerical confirmation of this argument is that @xmath55 converges to @xmath77 in the @xmath78 limits for @xmath76 , but @xmath79 for @xmath80 : the fraction of the volume @xmath63 for which @xmath81 is not negligible in the thermodynamic limit . in fig .
3 we report the energy , magnetisation and order parameter of local ordering as a function of @xmath29 for the @xmath4 dynamics in the @xmath80 , 3 and 2 cases .
it seems that @xmath55 presents the same power - law dependence on time as the energy in the @xmath80 case .
inverting the order of the limits and extrapolating to @xmath19 with a linear fit , as done for the energy in @xcite , we have @xmath82 , while for @xmath51 , @xmath55 converges to 1 . this local zero magnetisation for @xmath2 ( see fig 2 , right ) characterises the glassy phase also in the sense that it predicts a @xmath83 dependence of @xmath22 , which coincides in the @xmath84 limit with the result @xmath85 of @xcite .
let us show this fact in the approximation of domains of identical size @xmath45 @xcite .
in order to satisfy the local-@xmath39 condition there must be @xmath86 domains inside each cell of size @xmath60 , @xmath87 being an integer . in @xmath88 dimensions it is @xmath89 , where @xmath90 is the number of domains , and setting @xmath91 we have @xmath92 .
when @xmath0 approaches 4 , coalescence effects not considered in this argument become important , to which we attribute the slower dependence on @xmath0 of the energy , @xmath85 .
as we have seen there are evidences that , in absence of thermal fluctuations , the metropolis dynamics can not break the local symmetry of colours when @xmath2 .
this fact and the @xmath2 limit can be explained ( not yet in a rigorous fashion ) as a consequence of an assumption on the glassy phase symmetries . to introduce it
let us define the @xmath0 ` coarse - grained ' fields @xmath93 $ ] , in analogy with @xcite , as the fraction of colour @xmath33 in a cell centred in @xmath59 and of size @xmath60 .
clearly @xmath94 and the fields @xmath95 describe the configuration in a @xmath60-dependent way .
given the fact that the hamiltonian is invariant with respect to @xmath96 , the group of permutations of the @xmath0 colours , we assume that the @xmath39 symmetric configuration corresponding to @xmath74 , @xmath18 is maximally ordered in such a way that any permutation of the colours , @xmath97 , is equivalent to a @xmath98-dependent global spatial transformation , @xmath99 .
i.e. : @xmath100 if we assume the operator @xmath101 to be linear we have that @xmath102 is a representation of @xmath96 in the vector space of linear operators in the plane , @xmath103 . since there is no representation of @xmath96 in @xmath104 for @xmath2 different from the trivial representation ( @xmath105 , the identity operator on @xmath106 , @xmath107 ) , from the above assumption it follows the local - zero magnetisation of the glassy phase for @xmath2 : setting @xmath105 in ( 4 ) we obtain @xmath108 , that with the normalisation condition gives @xmath109 @xmath110 , i.e. , the local @xmath39 symmetry in the cell of size @xmath60 .
the assumption ( 4 ) can also ` justify ' the spatial symmetries of configurations with @xmath111 in the @xmath39 regime : for @xmath70 the group @xmath112 admits a nontrivial representation in the plane , @xmath113 , where @xmath114 is the identity and @xmath115 is the rotation of @xmath116 radians in the polar angle and , in fact , configurations with @xmath70 in the glassy phase seem to present this symmetry ( see fig .
2 , left ) , and a similar @xmath117 rotational symmetry seem to exist for @xmath118 configurations , even if it is often hindered by the two - preferred directions of the square lattice interaction @xcite and by the formation of structures locally stable under the @xmath4 dynamics @xcite .
moreover , in a general @xmath88-dimensional system , there exist no nontrivial representations of @xmath96 in @xmath119 for @xmath120 @xcite .
this proposition seems to be true for all @xmath2 , and we are working on the general proof .
if verified , this consequence of the assumption ( 4 ) would coincide with the result by lifshitz @xcite , who argued that a @xmath88-dimensional system quenched below its critical point does not necessarily equilibrate into an ordered state , in the presence of a ground state that is degenerated more than @xmath121 times .
we have studied the ordering dynamics of the 2-d potts model in a square lattice with @xmath51 and 7 by use of single - spin flip dynamic monte - carlo simulations . at positive temperature , @xmath26 , and for @xmath80
the system relaxes leaving the allen - cahn power law at a given time , which is independent on @xmath6 for large @xmath6 . at zero temperature , in the @xmath5 limit ,
systems with @xmath2 are not able to nucleate breaking the local symmetry , and hence they converge to a phase with nonzero energy .
we give a quantitative support for this argument in the @xmath80 case with the help of an order parameter of local ordering that we define .
the @xmath2 limit is presented as a consequence of a hypothesis on the symmetries of the glassy phase . for both temperatures the magnetisation at equal times decreases with increasing sizes and
it is argued to be zero in the thermodynamic limit at any time @xcite .
a generalisation of this study for different @xmath0 values and for larger @xmath6 is being performed @xcite together with an investigation of the relation between the ordering properties and the phase space structure .
+ the global magnetisation can be used to characterise the equilibrium of the potts model , e.g. in a variational mean - field approximation in terms of @xmath30 in : p. m. chaikin , p. lubensky , _ principles of condensed matter physics _ , cambridge university press , cambridge , 1995 . ) , magnetisation ( @xmath30 ) and order parameter of local ordering ( @xmath55 ) with respect to @xmath29 for the potts model with @xmath80 after a quench to temperature @xmath26 .
results are averaged over 200 , 60 and 30 realisations ( for @xmath122 and @xmath68 , respectively ) .
time is in mcs units .
error bars in the @xmath123-axes are the variances , @xmath124 , of @xmath16 and @xmath55 , corresponding to the average over different realisations of the quench ( shown every 2000 mcs only ) . in larger systems
the @xmath39 symmetry is broken later @xcite .
@xmath124 decreases with the size of the system at equal times .
the blue line is a fit to the @xmath47 system energy in the range @xmath125 $ ] , and the vertical dotted line is an estimation of the nucleation point , corresponding to the time at which @xmath16 differs from the fit more than @xmath126 . at that time , @xmath44 , the system nucleates leaving the allen - cahn power law , and @xmath55 increases towards 1.,title="fig:",height=321 ] + ) , magnetisation ( @xmath30 ) and order parameter of local ordering ( @xmath55 ) with respect to @xmath29 for the potts model after a quench at zero temperature .
the @xmath80 , 3 and 2 cases are presented , and three different sizes for the @xmath80 case .
magnitudes are an average over 240 , 60 and 40 realisations of the quench in systems with @xmath127 , 500 and @xmath68 , respectively . for the cases @xmath70 and 3 the energy converges to zero following the allen - cahn law , while in the @xmath80 case it converges to @xmath128 ( @xcite ) .
the global magnetisation decreases with @xmath0 at equal times and sizes , for all sizes and times studied .
it also decreases with @xmath6 for all times @xcite , and it is supposed to be zero in the thermodynamic limit . even with this constraint , systems with @xmath1 can form system - sized domains with zero energy , and in fact they order spatially , and @xmath129 , while the zero - temperature dynamics can not break the local zero magnetisation symmetry to form domains of the size of the system when @xmath2 , as illustrated by the fact that @xmath130 for @xmath80 , and this is the reason for which @xmath131 .
the extrapolated line is a fit to the @xmath47 data for @xmath132 mcs@xmath133 .
the fit coincides with the data within its variance limits .
the equation of the fit is @xmath134 with @xmath135 and @xmath136 .
on inset we present the variance @xmath137 of the energy , which is a decreasing quantity with @xmath6 , for all times and for the three sizes studied.,title="fig:",height=321 ] + | we have studied the ordering of the @xmath0-colours potts model in two dimensions on a square lattice . on the basis of our observations
we propose that if @xmath0 is large enough the system is not able to break global and local null magnetisation symmetries at zero temperature : when @xmath1 the system forms domains with a size proportional to the system size while for @xmath2 it relaxes towards a non - equilibrium phase with energy larger than the ground state energy , in agreement with the previous findings of de oliveira et al .
@xcite . |
cyclic phenomena play a very important role in different classes of processes in nature , particularly in epidemiological and ecological systems . in the epidemiological context , examples are diseases which do not leave us with permanent immunity @xcite . in ecology , cases when three variants of a species compete with one - another in a cyclic fashion have been observed @xcite .
another system of interest are cyclic food webs .
recently , kerr et al . obtained laboratory results for a system of three competing species that play rock - paper - scissors , when placed in a lattice - like spatial structure , and showed that there is agreement of the simulation results to the laboratory ones @xcite .
its alternative , as far as evolutionary processes are concerned , is the famous kimura - weiss model of neutral genetic drift @xcite . in a previous article @xcite
we have considered a non - spatial version of the @xmath0 model in both the cyclic competition and neutral genetic drift versions , and studied the evolutionary time behaviour of such a model .
the number of the @xmath16 species oscillates with an amplitude that drifts with time , until one of the species ( and then the second one ) goes extinct , i.e. biodiversity is lost .
the number of survivors vs. time plots show an exponential decay .
it is very interesting to note that there is no difference in the extinction time scale for the ensemble of competing species and the case of neutral drift .
this allows us to think of the neutral model as an `` adiabatic approximation '' of the cyclic system .
considering that the cyclic competition system would be a minimal model ( three alleles ) of darwinian evolution picture , it seems that , by just looking at the results , we can not tell which one must have been the mechanism for the evolution ! there is growing concern about the effects of habitat fragmentation in the survival of a species @xcite .
small , isolated forest fragments lose species we have many examples in the case of fast deforestation worldwide @xcite .
some species were lost within a few years of their isolation from the once - continuous forest @xcite .
oscillations in the number of population of one `` species '' are observed in the case of epidemics , as well as ecology . throughout the past century
we have observed changes in patterns of epidemics @xcite .
for some diseases , the major transitions have been between regular cycles and irregular epidemics , and from regionally synchronized oscillations to complex , spatially incoherent epidemics .
sinervo et al .
@xcite observed spatio - temporal oscillations in the number of lizards ( male and female ) that employ different mating strategies .
many population ecologists then conclude that a very important issue is the synchrony of population dynamics in different habitat patches @xcite .
if the population of a certain species goes extinct in one patch while it still survives in other patches , then the hope is that what is known as `` rescue effect '' can prevent global extinction @xcite .
otherwise , the population ecologists say , if there is synchrony of population dynamics between patches , the species is doomed to go extinct altogether . in this framework , there have been many proposed strategies for preservation , one of the most debated of which is the construction of `` conservation corridors '' that would make it possible for individuals to move along habitat patches @xcite .
considering that each habitat patch can be thought of as one copy of our non - spatial ensemble , it seemed reasonable to study the behaviour of the three - species cyclic system in the presence of mutations or migration . in the following section we introduce our model , and in the next ones
, we present the results of our calculations and simulations .
consider a non - spatial ( well - stirred ) system in which three species @xmath17 , @xmath18 , @xmath19 are competing in a way described by the reactions : @xmath20 at a rate @xmath21 , @xmath2 , @xmath7 at corresponding rates .
add to it the reactions : @xmath22 at rate @xmath23 , @xmath24 at rate @xmath23 , @xmath25 at rate @xmath26 , @xmath27 at rate @xmath26 , @xmath28 at rate @xmath29 , and @xmath30 at rate @xmath29 , where @xmath11 is the probability of mutations per individual particle in unit time .
the rate equations for this system will be : @xmath31 with @xmath32 const .
if the neutral drift system is considered instead , i.e. when @xmath33 or @xmath6 with equal probability ( and similarly for the two other reactions ) , the rate equations will only contain the terms that depend on @xmath11 , and the first two terms will be absent . in another version ,
mutations are replaced by migration into and out of the `` island '' . in other words
, the following migrations are happening : @xmath17 out at rate @xmath34 , @xmath18 out at rate @xmath35 , @xmath19 out at rate @xmath36 , @xmath17 , @xmath18 , or @xmath19 in at constant rate @xmath37 , where @xmath10 is the system size ( to make it comparable to the rate of leaving the island .
this would correspond to the average number of individuals in other patches , which are the source of our `` immigrants '' . )
the rate equations will be the same as those for the system with mutations .
the above equations have a fixed point at @xmath38 , and it is a stable solution .
however , the rate equations are just a `` mean field '' approximation ; they only describe the behaviour of the average values of the individual populations . in the real world , the system is subject to stochastic noise due to birth and death processes ( intrinsic noise ) , which we take to be poisson - distributed .
the random nature of these processes need be taken into consideration , if we want to get the correct picture of the evolution of the system . for that we ought to write the master equation , and then try to somehow expand it , obtaining a fokker - planck equation , and
eventually solve it .
the two most commonly used expansions of the master equation are the kramers - moyal expansion , which is essentially a taylor expansion in powers of system size @xmath10 , and the @xmath39-expansion of van kampen @xcite which is an expansion in powers of @xmath40 .
the first method produces a nonlinear fokker - planck equation .
the second method is quite systematic , and yields satisfactory results when the system has a single stable point , as the rate equations suggest is the case with the present model . in the absence of this stable solution ,
the task becomes very difficult @xcite .
using the `` shift '' operators notation : @xmath41 the master equation for the cyclic competition system with mutations reads : @xmath42+\]]@xmath43 @xmath44 \ } p(a , b , c , t)\ ] ] while that for the neutral drift system with mutations : @xmath45@xmath46+\]]@xmath43 @xmath47 \ } p(a , b , c , t)\ ] ] and that for the cyclic system with migrations : @xmath42+\]]@xmath48 @xmath49 \ } p(a , b , c , t)\ ] ]
in a previous work @xcite we studied the cyclic competition and neutral genetic drift systems in absence of mutations . in the mean - field approximation ( rate equations )
the cyclic system has an infinity of neutrally stable solutions : any trajectory that conserves the integral @xmath50 is a stable trajectory of the system .
on the other hand , the neutral drift system has an infinity of neutrally stable points : any state of the system is stable in the mean - field picture .
the story is different , when stochastic birth - and - death processes are taken into account : then the individual population distribution drifts , until one of the species , and then another one , go extinct . in other words , there is fixation of the population to one of the varieties , and the product @xmath50 becomes zero .
the survival probability versus time ( in units of @xmath10 ) plot exhibits exponential decay with slope @xmath51 , which was verified by numerically solving the corresponding fokker - planck equation .
this equation was obtained from a kramers - moyal expansion , since the infinite multiplicity of stable solutions ( points or limit cycles ) makes the van kampen method impractical to apply .
the first extinction time scales with the total population size @xmath10 .
this behaviour is the same for both the cyclic competition and the neutral drift models , the second one behaving as an `` adiabatic approximation '' to the first one .
for both models , the probability distribution function becomes uniform within a short time , i.e. any point inside the triangle that represents the phase space of the system is equally probable .
when @xmath11 is small ( the meaning of small will become clear in section 5 ) the system with mutations / migrations is in the `` fixation '' regime , in which the results of our previous study @xcite are applicable qualitatively . in that regime ,
the stochastic processes push the system towards the ( absorbing ) boundary , in which at least one of the species has met extinction .
this was verified by computer simulations of the master equations ( [ eq : mec ] , [ eq : nec ] , [ eq : mig ] ) .
these simulations started with equal individual populations of @xmath17 , @xmath18 , @xmath19 , ( i.e. the centre ) .
we generated times for the next possible reaction event with exponential distribution as @xmath52 , where the rate of the cyclic / neutral or mutations / migrations as in the master equations above ( [ eq : mec ] , [ eq : nec ] , [ eq : mig ] ) is substituted .
