article
stringlengths 0
745k
| abstract
stringlengths 0
92.7k
|
|---|---|
a novel influenza strain termed influenza a(h1n1)v , first identified in mexico in march 2009 , has rapidly spread to different countries and is currently the predominant influenza virus in circulation worldwide @xcite . as of april 11 , 2010
, it has caused at least 17798 deaths in 214 countries @xcite .
the first confirmed case in india , a passenger arriving from the usa , was detected on may 16 , 2009 in hyderabad .
the initial cases were passengers arriving by international flights
. however , towards the end of july , the infections appeared to have spread into the resident population with an increasing number of cases being reported for people who had not been abroad .
as of 11 april 2010 , there have been 30352 laboratory confirmed cases in india ( out of 132796 tested ) and 1472 deaths have been reported , i.e. , @xmath2 of the cases which tested positive for influenza a(h1n1)v @xcite . to devise effective strategies for combating the spread of pandemic influenza a(h1n1 ) , it is essential to estimate the transmissibility of this disease in a reliable manner
. this is generally characterized by the reproductive number @xmath3 , defined as the average number of secondary infections resulting from a single ( primary ) infection .
a special case is the basic reproductive number @xmath4 , which is the value of @xmath3 measured when the overall population is susceptible to the infection as is the case at the initial stage of an epidemic .
estimate of the basic reproduction number for influenza a(h1n1)v in reports published from data obtained for different countries vary widely .
for example , @xmath4 has been variously estimated to be between 2.2 to 3.0 for mexico @xcite , 1.72 for mexico city @xcite , between 1.4 and 1.6 for la gloria in mexico @xcite , between 1.3 to 1.7 for the united states @xcite and 2.4 for victoria state in australia @xcite .
the divergence in the estimates for the basic reproductive number may be a result of under - reporting in the early stages of the epidemic or due to climatic variations .
they may also possibly reflect the effect of different control strategies used in different regions , ranging from social distancing such as school closures and confinement to antiviral treatments . in this paper , we estimate the basic reproductive number for the infections using the time - series of infections in india extracted from reported data . by assuming an exponential rise in the number of infected cases @xmath5 during the initial stage of the epidemic when most of the population is susceptible
, we can express the basic reproductive number as @xmath6 ( see , e.g. , ref .
@xcite , p. 19 ) , where @xmath7 is the rate of exponential growth in the number of infections , and @xmath8 is the mean generation interval , which is approximately equal to 3 days @xcite . using the time - series data we obtain the slope @xmath7 of the exponential growth using several different statistical techniques .
our results show that this quantity has a value of around 0.15 , corresponding to @xmath1 .
we used data from the daily situation updates available from the website of the ministry of health and family welfare , government of india @xcite . in our analysis , data up to september 30 , 2009 was used , corresponding to a total of 10078 positive cases . note that , after september 30 , 2009 , patients exhibiting mild flu like symptoms ( classified as categories a and b ) were no longer tested for the presence of the influenza a(h1n1 ) virus .
as the data exhibit very large fluctuations , with some days not showing a single case while the following days show extremely large number of cases , it is necessary to smooth the data using a moving window average .
we have used an @xmath9-day moving average ( @xmath10 ) , which removes large fluctuations while remaining faithful to the overall trend .
the incidence data for the 2009 pandemic influenza data in india immediately reveals that the disease has been largely confined to the urban areas of the country . indeed , 6 of the 7 largest metropolitan areas of india ( which together accommodate about 5 % of the indian population @xcite )
account for 7139 infected cases up to september 30 , 2009 , i.e. , @xmath11 of the data - set we have used .
figure [ figure1 ] shows the daily number of confirmed infected cases , as well as , the 5-day moving average from june 1 to september 30 , 2009 , for the country as a whole and the six major metropolitan areas which showed the highest incidence of the disease : hyderabad , delhi , bangalore , mumbai , chennai and pune .
the adjoining map shows the geographic locations of these six cities . in the period up to july 2009 , infections were largely reported in people arriving from abroad .
there is a marked increase in the number of infections towards the end of july and the beginning of august 2009 in all of these cities ( note that the ordinate is in logarithmic scale ) .
this is manifest as a sudden rise in the number of infected cases for the country as a whole , implying that the infection started spreading in the resident population in the approximate period of 28 july to august 12 .
estimated from the time - series data of number of infected cases , @xmath12infected , averaged over a 5-day period to smoothen the fluctuations ( d , solid curve ) .
the slope @xmath7 is calculated by considering the number of infected cases over a moving window having different sizes ( @xmath13 ) , ranging between 7 days and 36 days . by moving the starting point of the window across the period 1st june-20th august ( in steps of 1 day ) and calculating the best fit linear slope of the data on a semi - logarithmic scale ( i.e. , time in normal axis , number of infections in logarithmic axis ) we obtain an estimate of @xmath7 .
the arrow indicates the region between july 28-august 12 ( region within the broken lines ) , which shows the largest increase in number of infections within the period under study , corresponding to the period when the epidemic broke out in the resident population . over this time - interval
, the average of @xmath7 is calculated for the set of starting dates and window sizes over which ( b ) the correlation coefficient @xmath14 between log(@xmath12infected ) and @xmath15 , is greater than @xmath16 ( we consider @xmath17 in our analysis ) and ( c ) , the measure of significance for the correlation @xmath18 . ]
figure [ figure2 ] ( a ) shows the exponential slope @xmath7 estimated in the following way . the time - series of the number of infections is first smoothed by taking a 5-day moving average .
the resulting smoothed time - series is then used to estimate @xmath7 by a regression procedure applied to the logarithm of the number of infected cases [ log(@xmath12infected ) ] across a moving window of length @xmath13 days .
the origin of the window is varied across the period 1st june to 20th august ( in steps of 1 day ) .
we then repeat the procedure by varying the length of the window over the range of 7 days to 36 days . to quantify the quality of regression we calculate the correlation coefficient @xmath14 [ fig .
[ figure2 ] ( b ) ] between log ( @xmath12infected ) and time ( in days ) , and its measure of significance @xmath19 [ fig .
[ figure2 ] ( c ) ] .
the correlation coefficient @xmath14 is bounded between @xmath20 and 1 , with a value closer to 1 indicating a good fit of the data to an exponential increase in the number of infections .
the measure of significance of the fitting is expressed by the corresponding @xmath19-value , which expresses the probability of obtaining the same correlation by random chance from uncorrelated data .
the average of the estimated exponential slope @xmath7 is obtained by taking the mean of all values of @xmath7 obtained for windows originating between july 28-aug 12 and of various sizes , for which the correlation coefficient @xmath21 ( we consider @xmath22 in our analysis ) and the measure of significance @xmath18 . for comparison
, we show again in figure [ figure2 ] ( d ) the number of infected cases of h1n1 in india ( dotted ) together with its 5-day moving average ( solid line ) .
the horizontal broken lines running across the figure indicate the period between july 28 and august 12 which exhibited the highest increase in number of infections within the period under study ( from 1st june to 30th september ) . of the variation in log(@xmath12infected ) with time @xmath15 , as a function of the threshold of correlation coefficient , @xmath16 , used to filter the data .
the averaging is performed for infections occurring within the period july 28-august 12 ( for details see caption to fig .
[ figure2 ] ) .
different symbols indicate the actual daily time - series data ( squares ) and the data smoothed over a moving @xmath9-day period , with @xmath9 = 2 ( right - pointed triangle ) , 3 ( diamond ) , 4 ( inverted triangle ) , 5 ( circle ) and 10 ( triangle ) . the significance of the correlation between log(@xmath12infected ) with time @xmath15 , @xmath23 for all data points used in performing the average .
note that for @xmath24 the data show very similar profiles for variation of @xmath25 with @xmath16 , indicating the robustness of the estimate with respect to different values of @xmath9 used .
the sudden increase in the value of the average slope around @xmath26 implies that beyond this region the slope depends sensitively on the cutoff value .
considering the region where the variation is more gradual gives us an approximate value of the slope @xmath27 , corresponding to a basic reproduction number @xmath1 . ]
figure [ figure3 ] shows the average exponential slope @xmath25 as a function of @xmath16 , calculated for the original data and for different periods @xmath9 over which the moving average is taken ( @xmath28 and 10 ) . for @xmath29 3 - 5
, the data show a similar profile indicating the robustness of the estimate of the average exponential slope @xmath25 with respect to different values of @xmath9 .
the sudden increase in @xmath30 around @xmath26 implies that beyond this region the slope depends sensitively on the cutoff value .
considering the region where the variation is smoother gives an approximate value @xmath27 , corresponding to a basic reproductive number for the epidemic @xmath31 , assuming the mean generation interval , @xmath32 days .
we compute the confidence bounds for the estimate of @xmath4 from the 5-day moving average time - series by using the _ confint _ function of the scientific software matlab @xcite .
this function generates the goodness of fit statistics using the solution of the least squares fitting of log(@xmath12infected ) to a linear function .
it results in a mean value @xmath33 , with the corresponding @xmath34 confidence intervals calculated as [ 0.116 , 0.206 ] , consistent with our previous estimate of @xmath35 . ,
calculated for different periods ( with the abscissa indicating the starting date and the symbol indicating the duration ) from the 5-day moving average time - series data of infected cases in india .
the curves corresponding to the periods of different durations ( 14 - 16 days ) intersect around july 31 , 2010 , indicating that the value of the average exponential slope is relatively robust with respect to the choice of the period about this date .
( b ) the distribution of bootstrap estimates of the exponential slope for the period july 31 to august 15 , 2009 .
the average slope @xmath36 obtained from 1000 bootstrap samples is 0.166 with a standard deviation of 0.024 , which agrees with the approximate value of @xmath37 ( corresponding to @xmath38 ) calculated in fig .
[ figure3 ] . ]
we have also used bootstrap methods to estimate the exponential slope , @xmath7 .
this involves selecting random samples with replacement from the data such that the sample size equals the size of the actual data - set .
the same analysis that was performed on the empirical data is then repeated on each of these samples .
the range of the estimated values @xmath39 calculated from the random samples allows determination of the uncertainty in estimation of @xmath7 .
[ figure4 ] ( a ) shows the average , @xmath40 , calculated for different periods ( with abscissa indicating the starting date and the symbol indicating the duration of the period ) from the 5-day moving average time - series data of infected cases . the curves corresponding to the periods of different durations ( 14 - 16 days ) intersect around july 31 , 2010 , indicating that the value of the average exponential slope is relatively robust with respect to the choice of the period about this date .
the average value of the bootstrap estimates @xmath39 at the intersection of the three curves is 0.15 , in agreement with our earlier calculations of @xmath7 .
[ figure4 ] ( b ) shows the distribution of the bootstrap estimates of the exponential slope for a particular period , july 31 to august 15 , 2009 .
the average slope @xmath40 obtained from 1000 bootstrap samples for this period is 0.166 with a standard deviation of 0.024 , which indicates that the spread of values around the average estimate of @xmath36 = 0.15 is not large .
this confirms the reliability of the estimated value of the exponential slope , and hence of our calculation of the basic reproductive number .
it may appear surprising that there was a very high number of infections in pune ( 1238 positive cases up to september 30 ) , despite it being less well - connected to the other major metropolitan cities of india , in comparison to urban centres that did not show a high incidence of the disease .
for example , the kolkata metropolitan area , which has a population around three times the population of the pune metropolitan area @xcite , had only 113 positive cases up to september 30 .
this could possibly reflect the role of local climatic conditions : pune , located at a relatively higher altitude , has a generally cooler climate than most indian cities .
in addition , the close proximity of pune to mumbai and the high volume of road traffic between these two cities could have helped in the transmission of the disease .
another feature pointing to the role of local climate is the fact that in chennai , most infected cases were visitors from outside the city , while in pune , the majority of the cases were from the local population , even though the total number of infected cases listed for the two cities in our data - set are comparable ( 928 in chennai and 1213 in pune ) .
this suggests the possibility that the incidence of the disease in pune could have been aided by its cool climate , in contrast to the hotter climate of the coastal city of chennai .
the calculation of @xmath4 for india assumes well - mixing of the population ( i.e. , homogeneity of the contact structure ) among the major cities in india .
given the rapidity of travel between the different metropolitan areas via air and rail , this may not be an unreasonable assumption .
however , some local variation in the development of the epidemic in different regions can indeed be seen ( fig .
[ figure1 ] ) . around the end of july ,
almost all the cities under investigation showed a marked increase in the number of infected cases - indicating spread of the epidemic in the local population .
this justifies our assumption of well - mixing in the urban population over the entire country for calculating the basic reproductive number . to conclude
, we stress the implications of our finding that the basic reproductive number for pandemic influenza a(h1n1)v in india lies towards the lower end of the values reported for other affected countries .
this suggests that season - to - season and country - to - country variations need to be taken into account in order to formulate strategies for countering the spread of the disease .
evaluation of the reproductive number , once control measures have been initiated , is vital in determining the future pattern of spread of the disease .
ministry of health and family welfare , government of india , situation update on h1n1 , 11 april 2010 .
available from : ` http://mohfw-h1n1.nic.in/documents/pdf/situational ` ` updatesarchives / april2010/situational%20updates%20 ` ` on%2011.04.2010.pdf ` bolle , p. y. , bernillon , p. and desenclos , j. c. , a preliminary estimation of the reproduction ratio for new influenza a(h1n1 ) from the outbreak in mexico , march - april 2009 .
_ euro surveill . _ , 2009 , * 14*(19 ) , pii : 19205 .
cruz - pacheco , g. , duran , l. , esteva , l. , minzoni , a. a. , lopez - cervantes , m. , panayotaros , p. , ahued ortega , a. and villasenor ruiz , i. , modelling of the influenza a(h1n1)v outbreak in mexico city , april - may 2009 , with control sanitary measures .
_ euro surveill . _ , 2009 , * 14*(26 ) , pii:19254 .
fraser , c. , donnelly , c. a. , cauchemez , s. , hanage , w. p. , van kerkhove , m. d. , hollingsworth , t. d. , griffin , j. , baggaley , r. f. , jenkins , h. e. , lyons , e. j. , jombart , t. , hinsley , w. r. , grassly , n. c. , balloux , f. , ghani , a. c. , ferguson , n. m. , rambaut , a. , pybus , o. g. , lopez - gatell , h. , alpuche - aranda , c. m. , chapela , i. b. , zavala , e. p. , guevara , d. m. , checchi , f. , garcia , e. , hugonnet , s. , roth , c. ; who rapid pandemic assessment collaboration , pandemic potential of a strain of influenza a(h1n1 ) : early findings .
_ science _
, 2009 , * 324 * pp .
1557 - 1561 .
yang , y. , sugimoto , j. d. , halloran , m. e. , basta , n. e. , chao , d. l. , matrajt , l. , potter , g. , kenah , e. and longini , i. m. , the transmissibility and control of pandemic influenza a(h1n1 ) virus .
_ science _ , 2009 , * 326 * , pp .
729 - 733 .
mcbryde , e. , bergeri , i. , van gemert , c. , rotty , j. , headley , e. , simpson , k. , lester , r. , hellard , m. and fielding , j. , early transmission characteristics of influenza a(h1n1)v in australia : victorian state , 16 may - 3 june 2009 .
_ euro surveill .
_ , 2009 , * 14*(42 ) , pii : 19363 . | we analyze the time - series data for the onset of a(h1n1)v influenza pandemic in india during the period june
1- september 30 , 2009 . using a variety of statistical fitting procedures
, we obtain a robust estimate of the exponential growth rate @xmath0 .
this corresponds to a basic reproductive number @xmath1 for influenza a(h1n1)v in india , a value which lies towards the lower end of the range of values reported for different countries affected by the pandemic . |
compact symmetric objects ( csos ) are characterized by compact bright radio emission at scales @xmath1kpc with lobes on both sides of a central engine ( wilkinson et al.@xmath01994 ) .
csos tend to be associated with bright elliptical galaxies , although some appear to be associated with quasars and even spiral galaxies ( readhead et al.@xmath01996 ; augusto et al.@xmath02006 ; perlman et al.@xmath01996 ) .
many csos have an overall gigahertz - peaked spectrum ( gps ) , and the cores are often undetected at frequencies of 5@xmath0ghz or higher ( taylor et al.@xmath01996 ) .
csos are very luminous radio sources , significantly brighter ( by more than a factor of 10 ) than the luminosity threshold between fr@xmath0i and ii radio galaxies ( i.e. , @xmath2w@xmath0hz@xmath3sr@xmath3 at 178@xmath0mhz , fanaroff & riley 1974 ; readhead et al.@xmath01996 ) .
it has been proposed that the small source size could be caused by `` frustration '' of the radio jets , which fail to escape from a dense galactic medium ( e.g. , van breugel et al.@xmath01984 ; odea et al.@xmath01991 ; de young 1993 ) . however
, a different scenario is currently preferred : csos may be young progenitors ( @xmath410@xmath5 10@xmath6years ) of powerful extended radio sources ( e.g. , phillips & mutel 1982 ; augusto et al.@xmath02006 ; augusto 2007 ) . based on xmm - newton observations of five csos , vink et al.@xmath0(2006 ) found that their absorption column density ( @xmath7@xmath8 ) is on average greater than that of broad - line radio galaxies , but similar to column densities observed toward narrow - line and other types of radio galaxies ( nlrgs , rg ; see sambruna et al.@xmath01999 for definitions ) .
these results argue against the frustration scenario because much higher x - ray absorption column densities than found by vink et al .