( here @xmath53 is a random variable with uniform distribution in @xmath54 $ ] . this way we get poisson distribution for the event times , i.e. really independent events @xcite . )
the reaction which occurs first is then picked and the system is updated accordingly . the process is repeated for a large number of events . we observed a `` fixation '' regime , in which one of the populations , and then the next one , go to zero quite fast . in this regime , the number of individual populations oscillates with an amplitude that drifts with time .
the main reaction is then the cyclic one , and there are only occasional mutations / migrations , which are not frequent enough to prevent fixation . in other words ,
the system is experiencing `` forces '' due to fluctuations , and `` forces '' due to mutations ( migrations ) .
the fluctuations are pushing the system toward the boundary , while the mutations ( migrations ) push toward the centre . for small mutation / migration rate
the `` fluctuation - generated forces '' are stronger .
[ fig : tsext ] shows the variation with time of the population of @xmath19 s for a realisation of the system in the `` extinction '' regime .
occasionally , an individual of the competing species ( the one next in the `` food chain '' ) is introduced in the system by mutations ( or migrations ) , and then it either causes an occasional spike in the time series ( like the one we see in fig . [
fig : tsext ] at @xmath55 , or it `` eats up '' the old species completely and replaces it in the system ( which is what happens between @xmath56 and @xmath57 in our plot ) .
we investigated the behaviour of the probability distribution for the first crossing of the boundary , i.e. when one of the populations becomes zero for the first time , as the mutation / migration rate increases .
for that we ran ten thousand copies of the system for each value of the mutation rate . the survival probability ( as defined above ) was plotted vs. time . in the `` fixation regime ''
the decay is still exponential , but the exponent varies with the mutation / migration rate , getting close to zero as the system approaches criticality . fig .
[ fig : expvsmu ] shows a plot of the exponent as a function of mutation probability @xmath11 for system size @xmath58 and @xmath59 .
if only mutations ( migrations ) are present , the system remains near the centre point ( @xmath38 ) , and the order parameter remains considerably above zero ; in other words , all three species are present in the system .
one can occasionally observe temporary extinctions , but this happens very seldom , and for very long times ; when the system size is very large , it almost never happens .
since the boundary is not absorbing , occasional mutations / migrations will return the system to the state with maximal symmetry ( biodiversity ) where all three varieties coexist .
when both the cyclic / neutral drift mechanism and mutations ( migrations ) are present , and the @xmath11 is above critical , the mutations manage to keep the system maximally disordered , since they are stronger than the fluctuations ( which , as we saw , try to drive the system toward the boundary , i.e. fixation , and keep it there ) .
compared to the situation with no / low mutations ( migrations ) , where the boundary is ( practically ) absorbing , and the final state of the system is a `` pure '' one ( with only one of the species present ) , the mutation / migration rate acts then as a `` temperature '' . work has been done on the two allele almost neutral drift model with mutations @xcite .
the almost neutral model with mutations , preserving the total number of individuals , has only one degree of freedom , and allows one to derive an `` effective potential '' from the fokker - planck equation , obtained by a kramers - moyal expansion of the master equation . for small mutation probabilities , such that @xmath60 , there is extinction of one species and fixation .
the effective potential is symmetric around the centre ( where both species are in equal numbers ) and the branches of the effective potential are down .
this allows for the system to quickly slip into the state where only one of the species is present .
otherwise , both species coexist forever in the high mutation regime , i.e. when @xmath61 . in that regime ,
the effective potential is symmetric around the centre point , but with branches upwards , which means that the centre point is a minimum potential point .
the system then remains in the vicinity of that point for very long times .
the effective potential `` flips '' from `` branches up '' to `` branches down '' at the point where @xmath62 .
the transition is second - order , and critical behaviour is observed .
the system behaves similarly when migrations are present , instead of mutations .
one aspect of migrations in a four - species system has been treated recently by togashi and kaneko @xcite .
since the rate equations have a stable solution , we employ the van kampen expansion @xcite .
the idea of this expansion is to split the variables of the problem into a non - fluctuating part , and a fluctuating one , i.e. deal separately with the mean - field solutions and the fluctuations . in this approach
, the numbers of the individual populations would be written as : @xmath63 here the @xmath64 are the concentrations of @xmath17 , @xmath18 , and @xmath19 species respectively ( which only depend on time ) , and the @xmath65 are the fluctuations ( proportional to the square root of system size ) .
the system size ( total population ) @xmath10 is conserved for the system with mutations , but not for that with migrations .
this conservation rule will cause trouble in the case of the system with mutations , and will require special attention . using the van kampen ansatz ,
the probability distribution @xmath66 is transformed into @xmath67 , and the following relations are true : @xmath68 and @xmath69 next we substitute everything into the master equation , leave only the term @xmath70 in the left hand side , and group the right hand side terms according to powers of @xmath40 .
the first term is of order @xmath71 , and it must be equal to zero , for an expansion in terms of @xmath71 to make sense .
that term for e.g. the cyclic model is : @xmath72 = 0,\ ] ] which reproduces the rate equations in terms of the concentrations @xmath64 , with steady state solution @xmath73 .
similarly , we get the rate equations for the other models .
the terms of order @xmath74 give a linear fokker - planck equation of the form : @xmath75\ ] ] we are going to solve the system with migrations first , since the absence of the conservation of total population @xmath10 makes this system easier to deal with .
for simplicity , let us limit our attention to fluctuations around the steady state @xmath73 .
the a - matrix for the system with migrations is : @xmath76 and the b - matrix : @xmath77 these equations are linear , and the coefficients depend on time through @xmath64 .
this approximation is otherwise known as `` linear noise approximation '' .
the solution is known to be a gaussian ; the problem represents itself as an ornstein - uhlenbeck process . for our purposes , it suffices to determine the first and second moments of the fluctuations . following van kampen @xcite
, we can multiply the fokker - planck equation by @xmath65 and integrate by parts to get : @xmath78 the eigenvalues of the @xmath17 matrix are of the form @xmath79 , @xmath80 .
the negativity of the eigenvalues guarantees the stability of the zero solutions to the first moments equations .
hence , the average of the fluctuations decays to zero and remains zero .
the equations for the second moments can be obtained similarly : @xmath81 by symmetry , in the steady state all the diagonal terms @xmath82 are equal , as well as off - diagonal terms ( correlations ) @xmath83 .
they depend on the migration probability @xmath11 alone .
the steady state solutions are : @xmath84 the diagonal terms coincide with the variance , since the @xmath85 .
these fluctuations were measured `` experimentally '' , i.e. calculated from the results of the simulations .
we simulated 1000 copies of the system ( size @xmath59 ) at different mutation rates , and calculated the mean and variance of the fluctuations , as well as correlations .
the results of those simulations are shown in fig .
[ fig : flucmig ] for the system with migrations .
they agree with the calculated values . in the case of the system with mutations ,
the total population @xmath10 is a conserved quantity ; this imposes restrictions on allowed fluctuations .
we avoid this problem by excluding one of the variables . for that
, we transform the problem into one with two variables , which can be done by putting : @xmath86 this corresponds to transforming to cartesian coordinates with the origin placed at the geometrical centre of the equilateral triangle , which constitutes the phase space of our system @xcite . the fokker - planck equation for the cyclic competition system becomes : @xmath87 the a - matrix has the form : @xmath88 with eigenvalues @xmath80 .
the mean value of the fluctuations decays to zero with an oscillating behaviour , and remains zero .
the b - matrix is diagonal , with both diagonal elements equal to @xmath89 .
the equations for the second moments become : @xmath90 with solutions that tend to @xmath91 as @xmath92 .
these solutions need to be transformed back in terms of @xmath82 and @xmath83 , and the results are : @xmath93 for the neutral drift case , the fokker - planck equations becomes : @xmath94 with the same steady state distributions as the cyclic competition model .
the qualitative change in the behaviour of the system , when the value of the parameter @xmath11 is varied , speaks of the presence of a phase transition .
the survival probability decays exponentially with time in the `` fixation '' regime , and the exponent goes down as the mutation probability increases , as shown in fig .
[ fig : expvsmu ] .
this speaks of a symmetry - breaking transition as the mutation probability goes through the critical value .
when the mutation / migration probabilities per particle approach zero , the leading term in the variances of the concentrations of individual populations is of order @xmath95 , and it becomes of the same order of magnitude as the macroscopic concentrations , when @xmath96 .
this gives us the critical mutation / migration probability dependence on the system size @xmath10 .
the critical @xmath11 was verified to be the same for both models with mutations present , and the critical @xmath11 for the system with migrations from the simulations is very close to that for the systems with mutations , which supports the calculations above . fig .
[ fig : mucrit ] shows plots of critical @xmath11 vs. @xmath10 , as well as a line of slope @xmath97 for comparison . to further verify this result
, we performed a kramers - moyal expansion the models with mutations and migrations , in which @xmath11 was posed as @xmath98 .
numerical solution of the resulting nonlinear fokker - planck equations gave the equation for the eigenvalue ( exponent of decay ) : @xmath99 , where @xmath100 is the eigenvalue . at criticality , @xmath101 , which yields @xmath102 or @xmath103 @xcite .
this is in excellent agreement with the results obtained from the analysis above , as well as the computer simulations .
it is worth noting that the neutral system essentially behaves like the cyclic one .
the probability that the system has never crossed the boundary , ( i.e. none of the populations has ever become zero , ) when the system is in the transition regime , still decays to zero , but not exponentially any more .
the transition is critical ; however , transitions in nonequilibrium ( steady - state ) systems are different from the thermodynamic phase transitions .
if the system size were infinite , one would introduce one infected individual ( mutant ) and measure the probability of survival of infection / mutation as time goes to infinity .
this would constitute the order parameter of our system in the thermodynamic limit . however , our systems are finite , and for finite @xmath10 the quantity which exhibits critical behaviour is the first passage time for crossing the boundary . in fig .
[ fig : plcrit ] we show plots of the first extinction times ( with their cummulative probability in the y - axis ) just below , just above , and at the critical @xmath11 ( system size @xmath104 ) .
( it is worth mentioning that we ran a statistics of 10000 copies of the system , but we are showing only 1 in 20 points in that plot , to prevent figures from becoming cluttered . ) the power - law decay behaviour was used as a criterion for determining the critical point .
the plot in fig .
[ fig : plcrit ] shows clearly that the survival probability decays as a power law at the critical point .
however , once the system is above the critical point , there is considerable probability for the system to have never `` gone dead '' , even at very long times . the extinction times at the critical point scale with the system size .
the power - law exponent is @xmath105 for cyclic competition model , @xmath106 for the neutral model with mutations ; and @xmath107 for the cyclic system with migrations .
this exponent close to 1 , found in simulations is compatible with the one expected from branching processes @xcite , as well as the one obtained for the two - species kimura - weiss model @xcite .
we have considered an @xmath0 model with cyclic competition / neutral drift and mutations ( migrations ) at a constant probability .
the system exhibits a critical transition from a `` fixation '' regime to one in which biodiversity is preserved over long time . in the `` fixation '' regime ,
the number of the @xmath16 species oscillates with an amplitude that drifts with time , until one of the species ( and then the second one ) goes extinct , i.e. the order parameter , defined as the product of @xmath108 goes to zero , and remains zero , except for occasional `` bursts '' . in the `` diversity '' regime
, the number of the @xmath16 varieties fluctuates around the centre point , and there are rare extinctions , but the order parameter remains nonzero almost always .
the survival probability decays exponentially below the transition point , but the exponent decreases as the mutation ( migration ) probability per particle increases , until it becomes zero at the critical point .
the critical mutation / migration probability depends on system size as @xmath15 , and the models have the same power - law exponent : -1 .
there is no difference in the behaviours of the neutral system and the cyclic system . also , there is no qualitative difference between the system with mutations and that with migrations .
these results address the concern about the effects of habitat fragmentation in the survival of the species @xcite .
if the population of a certain species goes extinct in one patch ( e.g. a herd , school , swarm ) while it still survives in other patches , then it is hoped that the `` rescue effect '' can prevent global extinction @xcite .
otherwise , the species is doomed to go extinct altogether .
our results suggest that the number of mutants / migrants necessary to preserve diversity is independent of system size , while in most realistic situations the number of mutations that happen would be proportional to system size . on the other hand ,
if one has in mind epidemiological systems , the important factor is to reduce migration probabilities below critical , which is exactly the purpose quarantines serve .
let s recall here the revival of sars epidemics in toronto , once the guard was let down , i.e. infected individuals were allowed to migrate from one community on to others . in that case , a large system size may be a disadvantage .
it takes only one bad apple ... in lattice models @xcite small amounts of migration bring about `` phase synchronization '' .
peak population abundances , however , are observed to be largely uncorrelated .
blasius et al .