( 2006 ) are expected if the expansion of the radio jets were confined ( see also guainazzi et al . 2006 ) .
in contrast , the dynamical age of csos measured from the expansion velocity of hotspots in the radio lobes supports the youth hypothesis ( taylor et al.@xmath02000 , polatidis & conway 2003 ) .
the youth interpretation is also supported by limits on molecular gas content ( odea et al.@xmath02005 ) , observations ( pihlstrm et al.@xmath02003 ) , and under - luminous o@xmath0iii line emission ( vink et al.@xmath02006 ) .
the study of csos is therefore fundamental for the understanding of the formation and evolution of radio galaxies .
in addition , csos are good candidates to investigate the fueling of supermassive black holes given that the compact radio emission enables very high angular resolution observations of circumnuclear gas via absorption studies . in this work we report high sensitivity and high angular resolution observations of toward b2352 + 495 ; an archetype of the cso class . according to curran et al.@xmath0(2008 ) , absorption associated with sources at redshifts greater than @xmath9 has been detected toward 37 sources , 13 of them classified as csos , gps , or hfp ( high frequency peaked ) sources .
an overview of absorption in compact radio sources was published by vermeulen et al.@xmath0(2003a ) .
they report an detection rate of @xmath10 based on westerbork synthesis radio telescope ( wsrt ) observations of 59 sources ( i.e. , 19 detections ) . of the 59 sources ,
11 were classified as cso by augusto et al.@xmath0(2006 ) and polatidis & conway ( 2003 ) . from this subsample of 11 csos ,
6 ( 54@xmath11 ) show absorption .
other studies show similar results , e.g. , peck & taylor ( 2001 ) report a cso detection rate of @xmath12 , and gupta et al.@xmath0(2006 ) report a detection rate of @xmath13 in gps sources ( see also vermeulen 2002 ) .
the optical depth range of the 6 csos with absorption from vermeulen et al.@xmath0(2003a ) is 0.2 to 1.7@xmath11 , with linewidths between 13 and 297 and column densities between @xmath14 and @xmath15@xmath8 . to estimate column densities , vermeulen et al.@xmath0(2003a )
assumed that the atomic gas was uniformly covering the radio continuum ( covering factor of 1 ) and a spin temperature of 100@xmath0k , that may underestimate the spin temperature by more than an order of magnitude if the gas were located at parsec - scales from the agn .
thus , the column densities reported by vermeulen et al.@xmath0(2003a ) are lower limits . including the results reported in this work , imaging at high angular resolution
has been reported toward five csos ( table 1 ) .
we also list in table 1 all non - cso agn and starbursts that have been imaged at high angular resolution in , i.e. , in interferometric observations using baselines greater than 100@xmath0 km .
in the rest of this section we review the most salient characteristics of the csos that have been studied in at high angular resolution : _ b1946 + 708 : _ high angular resolution observations were conducted by peck et al.@xmath0(1999 ; see also peck & taylor 2001 ) .
they found extended and multi - peaked absorption throughout the radio continuum source .
the peak optical depths range between 3 and 7@xmath11 . the maximum column density ( @xmath16@xmath8 ; peck & taylor 2001 )
was found toward the agn core , which also shows the greatest velocity dispersion ( @xmath17 ) .
the extent of the absorption in b1946 + 708 implies that the atomic gas is tracing a thick torus ( thickness @xmath18pc ) . _
[email protected] : _ the host galaxy is a bright elliptical at z = 0.0592 . it was classified as cso based on vlbi continuum observations that revealed a double radio lobe with a flat spectrum radio core ( cotton et al.@xmath01995 ; giovannini et al.@xmath0 2001 , giroletti et al.@xmath02003 ; see also augusto et al.@xmath01998 ) .
detection of in [email protected] was reported by mirabel ( 1990 ) based on arecibo observations .
mirabel ( 1990 ) found two absorption features , a broad ( fwhm = 133 ) component and a double peaked narrow component ( fwhm = 6 and 16 ) redshifted approximately 400 from the optical velocity .
subsequent vlbi observations by conway ( 1999 ) showed that the broad line covers one of the radio lobes and just partially the second , indicating a sharp edge of the atomic gas torus .
the narrow component is detected toward both radio lobes .
the maximum optical depth of the lines is @xmath19 to @xmath20 .
mirabel ( 1990 ) interpreted the narrow absorption as being caused by clouds similar to high velocity clouds in our galaxy . _
[email protected] : _ in contrast to the two previous csos , in [email protected] has been detected only toward one of the radio lobes .
the absorption is characterized by two broad ( fwhm@xmath21100 ) features from two separated ( in line - of - sight and velocity ) clouds .
the peak optical depth of the lines is @xmath22 .
rodrguez et al.@xmath0(2009 ) concluded that the absorption likely originates in a thick circumnuclear torus . _
pks@xmath23 : _ as in [email protected] , absorption is detected only toward one radio lobe ( perlman et al.@xmath02002 ) .
the optical depth is @xmath24 and the line is quite narrow , between 16 and 18 .
perlman et al.@xmath0(2002 ) concluded that the absorption is likely from a giant molecular cloud in the outer disk of the host galaxy .
although pks@xmath23 was classified by perlman et al.@xmath0(1996 ) and gugliucci et al.@xmath0(2005 ) as a cso , it has several characteristics that differ form a typical cso . in particular , it was originally classified as a bl lac object based on its optical spectrum ( bregman et al . 1981 ) , the nucleus dominates by more than 50@xmath11 of the total cm flux density , the radio emission is variable , and the host appears to be a spiral galaxy ( see perlman et al.@xmath01996 ) . finally , two additional sources ( b1934@xmath25638 and [email protected] ) are worth discussing in detail .
b1934@xmath25638 is a compact - double gps source that has been classified as a cso ( augusto et al .
2006 ) , however , no flat - spectrum radio core has been detected ( tzioumis et al . 2002 ) . even though no imaging has been reported at high angular resolution
, the absorption is against the compact radio continuum as shown by detection of absorption with three baselines of the lba ( vron - cetty et al .
the absorption is narrow ( fwhm = 18 ) and redshifted 260 from the optical systemic velocity , thus , the is probably not located in the nuclear region of the galaxy .
we include b1934@xmath25638 in table 1 although no vlbi imaging is available and the radio core has not been detected . in the case of [email protected] ,
high angular resolution absorption observations were reported by morganti et al.@xmath0(2004 ) .
they found absorption against a weak counterjet and concluded that the likely traces a jet / cloud interaction and not a circumnuclear torus or disk .
this source was classified as a cso by lister et al.@xmath0(2003 ) despite several atypical characteristics .
for instance , while the 1.3@xmath0ghz continuum image reveals a s - shaped distribution with a total extend of @xmath26pc ( morganti et al.@xmath02004 ) , the 15@xmath0ghz vlba observations are consistent with a core - jet morphology plus a weak counterjet ( i.e. , inconsistent with the typical symmetrical morphology of csos ) .
[email protected] has no clear hotspots , it is classified as a css ( compact steep spectrum ) , the radio jet exhibits superluminal motions , some isolated features have unusually high ( @xmath27 ) linear polarization in comparison to other csos , and more importantly , there appears to be continuity between the @xmath26pc compact radio jets and large scale ( @xmath28kpc ) radio lobes ( stanghellini et al.@xmath02005 ) .
therefore , we do not consider [email protected] a bonafide member of the cso class ( see table 1 ) .
b2352 + 495 has been studied in detail by a number of authors ( e.g. , readhead et al.@xmath01996 , taylor et al.@xmath01996 ) ; in this section we summarize the characteristics of this object : _ radio continuum : _
the most salient property of b2352 + 495 is its highly symmetric morphology at radio wavelengths .
it exhibits an s - shaped morphology with a total extent of @xmath29pc ( readhead et al.@xmath01996 ) .
high angular resolution radio continuum observations from 610@xmath0mhz to 15@xmath0ghz reveal symmetric hotspots and lobes ; remarkably similar to the morphology of fr ii objects ( but at much smaller scale ) .
the dynamical age of the system is @xmath30yr based on proper motion studies of the hotspots in the radio lobes ( taylor et al.@xmath02000 ) .
taylor et al.@xmath0(1996 ) detected the radio core of the system , i.e. , the putative location of the central engine .
the core is approximately located symmetrically between the radio lobes , and drives a collimated jet that is brighter toward the northern side of the core .
the radio jet is detected up to the brightest radio emission in the system ( b1 and b2 ; taylor et al 1996 ; see also @xmath31 ) which is also located between the two radio lobes .
the hotspots in the radio lobes show no superluminal motion ( conway et al.@xmath01992 ) .
multi - frequency vla observations give fractional polarization limits of @xmath32 for this source ( rudnick & jones 1983 ) .
the radio spectral energy distribution of b2352 + 495 peaks at @xmath33ghz , thus by definition , it is a gps object ( conway et al.@xmath01992 ) .
the total flux density of b2352 + 495 shows only modest ( @xmath34 ) variability on time scales of months ; no systematic long term variability was found over a period of @xmath35years ( waltman et al.@xmath01991 ; conway et al.@xmath01992 ) .
the radio luminosity is @xmath36w@xmath0hz@xmath3 ( @xmath375@xmath0ghz at the emitted frame ) , i.e. , some 20 times brighter than the luminosity boundary between fr is and fr iis ( readhead et al.@xmath01996 ) .
_ optical / infrared counterpart : _
b2352 + 495 was not detected by iras to sensitivity limits well below the corresponding luminosity of arp@xmath0220 ( at the redshifted frequency ) , thus b2352 + 495 is significantly less luminous than ultra - luminous infrared galaxies ( readhead et al.@xmath01996 ) .
snellen et al.@xmath0(2003 ) conducted imaging and spectroscopic optical observations of b2352 + 495 .
they found that the radio source resides in an optical host in the fundamental plane of elliptical galaxies , i.e. , the host is a normal elliptical.@xmath0snellen et al.@xmath0(2003 ) report a redshift of [email protected] , stellar velocity dispersion of 201@xmath3817 , and an effective radius of [email protected] ( [email protected]@xmath0kpc).mas correspond to 36.3@xmath0pc assuming a cosmology of @xmath39@xmath0mpc@xmath3 , @xmath40 , @xmath41 . ] based on the stellar velocity dispersion , the mass of the supermassive black hole at the center of b2352 + 495 is of the order of @xmath42 .
the absolute magnitude of the galaxy is @xmath43 with an apparent magnitude of @xmath44 ( readhead et al.@xmath01996 ) .
the optical luminosity of b2352 + 495 is @xmath45 , which is smaller than the typical optical luminosity of classical doubles ( fr iis ; @xmath46 ; owen & white 1991 ) and cd galaxies ( @xmath47 ; schechter 1976 ) .
the absence of balmer lines and strong blue continuum in the spectrum demonstrates no starburst activity .
in contrast , a number of emission lines are detected typical of active elliptical galaxies ( readhead et al.@xmath01996 ) . the agn contribution to the total optical continuum
is @xmath48 ( snellen et al.@xmath02003 ) . _
x - ray properties : _ high sensitivity xmm - newton observations of b2352 + 495 were conducted by vink et al.@xmath0(2006 ) .
they detected an x - ray source with a luminosity of @xmath49erg@xmath0s@xmath3 ( @xmath50kev band , ignoring absorption ) .
based on the x - ray detection , they estimated an intrinsic absorption column density of ( [email protected])@xmath51@xmath8 , which is similar to the column densities found toward extended radio galaxies .
_ prior observations : _ absorption of the transition in b2352 + 495 was first detected by vermeulen et al.@xmath0(2003a ) based on wsrt observations . the absorption in b2352 + 495 consists of a broad line ( fwhm = 82 ) and a redshifted narrow component ( fwhm = 13 ) .
the peak flux density of the two absorption components is similar ( @xmath52mjy ) .
preliminary results of the data reported in this paper were discussed by vermeulen ( 2002 ) .
very long baseline interferometer ( vlbi ) observations were conducted on december 01 and 02 , 1997 , with the nrao very long baseline array ( vlba ) and the wsrt .
the data were correlated with the vlba correlator at the p.v.d .
science operations center in socorro , new mexico .
the target source ( b2352 + 495 ) was observed during approximately 8.5@xmath0hours each day ( @xmath53hours total ) in the 21@xmath0 cm transition ( @xmath54mhz ) which is redshifted to a sky frequency of @xmath55mhz ( z = [email protected] , snellen et al.@xmath02003 ; v@xmath56 = 71321@xmath3848 , optical definition ) .
we observed with a bandwidth of 8@xmath0mhz , 1024 channels , dual circular polarization , and an initial channel width of @xmath57khz smoothed to a spectral resolution of 15.6@xmath0khz ( 4.1 ) .
only eight of the ten vlba antennas were used because of rfi filters that block access at fort davis and kitt peak .
we observed 3c84 for bandpass calibration .
all data reduction was done in the nrao package aips using standard vlba spectral - line calibration procedures .
the wsrt data were significantly affected by rfi and we decided not to use them after careful inspection of the results obtained including and excluding wsrt .
we used line - free channels of the b2352 + 495 observations for fringe fitting .
a continuum data set was obtained by averaging the line - free channels of the visibility file ; the continuum data were used for self - calibration .
the absolute astrometry of the data was therefore lost , and we simply assign the radio continuum peak to the phase tracking center ( ra = 225509.4581 , decl .
= + 495008.340 , j2000 ) .
given the non - standard frequency of the observations , a reliable flux density calibration using the t@xmath58 method was not possible , thus we scaled the flux density to a total radio continuum emission of 2.55@xmath0jy at 1.14@xmath0ghz .
this value was obtained by interpolating the flux density measurements reported by readhead et al.@xmath0 ( 1996 ) .
the use of the readhead et al.@xmath0(1996 ) values to interpolate the flux density is valid because the radio emission is dominated by extended radio lobes ( see below ) and not by a compact core ( and thus it is less likely to exhibit strong variability ) . despite the uncertainty on the flux density calibration , the peak flux density values of the narrow and broad lines detected in this work are very similar to those obtained with wsrt ( vermeulen et al.@xmath02003a ) .
in addition , our discussion of absorption is based on opacity considerations , which are not affected by systematic flux density calibration errors .
we transferred the self - calibration tables from the continuum data set to the line file , and subtracted the radio continuum using the task uvlin in aips .
we used a brigg s robust 0 weighting for the final imaging that resulted in a @xmath59mas ( @xmath60pc linear scale ) , pa = 0.5 , rms = 0.71 ( in continuum ) and 7.1 ( per frequency channel ) .
all velocities reported in the paper are in the rest frame of the cso ( assuming @xmath61 ) unless indicated otherwise .
we detected both radio continuum and compact absorption in b2352 + 495 . in figure 1
we show the spectrum ( top panel ) ; the bottom panel shows the 1.14@xmath0ghz continuum and the zero velocity moment ( integrated intensity ) map of the detection .
the radio continuum has a symmetric s - shaped distribution that is consistent with previous radio continuum images of the source ( e.g. , wilkinson et al.@xmath01994 , readhead et al.@xmath01996 , owsianik et al.@xmath01999 ) .
absorption is only detected toward the brightest continuum emission close to the center of the cso . in figure
2 we show the optical depth channel maps ; only signal above 4@xmath62 in the continuum and data were used to calculate the opacity .
the absorption spectrum has two features , a broad component ( @xmath63 , fwhm @xmath64 , @xmath65 ) that is almost centered at the systemic velocity of the galaxy ( centroid heliocentric velocity = @xmath66 ; systemic velocity v@xmath56 = 71321@xmath3848 , optical definition ) , and a narrow component ( fwhm @xmath67 ) redshifted from systemic by @xmath68 .
our measurements are therefore consistent with those of vermeulen et al.@xmath0(2003a ) within our velocity resolution.(2003a ) , it should read 1147.0 instead of 1147.2@xmath0mhz . ]
as in the case of the gas associated with the nucleus of b1946 + 708 ( peck et al.@xmath01999 ) , we estimate the column density assuming a covering factor of 1 and a spin temperature of 8000@xmath0k , i.e. , assuming that the gas is directly associated with the agn . using the standard column density / opacity relation ( e.g. , rohlfs & wilson 2000 ) , we estimate a total column density of @xmath69 and @xmath70@xmath8 for the broad and narrow lines , respectively ( only statistical errors from the fit were used to estimate the column density uncertainty ) .
the values are greater than the b2352 + 495 column densities reported by vermeulen et al.@xmath0(2003a ) , however , as mentioned above , their values are lower limits because of the assumption of uniform coverage of the radio continuum ( which is not the case , see figure 1 ) and the assumption of a lower spin temperature ( @xmath71k ) . if the clouds have physical dimensions comparable to the synthesized beam ( note that there appears to be structure in the opacity channels , figure 2 ) , then the density would be between @xmath72 and 10@xmath73@xmath74 , corresponding to total masses of @xmath75 and @xmath76 for the broad and narrow features , respectively .
if the absorption profile of the broad component is the superposition of many narrow features , and assuming a typical optical depth of @xmath77 for the narrow features , then we could not have detected absorption ( at a @xmath783@xmath62 level ) anywhere except toward the brightest continuum region at the center of the cso ( i.e. , the region where absorption was indeed detected ) .
this implies that , given our sensitivity limit , our data do not set significant constraints on the scale height of the distribution .
as discussed below , we favor the interpretation that the broad component is associated with rotation of a disk / torus around the supermassive black hole in b2352 + 495 .
the detection of a narrow feature redshifted with respect to the systemic velocity indicates infall .
however , we can not establish whether the feature is associated with circumnuclear material or with a foreground cloud in the host galaxy , e.g , whether it is located in the narrow - line region ( nlr ; e.g. , b2050 + 364 , vermeulen et al.@xmath02006 ) or associated with a high velocity cloud .
we note that other studies ( e.g. , van gorkom et al.@xmath01989 , peck & taylor 1998 , conway 1999 ) have also found evidence for infalling material in galaxies with compact radio cores ( see however vermeulen et al.@xmath02003a ) .