@xcite use a langevin type system of equations to introduce noise in the three - species spatially structured model , and obtain complex chaotic travelling - wave structures .
this work on three - species ecological or epidemiological systems relates to autocatalytic systems with a loop - like structure @xcite .
methods of analysis employed in this paper can be expanded to the loop - like autocatalytic systems .
we have work in progress regarding the four- or more - species autocatalytic loops @xcite . | we study the @xmath0 model in the cyclic competition ( @xmath1 , @xmath2 , @xmath3 ) and the neutral drift ( @xmath4 or @xmath5 , @xmath2 or @xmath6 , @xmath7 or @xmath8 ) versions , with mutations and migrations introduced into the model . when stochastic phenomena are taken into account , there are three distinct regimes in the model . @xmath9 in the `` fixation '' regime , the first extinction time scales with the system size @xmath10 and has an exponential distribution , with an exponent that depends on the mutation / migration probability per particle @xmath11 . @xmath12 in the `` diversity '' regime , the order parameter remains nonzero for very long times , and becomes zero only rarely , almost never for large system sizes .
@xmath13 in the critical regime , the first passage time for crossing the boundary ( one of the populations becoming zero ) has a power law distribution with exponent @xmath14 .
the critical mutation / migration probability scales with system size as @xmath15 .
the transition corresponds to a crossover from diffusive behaviour to gaussian fluctuations about a stable solution .
the analytical results are checked against computer simulations of the model . |
investigation of electromagnetic hadron decays plays an important role in elementary - particle physics .
these processes , governed by the interaction of real and virtual photons with the electric charges of quark fields , make it possible to obtain unique information on the character of various quark configurations in hadrons , the mechanism of mixing , the electromagnetic structure of strongly interacting particles , and on some phenomenological features of such particles ( magnetic moments , formfactors , polarizability , etc . ) .
recent years have seen a significant progress in this field .
radiative hadron decays of the type @xmath0 can be studied by both direct and indirect methods . in the former case , particle @xmath1
is produced in a reaction and its decay @xmath0 is detected directly in the experimental setup . among indirect methods
one should mention the process of interaction of primary particle @xmath2 with a virtual photon in the coulomb field of the nucleus @xmath3 as we will show later , at a very high energy @xmath4 coulomb reaction ( 1 ) is characterized by small values of the squared 4-momentum transfer @xmath5 which is identical with the squared 4-momentum @xmath5 of the virtual photon propagator .
thus , at @xmath6 the virtual photons are quasi - real . in this approximation
the cross section for this coulomb production process is proportional to the radiative width @xmath7 .
the main aim of my talk is to discuss these electromagnetic coulomb processes and their application to the study of radiative decays of hadrons and to the search for some exotic states .
actually , the type ( 1 ) reactions involving collisions with virtual ( quasireal ) photons are photoproduction reactions on primary hadrons @xmath2 .
with pion , kaon , or hyperon beams we have unique possibility to study photoproduction on unstable targets " ( @xmath8 ) .
the coulomb production mechanism was considered for the first time independently by h.primakoff [ 1 ] and by i. pomeranchuk and i. schmushkevich [ 2 ] .
it is often designated as the primakoff production or primakoff reaction .
in this type of experiments the particles under study ( @xmath1 ) are produced either in hadron reactions or in electromagnetic interactions ( photoproduction ; resonance production in @xmath9 collisions ) . as a rule , to select rare electromagnetic decays @xmath10 @xmath11 it is necessary to detect all the decay products charge particles as well as photons
to measure their momenta and energy and to reconstruct the effective mass of decay particles @xmath12 . in the directly detected
@xmath0 decays or more intricate electromagnetic processes , we must suppress a background from the decay @xmath13 with one photon being lost .
a similar situation prevails , for example , in searches for the decay @xmath14 against the background from the main decay mode @xmath15 , @xmath16 [ hereafter , ( @xmath17 ) stands for the lost photon ] .
reliable separation of rare radiative decays involving single photons requires a complete detection of all secondaries ( both charged and neutral ones ) , measurement of their momenta , and reconstruction of effective masses with the highest possible resolution .
the background from the lost photons can be minimized by equipping the setup with a veto system consisting of counters with a low threshold for photon detection and having the maximum possible coverage .
this system also includes a hodoscopic @xmath17 spectrometer detecting photons from radiative decays in the operating acceptance of the setup
. a strong kinematical constraint on the processes in question can also play an important role .
the point is that the background from the events with a lost photon does not yield a narrow peak at the mass of a particle under study and can therefore be considerably reduced owing to kinematical constraints .
the above is illustrated by searches for the radiative decay @xmath18 in the experiments with the lepton - f [ 3 ] and asterix [ 4 ] setups .
the background from the lost photons was substantially suppressed by appropriately chosen conditions and using a veto system in the former case and by imposing more stringent kinematical constraints in the latter case . as a result ,
the two experiments yield approximately the same upper limit on the probability of the above radiative decay : @xmath19 ( at a 95% c.l . )
this demonstrates that radiative decays can be sought directly even if their probabilities are less than 1% of the probabilities of the most dangerous background processes . at the same time , the decay @xmath20 , for which the background conditions proved to be much more favorable ( the process @xmath21 , which could mimic the decay @xmath22 because of the lost photon , is forbidden by @xmath23 and @xmath24 conservation in strong and electromagnetic interactions ) , was recorded at a significantly lower level of @xmath25 [ 5 ] .
the same is true for several other electromagnetic decays with a relatively small background .
the results of several experiments for the direct study of rare radiative decays of mesons are presented in table 1 and in [ 3 - 10 ] .
an overview of various methods for detecting electromagnetic decays is given , for example , in review papers [ 11 - 14 ] .
to determine the radiative widths of some excited hadrons for which direct measurements involve considerable difficulties or are impossible , it was proposed long ago to use coherent electromagnetic production of particles in the coulomb field of heavy nuclei ( [ 1,2 ] ) , see also [ 15 - 18 ] ) .
in the small - width approximation as applied to the resonance state @xmath1 in ( 1 ) , the cross section for electromagnetic production has the from @xmath26}{dq^2 } \right]_{\rm coul}=}\\ \displaystyle { = |t_{\rm coul}|^2=8\pi \alpha z^2 \cdot \frac{2j_a+1}{2j_h+1 } \cdot \gamma(a \to h \gamma)}\\ \displaystyle { \nonumber\times \left[\frac{(q^2 - q^2_{\min})}{q^4 } \right ] \cdot \left(\frac{m_a}{(m^2_a - m^2_h)^3}\right)^3 \cdot \end{array}\ ] ] where @xmath27 is the charge of the nucleus ; @xmath28 is the fine - structure constant ; @xmath29 is the radiative decay width of particle @xmath1 ; @xmath30 and @xmath31 are the spins of particles @xmath1 and @xmath2 , respectively ; @xmath32 and @xmath33 are their masses ; @xmath34 is the nuclear electromagnetic form factor ; @xmath35 is the minimum 4-momentum transfer squared ; and @xmath4 is the energy of the primary particle @xmath2 . if photons appear as primary particles , an additional factor of two must be included in the expression for the differential cross section . as it is seen from ( 4 ) the coulomb cross section increases
fast with decreasing @xmath5 .
it can easily be found that the cross section @xmath36_{\rm coul}$ ] attains a maximum at @xmath37 and that @xmath36_{\rm coul}^{\max } \propto q^{-2}_{\min } \propto e^{2}_h$ ] . as the primary momentum increases , the cross - section value at the maximum increases in proportion to @xmath38 , whereas the position of this maximum shifts toward lower values of @xmath5 .
the total cross section for the coulomb process increases as @xmath39 .
concurrently , the differential cross section for a coherent process that is governed by a strong interaction exhibits a much broader distribution in @xmath5 , and the region of small @xmath5 is dominated by the coulomb contribution .
the total coulomb production cross section has the form @xmath40_{\rm
coul}=\\ \displaystyle { = \frac{2j_a+1}{2j_h+1}\cdot 8\pi \alpha z^2 \gamma(a \to h \gamma ) \left(\frac{m_a}{m^2_a - m^2_h}\right)^3 } \times \\ \displaystyle { \nonumber \times \int^{q^2_{\max}}_{q^2_{\min } } \frac{(q^2 - q^2_{\min})}{q^4}|f_z(q^2)|^2 d(q^2)= \sigma_0 \cdot \gamma(a \to h \gamma ) . } \end{array}\ ] ] the quantity @xmath41 bounds the integration domain dominated by coulomb processes .
certainly , a more precise expression for @xmath42 can be obtained and used in the data analysis , in which @xmath43 decay is expressed more carefully , in a relativistic breit - wigner resonance form .
the coherent production of hadrons by a primary particle @xmath2 on nuclei @xmath44 is a rather complicated process affected by both electromagnetic and strong interactions .
the contribution of strong interactions leads to a broader distribution in momentum transfer squared @xmath5 .
the differential cross section for coherent particle production on nuclei can be represented as @xmath45_{\rm coherent}=\left|t_{\rm coul . }
+ e^{i\varphi } t_{\rm strong}\right|^2,\ ] ] where @xmath46 is the amplitude of coulomb production , @xmath47 is the amplitude of a coherent process that is governed by a strong interaction , and @xmath48 is the relative phase of the two amplitudes . for the understanding of a possibility to separate these coherent processes it is instructive to formulate their main characteristics which is done in table 2 .
it can be seen from table 2 that in accordance with quantum numbers of the produced hadrons in ( 1 ) two possible scenarios for the strong coherent background can take place : 1 .
coherent diffractive background is forbidden by selection rules for ( 1 ) and the leading coherent background is determine usually by @xmath49 reggeon exchange in t channel of the reaction .
this amplitude dies at a very high energy . in several cases
this background is also reduced in the region of small @xmath5 ( as @xmath50 ) .
the absence of diffractive coherent background significantly simplifies the separation of the coulomb process .
2 . coherent diffractive background plays a leading role in reaction ( 1 ) . in this case
the separation of the coulomb process is much more difficult and might be performed only in the region of very small @xmath51 gev@xmath52 ) and with a very large statistics .
let us complete the discussion of the coulomb production reactions making several conclusions : 1 .
the coulomb production reactions open a new possibility to study rare radiative decays @xmath53 if these processes can not be detected directly due to the background from the decay @xmath54 ; @xmath55 with lost photons or if the decay width is too small for a direct measurement ( the later case takes place , for example , for @xmath56 decays ) .
the coulomb production processes open a unique possibility to study the photoproduction reactions on unstable targets " ( like pions , kaons , hyperons ) , which can be very important for the search for exotic hadrons , for studying some electromagnetic properties of unstable hadrons ( their polarizability , for example ) and other phenomena .
3 . to accomplish this study it is necessary to separate the coulomb production processes and to make the absolute measurements of their cross sections .
all these require very precise and sophisticated experiments in the high energy region .
this is the reason for a relatively small amount of the coulomb production experiments which have been carried out in a 50 year period of time since the first theoretical work of h.primakoff [ 1 ] .
the main experimental results in this field at the high energies [ 1931 ] are summarized in table 3 .
in 1998 the first preliminary results of studying several coulomb production reactions in the selex experiment at the tevatron of fermilab were obtained and are presented here in short ( for more details see [ 32,33 ] ) .
the selex setup [ 34,35 ] , that is used in these measurements includes the three - stage magnetic spectrometer with proportional and drift chambers , vertex microstrip detector , trigger hodoscopes , some additional microstrip detectors , rich and trd detectors for particle identification , three photon lead glass multichannel spectrometers , neutron calorimeter . the main measurements in the selex ( e781 ) experiment were performed with a negative - charged beam of 600 gev momentum and with practically equal amount of @xmath58-mesons and @xmath59-hyperons ( they were identified with the beam trd detector ) .
the wide research program is performed by selex(e781 ) international collaboration ( with the participation of scientists from brasil , china , germany , israel , italy , mexico , russia , united kingdom , usa ) . the main part of this program involves the study of charmed and strange - charmed baryons , the search for exotic states , the investigation of electromagnetic properties of hadrons and the coulomb production processes , some other researches . a short summary of the preliminary coulomb production results in the selex experiment is presented below .
the coherent reactions @xmath62+(a , z ) \vspace*{-0.2cm}\ ] ] on carbon , copper and lead nuclei were separated in the analysis of the events with 3 charge particles in the final state in the interactions of identified primary @xmath58-mesons in one of the targets .
these events must satisfy the elastic conditions " the difference between the beam particle momentum and the sum of momenta of three secondary particles must be less than 17.5 gev .
the statistics of the selected events of ( 7 ) is tabulated in table 4 .
the @xmath63 distribution at a very high momentum . ] of the @xmath64-system in reaction ( 7 ) for the copper target is shown , as an example , in fig.1a .
this @xmath65 distribution was fitted by the sum of two exponents @xmath66 with the slopes @xmath67 gev@xmath68 and @xmath69 gev@xmath68 .
the second slope @xmath70 is in a good agreement with the slope for diffractive production coherent process ( 7 ) obtained from the data [ 36 ] at @xmath71 gev .
the slope @xmath72 corresponds to the monte carlo estimations for the coulomb mechanism ( with account for the @xmath63 resolution ) .
two @xmath63 regions were defined for @xmath65 distribution ( see fig.1a ) . in the first region ( @xmath73 gev@xmath52 ) the coulomb production process dominating .
the second region ( @xmath74 gev@xmath52 ) was used to estimate the coherent diffractive background .
the mass spectra @xmath75 for these two regions are presented in fig.1b .
the result of diffractive background subtraction ( after a proper normalization ) is presented in fig.1c . in this subtracted mass spectrum
the coulomb production of @xmath76 meson in the reaction @xmath77 is clearly seen .
the same results were also obtained from the measurements with carbon and lead targets .
after the absolute normalization of cross sections for the primakoff reactions @xmath78=\sigma_0\cdot \gamma [ a_2(1320)^- \to \pi^- \gamma]$ ] , the values for radiative width @xmath79 $ ] from all these measurements were obtained ( see table 4 ) . for the absolute normalization the cross sections of diffractive production reactions ( 7 ) for @xmath80 ; @xmath81 ; @xmath24 from [ 36 ] were used .
the data for the radiative width in the measurements with different targets are consistent with each other , which confirms the coulomb production of @xmath82 meson .
the average value of radiative width over all the targets is @xmath83=233 \pm 31 ( { \rm stat . } ) \pm 47({\rm syst.)~kev}. \vspace*{-0.2cm}\ ] ] the main source for the systematic uncertainty is the absolute normalization procedure .
this value of the radiative width is preliminary and might be obtained with a better accuracy in the future .
the comparison of value ( 10 ) with previous measurement [ 21 ] and with theoretical predictions [ 37 , 38 ] is also presented in table 4 .
@xmath84 + ( a , z ) , \\
\nonumber & & \hspace{17mm}^|\hspace{-2 mm } \rightarrow \gamma \gamma \vspace*{-0.7cm}\end{aligned}\ ] ] @xmath85 + ( a , z ) , \\
\nonumber & & \hspace{7mm}^|\hspace{-2 mm } \rightarrow \pi^+ \pi^-
\pi^0 \\ & \rightarrow & b_1(1235 ) + ( a , z ) , \\
\nonumber & & \hspace{2mm}^| \hspace{-2 mm } \rightarrow \omega \pi^- \\ & \rightarrow & [ \pi^- \eta ] + ( a , z ) , \\
\nonumber & & \hspace{6mm}^| \hspace{-2 mm } \rightarrow \pi^+ \pi^-
\pi^0 \\ & \rightarrow & a_2(1235 ) + ( a , z ) , \\
\nonumber & & \hspace{2mm}^| \hspace{-2 mm } \rightarrow \pi^- \eta \vspace*{-0.2cm}\end{aligned}\ ] ] were performed at @xmath86 gev on the lead , copper and carbon targets in the selex experiment .
the coulomb production of @xmath87 and @xmath88 was observed quite clearly in ( 12 ) and ( 14 ) under very favorable background conditions .
the coherent strong interaction background here is dominated by @xmath49 exchange and is small in the region of @xmath89 gev@xmath52 .
now the data from these primakoff reactions are in the course of analysis to obtain the absolute normalization of their cross sections and to determine the radiative widths @xmath90 $ ] and @xmath91.$ ] @xmath92 this primakoff process is determined by a very small width of su(3 ) forbidden radiative decay @xmath93 only a short dedicated run at the selex facility can be used for the study of primakoff reaction ( 15 ) .
the preliminary value for the upper limit of radiative width for decay ( 16 ) is obtained in this measurement @xmath94 < 7 { \rm kev}~~({\rm 95 c.l.})\ ] ] the theoretical expectations for this width are in the region of 1 10 kev ( see reviews [ 39,40 ] and the references herein ) . for comparison ,
the expected value of radiative width for the su(3 ) allowed decay @xmath95 is predicted at a level of 100 kev .
the coulomb production processes ( primakoff reactions " ) can be of great concern not only in the study of radiative hadronic decays , but in the search for new types of hadrons as well . in the last decade
the problem of existence of a novel form of hadronic matter exotic hadrons has become the leading direction in hadron spectroscopy .
a rapid development of this field led to a significant advance in the systematics of ordinary " hadrons ( with the valence quark structure @xmath96 or @xmath97 ) and to the observation of several unusual states which do not fit this simplest systematics .
these anomalous states are real candidates for exotic hadrons with a complicated valence internal structure ( multiquark formations @xmath98 , @xmath99 ; hybrid states with valence quarks and gluons @xmath100 , @xmath101 ; pure gluonic mesons glueballs @xmath102 ) .
the success of experiments aimed at the search for exotic hadrons and first of all for cryptoexotic states with usual quantum numbers but with anomalous dynamical properties ( hidden exotics " ) is , to a great extent , determined by the appropriate choice of the production processes for which some qualitative considerations can predict more distinct manifestations of exotic states . for example , it was emphasized in a number of studies that gluon - rich diffractive production reactions on nucleons or nuclei ( coherent reactions ) offer some favorable grounds for exotic hadron production [ 4144 ] .
it was also stated that for some cases the electromagnetic mechanisms can be very promising for these aims [ 4547 ] .
let us consider , for example , the exotic hadrons with hidden strangeness @xmath103 mesons or @xmath104 baryons ( here @xmath105- or @xmath106-quarks ) .
due to a significant coupling of photon with @xmath107 pairs through @xmath108 meson in the framework of vdm , electromagnetic interactions can provide a natural way to embed the @xmath109 pair into an intermediate hadron state and to produce the exotic hadron with hidden strangeness .
this possibility can be realized in the resonance @xmath110-channel reaction @xmath111 or in the coulomb production reaction @xmath112 as has been shown in section 3 , the coulomb production mechanism plays the leading role in the region of very small transfer momenta , where it is dominating over the strong interaction process .
thus , the coherent reactions at a very small @xmath113 gev@xmath52 ) can be used in the search for exotic states with hidden strangeness . as was stated before ,
the most interesting coulomb production processes were those initiated by unstable primary particles ( @xmath114-mesons , @xmath115-mesons , hyperons ) , because these processes give the only possibility to study the photoproduction reactions on unstable targets .
reaction ( 20 ) for protons @xmath116 is , in principle , the same as photoproduction process ( 19 ) which can be studied in a very detailed way on photon beams of electron accelerators .
but the existing data for these reactions are rather poor now and insufficient for such systematical studies .
one can hope that in the near future reliable data would be obtained in the experiments at strong current electron accelerators cebaf and elsa ( see , for example , [ 48 ] ) .
on the other hand , there are the data for coherent reactions @xmath117 + c\ ] ] obtained in the experiments of the sphinx collaboration which might be connected with the coulomb production mechanism , [ 47,49 ] .
the feasibility to separate the coulomb production processes in the coherent proton reactions at @xmath118 gev on the carbon target in the measurement with the sphinx setup has been recently demonstrated in the observation of the coulomb production of @xmath119 isobar with @xmath120 in the reaction @xmath121 ( see [ 47 ] ) .
the recent results of the coulomb production reactions study presented in this talk allow one to obtain the important information on the electromagnetic properties of hadrons and to search for new hadronic structures .
we continue the analysis of the data obtained in the selex experiment and we hope to improve the precision of these measurements and the number of primakoff reactions under study . 00 primakoff h. , phys .
, 1951 , vol .
81 , p.119 .
pomeranchuk i.ya . ,
shmushkevich i.m .
phys . , b , 1961 , vol . 23 , p. 452
bityukov s.i .
et al . , yad.fiz . , 1988 ,
47 , p. 1258 .
weidenauer , p. et al . , z.phys .
c , 1990 , vol .
47 , p.353 .
alde d. et al . ,
phys.lett.b , 1994 , vol .
340 , p. 122 .
binon f. et al . ,
fiz . , 1981 , vol .
203 , p. 327 .
bityukov , s.i .
phys.lett.b , 1988 , vol .
203 , p. 327 .
landsberg l.g .
chastits at.yadra , 1990 , vol .
21 , p. 1054 .
djhelyadin r.i .
phys.lett.b , 1980 , vol .
94 , p. 548 .
victorov v.a .
et al . , yad.fiz . , 1980 ,
32 , p. 1005 .
landsberg l.g .
, phys . , rep .
, 1985 , vol .
128 , p. 301 .
zielinski m. , acta phys .
polonica b , 1987 , vol .
18 , p. 455 .
odonnel p.j . ,
1981 , vol .
53 , p. 673 .
h.kolanoski , springer tracts mod.phys .
, 1984 , vol . 105 .
halpern a. et al . , phys.rev . , 1966 ,
152 , p. 1295 .
dreitlein j. , primakoff h. , phys.rev .
, 1962 , vol .
125 , p. 591 .
berlad g. et al .
, annals of phys . , 1973 , vol . 75 , p. 461 .
f@xmath122ldt g. et al . ,
, 1972 , vol .
41 , p. 125 ; vol .
43 , p. 591 .
jansen t. et al . , phys.rev.d , 1983 , vol .
27 , p. 26 .
huston j. et al . , phys.rev.d , 1986 , vol .
33 , p. 3199 .
cihangir s. et al . ,
phys.lett.b , 1982 , vol .
117 , p. 119 .
zeilinski m. et al .
, phys.rev.lett . , 1984 ,
52 , p. 1195 .
collick b. et al . , phys.rev.lett . , 1984 ,
53 , p. 2374 .
berg d. et al . ,
phys.lett.b , 1981 , vol .
98 , p. 119 .
carismith d. et al . , phys.rev.lett . , 1986 ,
56 , p. 18 .
carismith d. et al . ,
phys.rev.d , 1986 , vol .
36 , p. 3199 .
capraro l. et al . ,
nucl.phys.b , 1987 , vol.288 , p. 659 .
antipov yu.m .
phys.lett.b , vol .
121 , p. 445 .
antipov yu.m .
phys.rev.d , vol .
36 , p. 21 .
dydak f. , et al .
, nucl.phys.b , 1977 , vol .
118 , p.1 ; delvin t. et al .
, phys.rev.d , vol .
34 , p. 1626 .
peterson p.c .
phys.rev.lett . , 1986 , vol .
57 , p. 949 .
kubarowsky v.p .
, report on ichep-98 , vancouver , july , 1998 .
landsberg l.g . , yad.fiz .
( in press ) edelstein r. et al . ,
fermilab proposal e781 , 1987 ( revised 1993 ) ; russ j.s .
, nisl.phys , 1995 , vol .
585a , p. 39c .
smith v.j .
, hadron spectroscopy ( `` hadron-97 '' ) , seventh intern .
conf . , upton , ny , august 1997 , ( ed , chung , s - u . , willutzki , h.j . )
zielinski m. et al . , z. phys.c , 1983 , vol .
16 , p. 197 . babcock j. , rosner j. l. , phys.rev . , 1976 ,
14d , p. 1286 .
ishida s. et al . ,
phys.rev , 1989 , vol 40d , p. 1497 .
landsberg l.g . , yad.fiz . , 1996 ,
59 , p. 2161 .
landsberg l.g . ,
molchanov v.v .
, preprint ihep 97 - 42 , protvino , 1997 .
hogaasen h. , sorba p. , nucl.phys.b , 1978 , vol .
145 , p. 119 .
chan hoang - mo , tsou s.t .
, nucl.phys.b , 1977 , vol .
118 , p. 413 .
hirose t. et al . , nuov .
cim.a , 1979 , vol .
50a , landsberg l.g .
, yad.fiz .
1994 , vol .
57 , p. 47 ; ufn , 1994 , vol .
164 , p. 1129 .
zielinski m. et al .
, z.phys.c , 1986 , vol .
31 , p. 545 ; vol .
34 , p. 255 .
landsberg l.g .
, yad.fiz . , 1990 , vol .
52 , p. 192 ; landsberg l.g . , nucl.phys.,b ( proc.suppl . ) , 1991 , vol.21 , p. 179 .
vavilov d.v .
yad.fiz . , 1999 ,
vol.62 , p.501 .
shumacher r. , preprint cmu meg-96 - 007 , pittsburg , 1996 .
landsberg l.g .
, phys.rep .
( in press ) . | in this talk the coulomb production reactions at high energy are discussed as well as the study of electromagnetic properties of hadrons ( mesons and hyperons ) in these processes .
the results of previous investigations are summed together with some recent data of selex(fermilab ) and sphinx(ihep ) experiments .
xix international symposium on lepton and photon interaction at high energy + stanford university , august 9 - 14 , 1999 + l.g.landsberg + state research center , institute for high energy physics , + protvino , moscow region , russia , 142284 + * study of the primakoff reactions + at high energy * |
recent observations of type ia supernovae indicate that the expansion of the universe is accelerating [ 1 - 5 ] and lead to the search for a new type of matter which violates the strong energy condition , i.e. , @xmath0 .
the matter responsible for this condition to be satisfied at some stage of evolution of the universe is referred to as _ dark energy _ [ 6 - 8 ] .
several candidates to present dark energy have been suggested with observations : the cosmological constant [ 7 , 9 ] , quintessence [ 10 , 11 ] , phantom [ 12 , 13 ] , braneworld models [ 14 ] , pure chaplygin gas model [ 15 ] , generalized chaplygin gas ( gcg ) model [ 16 , 17 ] , modified chaplygin gas ( mcg ) model [ 19 , 20 ] . in the gcg and mcg
approach dark energy and dark matter can be unified by using an exotic equation of state ( eos ) .
interesting feature of mcg ( or gcg ) eos is that it shows radiation era ( or dust era ) in the past while a @xmath1cdm model in the future . + in 2003 ,
p. f. gonz@xmath2lez - diaz [ 21 ] have introduced the generalized cosmic chaplygin gas ( gccg ) model in such a way that the resulting models can be made stable and free from unphysical behaviours even when the vacuum fluid satisfies the phantom energy condition .
the eos of this model is @xmath3\ ] ] where @xmath4 with @xmath5 a constant which can take on both positive and negative values and @xmath6 , @xmath7 being a positive definite constant which can take on values larger than unity .
the eos reduces to that of current chaplygin unified models for dark matter and dark energy in the limit @xmath8 and satisfies the conditions : ( i ) it becomes a de sitter fluid at late time and when @xmath9 , ( ii ) it reduces to @xmath10 in the limit that the chaplygin parameter @xmath11 , ( iii ) it also reduces to the eos of current chaplygin unified dark matter models at high energy density and ( iv ) the evolution of density perturbations derived from the chosen eos becomes free from the pathological behaviour of the matter power spectrum for physically reasonable values of the involved parameters at late time .
this eos shows dust era in the past and @xmath1cdm in the future .