we now discuss the nature of the radio continuum having the absorption . in principle , the brightest 1.14@xmath0ghz continuum source ( located close to the center of the two radio lobes , figure 1 ) is suggestive of the active core , i.e. , the location of the supermassive black hole . however , comparing our continuum data with higher angular resolution observations reveal a different scenario . in figure 3 ( upper panel )
we show our 1.14@xmath0ghz continuum map superimposed on the 15.4@xmath0ghz observations of taylor et al.@xmath0(1996 ) .
both continuum data sets were self - calibrated , and thus no absolute astrometry is available . to overlay the two images
, we used an average offset computed from the peak in the northern and the southern radio lobe emission ; we assume that position shifts due to opacity effects are small there , since their measured separation is the same within the errors at 1.14@xmath0ghz and 15.4@xmath0ghz .
the brightest radio emission at the center of the radio lobes was _ not _ used to estimate the offset .
taylor et al.@xmath0(1996 ) found a compact radio source characterized by an inverted spectral index ( @xmath79 ; @xmath80 ) , which convincingly indicates the location of the active core ( see figure 3 ) . assuming a flat spectral index between 1 and 8@xmath0ghz , the flux density of the core at 1@xmath0ghz is @xmath81mjy .
our optical depth limit toward the radio core is @xmath820.1 ( 3@xmath62 ) .
a prominent radio jet is detected to the north of the core , and a weaker counterjet to the south ( taylor et al.@xmath01996 ; 2000 ) .
the northern edge of the jet is coincident with the brightest 1.14@xmath0ghz continuum source reported in this work , implying that the absorption is not against continuum from the core , but radio emission from the jet . among other possibilities ,
the detection of the radio jet and weaker counterjet could be due to relativistic beaming ( see taylor et al.@xmath0[1996 ] for this and other interpretations ) . the absorption in b2352
+ 495 is thus similar to the absorption observed in b0402 + 379 ( rodrguez et al.@xmath02009 ) and ngc@xmath04151 ( mundell et al.@xmath01995 , 2003 ) where is detected only toward one side of the radio jet .
it is interesting to note that the radio continuum source toward which the is detected ( figure 3 ) is the location where the radio jet begins to change its position angle .
this would suggest that , as for example in the case of 3c@xmath0236 ( conway 1999 ) , the absorption marks the location of a jet / cloud interaction responsible for disturbing ( and possibly bending ) the jet .
however , the b2352@xmath83495 morphology is quite symmetric in both radio lobes which suggests that the deviation of the radio jets is not due to local jet / cloud interactions but governed by the large scale magnetohydrodynamical properties of the galactic nucleus . in the next section ,
we propose a different interpretation for the nature of the absorption in b2352 + 495 .
it is difficult to reliably investigate the velocity field of the broad line given that the absorption is only detected toward the central continuum source and , as shown in figure 2 , the velocity field is complex .
nevertheless , a velocity gradient may be present : the two channels with stronger absorption at velocities above 10 from systemic appear to have stronger absorption toward the west , whereas most of the blueshifted absorption is stronger toward the east ( velocity channels @xmath84 from systemic , see figure 2 ) .
figure 4 shows the position - velocity diagram of the broad absorption line in the east - west direction . as also seen in the individual channel maps ,
there is a hint of a velocity gradient between blueshifted and redshifted material . to illustrate the possible gradient
, we show in figure 3 ( top panel ) the opacity channels that correspond to the rest frame velocities of 17.6 and @xmath85 ( fig . 2 ) .
the existence of a velocity gradient is tantalizing given that it is approximately centered at the systemic velocity of the galaxy , and the position angle of the gradient is almost perpendicular to the north - south jet , i.e. , the gas could be tracing some kind of rotating structure around the supermassive black hole .
we explore this hypothesis in the rest of this section , however , we stress that more sensitive global - vlbi observations are required to confirm the velocity gradient .
to investigate whether rotation around the supermassive back hole is physically plausible , we followed the formulation presented by rodrguez et al.@xmath0(2009 ; see their appendix ) . to model the rotating structure , we used the position and velocity of the peak optical depth of the 17.6 and @xmath2531.4 channels ( figure 2 ) .
the position of the radio core ( figure 3 upper panel ; see also taylor et al.@xmath01996 ) was assumed to be coincident with the center of mass of the system . for a black hole mass of 10@xmath86 ( snellen et al.@xmath02003 ) and assuming circular trajectories coplanar with the supermassive black hole , we obtain that the orbit of the clouds from the 17.6 and @xmath2531.4 channels would have a radius of @xmath87pc with an orbital inclination of @xmath88 with respect to the plane of the sky . such an orbit is reasonable
given the expected distribution of atomic clouds around agns ( e.g. , peck 2005 ) .
similar results are obtained in the case that the orbital plane of the clouds is not coplanar with the supermassive black hole but located above by distances of the order of the projected separation between the agn core and the absorption .
this estimate is clearly a zero - order approximation because non - radial inhomogeneities in the gravitational field due to stars and gas are neglected .
in addition , the radius of influence of the supermassive black hole ( @xmath89 , where @xmath90 is the gravitational constant , @xmath91 is the mass of the black hole , and @xmath92 is the stellar velocity dispersion ; e.g. , neumayer et al .
2007 ) is @xmath93pc , thus , the stellar mass enclosed by the atomic gas is likely comparable to the black hole mass . in figure 3 ( bottom panel )
we show two possibilities for the nature of the distribution , i.e. , a thick torus ( characterized by an scale height similar or greater than the length of the radio source ) or a thin ( most likely warped , e.g. , pringle 1996 ) disk .
the obvious method to discriminate between these possibilities is to check whether there is toward the southern radio lobe .
however , as mentioned above , our optical depth sensitivity limit does not set significant constraints on the distribution , i.e. , clouds like those detected toward the brightest radio continuum emission could be present throughout the region . an indirect approach to discriminate between the models
is to compare our observations with lower angular resolution data .
as mentioned in @xmath94 , the peak flux density measured with wsrt ( vermeulen et al.@xmath02003a ) is approximately the same peak intensity we measured with the vlba ( figure 1 ) and the linewidths are also consistent .
thus , in contrast to other systems where vlbi observations resolve / filter out significant signal ( e.g. , ic@xmath05063 , oosterloo et al.@xmath02000 ) , the in b2352 + 495 is compact and mostly distributed along the central continuum source , i.e. , consistent with the thin disk interpretation .
the thin disk interpretation is also circumstantially supported by x - ray observations .
vink et al.@xmath0(2006 ) reported a x - ray absorption column density of @xmath95@xmath8 , which is similar to the column density of the narrow component , but several times smaller than the value of the broad line .
if the absorption originates from an inclined ( and warped ) thin disk that does not substantially obscure the agn ( figure 3 , bottom panel ; see also fig . 5 of orienti et al.@xmath02006 )
then as observed , the column density toward the agn derived from x - ray observations is expected to be smaller than that of the neutral disk .
we note that the thin disk interpretation does not necessarily imply a misalignment between the jet and the disk , i.e. , the position angle of the jet changes as a function of distance from the core ( figure 3 ) and the disk could be warped , thus , the jet and disk could be perpendicular at ( sub)parsec scales .
in addition , given that the detection rate of absorption in csos is @xmath12 ( see @xmath96 ) , it is certainly possible that in some cases the atomic gas is distributed in a circumnuclear structure that is seen in projection only toward one radio lobe / jet . assuming the orbital parameters from the thin disk model , we now explore the stability of the disk with respect to gravitational fragmentation .
we used the toomre stability formulation as presented by gallimore et al.@xmath0(1999 ) , i.e. , @xmath97 where @xmath98 is the atomic hydrogen surface density , @xmath99 is the critical surface density above which the disk is unstable to fragmentation , @xmath100 is the column density in units of 10@xmath101@xmath8 , @xmath102 is the rotation velocity in , and @xmath103 is the orbital radius in parsecs . in the case of b2352 + 495
we obtain @xmath104 , thus , the disk is expected to be unstable to fragmentation .
we note that reducing the assumed spin temperature by an order of magnitude would not significantly modify this conclusion , i.e. , @xmath105 would still be @xmath106 .
this simple stability test indicates that the can not be distributed in a homogeneous disk .
indeed , the opacity variations from channel to channel ( figure 2 ) suggests that the gas is not homogeneously distributed .
we therefore concur with the standard schematic of clumpy atomic gas around agns ( e.g. , peck et al.@xmath01999 ) . in this section
we discuss the properties of absorption in b2352 + 495 with respect to other extragalactic absorbers that have been studied at very high angular resolution ( i.e. , interferometric observations with baselines @xmath107 km ; table 1 ) .
as is clear from table 1 and the discussion in @xmath96 , high angular resolution observations of in csos reveal that the absorption does not trace a specific kind of cloud , instead , the absorbing clouds appear to be in quite different locations throughout the host galaxy , from infalling high velocity clouds and giant molecular clouds at kiloparsec scales , to torus and disks closer ( @xmath108pc ) to the nucleus .
only two of the five csos imaged at high angular resolution exhibit absorption toward both radio lobes ( table 1 ; b1934@xmath25638 has not be imaged at high angular resolution , see @xmath1091.1 ) . in general
( four out of six , table 1 ) , csos studied at high angular resolution are characterized by a broad ( @xmath110 100 ) absorption component , and tend to be multi - peaked , with narrow ( @xmath111 100 ) lines detected in five of six cases .
high angular resolution observations of toward non - cso radio galaxies ( table 1 ) show a similar trend as the csos counterparts , i.e. , the absorption appears to trace a variety of environments , from disks directly associated with the agn to clouds at kiloparsec - scale distances from the active nucleus .
an apparent difference is the occurrence of clouds interacting with radio jets . in the case of several non - cso with double jet / lobe radio emission ,
the appears to trace regions of interaction between atomic gas and radio jets , whereas such an interpretation has not been preferred in any of the csos imaged in at high angular resolution .
this trend does not appear to be due to bias of the researchers but , in most cases , seems due to real differences in the samples .
for instance , the absorption associated with tori / disks in csos is found very close ( @xmath108pc ) to the radio core , close to the systemic velocity , and/or has evidence of rotation .
conversely , the in non - cso double jet / lobe sources with possible jet / cloud iteration is found at larger distances from the core ( @xmath112pc ) , shows no evidence of rotation , and/or has significantly different velocity from systemic .
although a more robust statistical sample is needed to achieve strong conclusions , the apparent absence of interaction with the radio jets supports the current view that csos are young radio galaxies instead of being confined by dense circumnuclear material as proposed in the frustration scenario .
arguments for the youth scenario are mostly based on the overall radio continuum properties of the sources , power budget estimates of the jets , expansion velocities , and properties of the nuclear molecular / atomic gas .
csos lack strong extended radio emission related to energy deposited by the jets over long periods , and vlbi studies reveal proper motions that imply short dynamic ages ( of the order of @xmath113yr , e.g. , gugliucci et al . 2005 ) .
in addition , if the advance of the radio lobes would be truly frustrated , then for them to be in ram - pressure equilibrium the density of the surrounding medium should be above @xmath114@xmath74 . however , if the high density material is confined to a disk / torus then the frustration scenario is not viable .
b2352 + 495 has evidence of being a young source based on all these arguments : it has no detected extended radio emission , the expansion velocity implies a dynamic age of @xmath30yr ( taylor et al.@xmath02000 ) , and , as discussed above , the high density atomic gas appears to be distributed along a disk or torus which does not interact with the radio jet .
hence , the evidence favors the youth interpretation in this case .
we report high angular resolution ( @xmath115mas ) observations of in the cso b2352 + 495 .
we detect two absorption features : a broad line almost centered at the systemic velocity of the host galaxy , and a redshifted narrow line indicating infall .
both lines are seen in absorption against the radio continuum of the central radio source of the cso , which is the brightest continuum source in the region at the observed frequency .
the radio continuum originates from the radio jet and not from the core . in the case of the narrow component
, our data can not establish the location of the cloud with respect to the agn , i.e. , the radial distance of the cloud to the galaxy center is unknown . on the other hand ,
the broad component is likely tracing circumnuclear material close ( @xmath116pc ) to the agn . even though we can not constrain the scale height of the gas ,
the broad absorption could originate from a clumpy circumnuclear disk .
all available data suggest that b2352 + 495 is a young radio source and not an older source whose radio lobes have been confined by high density circumnuclear material.@xmath0hence , this work supports the study of csos as a key to understand the origin of extended radio galaxies .
we acknowledge an anonymous referee for critical comments that significantly improved the manuscript .
was partially supported by a postdoctoral fellowship at the university of new mexico .
this research has made use of nasa s astrophysics data system , and the nasa / ipac extragalactic database ( ned ) which is operated by the jet propulsion laboratory , california institute of technology , under contract with nasa .
augusto , p. 2007 , in asp conf .
380 , at the edge of the universe : latest results from the deepest astronomical surveys , ed .
j. afonso , h. c. ferguson , b. mobasher , & r. norris ( san francisco : asp ) , 261 morganti , r. , oosterloo , t. , tadhunter , c. n. , wills , k. a. , tzioumis , a. , & reynolds , j. 2000 , in proc . of the 5th european vlbi network symp . , ed .
j. e. conway , a. g. polatidis , r. s. booth , & y. m. pihlstrm , isbn 91 - 631- 0548 - 9 , 111 ( astro - ph/0010482 ) polatidis , a. g. , & aalto , s. 2001 , in iau symp .
205 , galaxies and their constituents at the highest angular resolutions , ed .
r. t. schilizzi , s. vogel , f. paresce , & m. elvis ( san francisco : asp ) , 198 lllll + + 4c 31.04 , b0116 + 319 , j0119 + 3210 & both lobes & n , b , m & agn torus + & 1 + & & & infalling hvc & + 4c 37.11 , b0402 + 379 , j0405 + 3803 & one
lobe & b , m & agn torus & 2 + oq + 122 , pks 1413 + 135 , j1415 + 1320 & one lobe & n & gmc in kpc disk & 3 + b1934@xmath25638 , pks 1934@xmath2563 ,
j1939@xmath256342 & @xmath117 & n & infalling hvc ? & 4 + b1946 + 708 , j1945 + 7055 & both lobes & n , b , m & agn torus & 5,6 + b2352 + 495 , j2355 + 4950 & one lobe & n , b , m & agn disk + & this + & & & infalling cloud & work + + + + 3c@xmath049 , b0138 + 136 , j0141 + 1353 & one lobe & n , m & cloud / jet interaction & 7 + ngc@xmath01052 , b0238@xmath25084 , j0241@xmath250815 & both radio jets & n , m & nuclear and/or galactic ? & 8 + mrk 6 , iras 06457 + 7429 , j0652 + 7425 & one side of jet & n & kpc disk & 9 + hydra a , 3c@xmath0218 , b0915@xmath25118 , j0918@xmath251205 & both jets+core & n , m & agn disk & 10 + 3c@xmath0236 , b1003 + 351 , j1006 + 3454 & one lobe & b & cloud / jet interaction & 1 + ngc@xmath03894 , b1146 + 596 , j1148 + 5924 & lobes + core & n , m & agn torus & 11 + [email protected] , b1203 + 6430 , j1206 + 6413 & one lobe & n & cloud / jet interaction & 7 + ngc@xmath04151 , b1208 + 396 , j1210 + 394 & one radio jet & n , m & agn torus & 12 + ngc@xmath04261 , b1216 + 061 , j1219 + 0549 & counterjet+core ? &
n , m ? & agn disk & 13 + mrk 231 , iras 12540 + 5708 , j1256 + 5652 & diffuse cont . & b & @xmath118pc disk & 14 + 4c 12.50 , iras 13451 + 1232 , j1347 + 1217@xmath119 & counterjet & b , m & cloud / jet interaction & 15 + 3c@xmath0293 , b1350 + 3141 , j1352 + 3126 & both jets+core & b , n , m & outer ( 8@xmath0kpc ) and & 16 + & & & @xmath120pc disk & + ngc@xmath05793 , iras 14566@xmath251629 , j1459@xmath251641 & both lobes & n , m & @xmath1211@xmath0kpc torus ? & 17 + ngc@xmath05929 , b1524 + 4151 , j1526 + 4140 & one lobe & n & agn ring , or bar ? & 18 + pks@xmath01814@xmath2563 , j1820@xmath256343 & both lobes & m , n & agn disk + & 19 + & & & cloud / jet interaction ? & + ic 5063 , b2048@xmath25572 , j2052@xmath255704 & one lobe & b , m & cloud / jet interaction & 20 + txs 2226@xmath25184 , iras 22265@xmath251826 , j2229@xmath251810 & radio jet & b & nuclear region , & 21 + & & & cloud / jet interaction ? & + ngc@xmath07469 , b2300 + 0836 , j2303 + 0852 & radio core & n & nuclear or galactic ? & 22 + pks 2322@xmath25123 , j2325 - 1207 & one jet + core & b , m & agn disk + infall ? & 23 + ngc@xmath07674 , iras 23254 + 0830 , j2327 + 0846 & one lobe + core ? & b , n , m & agn disk / torus & 24 + + + + ngc@xmath0315 , b0055 + 300 , j0057 + 3021 & one jet + core & n & infalling cloud & 25 + ngc@xmath03079 , iras 09585 + 5555 , j1001 + 5540 & one jet + core & n , m & agn torus & 26 + cen a , ngc@xmath05128 , b1322@xmath25428 , j1325@xmath254301 & one jet & n , m & kpc disk or rings & 25 + pks 1549@xmath2579 , b1549@xmath25790 , j1557@xmath257913 & one jet + core & n & nuclear region ? & 27 + cygnus a , iras 19577 + 4035 , j1959 + 4044 & counterjet + core ? & b & torus or 2@xmath0kpc ring & 1 + da 529 , b2050 + 364 , j2052 + 3635 & one jet + core & n , m & nlr / blg ( @xmath1kpc ) & 28 + + + + ngc@xmath02146 , 4c+78.06 , b06106 + 7822 , j0618 + 7821 & starburst cont . & b , n , m & disk or ring & 29 + ic@xmath0694 , arp@xmath0299 , mrk 171 , b1125 + 588 , j1128 + 5834 & starburst cont . & & 250@xmath0pc disk & 30 + mrk 273 , b1342 + 5608 , j1344 + 5553 & starburst cont . & b , m & starburst cores & 31 + arp@xmath0220 , b1532 + 2339 , j1534 + 2330 & starburst cont . & n , b , m & disks in starburst cores & 32 + pgc 060189 , iras 17208@xmath250014 , j1723@xmath250016 & starburst cont . & b , n , m & nuclear @xmath33kpc & 33 + | b2352 + 495 is a prototypical example of a compact symmetric object ( cso ) .