+ since models trying to provide a description of the cosmic acceleration are proliferating , there exists the problem of discriminating between the various contenders . to this aim sahni
et al [ 22 ] proposed a pair of parameters @xmath12 , called _ statefinder _ parameters . in fact trajectories in the @xmath12 plane corresponding to different cosmological models demonstrate qualitatively different behaviour .
the above statefinder diagnostic pair has the following form : @xmath13 where @xmath14 and @xmath15 are the hubble parameter and the deceleration parameter respectively .
the new feature of the statefinder is that it involves the third derivative of the cosmological radius .
these parameters are dimensionless and allow us to characterize the properties of dark energy .
trajectories in the @xmath12 plane corresponding to different cosmological models , for example @xmath1cdm model diagrams correspond to the fixed point @xmath16 .
+ in this paper , we consider the universe is filled with the mixture of radiation and gccg in section ii .
we perform a statefinder diagnostic to this model without and with interaction in different cases in sections iii and iv respectively . from statefinder parameters we have shown graphically that the universe starts from radiation era instead of dust era .
different phases of the evolution of the universe have been shown graphically . with interaction case
, the model goes from radiation to @xmath1cdm era only and without interaction case the model goes from radiation to @xmath1cdm and further from @xmath1cdm to phantom era and then back to @xmath1cdm .
the paper ends with a short discussion in section v. +
the metric of a spatially flat isotropic and homogeneous universe in frw model is , @xmath17\ ] ] where @xmath18 is the scale factor . + the einstein field equations are ( choosing @xmath19 ) @xmath20 and @xmath21 the energy conservation equation ( @xmath22 ) is @xmath23 where , @xmath24 and @xmath25 are the total energy density and the pressure of the universe , given by , @xmath26 and @xmath27 with @xmath28 and @xmath29 are respectively the energy density and pressure due to the gccg satisfying the eos ( 1 ) and @xmath30 and @xmath31 are the energy density and the pressure corresponding to the radiation fluid with eos , @xmath32 where @xmath33 .
+ since gccg can explain the evolution of the universe starting from dust era to @xmath1cdm , considering the mixture of gccg with radiation would make it possible to explain the evolution of the universe from radiation to @xmath1cdm .
in this case gccg and the radiation fluid are conserved separately . conservation equation ( 6 )
yields , @xmath34 and @xmath35 from equations ( 1 ) , ( 9 ) , ( 10 ) , ( 11 ) we have @xmath36^{\frac{1}{1+\alpha}}\ ] ] and @xmath37 for the two component fluids , equation ( 2 ) takes the following forms : @xmath38\ ] ] and @xmath39\ ] ] also the deceleration parameter @xmath40 has the form : @xmath41 now substituting @xmath42 , equation ( 14 ) and ( 15 ) can be written as , @xmath43\ ] ] and @xmath44}\ ] ] normalizing the parameters we have shown the graphical representation of the @xmath45 parameters in figure 1 . from the figure we have seen that the universe starts from radiation era @xmath46 via dust stage @xmath47 to @xmath1cdm ( @xmath48 ) model for choices of @xmath49 . fig.1 fig .
1 shows the variation of @xmath50 against @xmath51 ( eqs .
( 17 ) and ( 18 ) ) for @xmath52 .
we consider the gccg interacting with radiation fluid through an energy exchange between them .
the equations of motion can be written as , @xmath53 and @xmath54 where @xmath55 is a coupling function .
+ let us choose , @xmath56 now equation ( 19 ) together with equation ( 1 ) gives , @xmath57^{\frac{1}{1+\alpha}}\ ] ] also equations ( 9 ) , ( 20 ) and ( 22 ) give @xmath58 with @xmath59 and @xmath60^{\frac{1}{1+w}}\ ] ] from ( 22 ) , we see that if @xmath61 , i.e. , @xmath62 , then the expression ( 22 ) reduces to the expression ( 12 ) . + +
now for the two component interacting fluids with equations of motion ( 19 ) and ( 20 ) , the @xmath45 parameters read : @xmath63\ ] ] and @xmath64 also the deceleration parameter @xmath40 has the form : @xmath65 now substituting @xmath42 , equation ( 14 ) and ( 15 ) can be written as , @xmath66\ ] ] and @xmath67 where , @xmath68\ ] ] @xmath69 @xmath70 @xmath71 and @xmath72 + now we find the exact solution for the @xmath45 parameters for the following particular choices of @xmath73 : + ( i ) if @xmath74 , i.e. , @xmath75 , equation ( 23 ) can be written as @xmath76 as @xmath77 . normalizing the parameters ,
the corresponding statefinder parameters are given in figure 2 . from the figure
we have seen that the universe starts from radiation era @xmath46 via dust stage @xmath78 to @xmath1cdm ( @xmath48 ) model and further from @xmath1cdm to phantom era ( @xmath79 ) and then back to @xmath1cdm for choices of @xmath80 , @xmath81 .
+ fig.2 fig .
2 shows the variation of @xmath50 against @xmath51 ( case ( i ) ) for @xmath80 , @xmath82
. fig.3 fig .
3 shows the variation of @xmath50 against @xmath51 ( case ( ii ) ) for @xmath83 , @xmath84 .
\(ii ) if @xmath85 , i.e. , @xmath86 , equation ( 23 ) can be written as @xmath87\ ] ] normalizing the parameters , the corresponding statefinder parameters are given in figure 3 . from the figure
we have seen that the universe starts from radiation era @xmath46 via dust stage @xmath88 to @xmath1cdm ( @xmath48 ) model for choices of @xmath83 , @xmath89 .
+ ( iii ) if @xmath90 , equation ( 23 ) can be written as @xmath91\ ] ] normalizing the parameters , the corresponding statefinder parameters are given in figure 4 . from the figure
we have seen that the universe starts from radiation era @xmath46 via dust stage @xmath92 to @xmath1cdm ( @xmath48 ) model and further from @xmath1cdm to phantom era ( @xmath79 ) and then back to @xmath1cdm for choices of @xmath83 , @xmath93
. fig.4 fig .
4 shows the variation of @xmath50 against @xmath51 ( case ( iii ) ) for @xmath83 , @xmath94 .
in this work , we have considered the matter in our universe as a mixture of the gccg and radiation as gccg can explain the evolution of the universe from dust era to @xmath1cdm .
these gases are taken both as non - interacting and interacting mixture . in the first case
we have considered a non - interacting model and plotted the @xmath45 parameters .
as expected this model represents the evolution of the universe from radiation era to @xmath1cdm with a discontinuity at dust . in the second case the interaction term is chosen in a very typical form to solve the corresponding conservation equations analytically . also the statefinder parameters are evaluated for various choices of parameters and the trajectories in the @xmath12 plane
are plotted to characterize different phases of the universe .
these trajectories show discontinuity at same @xmath51 in the neighbourhood of @xmath95 and have peculiar behaviour around @xmath96 .
the @xmath12 curves have two branches on two sides of the asymptote .
the branch on the right hand side of the asymptote corresponds to decelerating phase before ( or up to ) dust era , while the left hand side branch has a transition from decelerating phase upto @xmath1cdm era .
some peculiarity has been shown in figures 2 and 4 around @xmath96 . in these two cases ,
the model goes further from @xmath1cdm to phantom era and then back to @xmath1cdm .
moreover , in figure 4 , there is further transition from @xmath1cdm to decelerating phase and then then again back to @xmath1cdm .
thus we can conclude that the present model describes a number of transitions from decelerating to accelerating phase and vice - versa .
+ + * acknowledgement : * + the authors are thankful to iucaa , india for warm hospitality where part of the work was carried out .
also ud is thankful to ugc , govt . of india for providing research project grant ( no .
32 - 157/2006(sr ) ) .
+ * references : * + + @xmath97 $ ] s. j. perlmutter et al , _ bull .
* 29 * 1351 ( 1997 ) .
+ @xmath98 $ ] s. j. perlmutter et al , _ astrophys .
j. _ * 517 * 565 ( 1999 ) .
+ @xmath99 $ ] a. g. riess et al , _ astron .
j. _ * 116 * 1009 ( 1998 ) .
+ @xmath100 $ ] p. garnavich et al , _ astrophys
. j. _ * 493 * l53 ( 1998 ) .
+ @xmath101 $ ] b. p. schmidt et al , _ astrophys .
j. _ * 507 * 46 ( 1998 ) .
+ @xmath102 $ ] v. sahni and a. a. starobinsky , _ int . j. mod
. phys . a _ * 9 * 373 ( 2000 ) .
+ @xmath103 $ ] p. j. e. peebles and b. ratra , _ rev .
phys . _ * 75 * 559 ( 2003 ) .
+ @xmath104 $ ] t. padmanabhan , _ phys .
* 380 * 235 ( 2003 ) .
+ @xmath105 $ ] s. m. carrol , _ living rev .
rel . _ * 4 * 1 ( 2001 ) .
+ @xmath106 r. r. caldwell , r. dave and p. j. steinhardt , _ phys .
_ * 80 * 1582 ( 1998 ) .
+ @xmath107 $ ] v. sahni and l. wang , _ phys .
d _ * 62 * 103507 ( 2000 ) .
+ @xmath108 $ ] r. r. caldwell , _ phys .
b _ * 545 * 2 ( 2002 ) .
+ @xmath109 $ ] s. m. carrol , m. hoffman and m. trodden , _ phys .
d _ * 68 * 023509 ( 2003 ) .
+ @xmath110 $ ] u. alam , v. sahni , _ astro - ph_/ 0209443 .
+ @xmath111 $ ] a. kamenshchik , u. moschella and v. pasquier , _ phys .
b _ * 511 * 265 ( 2001 ) .
+ @xmath112 $ ] v. gorini , a. kamenshchik and u. moschella , _ phys .
d _ * 67 * 063509 ( 2003 ) .
+ @xmath113 $ ] u. alam , v. sahni , t. d. saini and a.a .
starobinsky , _ mon . not .
soc . _ * 344 * , 1057 ( 2003 ) .
+ @xmath114 $ ] m. c. bento , o. bertolami and a. a. sen , _ phys .
d _ 66 043507 ( 2002 ) .
+ @xmath115 $ ] h. b. benaoum , _ hep - th_/0205140 .
+ @xmath116 $ ] u. debnath , a. banerjee and s. chakraborty , _ class . quantum grav . _
* 21 * 5609 ( 2004 ) .
+ @xmath117 $ ] p. f. gonz@xmath2lez - diaz , _ phys .
d _ * 68 * 021303 ( r ) , ( 2003 ) .
+ @xmath118 $ ] v. sahni , t. d. saini , a. a. starobinsky and u. alam , _ jetp lett . _ * 77 * 201 ( 2003 ) . | recently developed generalized cosmic chaplygin gas ( gccg ) is studied as an unified model of dark matter and dark energy .
to explain the recent accelerating phase , the universe is assumed to have a mixture of radiation and gccg .
the mixture is considered for without or with interaction .
solutions are obtained for various choices of the parameters and trajectories in the plane of the statefinder parameters and presented graphically . for particular choice of interaction parameter
, we have shown the role of statefinder parameters in various cases for the evolution of the universe . |
high - redshift radio galaxies ( hzrgs ; @xmath4 ) play a fundamental role in our understanding of the formation and evolution of the most massive galaxies in the early universe . as inferred from the hubble @xmath5@xmath6 diagram ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , galaxies with powerful radio emission are among the most massive systems at each cosmic epoch , and are 23 mag brighter than optically - selected galaxies at @xmath7 . explaining how hzrgs formed is a crucial test for hierarchical models of galaxy formation ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) ; the discovery of quasars at @xmath8 ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) implies that the central supermassive ( @xmath9@xmath10 m@xmath11 ) black hole must be in place @xmath12 1 gyr after the big bang .
moreover , given the correlation between black hole and galaxy bulge mass ( e.g. * ? ? ?
* ; * ? ? ?
* ) , we can trace a direct evolutionary path from the most massive galaxies at high redshift to powerful radio galaxies in the local universe , which are well known to be giant ellipticals @xcite .
as hzrgs are rare objects , it is necessary to filter out low - redshift objects from large radio catalogues in order to maximize the efficiency of follow - up observations .
the most successful selection method has proven to be ultra - steep spectrum ( uss ) selection .
this technique is based on the results of @xcite and @xcite , who found that radio sources with the steepest spectral indices by the relation @xmath1 , where @xmath13 is the flux density at frequency @xmath14 . ]
were not detected on palomar observatory sky survey ( poss ) plates .
the implied correlation between redshift and spectral index ( the @xmath6@xmath15 correlation ) is regularly used to optimize hzrg searches ; uss samples with @xmath16 ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) have yielded over 30 @xmath17 radio galaxies , out to @xmath18 @xcite . given that malmquist bias is an inherent feature of any flux - limited sample , and an upper limit exists in radio luminosity , we must use radio surveys with faint flux density limits to find the most distant massive galaxies .
indeed , hzrg searches have been primarily limited to the northern hemisphere due to a paucity of sensitive radio surveys in the south . however , with the completion of the 843 mhz sydney university molonglo sky survey ( sumss ; * ? ? ?
* ; * ? ? ?
* ) , and a deeper re - analysis of its predecessor , the 408 mhz molonglo reference catalogue ( mrc ; * ? ? ?
* ) , this is no longer the case . using these two surveys , and the 1400 mhz nrao vla sky survey ( nvss ; * ? ? ?