it has a double radio lobe symmetrically located with respect to a central flat spectrum radio core ( the location of the agn ) and has a physical extent of less than 200@xmath0pc . in this work
we report vlba observation of 21@xmath0 cm absorption toward b2352 + 495 to investigate the properties of this remarkable radio source , in particular , to explore whether the radio emission can be confined by circumnuclear material ( frustration scenario ) or whether the source is likely to be young .
we confirmed the two absorption features previously detected toward b2352 + 495 a broad line nearly centered at the systemic velocity of the galaxy and a narrow redshifted component .
the atomic gas from the broad absorption component is likely associated with circumnuclear material , consistent with the current paradigm of clumpy distribution in toroidal structures around supermassive black holes . |
consider a cubic ( i.e. , 3-regular ) plane graph , and let @xmath3 ( @xmath12 ) denote the number of its @xmath4-gonal faces .
it is a simple corollary of euler s formula that @xmath13 it is natural to ask for which sequences @xmath14 satisfying ( [ eq : basic ] ) there exists a cubic plane graph whose face sizes comply with the sequence @xmath15 .
this question is even more interesting when additional restrictions on the graph are given .
the most important case is to consider graphs of 3-dimensional convex polyhedra , so called _ polyhedral graphs_. by steinitz s theorem , this is the same as requiring the graphs to be 3-connected .
an important subcase is when the polyhedra are _ simple _ , in other words , when the corresponding graphs are cubic .
the general problem about the existence of polyhedral graphs with given face sizes is still wide open .
however , there are many special cases that have been solved . for example
* theorem 13.4.1 ) , it is known that there exists a simple polyhedron with six quadrangular faces and @xmath2 faces of size six if and only if @xmath16 ; and there exists a simple polyhedron with twelve pentagonal faces and @xmath2 faces of size six ( a `` fullerene '' graph ) if and only if @xmath16 . a similar case of four triangular faces and @xmath2 faces of size 6 has infinitely many exceptions : such a polyhedron exists if and only if @xmath2 is even
. we refer to @xcite for a complete overview .
the most fundamental result in this area is the following classical theorem of eberhard @xcite , stating that there is always a solution provided we are allowed to replace @xmath2 ( whose value does not affect the satisfaction of ) by a large enough integer .
[ thm : eberhard ] for every sequence @xmath0 of non - negative integers satisfying @xmath17 , there are infinitely many values @xmath2 such that there exists a simple convex polyhedron having precisely @xmath3 faces of size @xmath4 for every @xmath5 , where @xmath6 if @xmath7 .
eberhard s proof is not only long and messy but also some of its parts may not satisfy today s standards of rigor .
grnbaum @xcite gave a simpler complete proof utilizing graphs and steinitz s theorem .
this result was strengthened by fisher @xcite who proved that there is always a value of @xmath2 that satisfies @xmath18 .
grnbaum also considered a 4-valent analogue of eberhard s theorem .
fisher @xcite proved a similar result for 5-valent polyhedra , establishing existence for all sequences of face sizes with @xmath19 that comply with euler s formula .
various other generalizations of eberhard s theorem have been discovered .
papers by jendrol @xcite give a good overview and bring some of today s most general results in this area .
some other relevant works include @xcite .
several papers treat extensions of eberhard s theorem to the torus @xcite and more general surfaces @xcite .
it is worth pointing out that on the torus there is precisely one admissible sequence ( namely @xmath20 and @xmath21 for @xmath22 ) , for which an eberhard - type result with added hexagons does not hold @xcite . in this paper
we consider a similar problem that is also motivated by ( [ eq : basic ] ) .
let us suppose that we are given face sizes as before but we are only allowed to change @xmath23 and @xmath24 ( or @xmath25 and @xmath26 for some @xmath27 , @xmath28 ) . in this case
, we think of @xmath3 ( for @xmath29 , @xmath30 ) as being fixed and @xmath11 as being free to choose .
equation ( [ eq : basic ] ) determines the difference @xmath31 , and we are asking if there exist @xmath23 and @xmath32 with a polyhedral realization .
we give an affirmative answer to this question , and derive an extension solving the corresponding problem on an arbitrary closed surface .
our construction gives simple ( i.e. , 3-regular ) polyhedral maps on a surface , and one can impose the additional conditions that these maps have large face - width and their graphs be 3-connected .
more precisely , we prove [ thintro ] let @xmath33 be a sequence of non - negative integers , let @xmath34 be a closed surface , and let @xmath35 be a positive integer . then there
exist infinitely many pairs of integers @xmath23 and @xmath24 such that there is a 3-connected cubic map realizing @xmath34 , with face - width at least @xmath35 , having precisely @xmath3 faces of size @xmath4 for every @xmath36 .
it is worth observing that we also fix the number @xmath2 of hexagonal faces .
secondly , observe that the extension of eberhard s theorem to a surface @xmath34 other than the sphere needs an adjustment in ( [ eq : basic ] ) ; the right hand side has to be replaced by @xmath37 where @xmath38 is the euler characteristic of @xmath34 . however , in our setting the formula adjusts itself by using an appropriate number of pentagons and heptagons .
finally , as we point out in section [ other ] , our proof suggests a general method for obtaining results of this kind .
a finite sequence @xmath39 is _ plausible _ for a closed surface @xmath34 if @xmath40 where @xmath38 is the euler characteristic of @xmath34 . by euler s formula ,
is a necessary condition for the existence of a cubic graph embeddable in @xmath34 with precisely @xmath3 @xmath4-gons for @xmath41 and no other faces .
if there exists a cubic graph which is 2-cell embeddable in @xmath34 with precisely @xmath3 faces of size @xmath4 for @xmath41 and no other faces , then we say that @xmath42 is _ realizable _ in @xmath34 .
if @xmath43 , then we call @xmath42 a _ neutral _ sequence . for any two such sequences
, one can consider their sum which is defined in the obvious way .
let us observe that the sum of a neutral sequence and a plausible sequence is a plausible sequence .
we would like to understand in this context which plausible sequences are realizable , and try to do so by asking when a sum of a plausible sequence with an appropriate neutral sequence is realizable . for the neutral sequence @xmath44 this is eberhard s theorem .
the most important building block in both eberhard s as well as our proofs is a construction called a triarc .
a _ triarc _ is a plane graph @xmath45 such that the boundary @xmath46 of the outer face of @xmath45 is a cycle , and moreover the following conditions are satisfied ( examples are the graphs in figure [ 443224tr ] with the half - edges in the outer face removed ) : * every vertex of @xmath47 has degree 3 in @xmath45 ; * @xmath46 contains distinct vertices @xmath48 of degree 2 ( called the _
corners _ of the triarc ) such that the degrees ( in @xmath45 ) of the vertices on each of the three paths in @xmath49 alternate between 2 and 3 , starting and ending with a vertex of degree 2 .
a _ side _ of a triarc @xmath45 as above is a subpath of @xmath46 that starts and ends at distinct corners of @xmath45 and does not contain the third corner .
the _ length _ of a side @xmath50 of @xmath45 is the number of inner vertices of degree 2 on @xmath50 ; note that although the corners of a triarc have degree 2 , they are not counted when calculating the lengths of its sides . a triarc with sides of lengths @xmath51 , @xmath52 , @xmath53 is called an _ @xmath54-triarc_. of course , we can flip or rotate such a triarc and consider it , for example , as a @xmath55-triarc .
triarcs are very versatile tools .
firstly , if the length of some side of a triarc @xmath45 equals the length of some side of another triarc @xmath56 , then @xmath45 and @xmath56 can be glued together along those sides to yield a new plane graph with all inner vertices having degree 3 ; see for example figure [ fig0 ] .
secondly , every triarc @xmath45 has zero total curvature ; to see this , take two copies of @xmath45 , turn one of them upside down , glue them along a common side to obtain a ` parallelogram ' ( see figure [ fig0 ] again ) , and identify opposite sides of this parallelogram to obtain a graph embeddable in the torus .
but perhaps the most important property of triarcs is the possibility to ` glue ' them together to obtain larger triarcs ; we describe this operation below .
.,width=264 ] .,width=151 ] suppose we have an @xmath57-triarc and an @xmath58-triarc such that @xmath59 and @xmath60 are even
. then , we may combine these triarcs ( and several pentagons and heptagons ) to construct an @xmath61-triarc . to do this
, we identify a corner ( and an incident edge ) of the first triarc with a corner ( and an edge ) of the second triarc see figure [ glueing57] so that the two identified corners yield a vertex of degree 3 on a side of length @xmath62 in a new triarc .
then , we can add a _
`` parallelogram '' _ consisting of hexagons to obtain an @xmath61-triarc .
however , we do not want to add hexagons . instead
, we decompose the parallelogram into tiles each consisting of four hexagons as depicted in figure [ glueing57 ] , and replace each of these tiles by two pentagons and two heptagons as indicated in figure [ 2x2tile ] . the `` tile '' on the right of figure [ 2x2tile ]
will be used several times in the sequel , and we shall refer to it as a _
h - tile_. we are going to use this operation of glueing two triarcs into a larger one several times in the following section .
we are ready to state and prove our main result .
let us observe that , unlike eberhard s theorem , we do not need to assume that the given face - sizes form a plausible sequence ( although we make this assumption in the formulation of the theorem ) because given a sequence @xmath33 , the sequence can always be appended by appropriate values @xmath23 and @xmath24 to become plausible . [ ebe57 ]
let @xmath63 be a plausible sequence for the sphere . then there exist infinitely many integers @xmath64 such that the sequence + @xmath65 is realizable in the sphere .
we will give an explicit construction of a cubic graph embeddable in the sphere whose face sequence is of the form @xmath65 .
the rough plan for this is as follows . for each face imposed by the sequence @xmath42
, we create a _ basic triarc _ containing this face as well as some pentagons and heptagons .
then , we glue all these triarcs together and extend to a triarc with sides of suitable lengths . finally , we construct a new triarc having the same side lengths , and glue these two triarcs together ( as explained later ) to obtain the desired graph embedded in the sphere . to construct a basic triarc for a @xmath4-gon ( we will make @xmath3 copies of it ) , we surround the @xmath4-gon by three heptagons and @xmath66 pentagons as shown in the right half of figure [ 443224tr ] ( where the @xmath4-gon we are surrounding happens to be a pentagon ) .
note that we can always make the basic triarc isosceles with the equal sides having even length .
we call the @xmath4-gon we started with the _ nucleus _ of this triarc .
having constructed all basic triarcs , our next step is to glue them all together to obtain a single triarc @xmath45 containing them all .
we do so recursively , attaching one basic triarc at a time as shown in figure [ glueing57 ] , where we use many copies of the h - tile in order to build the parallelogram needed .
each time we use this glueing operation we are assuming that both triarcs in figure [ glueing57 ] are isosceles , with the equal sides having even length , and align them so that the two equal even sides are the upper left and upper right side .
note that the resulting triarc is also isosceles with two equal sides of even length .
thus , we can continue recursively to glue all basic triarcs into one isosceles triarc @xmath45 .
our next aim is to enlarge @xmath45 into an equilateral triarc @xmath67 with sides of length @xmath68 , where @xmath68 is a multiple of 8 and satisfies @xmath69 , using only pentagons and heptagons . to this end
, we will use the glueing operation of figure [ glueing57 ] and many copies of a @xmath70-triarc and a @xmath71-triarc .
figure [ 443224tr ] shows how to construct those triarcs with pentagons and heptagons only .
-triarc and a @xmath71-triarc.,width=377 ] note that glueing @xmath67 with a @xmath70-triarc ( as in figure [ glueing57 ] ) keeps it isosceles and decreases the difference of lengths between the `` base '' and the other two sides by 1 , while glueing with a @xmath71-triarc increases that difference by 2 .
thus , recursively glueing with such triarcs we can enlarge @xmath45 into an equilateral triarc @xmath34 with sides of even length .
moreover , using the glueing operation of figure [ glueing57 ] three times , once with a @xmath71-triarc and twice with a @xmath70-triarc , we can increase the side - lengths by @xmath72 .
thus we can increase the length of each side of @xmath34 by 10 . since @xmath73 , we can use this operation to enlarge @xmath34 into an equilateral triarc @xmath74 with even sides of length @xmath75 .
moreover , since performing this operation three times increases the length of each side by 30 , and @xmath76 , we can enlarge @xmath74 into an equilateral triarc @xmath67 with the length of each side being a multiple of @xmath77 and congruent to @xmath78 modulo @xmath79 .
next , we are going to construct a triarc @xmath56 that has the same side lengths as @xmath67 but consists of pentagons and heptagons only . by glueing together a @xmath71-triarc , a @xmath80-triarc and a @xmath81-triarc ( that is , the same triarc in three different rotations ) we get an @xmath82-triarc , which we will call @xmath83 .
since the sides of @xmath67 have length a multiple of 8 , by glueing copies of @xmath83 together recursively as in figure [ glueing57 ] we can indeed construct a triarc @xmath56 that has the same dimensions as @xmath67 .
we can now combine @xmath56 and @xmath67 together to produce a cubic graph tiling the sphere as shown in figure [ cyl57 ] . by construction
, this graph has for every @xmath84 precisely @xmath3 faces of size @xmath4 , and moreover it has at least @xmath23 pentagons and at least @xmath24 heptagons .
thus its face sequence is of the form @xmath85 for some @xmath86 .
since both @xmath42 and @xmath85 satisfy euler s formula ( the former by assumption , the latter because the plane graph we just constructed implements it ) , we have @xmath87 .
this completes the construction and shows the existence of one particular value of @xmath68 as desired . however , observe that the construction of @xmath67 and @xmath56 allows us to make the side lengths of these triarcs arbitrarily large .
this shows that we can get examples for infinitely many values of @xmath68 and thus completes the proof .
we now turn from planar graphs to maps on arbitrary ( compact ) surfaces .
map _ on a surface @xmath34 is a graph together with a 2-cell embedding in @xmath34 .
a map is _ polyhedral _
if all faces are closed disks in the surface and the intersection of any two faces is either empty , a common vertex or a common edge .
if the graph of the map is cubic , then we say that the map is _
simple_. a cycle contained in the graph of a map is _ contractible _ if it bounds a disk on the surface .
the _ edge - width _ of a map @xmath88 is the length of a shortest non - contractible cycle in @xmath88 .
the _ face - width _ of @xmath88 is the minimum number of faces , the union of whose boundaries contains a non - contractible cycle .
we refer the reader to @xcite for more about the basic properties and the importance of these parameters of maps . at this point
we only note that a map is polyhedral if and only if its graph is 3-connected and its face - width is at least three , see ( * ? ? ?
* proposition 5.5.12 ) .
we also note that if @xmath89 is the largest size of a face of @xmath88 , then the edge - width of @xmath88 can not exceed @xmath90 times the face - width of @xmath88 .
we now restate and prove our main result , theorem [ thintro ] .