* ) , we are carrying out extensive searches for southern hzrgs , which will be used to better constrain galaxy formation and evolution theories .
we select samples of uss sources from the revised mrc ( henceforth referred to as mrcr ) , sumss and nvss in the overlap region @xmath3 .
initially , we @xcite defined a pilot sample of 76 uss sources with @xmath19 and @xmath20 mjy by cross - correlating the sumss and nvss catalogues .
follow - up spectroscopy of this sample ( henceforth referred to as the sumss
nvss sample ) resulted in the discovery of five radio galaxies at @xmath17 , the highest redshift being @xmath21 @xcite .
a further seven galaxies may be at @xmath22 or heavily obscured by dust @xcite . in this paper
, we define the follow - up mrcr
sumss sample , which consists of 234 uss sources with @xmath0 and @xmath2 mjy selected by cross - matching the mrcr , sumss and nvss catalogues .
our observing strategy is the same as that used previously for the sumss
nvss sample .
after defining our sample through uss selection , we use high - resolution radio observations from the australia telescope compact array ( atca ; * ? ? ?
* ) in conjunction with near - infrared ( nir )
@xmath5-band imaging to unambiguously identify the host galaxy of each radio source .
redshifts are then determined by optical or nir spectroscopy . in this paper , we focus on the radio properties of the mrcr
sumss sample .
@xmath5-band and spectroscopic data will be presented in a subsequent paper ( bryant et al .
, in preparation ) , henceforth referred to as paper ii .
this paper is set out as follows . in section [ 408_paper1_sample_definition1 ]
, we define the mrcr
sumss sample . in section [ 408_paper1_atca_observations ]
, we describe the high - resolution atca observations of the sample .
we analyse the radio properties of the sample in section [ 408_paper1_analysis_radio_properties ] , before discussing in section [ 408_paper1_discussion ] both the efficiency of our search and the implications of the radio properties .
finally , we present our conclusions in section [ 408_paper1_conclusions ] . throughout
we assume a flat @xmath23 cold dark matter cosmology with @xmath24 km s@xmath25 mpc@xmath25 , @xmath26 and @xmath27 @xcite .
the mrc @xcite is the product of a 408 mhz continuum survey undertaken with the molonglo cross telescope @xcite .
the survey was conducted in the region @xmath28 for galactic latitudes @xmath29 . as a result of a re - analysis of the digitised data from the molonglo cross , a much deeper version of the mrc
is now available , with the flux density limit lowered from 1 jy to @xmath9200 mjy ( @xmath95@xmath30 ) .
the mrcr comprises @xmath977000 sources with @xmath2 mjy .
the angular resolution is @xmath31 arcmin@xmath32 , the typical positional uncertainty near the flux density limit is @xmath920 arcsec , and the source density is @xmath92.5 deg@xmath33 .
we stress that the mrcr is incomplete : only data from the best - quality scans have been included .
details of this catalogue will be published elsewhere ( crawford , in preparation ) .
sumss @xcite is an 843 mhz continuum survey of the entire sky south of declination @xmath34 with the molonglo observatory synthesis telescope ( most ; * ? ? ?
* ; * ? ? ?
the angular resolution is @xmath35 arcsec@xmath32 . for @xmath3 ,
the median rms noise level is 1.9 mjy beam@xmath25 , and the completeness limit is 18 mjy .
positional uncertainties are 12 arcsec for sources with peak brightnesses greater than 20 mjy beam@xmath25 . in this paper , we use version 2.0 of the sumss catalogue has been rectified . ] , which consists of @xmath9210000 sources , with surface density @xmath920 deg@xmath33 above 10 mjy beam@xmath25 .
nvss @xcite is a 1400 mhz continuum survey of the entire sky north of @xmath36 with the very large array ( vla ; * ? ? ?
the angular resolution is @xmath37 arcsec@xmath32 , the rms noise level is 0.45 mjy , and the completeness threshold is 2.5 mjy .
the rms positional uncertainties in right ascension and declination are @xmath38 arcsec for sources stronger than 15 mjy . in this paper , we use version 41 of the nvss catalogue , which consists of @xmath9@xmath39 sources with surface density @xmath9@xmath40 deg@xmath33 . & & & & & & + nvss j011606@xmath41331241 & @xmath42 & @xmath43 & @xmath44 & @xmath45 & @xmath46 & @xmath47 + nvss j202856@xmath41353709 & @xmath48 & @xmath49 & @xmath50 & @xmath51 & @xmath52 & obscured by galactic star + nvss j202945@xmath41344812 & @xmath53 & @xmath54 & @xmath55 & @xmath56 & @xmath57 & @xmath58 + nvss j225719@xmath41343954 & @xmath59 & @xmath60 & @xmath61 & @xmath62 & @xmath63 & @xmath64 + nvss j230527@xmath41360534 & @xmath65 & @xmath66 & @xmath67 & @xmath68 & @xmath69 & not detected ; @xmath22 candidate + nvss j230846@xmath41334810 & @xmath70 & @xmath71 & @xmath72 & @xmath73 & @xmath74 & not observed + nvss j231727@xmath41352606 & @xmath75 & @xmath76 & @xmath77 & @xmath78 & @xmath79 & @xmath80 + nvss j232058@xmath41365157 & @xmath81 & @xmath82 & @xmath83 & @xmath84 & @xmath85 & not observed + nvss j233729@xmath41355529 & @xmath86 & @xmath87 & @xmath88 & @xmath89 & @xmath90 & not observed + nvss j234145@xmath41350624 & @xmath91 & @xmath92 & @xmath93 & @xmath94 & @xmath95 & @xmath96 + + we cross - matched the mrcr , sumss and nvss catalogues in the overlap region @xmath3 in order to define a sample of uss sources with @xmath0 and @xmath2 mjy .
we chose to use three low - frequency catalogues in order to investigate the spectral energy distributions ( seds ) of our targets . while the spectral index cutoff is flatter than the @xmath19 condition used in our pilot study , we note that a number of @xmath4 radio galaxies in the sumss nvss sample were found to have spectral indices as flat as @xmath97 after seds were constructed over a wider frequency range @xcite .
the galactic plane was avoided by restricting the cross - matching to the right ascension regions 204@xmath98 and 1016@xmath98 ( @xmath99 ) .
note that the former right ascension zone could in principle be extended from 4@xmath98 to 7@xmath98 but the sumss catalogue did not cover this region when we defined our sample .
we adopted a 60 arcsec search radius when searching for sumss matches around the position of each mrcr source .
as the sumss source density is @xmath920 deg@xmath33 , the probability of an mrcr sumss match occurring by chance within 60 arcsec is roughly 2 per cent .
we then used a 10 arcsec radius to search for sumss
nvss matches ; the probability of a sumss
nvss match occurring by chance within 10 arcsec is @xmath90.1 per cent .
thus , the probability of a chance match among all three catalogues is @xmath90.002 per cent , which implies that the expected number of spurious mrcr sumss
nvss matches in a sample of 234 sources is much less than one . due to the difference in resolution of the mrcr , sumss and nvss catalogues
, source confusion will affect the reliability of the spectral indices .
therefore , we visually inspected each individual sumss field and removed any sources that were obviously blended at 408 mhz .
in addition , as the resolution of sumss varies from @xmath100 arcsec@xmath32 to @xmath101 arcsec@xmath32 over the declination range @xmath102 to @xmath103 , we removed any sumss nvss matches with an additional nvss match within 100 arcsec , except for five cases where high - resolution atca imaging confirmed that multiple nvss components were indeed matched to a single sumss source ( see table 3 ) .
while the removal of such sumss nvss matches imposes an implicit angular size cutoff in the sample , we do not expect a significant number of hzrgs to be excluded , if any .
studies of powerful radio sources have found that linear size decreases with increasing redshift ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , and no hzrgs with angular sizes greater than 1 arcmin have been reported in the literature . to optimize the optical / nir spectroscopy of our uss sample , it was essential to remove any low - redshift interlopers .
we used supercosmos sky survey @xcite scans of uk schmidt iiiaj ( blue , ukj ) and iiiaf ( red , ukr ) plates ( limiting magnitudes @xmath104 and @xmath105 , respectively ) to identify any sources with bright optical identifications , which are most likely at @xmath106 .
however , for 32 sources , candidate optical counterparts could only be unambiguously identified after high - resolution atca images were obtained , leading to the inclusion of some low - redshift sources in the sample ( see section [ 408_paper1_morphologies ] ) .
sumss sample consists of 234 sources for which we have obtained follow - up atca imaging at both 1384 and 2368 mhz .
the sky area covered by our sample is 0.35 sr .
as the sample was assembled over a three year period , revisions to the 408 , 843 and 1400 mhz catalogues have meant that a number of sources no longer meet our original selection criteria . using a preliminary version of the mrcr , version 2.0 of the sumss catalogue and version 41 of the nvss catalogue
, we find that 205 sources still have @xmath0 .
we decided to retain the remaining 29 sources ( @xmath107 ) because of the broadness of the @xmath6@xmath15 correlation , and also to increase our spectral index baseline for seeking trends with other radio properties . the mrcr sumss and sumss
nvss samples have been defined over the same declination range , and therefore some uss sources satisfy the selection criteria of both samples .
note that both samples do not fully overlap because the sources in the sumss
nvss sample are fainter ( see section [ 408_paper1_fluxes ] ) . in table
[ 408_paper1_table : sources_common_with_sumss - nvss ] , we list the flux densities , spectral indices and redshifts for an additional ten sources from the sumss
nvss sample that also have @xmath0 and @xmath2 mjy .
in this paper , we restrict our analysis to the 234 new targets for which high - resolution radio data have been obtained .
a full description of the sources in table [ 408_paper1_table : sources_common_with_sumss - nvss ] is given in @xcite , @xcite and @xcite .
table [ 408_paper1_table : atca_log ] contains a summary of the high - resolution radio observations of the mrcr
sumss sample obtained with the atca . for each session
, we list the date of observation , the number of sources observed , the array configuration ( including the baselines spanned by the array ) , and the secondary calibrators observed . in each session , we performed dual - frequency observations at 1384 and 2368 mhz ( wavelengths 20 and 13 cm ) using a @xmath108 mhz bandwidth correlator configuration .
we obtained 510 cuts of 35 min duration of each source using the snapshot imaging technique .
typically , the cuts were spread out over a @xmath912 h period to optimize the @xmath109 coverage .
the median integration time per source was @xmath925 min .
pks b1934@xmath41638 was used as the primary calibrator in each session .
data reduction followed standard procedures in miriad @xcite .
the sensitivity was optimized by using natural weighting to construct total intensity ( stokes @xmath110 ) dirty images . to ensure that the sidelobes of any nearby bright sources were properly deconvolved
, the dirty images were made at least twice the size of the primary beam full - width at half - maximum ( fwhm ) , which is 33 arcmin at 1384 mhz and 22 arcmin at 2368 mhz .
it is well known that running the clean deconvolution algorithm unconstrained on such images can lead to flux being subtracted from the target source .
this effect , known as clean bias , is discussed by @xcite in relation to nvss data . to resolve the problems encountered with unconstrained clean ,
we have developed a technique using miriad that automatically constrains clean by selecting clean regions around the positions of known sources in the nvss and/or sumss catalogues .
this is analogous to the faces task in the aips data reduction package .
using our automatically constrained clean technique , we defined clean regions around all of the nvss sources in each field with flux density @xmath111 mjy .
we chose nvss rather than sumss for this task because of its lower flux density limit .
after the clean regions had been defined , clean was run for 2500 iterations or until the peak residuals were no larger than 0.5 mjy .
the default loop gain of 0.1 was used .
we then performed five iterations of phase - only self - calibration to improve the image dynamic range .
the median angular resolution is @xmath112 arcsec@xmath32 at 1384 mhz and @xmath113 arcsec@xmath32 at 2368 mhz , while the median rms noise levels are 0.32 mjy beam@xmath25 at 1384 mhz and 0.27 mjy beam@xmath25 at 2368 mhz .
we estimate the astrometry to be accurate to @xmath114 arcsec .
the resolution and positional accuracy are sufficient to unambiguously identify the host galaxy of each radio source in our @xmath5-band images ( paper ii ) . to enable a polarimetric analysis of the sources in the mrcr
sumss sample , we produced stokes @xmath115 and @xmath116 images from the self - calibrated visibilities at 1384 and 2368 mhz .
all dirty images were cleaned for a maximum of 100 iterations .
the miriad task impol was then used to make total polarized intensity images [ flux density = @xmath117 $ ] that were corrected for ricean bias .
the median rms noise levels in the @xmath115 and @xmath116 images , @xmath118 , are 0.20 and 0.23 mjy beam@xmath25 at 1384 and 2368 mhz , respectively .
observations & sources & configuration & + & observed & ( baselines ) & + 2003 & 26 & 6a & pks b0153@xmath41410 + december 1112 & & ( 3375939 m ) & pks b2339@xmath41353 + + 2004 & 48 & 6d & pks b0153@xmath41410 + december 89 & & ( 775878 m ) & + + 2005 & 72 & 6a & pks b1232@xmath41416 + april 810 & & ( 3375939 m ) & pmn b1458@xmath41391 + + 2006 & 88 & 6c & pks b0008@xmath41421 + april 1417 & & ( 1536000 m ) & pks b0153@xmath41410 + & & & pks b1232@xmath41416 + & & & pmn b1458@xmath41391 + & & & pks b2211@xmath41388 +
[ fig:408_paper1_table : sample_definition ] the radio properties of each source in the mrcr sumss sample are summarised in tables 3 and 4 , and are discussed in the following sections .
we define our sample in table 3 , which contains flux density and spectral index properties from 408 to 2368 mhz , as well as morphological and polarization information . in table 4
, we list the atca properties of each source in our sample , in particular the flux density , spectral index and polarization properties of the individual radio components.in table 3 , the columns are : ( 1 ) name of the source in the nvss catalogue in iau j2000 format.(2 ) the ra and dec j2000 coordinates from the nvss catalogue .
( 3 ) the 408 , 843 , 1384 , 1400 and 2368 mhz integrated flux densities .
the 408 mhz flux densities are from a preliminary version of the mrcr , while the 843 and 1400 mhz flux densities are from versions 2.0 and 41 of the sumss and nvss catalogues .