[ cor ] let @xmath33 be a sequence of non - negative integers , let @xmath34 be a closed surface , and let @xmath35 be a positive integer .
then there exist infinitely many pairs of integers @xmath23 and @xmath24 such that the sequence @xmath91 is realizable in @xmath34 and there is a 3-connected realizing cubic map of face - width at least @xmath35 .
let us first describe a construction that does not necessarily achieve the desired face - width ; we will later explain how to modify this construction in order to get large face - width .
the rough sketch of this construction is as follows .
firstly , we increase the number of hexagons in the sequence @xmath15 to @xmath92 , where @xmath93 is the number of handles of @xmath34 and @xmath53 the number of its crosscaps ( by the surface classification theorem we may assume that one of @xmath94 is zero , but we do not have to ) .
it follows from theorem [ ebe57 ] that we can increase the numbers @xmath23 and @xmath24 of this sequence to some appropriate values so that the resulting sequence @xmath95 is realized by a map on the sphere .
we will then use the @xmath96 auxilliary hexagons of this map we added above to introduce some handles and/or crosscaps .
after doing so , all auxilliary hexagons will have disappeared , and we obtain a map on @xmath34 whose sequence of faces differs from @xmath15 by some pentagons and heptagons only .
more precisely , similarly to the proof of theorem [ ebe57 ] , we construct a basic triarc for each face in @xmath95 , but with one modification : for each hexagon we construct a triarc like the one in figure [ newhex ] ( on the left ) rather than one with two even sides of equal lengths ( in fact , we need this modification for the auxiliary hexagons only , but we might as well use it for the original hexagons in @xmath42 as well ) . next , we proceed as in theorem [ ebe57 ] to glue all basic triarcs together into one triarc @xmath45 . however , since we now have basic triarcs with all sides odd ( the ones of figure [ newhex ] ) , the glueing operation of figure [ glueing57 ] will not work for these triarcs . for this reason
, we first extend each such triarc into an equilateral triarc with even sides using three copies of the @xmath71-triarc of figure [ 443224tr ] as shown in figure [ newhex ] ( right ) .
consists of the dashed and the thick edges.,width=188 ] we continue imitating the proof of theorem [ ebe57 ] to obtain a cubic graph @xmath97 embedded in a homeomorphic copy @xmath74 of the sphere that contains all basic triarcs .
we will now perform some cutting and glueing operations on both @xmath74 and @xmath97 to obtain a new surface , homeomorphic to @xmath34 , with a cubic graph @xmath98 embedded in it .
suppose that @xmath99 .
then , pick @xmath93 pairs @xmath100 of hexagonal faces of @xmath97 , such that all the faces @xmath101 and @xmath102 are distinct ( there are enough hexagonal faces by our choice of the sequence @xmath95 ) .
now for each pair @xmath103 perform the following operations .
cut out the two discs of @xmath74 corresponding to @xmath104 , and glue their boundaries together with a half - twist ; that is , each vertex of the boundary of @xmath101 is identified with the midpoint of an edge of @xmath102 and vice - versa .
this operation creates a handle in @xmath74 , and the embedded graph remains cubic ; however , it also gives rise to some unwanted faces : the size of each face that was incident to @xmath101 or @xmath102 has now been increased by 1 .
we thus have the situation depicted in figure [ handle ] , where @xmath46 is the cycle of length 12 resulting from the boundaries of @xmath101 and @xmath102 . recall that since every hexagon is put in a basic triarc like the one in figure [ newhex ] , the sizes of the faces on each side of @xmath46 alternate between 6 and 8 as shown in figure [ handle ] .
but now , contracting and uncontracting each of the three thick edges ( in the way explained in figure [ 2x2tile ] ) turns each of the faces incident with @xmath46 into a heptagon . on the other hand ,
if @xmath105 , then pick @xmath53 distinct hexagonal faces @xmath106 , and for every @xmath107 cut out the disc corresponding to @xmath101 and glue in its place the outside of the hexagon of figure [ crosscap ] with a half twist .
each such operation gives rise to a new crosscap , but also to unwanted faces just like in figure [ handle ] .
but again , contracting and uncontracting each of the three thick edges we can turn all these unwanted hexagons and octagons into heptagons .
thus , after all these operations have been completed , we obtain a surface with @xmath93 handles and @xmath53 crosscaps with a cubic graph embedded in it whose face sequence is @xmath108 for some @xmath86 .
note that all auxiliary hexagons in @xmath109 have disappeared after the above operations .
it is easy to check that our maps are 3-connected .
indeed , our `` building blocks '' the basic triarcs and the triarcs of figure [ 443224tr ] are 3-connected after suppressing the degree 2 vertices .
this property is also true for triarcs in figure [ newhex ] , and it is maintained by the glueing operation of figure [ glueing57 ] . by glueing two triarcs along a `` circumference '' , using the ring in figure [ cyl57 ] ,
we get a 3-connected graph .
the gadget we used for introducing cross - caps ( figure [ crosscap ] ) is 3-connected , and it is not hard to see that we maintain 3-connectivity when adding this gadget or when creating a handle as depicted in figure [ handle ] .
it remains to discuss how to modify this construction to obtain maps with arbitrarily large face - width . by the remark preceding corollary [ cor ]
, it suffices to construct maps with arbitrarily large edge - width @xmath110 since the face sizes are bounded from above by @xmath89 .
this is achieved as follows .
first of all , we make every basic triarc used in the construction large enough that the distance from its nucleus to the boundary of the triarc is at least @xmath110 and the length of each side of each triarc is at least @xmath111 .
this can be achieved by the method we used in the proof of theorem [ ebe57 ] to enlarge @xmath45 into an equilateral triarc @xmath67 . to the boundary of the triarc
is at least @xmath110 and the length of each side of each triarc is at least @xmath111 . to achieve this
, we first glue the triarc with several @xmath71-triarcs ( or any other triarcs ) both on the left and on the right , to obtain a large triarc with the nucleus in the middle of the bottom side .
then we possibly glue it with a @xmath70-triarc to create a triarc with all sides even .
finally , we rotate the triarc by @xmath112 and perform more glueing with @xmath71-triarcs to get the nucleus away from the boundary .
( the notions `` left '' , `` right '' , and `` @xmath112 '' in this paragraph refer to the glueing operation of fig .
[ glueing57 ] . )
next , we replace the auxiliary hexagons used in order to add handles and crosscaps with @xmath113-gons , where @xmath114 is odd and greater than @xmath115 .
of course , this will force us to add some more pentagons to our sequence @xmath3 to make it plausible . note that we can generalize the triarc on the left of figure [ newhex ] so that the inner 6-gon is replaced by a @xmath113-gon surrounded by three heptagons and @xmath116 pentagons , arranged in a symmetric way so that any two heptagons separate @xmath117 pentagons from the rest .
we will make use of the fact that @xmath117 is odd .
we need to adapt the right half of figure [ newhex ] as well , since the inner triarc has now grown larger . for this , note that each side of the inner triarc has now length @xmath118 , and so in order to use the method of the right half of figure [ newhex ] the three peripheral triarcs must have a base of length @xmath119 ( in addition to having their other two sides of equal length ) .
since we chose @xmath114 to be odd , it turns out that @xmath119 is a multiple of four , and so we can construct the required peripheral triarcs by glueing several @xmath71-triarcs together using figure [ glueing57 ] into a @xmath120-triarc .
moreover , the crosscap gadget shown in figure [ crosscap ] can be generalized so that the inner 6-gon is replaced by a @xmath113-gon that is surrounded by @xmath121 heptagons and @xmath121 pentagons , arranged alternatingly around the @xmath113-gon ( here it is also important that we chose @xmath114 to be odd ) . when the time comes to insert crosscaps or glue pairs of such @xmath113-gons together ( after a half - twist ) , we obtain a similar configuration as in figure [ handle ] , but with @xmath121 thick edges .
some of these thick edges are surrounded by faces of sizes @xmath122 ( as in figure [ handle ] ) , while others are surrounded by four hexagons or by one octagon and three hexagons .
note , however , that for parity reasons we can make sure that every octagon is incident with a thick edge , and still every fourth edge on the dashed cycle is thick .
finally , the contract - uncontract operation of figure [ 2x2tile ] turns these faces into pentagons and heptagons only .
it is easy to see that these changes did not hurt 3-connectivity .
let us now argue that the resulting map @xmath97 has edge - width at least @xmath110 .
recall that the surface @xmath34 is obtained from a plane graph @xmath98 , embedded in the sphere , that is composed of large basic triarcs @xmath123 , some large parallelograms used to glue the basic triarcs together into a large triarc @xmath45 , and a remainder @xmath124 comprising the material we used to enlarge @xmath45 into @xmath67 , the ring of figure [ cyl57 ] , and the triarc @xmath56 .
let @xmath125 be the nucleus of @xmath126
. then @xmath34 is obtained from @xmath98 by glueing the crosscap gadget into some of the @xmath113-gons @xmath125 , and/or by identifying some pairs @xmath127 of the @xmath113-gons to create handles .
we claim that for every basic triarc @xmath126 such that the nucleus @xmath125 of @xmath126 is a @xmath113-gon , and @xmath128{0.9\textwidth } for every side $ p$ of $ t_i$ , there is a set of $ z$ pairwise disjoint $ { \ensuremath{l_i}\text{--}\ensuremath{p}~paths}$. \end{minipage } \ignorespacesafterend\ ] ] indeed , recall that in order to construct @xmath126 , we first surrounded @xmath125 by several pentagons and heptagons , @xmath113 in total , to obtain a triarc @xmath129 , then we performed the operation of the right half of figure [ newhex ] to obtain a triarc @xmath130 , and finally we enlarged this into a larger triarc @xmath131 using the operation of figure [ glueing57 ] several times ( this final step was described later , in the part of the current proof concerning large face - width ) .
now given any side @xmath132 of @xmath130 it is possible to find , within @xmath130 , a set of @xmath110 pairwise disjoint @xmath133 , see figure [ ebelink ] .
disjoint @xmath133 , in the case that the auxilliary hexagons are replaced with 42-gons ( @xmath113 for @xmath134 ) .
in light gray are the @xmath71-triarcs , in dark gray the h - tiles resulting from the glueing operation of figures [ glueing57 ] and [ 2x2tile ] .
the empty triangle at the bottom is a part of the triarc , isomorphic to the top - left and top - right ones . as our paths do not use the bottom part
, we do nt show the details in the figure .
the white triangular shape in the middle represents the nucleus .
the 16 thick paths are the ones we need in order to prove that our graphs have large face - width.,width=283 ] then , every time we use the operation of figure [ glueing57 ] while enlarging @xmath130 into @xmath135 , it is possible to recursively propagate those paths to reach the side of @xmath135 corresponding to @xmath132 ; if @xmath132 is included within a side of @xmath135 then nothing needs to be done , and if not then we can propagate our paths through the parallelogram of figure [ glueing57 ] while keeping them disjoint ( this is true even after performing the contract - uncontract operations of figure [ 2x2tile ] ) .
this proves our claim .
next , we claim that any two nuclei @xmath136 can be joined by @xmath110 pairwise disjoint paths in @xmath98 .
indeed , this follows easily from and the fact that whenever we glue two triarcs @xmath137 together as in figure [ glueing57 ] by a parallelogram @xmath56 with side - lengths @xmath138 , then we can find a set of @xmath139 pairwise disjoint paths within @xmath56 joining its two opposite sides of length @xmath139 , as well as a set of @xmath140 pairwise disjoint paths within @xmath56 joining the sides of @xmath45 and @xmath67 incident with @xmath56 .
we distinguish two cases .
[ [ case-1-the - surface - s - is - orientable ] ] case 1 : the surface @xmath34 is orientable + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + there are three types of non - contractible cycles in @xmath97 .
the first one comes from a path @xmath50 in @xmath98 that connects two nuclei @xmath125 , @xmath141 such that these nuclei are glued to create a handle .
as the distance from each nucleus to the boundary of the corresponding triarc @xmath126 is at least @xmath110 , the length of @xmath50 is at least @xmath110 as well ( even at least @xmath142 ) .
the second type of non - contractible cycle @xmath46 comes from a cycle @xmath143 in @xmath98 such that @xmath144 and @xmath143 separates some nucleus @xmath125 from some other nucleus @xmath141 in @xmath98 .
we use the above construction of @xmath110 pairwise disjoint paths from @xmath125 to @xmath141 to conclude that @xmath145 as desired ( in fact , we have @xmath146 because the graph is cubic and so any two paths that have a common inner vertex must have a common edge ) .
the last type is similar to the second one : it is a cycle @xmath46 that crosses some cycle @xmath147 of @xmath97 obtained by glueing two nuclei @xmath125 , @xmath141 to introduce a handle .
such a cycle @xmath46 comes from a @xmath4-tuple of paths in @xmath98 , where @xmath4 is the number of times that @xmath46 crosses @xmath147 , half of which paths have ends on @xmath125 , and the other half of them on @xmath141 .
we may assume that none of these paths @xmath50 leaves the triarc containing the endpoints of @xmath50 , for otherwise @xmath148 holds .
we will consider again the @xmath110 pairwise disjoint paths connecting @xmath125 to @xmath141 .
in fact , we only need to consider their parts that are contained in the triarcs @xmath126 , @xmath149 : let these parts be @xmath150 , , @xmath151 ( connecting @xmath125 to the boundary of @xmath126 ) and @xmath152 , , @xmath153 ( connecting @xmath141 to the boundary of @xmath149 ) .
as the paths @xmath154 start regularly along two thirds of the nucleus @xmath125 and the same holds for @xmath141 , we can use them to create @xmath115 pairwise disjoint paths @xmath155 , , @xmath156 in @xmath97 connecting the boundary of @xmath126 to the boundary of @xmath149 . note that the part of the surface @xmath34 containing @xmath126 and @xmath149 is a cylinder , and
the cycle @xmath46 goes around this cylinder .
thus @xmath46 must intersect all of the paths @xmath157 .
as the graph is cubic , each intersection with a path has to use at least two vertices , proving again that @xmath158 .
[ [ case-2-s - is - non - orientable ] ] case 2 : @xmath34 is non - orientable + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + in this case a non - contractible cycle @xmath46 in @xmath97 will either yield a cycle @xmath143 as above , in which case the same argument applies , or it will yield a path @xmath132 in @xmath98 whose endpoints were identified when introducing crosscaps . recall that we made every basic triarc used in the construction large enough that the distance from its nucleus to the boundary of the triarc is at least @xmath110 , thus @xmath132 is , without loss of generality , contained within one of the triarcs in which a crosscap was introduced . with the help of figure [ ebelink ] and
figure [ crosscap ] ( modified with a @xmath113-gon replacing the hexagon as described above ) it is now not hard to see that @xmath159 as desired .
in this paper we concentrated on the neutral sequence @xmath160 , but we believe that our methods and results apply in a much more general setting see also section [ outl] and it is the purpose of this section to explain this .
in section [ main ] we showed that every plausible sequence can be extended into a realizable one by adding pentagons and heptagons only . in
what follows we are going to give a rough sketch of a proof that an arbitrary neutral sequence @xmath161 can be used to extend any plausible sequence into a realizable one under the assumption that a couple of basic building blocks can be constructed using precisely the faces that appear in some multiple of @xmath161 .
we expect that our construction will help yield more general results in the future , by showing that these building blocks can indeed be constructed .
so let @xmath63 be a plausible sequence for the sphere or the torus , and let @xmath162 be a neutral sequence .
in order to prove that there is some @xmath68 so that @xmath163 is realizable , it suffices to find some @xmath164 for which it is possible to construct the following building blocks using precisely the faces that appear in some multiple of @xmath161 : 1 .
[ ii ] a @xmath165-triarc ; 2 .
[ iii ] a @xmath166-triarc ; 3 .
[ i ] for every non - zero entry @xmath167 in @xmath42 , a triarc containing a face of size @xmath168 , such that the length of two of the sides of this triarc is a multiple of @xmath4 ; 4 .
[ iv ] a `` ring '' like the one in figure [ cyl57 ] ( using the faces from @xmath161 in the right proportion rather than pentagons and heptagons ) for combining two equilateral triarcs .
-triarcs.,width=396 ] indeed , to begin with , construct a parallelogram with all sides of length @xmath4 out of two @xmath166-triarcs ( supplied by [ iii ] ) as shown in figure [ 2triarcs ] .
( in figures explaining our construction , we shall use triarcs made of hexagonal faces , but this is for illustration purposes only ; in fact they have to be made of multiples of @xmath161 . )
this also allows us to construct any parallelogram with dimensions @xmath169 for every @xmath170 .
next , similarly to the construction in theorem [ ebe57 ] , construct a ` basic ' triarc as in [ i ] for each face - size @xmath168 for which @xmath171 ; in fact , we construct @xmath167 copies of this basic triarc for every @xmath168 . then ,
using the parallelograms we constructed earlier , we recursively glue all those triarcs together into a single triarc @xmath45 , in a manner very similar to the operation of figure [ glueing57 ] . by recursively glueing the resulting triarc with a @xmath166-triarc provided by [ iii ] using the glueing operation of figure [ glueing57 ] , we can transform @xmath45 into an equilateral @xmath172-triarc @xmath67 for some ( large ) @xmath173 . using the glueing operation of figure [ glueing57 ] it is possible to construct a triarc @xmath56 with the same side - lengths as @xmath67 , using only @xmath165-triarcs ( provided by [ ii ] ) and the above parallelograms . in the case of the sphere
, we can combine @xmath56 and @xmath67 by using the `` ring '' provided by [ iv ] to complete the construction . and
@xmath67 together .
the black dots depict the faces imposed by the sequence @xmath42.,width=241 ] if the underlying surface @xmath34 is the torus , we glue @xmath56 and @xmath67 together along one of their sides to obtain a parallelogram , and glue two opposite sides of this parallelogram together to obtain a cylinder @xmath46 both of whose bounding cycles are in - out alternating cycles of length @xmath174 , see figure [ cylinder ] .
we then glue the two bounding cycles of @xmath46 together to obtain a realization of a torus .
if @xmath42 is plausible for some other surface @xmath34 , then we would need additional gadgets like those used in the proof of corollary [ cor ] .
trying to achieve a better understanding of the implications of euler s formula , we studied the question of whether , given a plausible sequence @xmath42 , and a neutral sequence @xmath175 , it is possible to combine @xmath42 and @xmath175 into a realizable sequence @xmath176 , but we did so in very restricted cases .
the general problem remains wide open ; in particular , we would be interested to see an answer to the following problem . given a closed surface @xmath34 , is it true that for every plausible sequence @xmath42 for @xmath34 , and every neutral sequence @xmath175 , there is an @xmath177 such that @xmath176 is realizable in @xmath34 with the exception of only finitely many pairs @xmath178 ? | eberhard proved that for every sequence @xmath0 of non - negative integers satisfying euler s formula @xmath1 , there are infinitely many values @xmath2 such that there exists a simple convex polyhedron having precisely @xmath3 faces of size @xmath4 for every @xmath5 , where @xmath6 if @xmath7 . in this paper
we prove a similar statement when non - negative integers @xmath3 are given for @xmath8 , except for @xmath9 and @xmath10 ( but including @xmath2 ) .
we prove that there are infinitely many values @xmath11 such that there exists a simple convex polyhedron having precisely @xmath3 faces of size @xmath4 for every @xmath5 .
we derive an extension to arbitrary closed surfaces , yielding maps of arbitrarily high face - width .
our proof suggests a general method for obtaining results of this kind . |
lhc in @xmath3 collision at 7 tev has already delivered a integrated luminosity of more than 5 fb@xmath4 . based on this
delivered integrated luminosity recently atlas @xcite and cms @xcite have presented their combined and updated results of sm higgs boson searches .
both the collaborations attempted to search for the sm higgs boson in mass range 110 - 600 gev , the main channels used by them for the analysis are : * * atlas @xcite * @xmath5 and with full data set of 4.8 and 4.9 respectively .
update of @xmath6 , @xmath7 , @xmath8 with 2.1 .
they reported an excess of events around the higgs mass of 126 - 127 gev with the maximum local significance level of 2.6 @xmath9 . * * cms @xcite :* , @xmath10 , @xmath11 , @xmath12 at 4.7 and @xmath13 , @xmath14 , @xmath15 , @xmath16 at 4.6 .
they reported a local significance of 2.4 @xmath9 around the higgs mass of 124 gev .
note that both cms and atlas report some excess but with lower statistical significance , in the @xmath17 and @xmath18 channels .
moreover , from the di - photon channel , atlas and cms have excluded a sm higgs in a small portions of this mass range , @xmath19@xmath20 gev for atlas and @xmath21@xmath22 gev for cms , at the @xmath23c.l .
+ with @xmath24 datasets using the combined channels , both atlas by cms , have narrowed further down the mass window for a light sm higgs , excluding respectively the mass ranges @xmath22@xmath25 gev and @xmath26@xmath27 gev @xcite , and @xmath21@xmath28 gev @xcite at the @xmath23c.l .
the effective cross - section of di - photon ( @xmath29 ) mode can be estimated by inclusive process @xmath30 .
this ( ) could provide possibly the best mode to search for light higgs boson in mass range 110 - 140 gev .
atlas @xcite reported 95% cl exclusion limit of @xmath31 in mass range 110 - 130 gev .
cms @xcite on the other hand reported the exclusion limit of @xmath32 in mass range 110 - 140 gev .
dark matter ( dm ) and electroweak symmetry breaking ( ewsb ) are one of the most important areas of research in particle physics and cosmology .
one of the main goal of lhc is to discover the higgs boson and hence provide the information about the ewsb mechanism .
a dm particle is expected to be a weakly interacting massive particle ( wimp ) with mass around ewsb scale . in sm
the ewsb is achieved by a higgs doublet developing a vacuum expectation value ( vev ) .
inert higgs doublet model ( ihdm ) is a very simple extension of the sm proposed by deshpande and ma @xcite to explain dm .
ihdm is basically a two higgs doublet model with imposed @xmath33 symmetry .
ihdm due to the imposed @xmath33 symmetry exhibits very interesting phenenomenology .
it predicts the existence of a heavy scalar field as a wimp candidate .
the rich phenomenology of ihdm had been extensively discussed in the context of dm phenomenology @xcite , neutrino mass @xcite , naturalness @xcite and colliders @xcite . in this work
we will analyze the effect of ihdm on in the light of recent results on the higgs boson searches from lhc .
this effect will mainly come from charged higgs boson contributions as well as from the total decay width of the higgs boson in case that the invisible decay of the higgs into dark matter is open .
we will show in this study that the ihdm can account for the excess in the di - photon cross - section reported by atlas / cms but it can also account for a deficit in the di - photon cross - section without modifying the gluon fusion rate as well as the other channels like @xmath34 .
the paper is organized as follow ; in section [ section:2 ] we will give the details of the ihdm .
section [ section:3 ] is devoted to theoretical and experimental constraints while in [ section:4 ] we give detail of the evaluation of @xmath35 as well as phenomenological observable at lhc . in section [ section:5 ]
we will present our numerical analysis and finally we conclude in section [ section:6 ] .
the inert higgs doublet model ( ihdm ) @xcite is an extension of the sm higgs sector that could provide dm particles .
apart from the sm higgs doublet @xmath36 it has an additional higgs doublet @xmath37 .
in addition there is a @xmath33 symmetry under which all the sm fields and @xmath36 are even while @xmath38 under @xmath33 .
we further assume that @xmath33 symmetry is not spontaneously broken _
i.e. _ @xmath37 field does not develop vacuum expectation value ( vev ) .
these doublets in terms of physical fields can be parameterized as : h_1 = ( c _ 1^+ +
v/ + ( h + i ) / ) , h_2 = ( c _ 2^+ + ( s + i a)/ ) the @xmath33 symmetry naturally imposes the flavor conservation .
the scalar potential allowed by @xmath33 symmetry can be written as : v = _ 1 ^ 2 |h_1|^2 + _ 2 ^ 2 |h_2|^2 + _ 1 |h_1|^4 + _ 2 [ potential ] the electroweak gauge symmetry is broken by : h_1 = ( c 0 + v/ ) , h_2 = ( c 0 + 0 ) this pattern of symmetry breaking ensures unbroken @xmath33 symmetry and results in two cp even neutral scalars ( @xmath39 , @xmath40 ) one cp odd neutral scalar ( @xmath41 ) in addition to a pair of charged scalars ( @xmath42 ) .
there is no mixing between the two doublets and hence @xmath39 plays the role of the sm higgs boson .
note the remaining higgs bosons namely @xmath43 and @xmath42 are `` inert '' and they do not have any interaction with quarks and leptons .
the @xmath33 symmetry also ensures the stability of the lightest scalar ( @xmath40 or @xmath41 ) that can act as a dark matter candidate .
this aspect has been extensively analyzed in many works while exploring dm phenomenology of ihdm @xcite .
the masses of all these six scalars can be written in terms of six parameters is constrained by ewsb condition @xmath44 namely \{_2 ^ 2 , _ 1 ,
_ 2 , _ 3,_4,_5 } [ eq : input_par_1 ] it is possible to write the quartic coupling @xmath45 in terms of physical scalar masses and @xmath46 as follow : _ 1 & = & , _ 3 = ( m_h^^2 - _ 2 ^ 2 ) , + _ 4 & = & , _ 5 = [ lambds ] we are then free to take as 6 independent parameters @xmath47 and @xmath46 or equivalently the four physical scalar masses , @xmath48 and @xmath46 , namely : \{_2 ^ 2 , m_h , m_s , m_a , m_h^ , _ 2 } [ eq : input_par_2 ]
the parameter space of the scalar potential of the ihdm is reduced both by theoretical constraints as well as by the results of experimental searches . from the theoretical constraints which the ihdm is subjected to ,
the most important are the ones that insure tree - level unitarity and vacuum stability of the theory : * we force the potential to be perturbative by requiring that all quartic couplings of the scalar potential eq .
( [ potential ] ) , obey |_i| 8 * to get a potential @xmath49 bounded from below we obtain the following constraints on the idhm parameters : @xmath50 * to constrain the scalar potential parameters of the ihdm one can demand that tree - level unitarity is preserved in a variety of scattering processes : scalar - scalar scattering , gauge boson - gauge boson scattering and scalar - gauge boson scattering .
we will follow exactly the technique developed in @xcite and therefore we limit ourselves to pure scalar scattering processes dominated by quartic interactions .
+ the full set of scalar scattering processes can be expressed as an @xmath40 matrix @xmath51 composed of 4 submatrices which do not couple with each other due to charge conservation and cp - invariance @xcite .
the entries are the quartic couplings which mediate the scattering processes . + the eigenvalues are : @xmath52 we impose perturbative unitarity constraint on all @xmath53 s .
@xmath54 comes from @xmath55 which gives : _ 1 + _ 2 *
a common approach to constrain physics beyond sm is using the global electroweak fit through the oblique @xmath40 , @xmath56 and @xmath57 parameters @xcite . in the sm
the ewpt implies a relation between @xmath58 and @xmath59 . in this model , there is also a relation among the masses .
it follows from the expression for @xmath40 and @xmath60 that : @xmath61\end{aligned}\ ] ] and , @xmath62\end{aligned}\ ] ] where the function @xmath63 is defined by @xmath64 for the purpose of this paper , we will use the pdg values of @xmath40 and @xmath56 with @xmath57 fixed to be zero @xcite .
we allow @xmath40 and @xmath56 parameters to be within 95% cl .
the central value of @xmath40 and @xmath56 , assuming a sm higgs mass of @xmath65 gev , are given by @xcite : s = 0.03 0.09 , t = 0.07 0.08 with a fit correlation of 87% .
it appears that when the unitarity constraints and vacuum stabilities are applied , a bound on @xmath66 may be obtained .
note that we can restore custodial symmetry in the scalar potential of ihdm by taking @xmath67 .
* here we will discuss the experimental constraints from direct searches on the masses of the ihdm . in the case of the sm higgs ( @xmath39 )
, we can use cms and atlas constraints discussed in section [ section:1 ] when the non sm higgs decays such as @xmath68 , @xmath69 , @xmath70 are kinematically forbidden . in the case where one of these decays is kinematically allowed , it will have a substantial branching ratio .
therefore , it will suppress the other sm decays and hence one can evade the present constraints on sm higgs which are based on conventional sm higgs decays like @xmath71 , @xmath72 , @xmath73 and @xmath74 .
( see fig .
[ fig:2 ] ) .
+ from the precise measurement of w and z widths , one can get some constraints on the higgs masses by demanding that the decays @xmath75 and/or @xmath76 are forbidden .
this leads to the following constraints : @xmath77 , @xmath78 , @xmath79 and @xmath80 @xcite . + additional constraints on the charged higgs @xmath42 and cp - odd @xmath81 masses can be derived .
note that lep , tevatron and lhc bounds on @xmath42 and @xmath81 can not apply because the standard search channels assumes that those scalars decays into a pair of fermions which are absent in the ihdm due to @xmath33 symmetry . in the ihdm
, the charged higgs @xmath42 can decay into @xmath82 followed by @xmath83 or @xmath84 .
therefore the decay product of the production processes @xmath85 , @xmath86 would be missing energy and multi - leptons or multi - jets depending on the decay product of w and z. such signature would be similar to some extent to the supersymmetric searches for charginos and neutralinos at @xmath87 or at hadron colliders @xcite . taking into account those considerations , we will assume that @xmath88 gev ( see @xcite for more details ) .
it is well known that for the sm higgs searches , the low mass @xmath90 $ ] gev , is the most challenging for lhc searches . in this low mass region ,
the main search is through the di - photons which can be complemented by the @xmath91 mode and potentially with the @xmath92 mode , while the @xmath93 channels are already competitive in the upper edge ( @xmath94@xmath95 gev ) of this mass range @xcite .
the sm predictions for the one - loop induced @xmath35 have been worked out since many years @xcite .
it is well known that the sm rate for @xmath35 is dominated by the w loops and the branching ratio of this channel is of the order of @xmath96 .
several studies have been carried out looking for large loop effects beyond sm .
such large effects can be found in various extensions of the sm , such as the minimal supersymmetric standard model ( mssm ) @xcite , the next - to - mssm @xcite , the two higgs doublet model @xcite , the little higgs models @xcite , extra - dimensions @xcite and in models with triplet higgs @xcite . in the ihdm
, the partial width of @xmath89 receives an additional contribution from the charged higgs boson loop which can both lower and raise the width compared to the sm .
it can be expressed as @xcite : @xmath97 where @xmath98 , @xmath99 are the color factor and the electric charge respectively for a particle @xmath100 running in the loop .
the dimensionless loop factors for particles of spin given in the subscript are : @xmath101 , \quad f_0 = \tau[1-\tau f(\tau)],\end{aligned}\ ] ] with @xmath102 ^ 2 , & \tau \geq 1 \\ -\frac{1}{4 } [ \ln(\eta_+/\eta_- ) - i \pi]^2 , & \ , \tau < 1 \end{array } \right.\ ] ] and @xmath103 in eq .
( [ widths ] ) , the coupling @xmath104 is given by @xmath105 it is clear from the above eq .
( [ hgaga_coupl ] ) , the coupling of the sm higgs boson to a pair of charged higgs is completely fixed by @xmath106 parameter . as we will see later
, the sign of @xmath106 will play an important role in the evaluation of the partial width of @xmath35 .
more important than the branching ratios , however , is the total cross - section of @xmath107 , since that is what is measured at the collider .
the largest contribution to the production cross - section for this observable @xmath108 is through gluon fusion , @xmath109 . for phenomenological purpose
, we define the ratio of the di - photon cross section normalized to sm rate as follow : @xmath110 where we have used the narrow width approximation in the first line of eq .
( [ ratio ] ) while we have used the fact that @xmath111 is the same both in the ihdm and sm .
one conclude that the ratio @xmath112 in the ihdm depend only on the branching ratio of @xmath35 . in the evaluation of the branching ratios , we use for the total decay widths the following expressions : @xmath113 where the expressions for the scalar decay widths are taken from @xcite .
note that the decays @xmath68 , @xmath114 and @xmath115 might be or not kinematically open . in the case where the dark matter particle is lighter than @xmath116 ,
the decay @xmath68 is open and could give substantial contribution to the total width of the higgs . for future use , we give here the analytical expression for @xmath117 coupling in the ihdm : @xmath118 which is proportional to @xmath119 , with @xmath120 .
before presenting our numerical results we would like to point out that in ref .
@xcite @xmath35 has been studied in 2hdm type i as well as in the ihdm .
but , ref .
@xcite only focused on the parameter space region where only sm decays namely @xmath121 decays of the sm higgs are kinematically allowed . in this case , the total width of the higgs boson is the same in sm and in ihdm and therefore our ratio @xmath112 given in eq .
( [ ratio ] ) reduce to : @xmath122 defined in ref .
our results agree with the results given in @xcite . in the case where @xmath68 is open , the ratio @xmath122 is not the appropriate one to be compared with cms and atlas data but rather the ratio of branching ratio as defined in eq .
( [ ratio ] ) .
moreover , as we will show , in our study we discuss the effect of all parameters that have some impact on the ratio @xmath112 like charged higgs mass , dark matter particle mass as well as the sign of the coupling @xmath123 .
we will also comment on constraints coming from wimp relic density . + as shown in fig .
[ fig:2 ] we can have a very interesting situation of evading the lhc bounds on sm higgs in the case where the invisible decay of higgs ( @xmath124 ) is kinematically allowed .
this issue will be discussed in details in forthcoming work @xcite . in our numerical analysis
, we perform a systematic scan over the parameter space of the ihdm .
we vary the ihdm parameters in the range @xmath125 in addition we have imposed @xmath126 and @xmath127 and @xmath128 .
this mass hierarchy ensures that @xmath129 could be the wimp dm candidate .
these values cover essentially the entire physically interesting range of parameters in the ihdm . for sm higgs ( @xmath39 )
we have specifically chosen a range where @xmath89 could be a important channel ( light higgs boson mass ) and the region that shows some deviations from sm as reported in recent lhc results @xcite,@xcite .
we have imposed the theoretical constraints mentioned above as well as constraints form oblique parameters @xmath40 and @xmath56 . + in addition , we would like to stress in passing that the coupling @xmath117 , which is proportional to @xmath130 ( see eq .
( [ hss_coupl ] ) ) , is an important piece for the calculation of the wimp relic density @xcite .
it has been show in @xcite , that with light higgs boson @xmath131 gev and @xmath132@xmath133 gev a relic density would be consistent with experimental data for @xmath134 . in the following numerical analysis , with low higgs mass 115140 gev , we will impose rather conservative limit on @xmath135 .
with all the above constraints discussed , we get the following limits : @xmath136 , @xmath137 gev and @xmath138 gev . in all plots , the coding color are : red ( ) and blue ( ) dots in scatter plots shown in figures [ fig:1 ] , [ addfig ] , [ fig:3],[fig:4 ] represents @xmath139 and @xmath140 respectively . in figure .
[ fig:1 ] , we show the allowed region in the @xmath141 ( left panel ) and @xmath142 ( right panel ) .
the perturbativity and vacuum stability constraints together dramatically reduce the allowed parameter space of the model .
in particular , the perturbativity and vacuum stability constraints excludes a large values of charged higgs mass @xmath66 and cp - odd mass @xmath143 while ewpt measurement constraint mainly the splitting between the scalar masses .
accordingly , an enhancement in is possible for relatively light charged higgs mass . + in the right panel of figure .
[ fig:1 ] we shown the scatter plot in @xmath144 space .
as can be seen again from this figure the enhancement in is possible only for negative values of @xmath106 .
note that the plots are symmetric under @xmath145 . in fig .
( [ addfig ] ) , we illustrate as a function of @xmath135 which is the main parameter contributing to the wimp relic density calculation . for large and negative @xmath146
one can see that can reach 1.6 while for large and positive @xmath146 , can be of the order 0.7 .
it is clear from this plot that even for small @xmath134 , which might be needed to accommodate wimp relic density @xcite , we still have both cases with @xmath147 and @xmath148 . as discussed in section.[section:1 ] due the presence of @xmath33 symmetry in ihdm the lightest higgs boson will be a stable particle . with the spectrum we have chosen
the lightest @xmath33 odd higgs boson is the neutral scalar @xmath40 and hence sm higgs will be having an invisible decay mode namely @xmath149 . for illustration , in figure .
[ fig:2 ] , we have fixed @xmath150 gev and shown the branching ratios as a function of @xmath46 for @xmath151 gev ( left plot ) and @xmath152 gev ( right plot ) . in the left panel of figure .
[ fig:2 ] , with @xmath153 gev , the invisible decay @xmath68 is open and dominate over all other sm decays except around @xmath154 where the coupling @xmath117 which is proportional to @xmath119 ( eq . [ hss_coupl ] ) gets suppressed and hence the situation becomes similar to sm . in the case where @xmath68 dominate , the partial width of @xmath68 contribute significantly to the total width of the higgs which becomes substantial resulting in a suppression of the @xmath155 which is always smaller than its sm value . for @xmath156 , the @xmath155 can reach the full sm value .
we can observe from figure .