( 4 ) the two - point observed - frame spectral indices @xmath119 , @xmath120 and @xmath121 . ( 5 )
the observed - frame spectral index based on a linear fit to all five flux density data points . for those sources that have curved spectra that flatten ( f ) or steepen ( s ) at high frequency , we use a quadratic fit to all five flux density data points to derive the observed - frame spectral index at 1400 mhz .
( 6 ) the nvss fractional linear polarization , determined from the integrated polarized flux densities in version 41 of the nvss catalogue .
we ignore those cases where the polarized flux density has a negative estimated value ( see * ? ? ?
( 7 ) the largest angular size ( las ) at 2368 mhz . for single - component resolved sources , this is the deconvolved major axis of the elliptical gaussian used to fit the source . for unresolved sources ,
we have used the distribution of deconvolved sizes to estimate a las upper limit of 2 arcsec . for multi - component sources ,
the las is the angular separation of the most widely - separated components.(8 ) the deconvolved position angle ( pa ) of the radio structure at 2368 mhz , measured from the north to the east . for multi - component sources ,
this is the orientation of the most widely - separated components used to calculate the las.(9 ) the radio morphology at 2368 mhz : single - component ( s ) , double ( d ) , triple ( t ) , head - tail ( ht ) , narrow - angle tail ( nat ) , wide - angle tail ( wat ) or complex ( c ) .
incipient doubles in which the components are not fully resolved are labelled as di . in table 4 , the columns are : ( 1 ) name of the source in the nvss catalogue in iau j2000 format.(2 ) 1384 and 2368 mhz integrated flux densities of the radio components in each atca for the central component , north lobe and south lobe , assuming that each component is not blended at 1384 mhz.(4 ) 1384 and 2368 mhz fractional linear polarization for the central , north lobe and south lobe components , measured at the position of peak intensity . a @xmath122 upper limit is given for source components with undetected polarization.(5 ) the observed - frame faraday rotation measure ( rm ) , not corrected for the galactic faraday screen , for the central , north lobe and south lobe components.(6 ) the galactic rm determined from the all - sky rm map of @xcite.note that for the head - tail , narrow - angle tail and wide - angle tail sources , the properties of the core are given under the c label , and the properties of the tail(s ) are given under the n and s labels ( assuming that these components are resolved ) . at 2368 mhz , 39 per cent of the sources in the mrcr
sumss sample have a single - component morphology , 52 per cent are fanaroff riley type ii ( fr ii ; * ? ? ?
* ) doubles , and 7 per cent are fr ii triples .
interestingly , many of the doubles and triples have asymmetric lobe flux densities , which suggests a corresponding asymmetry in the density of the intergalactic medium through which the radio jets propagate .
in addition , there are a few fr i sources that have either a head - tail , narrow - angle tail , wide - angle tail or complex morphology ; these sources are the expected principal contaminants in a uss - selected sample .
the largest angular sizes range from 176 arcsec at 2368 mhz , with a median of 11 arcsec . given that the relationship between angular size and projected linear size is relatively flat over the redshift interval @xmath914 ( @xmath978.5 kpc arcsec@xmath25 ) in our adopted cosmology , around 20 per cent of the sample would be classified as compact steep - spectrum sources ( css ; see e.g. the review by * ? ? ?
* ) , with projected linear sizes @xmath123 20 kpc . in section [ 408_paper1_sample_definition ]
, we noted that 32 sources ( 14 per cent of the sample ) were found to have likely optical identifications in supercosmos after atca imaging . in table
[ 408_paper1_table : optical ids ] , we give the coordinates and optical magnitudes for each of the potential host galaxies .
typical uncertainties in the optical magnitudes are a few tenths of a magnitude .
most of the optical counterparts are very faint ; the median supercosmos magnitudes are @xmath124 and @xmath125 . in fig .
[ 408_paper1_fig : morph_examples ] , atca contours have been overlaid on supercosmos images for sources with a possible optical identification , an unusual radio morphology , and/or information available from the literature . as the optical counterparts are faint , we smoothed the supercosmos images with a gaussian kernel with fwhm = 2.1 arcsec ( 3 pixels ) .
@xmath5-band - radio overlays for the remaining sources without optical counterparts are presented in paper ii .
notes on individual sources in fig .
[ 408_paper1_fig : morph_examples ] are given below : narrow - angle tail source with partially detached tails ( e.g. similar morphology to j1324@xmath413138 in the abell cluster a3556 ; * ? ? ?
the optical counterpart is located at the centroid of the cluster edcc 0523 @xcite ; using data from the 2df galaxy redshift survey ( 2dfgrs ; * ? ? ?
* ) , @xcite determined the cluster centroid to be located at @xmath126 , @xmath127 with a redshift of @xmath128 .
the 2dfgrs redshift of the optical counterpart is @xmath129 .
wide - angle tail source with a clear optical identification .
the 2dfgrs redshift of the optical counterpart is @xmath130 .
two independent , closely - spaced doubles . while the 408 , 843 and 1400 mhz flux densities are blended , both doubles are still hzrg candidates : @xmath131 for the north double , and @xmath132 for the south double .
this source has a complex , diffuse , multi - component morphology .
there are a number of faint confusing sources at 1384 mhz that are not resolved in nvss ( these are included in the total 1384 mhz flux density in table 3 ) .
both the southern lobe and most of the confusing sources are either marginally detected or not detected at 2368 mhz , implying that they have very steep spectra .
the 2dfgrs redshift of the optical counterpart is @xmath133 .
this single - component source is also in the mrc pmn uss sample , with spectral index @xmath134 @xcite .
it was observed with the atca at 1420 mhz by @xcite , who obtained a flux density of @xmath135 mjy .
this is consistent with our 1384 mhz flux density of @xmath136 mjy .
double source that is also in the molonglo southern 4 jy sample ( ms4 ; * ? ? ?
* ; * ? ? ?
* ) and the mrc pmn uss sample . @xcite used the anglo - australian telescope ( aat ) to detect an optical counterpart between the lobes at @xmath137 , @xmath138 with @xmath139 mag .
the supercosmos @xmath140 magnitude of the optical counterpart is @xmath141 mag .
two independent , closely - spaced wide - angle tail ( north ) and head - tail ( south ) sources that are potentially members of the same distant cluster ( e.g. similar morphology to mrc b1610@xmath41605 + mrc b1610@xmath41608 in * ? ? ?
* ) ; @xmath142 for the wide - angle tail source , and @xmath143 for the head - tail source . as the three low - redshift fr
i sources in the mrcr
sumss sample are interlopers , we exclude nvss j010619@xmath41325713 , nvss j014353@xmath41331041 and nvss j020827@xmath41342635 @xmath144 nvss j020828@xmath41342520 from any subsequent analysis in this paper .
& & + & & & & + & & & & + nvss j001737@xmath41364710 & 00 17 37.66 & @xmath4136 47 11.5 & 21.2 & 20.6 + nvss j002956@xmath41313701 & 00 29 56.76 & @xmath4131 37 02.7 & 22.6 & 19.8 + nvss j003445@xmath41372348 & 00 34 45.59 & @xmath4137 23 48.9 & @xmath141 & 20.6 + nvss j004206@xmath41361329 & 00 42 06.60 & @xmath4136 13 29.1 & @xmath141 & 20.9 + nvss j004851@xmath41324623 & 00 48 51.16 & @xmath4132 46 23.1 & @xmath145 & 20.7 + nvss j010619@xmath41325713 & 01 06 20.26 & @xmath4132 57 32.2 & 18.3 & 17.0
+ nvss j011732@xmath41383739 & 01 17 32.77 & @xmath4138 37 38.5 & 22.1 & 19.7 + nvss j012443@xmath41310820 & 01 24 43.73 & @xmath4131 08 22.0 & 22.2 & @xmath146 + nvss j014353@xmath41331041 & 01 43 54.11 & @xmath4133 10 47.0 & 18.4 & 16.8 + nvss j020827@xmath41342635 @xmath147 & & & & + nvss j020828@xmath41342520 & & & & + nvss j023601@xmath41314204 @xmath147 & & & & + nvss j023605@xmath41314235 & & & & + nvss j025521@xmath41333212 & 02 55 21.09 & @xmath4133 32 11.4 & 21.8 & 19.5 + nvss j101215@xmath41394939 & 10 12 15.19 & @xmath4139 49 22.6 & 20.8 & 19.8 +
nvss j103441@xmath41394957 & 10 34 42.10 & @xmath4139 49 58.2 & 20.3 & 19.6 + nvss j111539@xmath41375633 & 11 15 39.36 & @xmath4137 56 33.2 & 22.9 & 20.9 + nvss j111546@xmath41391410 & 11 15 46.47 & @xmath4139 14 10.5 & @xmath145 & 20.8 + nvss j132952@xmath41381251 & 13 29 52.56 & @xmath4138 12 53.9 & 22.6 & 19.8 + nvss j133317@xmath41383006 & 13 33 17.48 & @xmath4138 30 16.2 & 21.2 & 20.3 + nvss j134502@xmath41395932 & 13 45 02.16 & @xmath4139 59 28.9 & 21.8 &
@xmath146 + nvss j140330@xmath41394130 & 14 03 30.61 & @xmath4139 41 35.1 & @xmath145 & 20.0 + nvss j141428@xmath41320637 & 14 14 28.28 & @xmath4132 06 38.8 & 19.4 & 18.1 + nvss j144503@xmath41353542 & 14 45 03.24 & @xmath4135 35 44.6 & 22.2 & @xmath146 + nvss j151215@xmath41382220 & 15 12 15.51 & @xmath4138 22 22.2 & 21.4 & 19.7 + nvss j152921@xmath41362209 & 15 29 21.78 & @xmath4136 22 14.4 & 21.6 & 19.9 + nvss j201943@xmath41364542 & 20 19 43.56 & @xmath4136 45 43.4 & 22.7 & @xmath146 + nvss j204308@xmath41302527 & 20 43 08.56 & @xmath4130 25 28.8 & 22.2 & 19.9 + nvss j214522@xmath41361730 & 21 45 21.90 & @xmath4136 17 30.7 & 21.3 & 19.1
+ nvss j215029@xmath41310457 & 21 50 29.80 & @xmath4131 04 59.6 & @xmath145 & 20.1 + nvss j220651@xmath41364752 & 22 06 51.47 & @xmath4136 47 51.4 & 22.6 & @xmath146 + nvss j221805@xmath41363007 & 22 18 05.61 & @xmath4136 30 06.0 & 22.9 & @xmath146 + nvss j230004@xmath41304711 & 23 00 04.25 & @xmath4130 47 08.6 & @xmath145 & 20.8 + nvss j235104@xmath41325444 & 23 51 04.66 & @xmath4132 54 44.8 & @xmath145 & 20.7 + + the statistical properties of the flux densities of the sources in the mrcr
sumss sample are presented in table [ 408_paper1_table : flux_properties ] .
typical uncertainties are 10 per cent in the mrcr , and 34 per cent in sumss and nvss .
the atca 1384 and 2368 mhz flux densities were measured by fitting elliptical gaussians to source components with the miriad task imfit .
the fitting errors were calculated following @xcite and combined in quadrature with the calibration uncertainty to obtain a total flux density uncertainty . in general , the fitting uncertainty is much smaller than the calibration uncertainty , which is @xmath93 per cent at 1384 mhz and @xmath95 per cent at 2368 mhz .
thus , the total flux density uncertainties are 34 per cent at 1384 mhz and 56 per cent at 2368 mhz .
the sources in the mrcr
sumss sample are stronger than in the sumss nvss sample .
the median sumss and nvss flux densities in the mrcr
sumss sample are 134 and 80 mjy respectively ( table [ 408_paper1_table : flux_properties ] ) , compared with 47 and 22 mjy respectively in the sumss
nvss sample @xcite .
this is a consequence of the @xmath2 mjy cutoff which defines the mrcr
sumss sample , as opposed to the @xmath20 mjy cutoff used in the sumss
nvss sample ; the equivalent cutoff in the sumss
nvss sample at 408 mhz is 74.5 mjy for @xmath148 . the effectiveness of our automatically constrained cleaning procedure is illustrated in fig .
[ 408_paper1_fig : flux_ratio_hist ] , which shows a histogram of the atca / nvss 20 cm flux density ratio . as the sources in the mrcr
sumss sample have ultra - steep spectra , we would expect the atca 1384 mhz flux densities to be 12 per cent higher than the nvss 1400 mhz flux densities , due to the slight difference in centre frequency .
after automatically constrained cleaning , the agreement between the atca and nvss flux densities is excellent : @xmath995 per cent of the sources have flux densities that agree to within @xmath149 per cent , and @xmath975 per cent agree to within @xmath150 per cent .
however , on average , the atca flux densities tend to be slightly lower than in nvss by @xmath93 per cent .
this is evident in the statistical properties in table [ 408_paper1_table : flux_properties ] .
one possible explanation for the slight underestimation of the atca flux densities is that the vla array configuration used to construct nvss ( d - array , spanning baselines 351030 m ) is more compact than the atca configurations used in this study , resulting in a greater sensitivity to low - surface - brightness extended features .
indeed , we find that sources with angular sizes @xmath151 arcsec are underestimated by @xmath94 per cent on average , compared with @xmath92 per cent for sources with angular sizes @xmath152 arcsec
. source variability may also affect the flux densities of some sources . even though the atca flux densities are still marginally underestimated after automatically constrained cleaning , we emphasize that , for the vast majority of sources , these flux densities are still sufficiently accurate to permit an investigation of seds in the frequency range 4082368 mhz .
.flux density properties of the mrcr
sumss sample .
@xmath153 is the median flux density , @xmath13 ( min . )
is the minimum flux density and @xmath13 ( max . )
is the maximum flux density .