[ fig:2 ] ( left ) that the branching fraction of the invisible decay of sm higgs ( @xmath124 ) could be very large resulting on a suppression of the other modes such as @xmath157 , @xmath158 , @xmath159 and @xmath160 and hence one can evade the present experimental constraints on the sm higgs mass based on @xmath158 , @xmath159 and @xmath160 . we will discuss this in future work @xcite .
the invisible decay of sm higgs could evade some of the constraints on sm higgs boson this has been extensively studied in many phenomenological studies @xcite . in the right panel of figure .
[ fig:2 ] , we take @xmath161 gev and then the decay @xmath68 is close . in this case , the total decay width of the higgs boson is similar in both sm and ihdm .
therefore , the partial decay width @xmath162 will receive only smooth variation through the charged higgs contribution . so that an enhancement of the branching ratio @xmath155 with up to a factor of 2 over the sm is possible .
in fact , in our parameterization of ihdm given in eq .
( [ eq : input_par_2 ] ) , @xmath106 is fixed by the charged higgs mass and @xmath46 parameter through eq .
( [ lambds ] ) .
the sign of @xmath106 is then completely fixed by the sign of @xmath163 .
hence , for small @xmath164 , the sign of @xmath106 is positive . in this case , the charged higgs contribution to @xmath162 is totally destructive with the sm . while for large @xmath165 , @xmath106 becomes negative and the charged higgs contribution to @xmath162 becomes constructive with sm and gets substantial enhancement . in fig .
[ fig:3 ] ( left panel ) we have shown @xmath112 as a function of @xmath106 .
the other parameters are taken as specified in eq .
( [ eq : para_scan ] ) . as can be seen from the left plot , as seen previously
, ihdm can increase the value of ( @xmath148 ) only for negative values of @xmath106 where the charged higgs contribution is constructive with the w loops . for positive @xmath106 ,
the charged higgs contribution is destructive with the w loops resulting in a suppression of @xmath166 rate .
+ the dependence of @xmath112 on the charged higgs mass is illustrated in the right panel of figure.[fig:3 ] .
the variation of @xmath112 as a function of @xmath167 scales almost like @xmath168 .
varying @xmath66 between 70 gev and 190 gev results in dramatic change of @xmath112 from 1.5 to 1 .
we stress that even for light charged higgs @xmath169 $ ] gev , we could have @xmath170 .
this could be due to the possible opening of the invisible decay @xmath68 which could reduce significantly the branching fraction of @xmath35 or to the fact that @xmath46 is rather small making @xmath106 positive .
+ note that if we relax the constraint on @xmath146 discussed above , we can get large @xmath106 in the following range : @xmath171 $ ] .
a large and negative @xmath172 would give a constructive charged higgs contribution with the w loops and therefore amplify which can reach 1.62.2 for light charged higgs @xmath173 gev . in fig .
[ fig:4 ] , we show @xmath112 as a function of @xmath129 and @xmath46 . from figure .
[ fig:4 ] ( left panel ) one can observe that an enhancement compared to sm value in is only possible for @xmath174 while for @xmath175 a suppression of is guaranteed . similarly one can observe that can be enhanced with respect to sm value for relatively large value of @xmath46 while for small @xmath176 gev is suppressed .
to summarize , in this work we study @xmath35 in the ihdm by imposing vacuum stability , perturbativity , unitarity and precision electroweak measurements .
we have shown that within the allowed range of the ihdm model @xmath89 could show substantial deviation from the sm result .
hence the observation of higgs boson in @xmath89 could help us in constraining the parameter space of the model .
we have also shown that observation of @xmath177 or @xmath178 could rule out a large portion of the allowed parameter space of ihdm .
+ taking into account all the constraints defined in section [ section:3 ] there is an upper bound on @xmath66 and @xmath179 as evident from figure [ fig:1 ] ( left panel ) .
this bound essentially comes from unitarity of the model .
if the cms and atlas excess in the di - photon channel is confirmed with more data , having @xmath148 would favor the following scenarios : * @xmath180 , i.e @xmath165 * charged higgs boson mass ( @xmath66 ) will be bounded ( @xmath181 gev ) . on the other hand ,
if with more data we have @xmath170 this scenario would favor either a light dm particle @xmath175 such that @xmath68 is open and/or a positive @xmath106
_ i.e. _ @xmath182 .
we would like to thank chuan - hung chen and gilbert moultaka for useful discussions .
a.a would like to thank nsc - taiwan for partial support during his stay at ncku where part of this work has been done .
the work of r.b was supported by the spanish consejo superior de investigaciones cientificas ( csic ) .
the work of n.g . is supported by grants from department of science & technology ( dst ) , india under project number sr / s2/hep-09/10 and university grants commission ( ugc ) , india under project number 38 - 58/2009(sr ) .
m. gustafsson , e. lundstrom , l. bergstrom and j. edsjo , phys .
lett . *
99 * , 041301 ( 2007 ) [ astro - ph/0703512 [ astro - ph ] ] ; t. hambye and m. h. g. tytgat , phys .
b * 659 * , 651 ( 2008 ) [ arxiv:0707.0633 [ hep - ph ] ] ; e. lundstrom , m. gustafsson and j. edsjo , phys .
d * 79 * , 035013 ( 2009 ) [ arxiv:0810.3924 [ hep - ph ] ] ; p. agrawal , e. m. dolle and c. a. krenke , phys .
d * 79 * , 015015 ( 2009 ) [ arxiv:0811.1798 [ hep - ph ] ] ; e. nezri , m. h. g. tytgat and g. vertongen , jcap * 0904 * , 014 ( 2009 ) [ arxiv:0901.2556 [ hep - ph ] ] ; s. andreas , m. h. g. tytgat and q. swillens , jcap * 0904 * , 004 ( 2009 ) [ arxiv:0901.1750 [ hep - ph ] ] ; c. arina , f. -s .
ling and m. h. g. tytgat , jcap * 0910 * , 018 ( 2009 ) [ arxiv:0907.0430 [ hep - ph ] ] ; l. lopez honorez and c. e. yaguna , jhep * 1009 * , 046 ( 2010 ) [ arxiv:1003.3125 [ hep - ph ] ] ; a. melfo , m. nemevsek , f. nesti , g. senjanovic and y. zhang , phys .
d * 84 * , 034009 ( 2011 ) [ arxiv:1105.4611 [ hep - ph ] ] .
e. m. dolle and s. su , phys .
d * 80 * , 055012 ( 2009 ) [ arxiv:0906.1609 [ hep - ph ] ] ; e. ma , phys .
d * 73 * , 077301 ( 2006 ) [ hep - ph/0601225 ] .
r. barbieri , l. j. hall and v. s. rychkov , phys .
d * 74 * , 015007 ( 2006 ) [ hep - ph/0603188 ] .
e. lundstrom , m. gustafsson and j. edsjo , phys .
d * 79 * , 035013 ( 2009 ) [ arxiv:0810.3924 [ hep - ph ] ] .
q. -h .
cao , e. ma and g. rajasekaran , phys .
d * 76 * , 095011 ( 2007 ) [ arxiv:0708.2939 [ hep - ph ] ] ; e. dolle , x. miao , s. su and b. thomas , phys .
d * 81 * , 035003 ( 2010 ) [ arxiv:0909.3094 [ hep - ph ] ] ; x. miao , s. su and b. thomas , phys .
d * 82 * , 035009 ( 2010 ) [ arxiv:1005.0090 [ hep - ph ] ] .
b. w. lee , c. quigg and h. b. thacker , phys .
d * 16 * ( 1977 ) 1519 .
r. casalbuoni , d. dominici , r. gatto and c. giunti , phys .
b * 178 * ( 1986 ) 235 .
r. casalbuoni , d. dominici , f. feruglio and r. gatto , nucl .
b * 299 * ( 1988 ) 117 .
s. kanemura , t. kubota and e. takasugi , phys .
b * 313 * , 155 ( 1993 ) [ arxiv : hep - ph/9303263 ] .
a. g. akeroyd , a. arhrib and e. m. naimi , phys .
b * 490 * , 119 ( 2000 ) [ arxiv : hep - ph/0006035 ] .
a. arhrib , arxiv : hep - ph/0012353 .
j. horejsi and m. kladiva , eur . phys .
j. c * 46 * , 81 ( 2006 ) [ arxiv : hep - ph/0510154 ] .
b. gorczyca and m. krawczyk , arxiv:1112.5086 [ hep - ph ] .
i. f. ginzburg and i. p. ivanov , phys .
d * 72 * ( 2005 ) 115010 [ arxiv : hep - ph/0508020 ] .
i. f. ginzburg and i. p. ivanov , arxiv : hep - ph/0312374 .
m. e. peskin and t. takeuchi , phys .
d * 46 * ( 1992 ) 381 .
k. nakamura _ et al . _
[ particle data group collaboration ] , j. phys .
g g * 37 * , 075021 ( 2010 ) .
s. kanemura , y. okada , h. taniguchi and k. tsumura , phys .
b * 704 * , 303 ( 2011 ) [ arxiv:1108.3297 [ hep - ph ] ] . the atlas & cms collaborations ; combined standard model higgs boson searches with up to @xmath183 of pp collision data at @xmath184 tev at the lhc `` ; atlas - conf-2011 - 157/cms - pas - hig-11 - 023 '' . j. r. ellis , m. k. gaillard and d. v. nanopoulos , nucl .
b * 106 * , 292 ( 1976 ) .
m. a. shifman , a. i. vainshtein , m. b. voloshin and v. i. zakharov , sov .
j. nucl .
phys . * 30 * , 711 ( 1979 ) [ yad .
fiz . * 30 * , 1368 ( 1979 ) ] .
a. djouadi , v. driesen , w. hollik and j. i. illana , eur .
j. c * 1 * , 149 ( 1998 ) [ arxiv : hep - ph/9612362 ] .
m. carena , s. gori , n. r. shah and c. e. m. wagner , arxiv:1112.3336 [ hep - ph ] .
s. moretti and s. munir , eur .
j. c * 47 * , 791 ( 2006 ) [ arxiv : hep - ph/0603085 ] .
u. ellwanger , phys .
b * 698 * , 293 ( 2011 ) [ arxiv:1012.1201 [ hep - ph ] ] .
p. posch , phys .
b * 696 * , 447 ( 2011 ) [ arxiv:1001.1759 [ hep - ph ] ] .
a. arhrib , w. hollik , s. penaranda and m. capdequi peyranere , phys .
b * 579 * , 361 ( 2004 ) .
i. f. ginzburg , m. krawczyk and p. osland , nucl .
instrum .
a * 472 * , 149 ( 2001 ) [ arxiv : hep - ph/0101229 ] .
d. lopez - val and j. sola , phys .
b * 702 * , 246 ( 2011 ) [ arxiv:1106.3226 [ hep - ph ] ] .
n. bernal , d. lopez - val and j. sola , phys .
b * 677 * ( 2009 ) 39 [ arxiv:0903.4978 [ hep - ph ] ] .
p. m. ferreira , r. santos , m. sher and j. p. silva , arxiv:1201.0019 [ hep - ph ] ; p. m. ferreira , r. santos , m. sher and j. p. silva , arxiv:1112.3277 [ hep - ph ] .
t. han , h. e. logan , b. mcelrath and l. t. wang , phys .
b * 563 * , 191 ( 2003 ) [ erratum - ibid .
b * 603 * , 257 ( 2004 ) ] [ arxiv : hep - ph/0302188 ] .
l. wang and j. m. yang , phys .
d * 84 * , 075024 ( 2011 ) [ arxiv:1106.3916 [ hep - ph ] ] .
k. cheung and t. c. yuan , arxiv:1112.4146 [ hep - ph ] .
a. arhrib , r. benbrik , m. chabab , g. moultaka and l. rahili , arxiv:1112.5453 [ hep - ph ] .
fileviez perez , h. h. patel , m. j. ramsey - musolf and k. wang , phys .
d * 79 * ( 2009 ) 055024 [ arxiv:0811.3957 [ hep - ph ] ] .
a. djouadi , phys .
* 457 * ( 2008 ) 1 [ arxiv : hep - ph/0503172 ] .
a. djouadi , phys .
rept . , 459 , 2008 , pages 1 - 241 , arxiv : hep - ph/0503173 .
a. arhrib , r. benbrik , n. gaur , _ work under progress_. k. belotsky , d. fargion , m. khlopov , r. konoplich and k. shibaev , phys .
d * 68 * , 054027 ( 2003 ) [ hep - ph/0210153 ] ; y. mambrini , phys .
d * 84 * , 115017 ( 2011 ) [ arxiv:1108.0671 [ hep - ph ] ] ; m. raidal and a. strumia , phys .
d * 84 * , 077701 ( 2011 ) [ arxiv:1108.4903 [ hep - ph ] ] ; x. -g .
he and j. tandean , phys .
d * 84 * , 075018 ( 2011 ) [ arxiv:1109.1277 [ hep - ph ] ] ; i. low , p. schwaller , g. shaughnessy and c. e. m. wagner , arxiv:1110.4405 [ hep - ph ] ; c. englert , j. jaeckel , e. re and m. spannowsky , arxiv:1111.1719 [ hep - ph ] ; y. bai , p. draper and j. shelton , arxiv:1112.4496 [ hep - ph ] ; x. -g .
he , b. ren and j. tandean , arxiv:1112.6364 [ hep - ph ] ; c. englert , t. plehn , m. rauch , d. zerwas and p. m. zerwas , arxiv:1112.3007 [ hep - ph ] ; a. djouadi , o. lebedev , y. mambrini and j. quevillon , arxiv:1112.3299 [ hep - ph ] . | motivated by the recent result reported from lhc on the di - photon search for a standard model ( sm ) higgs - like boson .
we discuss the implications of this possible signal in the framework of the inert higgs doublet model ( ihdm ) , taking into account previous limits from higgs searches at lep , the tevatron and the lhc as well as constraints from unitarity , vacuum stability and electroweak precision tests .
we show that the charged higgs contributions can interfere constructively or destructively with the w gauge bosons loops leading to enhancement or suppression of the di - photon rate with respect to sm rate .
we show also that the invisible decay of the higgs , if open , could affect the total width of the sm higgs boson and therefore suppress the di - photon rate . * in inert higgs doublet model * * abdesslam arhrib @xmath0 , rachid benbrik @xmath1 , naveen gaur @xmath2 , * 1.5 cm |
the large hadron collider ( lhc ) being constructed at cern , will generate proton - proton collisions with total c.m .
energy of 14 tev with a design luminosity @xmath8 @xmath9 @xmath10 s@xmath11 .
the experiments at this new facility will have a high potential to discover new physics and to make various precision measurements , see , for example @xcite .
both the general purpose @xmath0 experiments , atlas @xcite and cms @xcite , will provide high statistics data samples , and the accuracy of the precision measurements will be limited by systematic effects and , in many cases , by the uncertainty in the measurement of the luminosity @xmath8 .
for example , precision measurements in the higgs sector of typical accuracy of about 7% , for a wide range of possible higgs mass , require the uncertainty in the luminosity to be @xmath12 @xcite .
both lhc experiments are aiming at measuring the luminosity to 5% .
recall also that the enhancement in luminosity that will be achieved in run ii of the fermilab tevatron @xmath1 - collider will herald a new era of precision studies @xcite .
an obvious requirement for the success of these precision measurements is that the uncertainties in the theoretical calculations of the cross sections for the basic processes , which are to be used to determine the luminosity , should match the desired experimental accuracies . in general , there are two possibilities to determine the luminosity either ( i ) to measure a pair of cross sections which are connected quadratically with each other , or ( ii ) to measure a cross section whose value is well known or which may be calculated with good accuracy .
the well - known example of the first possibility is the measurement of the total @xmath13 and differential forward elastic cross sections which are related by the optical theorem ; see , for example , @xcite for recent experimental discussions .
this method is discussed further in section 2 .
two types of processes stand out as examples of the second possibility to measure the luminosity .
first there is exclusive lepton - pair production via photon - photon fusion [ eq : g1 ] + l^+l^- + where @xmath14 or @xmath15 . to the best of our knowledge this proposal originated in @xcite .
a luminometer for the lhc based on measuring forward @xmath16 pairs ( of invariant mass @xmath17 mev and transverse momentum @xmath18 mev ) was proposed in @xcite , while ref .
@xcite concerns the central production of @xmath19 pairs ( with @xmath20 20 gev and @xmath21 1050 mev ) .
lepton - pair production is the subject of section 3 .
nowadays attention has also focussed on @xmath2 and @xmath3 production as a possible luminosity monitor , both for run ii at the tevatron and for the lhc see , for example , @xcite .
the reason is that the signal is clean , and the production cross sections are large and can now be calculated with considerable theoretical accuracy .
we discuss this possibility further in section 4 . in principle
, we may _ monitor _ the luminosity using _ any _ process , with a significant cross section , which is straightforward to detect .
for example , it could be single- ( or two- ) pion inclusive production in some rapidity and @xmath22 domain ( for example @xmath23 ) or inclusive @xmath24 production in a well - defined kinematic domain , etc . in this way
we may control the relative luminosity and then calibrate the monitor by comparing the number of events detected for the monitor reaction with the number of events observed for a process with a cross section which is already known , or which may be calculated with sufficient accuracy .
for some applications better accuracy can be achieved by measuring the parton - parton luminosity .
a discussion is given in section 5 .
finally , section 6 contains our conclusions .
first we discuss the classic method to measure the luminosity , that is using the observed total and forward elastic @xmath0 ( or @xmath1 ) event rates . neglecting coulomb effects should be multiplied by the well - known bethe phase @xcite @xmath25 . the uncertainty in bethe phase @xmath26 due to the non - point - like structure ( electric charge distribution ) of the proton is about @xmath27 .
it leads to an uncertainty of about @xmath28 in the re / im ratio , and hence to less than a 0.1% correction for the imaginary part of the strong amplitude @xmath29 .
a recent discussion of the coulomb phase for @xmath30 can be found in @xcite .