[ cols="^ , > , > , > " , ] while a large fraction of the seds in the mrcr
sumss sample are described by a single power law above 408 mhz , other studies have found that the spectra of most uss sources tend to flatten below our selection frequencies due to the effects of synchrotron self - absorption ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) . at these frequencies ,
high - redshift sources should be more easily distinguished from the low - redshift population because they are sampled at higher rest - frame frequencies and therefore exhibit less spectral curvature .
thus , it is thought that uss samples selected at very low frequencies ( e.g. * ? ? ?
* also see table [ 408_paper1_table : surface_densities ] ) might be more efficient at finding the highest - redshift radio galaxies . in order to investigate the amount of low - frequency spectral curvature present in the mrcr
sumss sample , we used data from the 74 mhz vla low - frequency sky survey ( vlss ; * ? ? ? * ) , which has a resolution of @xmath154 arcsec@xmath32 and a typical rms noise level of @xmath9100 mjy beam@xmath25 . however , as the southern boundary of the vlss is at @xmath155 , our analysis is limited by the small amount of overlap with the mrcr sumss sample . using the 2007 june 18 vlss catalogue
, we obtained 74 mhz flux densities for 32 sources ; these are listed in table [ 408_paper1_table : vlss ] .
using data from the vlss , the 352 mhz westerbork in the southern hemisphere survey ( wish ; * ? ? ?
* ) and nvss , @xcite found significant curvature below 352 mhz in a subset of 12 uss sources with @xmath156 ( wish nvss selection , see table [ 408_paper1_table : surface_densities ] ) : the median values of the two - point spectral indices are @xmath157 and @xmath158 .
we can test whether we see the same amount of curvature in the mrcr
sumss sample over a similar frequency range by using our 74 , 408 and 1400 mhz data .
we derive median spectral indices of @xmath159 and @xmath160 ; individual @xmath161 values are shown in table [ 408_paper1_table : vlss ] . while the flatter median spectral index between 74 and 408 mhz implies the presence of spectral curvature , the difference between the spectral index medians is about half that found by @xcite .
this suggests that low - frequency spectral curvature is not as significant in the mrcr
sumss sample .
we now examine whether there is sufficient flattening below 408 mhz such that the seds can not be fitted by a single power law . to do this , we computed sed fits similar to those described in section [ 408_paper1_spectral_indices ] , but extended each fit to 74 mhz .
the spectral indices obtained from these fits are also listed in table [ 408_paper1_table : vlss ] .
we find that 20 of the seds are still well described by a single power law below 408 mhz ( e.g. fig .
[ 408_paper1_fig : sed_examples2 ] ) .
eight seds flatten below 408 mhz ; of these , only nvss j215029@xmath41310457 previously showed evidence of spectral curvature between 408 and 2368 mhz ( see table 3 ) .
the remaining four seds have poor goodness - of - fit statistics between 408 and 2368 mhz , and this remains so after each fit is extended to 74 mhz ( see discussion in section [ 408_paper1_spectral_indices ] ) .
thus , while the relative fraction of sources with concave spectra has increased with the inclusion of vlss data ( 25 per cent compared with 2 per cent in section [ 408_paper1_spectral_indices ] ) , the majority of sources have negligible spectral curvature between 74 and 2368 mhz .
this may suggest that the sources with straight spectra are at higher redshift .
we investigate this issue further in paper ii .
we have defined a sample of 234 uss sources with @xmath0 and @xmath2 mjy , assembled to find distant radio galaxies in the region @xmath3 .
we have drawn the following conclusions based on our investigation of the radio properties of the sample : 1 .
198 sources ( 85 per cent ) have spectral energy distributions that remain straight from 408 mhz to 2368 mhz in the observed frame .
where 74 mhz data are available , we find that 20 sources ( 63 per cent ) remain well fitted by a single power law .
2 . with decreasing angular size ,
the average spectral index steepens slightly and the average fractional polarization decreases . both effects are possibly due to the smallest sources residing in the densest environments .
3 . the median fractional polarization varies from 3.5 per cent at 1400 mhz to 8.1 per cent at 2368 mhz .
the fractional polarization is found to be anti - correlated with flux density at both 1400 and 2368 mhz , but not at 1384 mhz .
higher - resolution observations with high s / n are needed to obtain a clearer picture of the fractional polarization properties .
observed - frame rotation measures have been determined for half of the sample .
the maximum observed - frame rm magnitude is 165 rad m@xmath33 .
5 . the surface density of the mrcr
sumss sample is 577 sr@xmath25 .
14 per cent of the sources have candidate optical identifications in the supercosmos sky survey .
jwb acknowledges the receipt of both an australian postgraduate award and a denison merit award .
rwh and jjb acknowledge support from the australian research council .
ems acknowledges support from the australian research council through the award of an arc australian professorial fellowship .
we thank david crawford for providing the mrcr , bryan gaensler for useful discussions and the anonymous referee for helpful suggestions .
the australia telescope compact array is part of the australia telescope which is funded by the commonwealth of australia for operation as a national facility managed by csiro .
supercosmos sky survey material is based on photographic data originating from the uk , palomar and eso schmidt telescopes and is provided by the wide - field astronomy unit , institute for astronomy , university of edinburgh .
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 .
athreya r. m. , kapahi v. k. , mccarthy p. j. , van breugel w. , 1998 , a&a , 329 , 809 barthel p. d. , miley g. k. , 1988 , nat , 333 , 319 baugh c. m. , benson a. j. , cole s. , frenk c. s. , lacey c. , 2003 , in bender r. , renzini a. , eds , the mass of galaxies at low and high redshift , eso springer , 91 blumenthal g. , miley g. , 1979 , a&a , 80 , 13 blundell k. m. , rawlings s. , eales s. a. , taylor g. b. , bradley a. d. , 1998 , mnras , 295 , 265 blundell k. m. , rawlings s. , willott c. j. , 1999 , aj , 117 , 677 bock d. c .- j . , large m. i. , sadler e. m. , 1999 , aj , 117 , 1578 bornancini c. g. , de breuck c. , de vries w. , croft s. , van breugel w. , rttgering h. , minniti d. , 2007 , mnras , 378 , 551 bower r. g. , benson a. j. , malbon r. , helly j. c. , frenk c. s. , baugh c. m. , cole s. , lacey c. g. , 2006 , mnras , 370 , 645 broderick j. w. , de breuck c. , hunstead r. w. , seymour n. , 2007 , mnras , 375 , 1059 burgess a. m. , hunstead r. w. , 2006a , aj , 131 , 100 burgess a. m. , hunstead r. w. , 2006b , aj , 131 , 114 carilli c. l. , taylor g. b. , 2002 , ara&a , 40 , 319 carilli c. l. , rttgering h. j. a. , van ojik r. , miley g. k. , van breugel w. j. m. , 1997 , apjs , 109 , 1 chambers k. c. , miley g. k. , van breugel w. j. m. , huang j .- s . , 1996 , apjs , 106 , 215 cohen a. s. , rttgering h. j. a. , jarvis m. j. , kassim n. e. , lazio t. j. w. , 2004 , apjs , 150 , 417 cohen a. s. , lane w. m. , cotton w. d. , kassim n. e. , lazio t. j. w. , perley r. a. , condon j. j. , erickson w. c. , 2007 , preprint , astro - ph/0706.1191 cole s. , lacey c. g. , baugh c. m. , frenk c. s. , 2000 , mnras , 319 , 168 colless m. et al .
, 2001 , mnras , 328 , 1039 condon j. j. , 1997 , pasp , 109 , 166 condon j. j. , cotton w. d. , greisen e. w. , yin q. f. , perley r. a. , taylor g. b. , broderick j. j. , 1998 , aj , 115 , 1693 cruz m. j. et al . , 2006 , mnras , 373 , 1531 de breuck c. , van breugel w. , rttgering h. j. a. , miley g. , 2000 , a&as , 143 , 303 de breuck c. , van breugel w. , stanford s. a. , rttgering h. , miley g. , stern d. , 2002a , aj , 123 , 637 de breuck c. , tang y. , de bruyn a. g. , rttgering h. , van breugel w. , 2002b , a&a , 394 , 59 de breuck c. , hunstead r. w. , sadler e. m. , rocca - volmerange b. , klamer i. , 2004 , mnras , 347 , 837 de breuck c. , klamer i. , johnston h. , hunstead r. w. , bryant j. , rocca - volmerange b. , sadler e. m. , 2006 , mnras , 366 , 58 de propris r. et al . , 2002 , mnras , 329 , 87 eales s. , rawlings s. , law - green d. , cotter g. , lacy m. , 1997 , mnras , 291 , 593 fan x. et al . , 2004 ,
aj , 128 , 515 fan x. et al .
, 2006 , aj , 131 , 1203 fanaroff b. l. , riley j. m. , 1974 , mnras , 167 , 31p frater r. h. , brooks j. w. , whiteoak j. b. , 1992 , j. electrical electron . eng .
australia , 12 , 103 gardner f. f. , whiteoak j. b. , 1966 , ara&a , 4 , 245 gebhardt k. et al .
, 2000 , apj , 539 , l13 hambly n. c. et al . , 2001 ,
mnras , 326 , 1279 inskip k. j. , best p. n. , longair m. s. , mackay d. j. c. , 2002 , mnras , 329 , 277 jarvis m. j. , rawlings s. , eales s. , blundell k. m. , bunker a. j. , croft s. , mclure r. j. , willott c. j. , 2001 , mnras , 326 , 1585 johnston - hollitt m. , hollitt c. p. , ekers r. d. , 2004 , in uyaniker b. , reich w. , wielebinski r. , eds , the magnetized interstellar medium .
copernicus gmbh , katlenburg - lindau , p. 13
jones p. a. , mcadam w. b. , 1992 , apjs , 80 , 137 kauffmann g. , haehnelt m. , 2000 , mnras , 311 , 576 klamer i. j. , ekers r. d. , bryant j. j. , hunstead r. w. , sadler e. m. , de breuck c. , 2006 , mnras , 371 , 852 krolik j. h. , chen w. , 1991 , aj , 102 , 1659 large m. i. , mills b. y. , little a. g. , crawford d. f. , sutton j. m. , 1981 , mnras , 194 , 693 lumsden s. l. , nichol r. c. , collins c. a. , guzzo l. , 1992 , mnras , 258 , 1 matthews t. a. , morgan w. w. , schmidt m. , 1964 , apj , 140 , 35 mauch t. , murphy t. , buttery h. j. , curran j. , hunstead r. w. , piestrzynski b. , robertson j. g. , sadler e. m. , 2003 , mnras , 342 , 1117 mcgreer i. d. , becker r. h. , helfand d. j. , white r. l. , 2006 , apj , 652 , 157 mclure r. j. , dunlop j. s. , 2002 , mnras , 331 , 795 mesa d. , baccigalupi c. , de zotti g. , gregorini l. , mack k .- h . , vigotti m. , klein u. , 2002 , a&a , 396 , 463 mills b. y. , 1981 , pasa , 4 , 156 mills b. y. , aitchison r. e. , little a. g. , mcadam w. b. , 1963 , proc .
radio engrs aust . , 24 , 156 neeser m. j. , eales s. a. , law - green j. d. , leahy j. p. , rawlings s. , 1995 , apj , 451 , 76 odea c. p. , 1998 ,
pasp , 110 , 493 pentericci l. , van reeven w. , carilli c. l. , rttgering h. j. a. , miley g. k. , 2000 , a&as , 145 , 121 robertson j. g. , 1991 , aust . j. phys .
, 44 , 729 rocca - volmerange b. , le borgne d. , de breuck c. , fioc m. , moy e. , 2004 , a&a , 415 , 931 rttgering h. j. a. , lacy m. , miley g. k. , chambers k. c. , saunders r. , 1994 , a&as , 108 , 79 sault r. j. , teuben p. j. , wright m. c. h. , 1995 , in shaw r. a. , payne h. e. , hayes j. j. e. , eds , asp conf .
ser . 77 : astronomical data analysis software and systems iv , 77 , 433 somerville r. s. et al . , 2004 ,
apj , 600 , l135 spergel d. n. et al . , 2003 ,
apjs , 148 , 175 thompson a. r. , clark b. g. , wade c. m. , napier p. j. , 1980 , apjs , 44 , 151 tielens a. g. g. m. , miley g. k. , willis a. g. , 1979 , a&as , 35 , 153 tucci m. , martnez - gonzlez e. , toffolatti l. , gonzlez - nuevo j. , de zotti g. , 2004 , mnras , 349 , 1267 van breugel w. j. m. , stanford s. a. , spinrad h. , stern d. , graham j. r. , 1998 , apj , 502 , 614 van breugel w. , de breuck c. , stanford s. a. , stern d. , rttgering h. , miley g. , 1999 , apj , 518 , l61 venturi t. , bardelli s. , morganti r. , hunstead r. w. , 1998 , mnras , 298 , 1113 willott c. j. , rawlings s. , jarvis m. j. , blundell k. m. , 2003 , mnras , 339 , 173 | this paper introduces a new program to find high - redshift radio galaxies in the southern hemisphere through ultra - steep spectrum ( uss ) selection .
we define a sample of 234 uss radio sources with spectral indices @xmath0 ( @xmath1 ) and flux densities @xmath2 mjy in a region of 0.35 sr , chosen by cross - correlating the revised 408 mhz molonglo reference catalogue , the 843 mhz sydney university molonglo sky survey and the 1400 mhz nrao vla sky survey in the overlap region @xmath3 .
we present australia telescope compact array ( atca ) high - resolution 1384 and 2368 mhz radio data for each source , which we use to analyse the morphological , spectral index and polarization properties of our sample .
we find that 85 per cent of the sources have observed - frame spectral energy distributions that are straight over the frequency range 4082368 mhz , and that , on average , sources with smaller angular sizes have slightly steeper spectral indices and lower fractional linear polarization .
fractional polarization is anti - correlated with flux density at both 1400 and 2368 mhz .
we also use the atca data to determine observed - frame faraday rotation measures for half of the sample .
-.65 in surveys galaxies : active radio continuum : galaxies radio continuum : general polarization . |
Subsets and Splits