] , the elastic cross section is given by [ eq:2a ] .
|_t = 0 = ( 1 + ^2 ) .
the ratio @xmath31 of the real to the imaginary parts of the forward amplitude is small at tevatron - lhc energies and can be estimated via dispersion relation techniques [ eq:2b ] .
for example , at the lhc energy , @xmath32 , it is predicted to be @xmath33 ; for a recent estimate , see , for example , ref .
even the largest conceivable uncertainty in @xmath34 that is @xmath35 , leads only to an uncertainty of less than 0.5% in @xmath36 .
thus if we measure the number of events corresponding to the elastic scattering and to the total cross section , then we may determine both the luminosity @xmath37 and @xmath38 ( since @xmath39 , whereas @xmath40 ) .
the main problem in that it is extremely difficult to make measurements at the lhc in the very forward region , and so it is necessary to extrapolate elastic data from , say , @xmath41 to @xmath42 . on the other hand it was found at the isr that the observed local slope [ eq:2c ] b ( t ) depends on @xmath43 . in particular at @xmath44 @xcite [ eq:2d ] b = b ( 0 ) - b ( |t| = 0.2 gev^2 ) 2 gev^-2 .
note that if this @xmath43 dependence were to be solely due to a non - linear pomeron trajectory @xcite , then the difference ( [ eq:2d ] ) would increase as the logarithm of the energy @xmath45 ; giving , for example , @xmath46 at the lhc energy .
the @xmath43-dependence of the local slope @xmath47 has recently been determined @xcite using a model which incorporates all the main features of high energy soft diffraction .
that is the model embodies * the _ pion - loop _ contribution to the pomeron pole ( which is the main source of the non - linearity of the pomeron trajectory @xmath48 ) , * a _ two - channel _ eikonal to include the pomeron cuts which are generated by elastic and quasi - elastic ( with @xmath49 intermediate states ) @xmath50-channel unitarity , * the effects of high - mass _ diffractive dissociation_. the parameters of the model are @xmath51 and @xmath52 of the _ bare _ pomeron trajectory , two parameters describing the elastic proton - pomeron vertex , and the triple - pomeron coupling and its slope @xcite .
the values of the parameters were tuned to describe the observed @xmath0 ( or @xmath1 ) elastic differential cross sections throughout the isr - tevatron range , restricting the description to the forward region @xmath53 .
note that there are two effects , leading to a @xmath43 dependence of @xmath54 , which act in opposite directions .
first , the non - linear pion - loop contributions @xcite to the pomeron trajectory , @xmath55 , lead to a contribution @xmath56 in single pomeron exchange , which increases as @xmath57 . on the other hand the absorptive ( rescattering ) corrections , associated with eikonalization , lead to a dip in @xmath58 ( whose position moves to smaller @xmath59 as the collider energy , @xmath60 , increases ) , with the result that the local slope @xmath47 grows as @xmath61 approaches the position of the diffractive minimum ; that is @xmath47 decreases as @xmath57 .
fortunately , just at the lhc energy , these two effects almost compensate each other .
so far it has proved impossible to sum up in a consistent way all the multi - pomeron diagrams .
for this reason two versions of the model were studied in ref .
@xcite with _ maximal _ and _ minimal _ contributions from high - mass diffractive dissociation . after tuning the values of the parameters to describe the @xmath38 and the forward @xmath58 data ,
the two versions of the model predict different values of the slope at @xmath62 .
the maximal choice gives a larger value @xmath63 , while for the minimal case we obtain @xmath64 .
however in both cases the variation of @xmath65 satisfies @xmath66 for @xmath67 , or @xmath68 for the more restricted interval @xmath69 .
it means that we may neglect such a small variation of @xmath47 with @xmath43 and safely extrapolate the observed cross section @xmath58 down to @xmath42 , using the slope measured , say , in the interval @xmath70 .
the error , due to the variation of @xmath54 with @xmath43 , is expected to be less than @xmath71 . in the very forward region , @xmath72
, it may only be possible to measure the elastic cross section in an lhc run with a rather low luminosity .
for the high luminosity runs one would then have to use another monitor .
one possibility is to choose a single particle inclusive process , with a significant cross section , and calibrate it in the low luminosity run by comparing with the total and elastic cross sections discussed above
. it would be even better if we were able to calibrate several different reactions .
then to use , as a luminosity monitor , the reaction with the closest topology ( or kinematic configuration ) to the process that we are to study in the high luminosity run .
at first sight any qed process , with sufficient event rate , may be used as a luminosity monitor . up to isr energies the luminosity was measured via coulomb elastic @xmath0 scattering , where at very small @xmath59 the influence of the dominant single photon exchange contribution is evident .
unfortunately , in the lhc environment , the coulomb interference region , @xmath73 is not likely to be experimentally accessible . other possibilities to consider are @xmath74 isobar coulomb excitation , ( @xmath75 ) , or @xmath76 and @xmath77 production , which are all mediated by photon exchange , as shown in figs .
1(a ) and 2(a ) respectively .
surprisingly , the electromagnetic widths @xmath78 and @xmath79 are known only to 710% accuracy @xcite .
moreover , strong interaction effects in the initial and final states give non - negligible corrections , see figs .
1(b ) and 2(b ) .
lepton pair production looks much more promising as a luminosity monitor .
the born amplitude of fig .
3(a ) may be calculated within pure qed ( see , for example , the reviews in @xcite ) , and there are no strong interactions involving the leptons in the final state .
the only question is the size of the absorptive corrections arising from inelastic proton - proton rescattering , sketched in fig .
fortunately the rescattering correction is suppressed for two reasons .
first , the main part of the born cross section ( fig .
3(a ) ) comes from the peripheral region with large impact parameter @xmath80 , where the strong amplitude @xmath81 is small .
second , even in the small @xmath80 domain , the rescattering correction is greatly suppressed due to the angular integration .
before we discuss the last point , we note that , in practice , it is difficult to exclude contributions coming from the reactions [ eq : f1 ] pp x + l^+ l^- + p + pp x + l^+ l^- + y where @xmath82 and @xmath83 are baryon excitations , that is @xmath49 or @xmath74 isobars .
of course the matrix elements of the corresponding processes ( such as fig .
3(c ) ) can , in principle , be determined from photoproduction and deep inelastic data , and may be taken into account . however the matrix elements of fig .
3(c ) are not known to sufficient accuracy and it is better to suppress the contributions of ( [ eq : f1 ] ) by experimental cuts .
the procedure is as follows .
recall that , due to gauge invariance , inelastic vertices of the type @xmath84 vanish like @xcite [ eq : f2 ] v(p x ) q_1 t as the photon transverse momentum @xmath85 .
unfortunately it is difficult to measure a leading proton with very small transverse momentum , that is @xmath86 .
so to take advantage of the behaviour of ( [ eq : f2 ] ) , it was proposed @xcite to select events with very small transverse momentum of the lepton pair [ eq : f3 ] p_t | _ l^+t + _
l^-t | < 10 - 30 mev , in order to suppress @xmath49 and @xmath74 production .
in such a case the integral over the transverse momenta of the photons in the born cross section of fig .
3(a ) takes the form [ eq : f4 ] ^(2 ) ( _ t - _ 1 t - _ 2 t ) d^2 p_t , where the dominant contributions come from the regions @xmath87 .
the values of the longitudinal components are [ eq : f5 ] t_i | t_i , min | = x^2_i m^2_p , where @xmath88 are the fractions of the momenta of the incoming protons carried by the @xmath89 pair , and @xmath90 is the mass of the proton .
to identify muons ( and to separate them from @xmath91 mesons ) it is necessary to observe charged particles after they have transversed a thick ( iron ) absorber .
it means that we consider only muons which have a rather large transverse energy @xmath92 gev .
however , it is still possible to select events where the sum of their transverse momenta is small , @xmath93 mev . in this case
[ eq : f6 ] x_i ( 2 - 3 ) e_t/ 10 ^ -3 , t_i 10 ^ -6 gev^2 , at the lhc energy . in such a configuration the main contribution to the integral ( [ eq : f4 ] ) comes from the domains @xmath94 and @xmath95 .
performing the @xmath96 integrations in these two domains gives @xmath97 and @xmath98 respectively , and so the born cross section behaves as [ eq : f7 ] ln ( ) in addition to the use of the small @xmath22 cut in order to separate the elastic process ( [ eq : g1 ] ) from the excitation process ( [ eq : f1 ] ) , we can also exploit the different kinematics of the processes .
for example in @xcite it was proposed to fit the observed distribution in the muon acoplanarity angle @xmath99 in order to isolate the elastic mechanism via its prominent peak at @xmath100 .
to determine the effect of the absorptive , or re - scattering , correction , we calculate diagram 3(b ) with an extra loop integration over the momentum @xmath102 transferred via the strong interaction amplitude shown by the blob " .
the relative size of the correction , to the amplitude for lepton - pair production of fig .
3(a ) , is [ eq : f8 ] = , where @xmath103 represents the collective effect of the appropriate proton electromagnetic form factors @xmath104 and the @xmath105 dependence of the strong interaction amplitude [ eq : f8b ] f ( q ) = f_n ( q_1 ^ 2 ) f_n ( q_2 ^ 2 ) f_n ( ( q + q_1)^2 ) f_n ( ( q + q_2)^2 ) e^b q^2/2 .
@xmath106 is the slope of the @xmath0 elastic cross section , that is @xmath107 .
we can interpret the correction @xmath108 in terms of @xmath109 , where @xmath110 is the survival probability amplitude that the secondary hadrons produced in the soft rescattering do not accompany @xmath111 production .
of course this effect is due to the inelastic strong interaction only . consider the hypothetical case with pure elastic @xmath0 rescattering ( at fixed impact parameter @xmath80 , that is for fixed partial wave @xmath112 ) .
then elastic rescattering only changes the phase of the qed matrix element , @xmath113 , and does not alter the qed cross section @xmath114 .
thus we have to account only for the inelasticity of the strong interaction , that is @xmath115 as in ( [ eq : f8 ] ) is that we have to sum up multiple @xmath0 rescattering . on resumming all the eikonal graphs ,
it turns out that one obtains ( [ eq : f8 ] ) with @xmath116 . ] . throughout this paper
we do not discuss the effects of pile - up events .
thus there will be an additional factor @xmath117 which represents the probability not to include events with extra secondaries coming from an almost simultaneous interaction of another pair of protons .
experimentally the depletion due to such pile - up events may be overcome , for example , by cleanly observing the vertex of @xmath24 production . the cross section @xmath118 in ( [ eq : f8 ] )
plays the role of the cross section for the qed subprocess of @xmath119 scattering through the lepton box . for the absorptive process
, the photons have momenta @xmath120 and @xmath121 in the left `` amplitude @xmath122 shown in fig .
3(b ) , and @xmath123 in the right '' amplitude @xmath124 , which is not shown . here
we have assigned the absorptive effect to @xmath122 .
actually the numerator of ( [ eq : f8 ] ) , and the symbolic ) the only qualification is that when the photon is emitted inside the ladder , the form factor @xmath104 may have different behaviour at large @xmath125 . in our case , when @xmath125 is very small , this effect is negligible . ]
diagram 3(b ) , summarize a set of feynman graphs which describe the effects of the strong interaction between the protons in the amplitude @xmath122 .
then we have to add the equivalent set of diagrams in which the rescattering effects occur in the amplitude @xmath124 . the total correction to ( [ eq : f7 ] )
is therefore @xmath126 .
the rescattering correction is due to the imaginary part of the strong amplitude only .
the effects coming from the real part in @xmath122 cancel with those in @xmath124 . at first sight
the largest absorptive effects , ( [ eq : f8 ] ) , appear to come from the largest values of @xmath127 , where @xmath128 is the proton radius ; higher values of @xmath125 are cut - off by the proton form factor .
however when @xmath129 an interesting suppression occurs . in this domain @xmath130 occurs dominantly in the @xmath131 two - photon state .
the projection of the orbital angular momentum on the longitudinal @xmath132 axis is clearly zero , while the spin ( which originates from the photon polarisations ) is described by the tensor @xmath133 .
after the azimuthal integration , the tensor takes the form @xmath134 , leading to @xmath131 .
if we neglect the mass of the lepton , @xmath135 , and higher order qed corrections , then the born amplitude , with @xmath131 , vanishes by pumplin @xcite , see also @xcite . ] .
this is a well known result , see , for example , @xcite . as shown in ref .
@xcite , the physical origin of the suppression is related to the symmetry properties of the @xmath131 born amplitude .
the cross section @xmath136 for our qed subprocess depends on the transverse momenta of four virtual photons .
the full expression for @xmath118 is rather complicated .
however in the limit @xmath137 we can integrate over the direction of the lepton transverse momentum @xmath138 ( with @xmath139 ) , and the formula reduces to the simple form [ eq : f9 ] where the notation @xmath140 is used for the scalar product of two _ transverse _ vectors , and where the rapidity difference [ eq : f10 ] | ( l^+ ) - ( l^- ) | . it is interesting to note that the dependence of the cross section @xmath118 on @xmath141 and @xmath125 follows directly from the symmetry properties of the process .
first , gauge invariance implies that @xmath118 must vanish when the transverse momentum of any photon goes to zero . that is @xmath118 must contain a factor @xmath142 .
next , it has to be symmetric under the interchanges @xmath143 and @xmath144 . finally , as discussed above , in the limit @xmath145
, it must vanish after the azimuthal angular integration of @xmath146 has been performed . the only possibility to satisfy these conditions , to lowest order in @xmath147 and @xmath148 , is given by the expression in square brackets in ( [ eq : f9 ] ) . to obtain the @xmath149 and @xmath150 behaviour , shown in the first factor in @xmath118 of ( [ eq : f9 ] )
, it is sufficient to put @xmath151 and to recall the well known pure qed cross section ) for @xmath118 . ] .
recall that for @xmath19 production we select events with small transverse momentum of the lepton pair , @xmath22 . now
if @xmath152 , that is if @xmath153 , then the last factor in expression ( [ eq : f9 ] ) of the absorptive cross section simplifies to [ eq : f11 ] , which , upon integration over the azimuthal angle of the vector @xmath154 , gives zero . therefore the main contribution to ( [ eq : f8 ] ) comes from values of @xmath155 , and the absorptive correction [ eq : f12 ] p^2_t c. the coefficient @xmath156 is numerically small for two reasons . first , the integration over the loop momentum @xmath125 kills the logarithmic factor , ln@xmath157 , which enhanced the original cross section ( [ eq : f7 ] ) in the absence of absorptive corrections . to see this , note that the @xmath158 came from the @xmath159 integrations , for both @xmath160 and @xmath161 , of ( [ eq : f4 ] ) , see ( [ eq : f7 ] )
however when @xmath162 the integrands are of the form @xmath163 , and there is no logarithmic singularity for either @xmath164 or @xmath165 .
second , there is a suppression of @xmath156 from cancellations which occur after integration over the remaining azimuthal angles .
for example , if the invariant mass of the @xmath24 pair produced at zero rapidity is @xmath166 , then the value of the coefficient @xmath167 and 0.08 for @xmath168 and 50 mev respectively .
this leads to a negligible correction to the cross section ( [ eq : f7 ] ) , for example [ eq : f13 ] 2 p^2_t c < 0.02% ( 0.13% ) for @xmath169 mev
( 30 mev ) .
note that the rescattering contribution of fig .
3(b ) is less singular as @xmath170 than the pure qed term of fig .
therefore the rescattering correction does not induce a sharp peak at @xmath100 in the muon acoplanarity distribution . thus removing the excitation processes ( [ eq : f1 ] ) by fitting the @xmath99 distribution ,
will automatically suppress the rescattering correction .
unlike @xmath24 production , for @xmath171 production we do not need to select events with electrons of large transverse momentum @xmath172 . providing the energy of the electrons @xmath173
, we may use , say , an electromagnetic calorimeter to identify them .
therefore we may consider the small @xmath172 domain where the @xmath171 production cross section is much larger .
the dominant contribution to the cross section comes from the region of @xmath172 of the order of the electron mass ( @xmath174 mev ) . as @xmath172 is small
, we can not neglect @xmath125 in the @xmath43 channel electron propogator . as a consequence the integral over @xmath125 becomes super - convergent [ eq : f14 ] .
it is informative to note the origin of this form . due to gauge invariance ,
the numerator vanishes when the photon transverse momenta @xmath175 .
the first two factors in the denominator arise from the photon propagators .
finally we have the factor due to the electron propagator , with @xmath176 .
contrary to @xmath24 production , where the lepton propagator was driven by the large @xmath177 of the muon , here the momentum @xmath125 is not negigible in comparison with @xmath172 and the dominant contribution comes from the region @xmath178 gev@xmath179 .
hence the expected absorptive correction due to strong interaction rescattering is [ eq : f15 ] 2 q^2_t ~ 10 ^ - 6 gev^2 10 ^ -5 we see that both @xmath19 and @xmath16 pair production may be used as monitor processes in high luminosity lhc collisions .
however we must employ special selection cuts on the @xmath101 data . to suppress @xmath49 contamination
we impose small transverse momentum @xmath22 of the lepton pair , moreover for @xmath16 production we require the individual electron @xmath172 to be small .
we also require the leptons to be sufficiently energetic in order to identify them .
then the cross section for @xmath101 production at the lhc @xmath180 can be calculated within pure qed and , _ importantly _ , we may neglect the strong rescattering effects between the protons up to @xmath181 accuracy .
of course by requiring the muons to have high @xmath150 we have a rather small cross section .
however here we may trace back the muon tracks and determine the interaction vertex , and hence isolate the interaction in pile - up events .
so , in principle , @xmath19 pair production , with high @xmath150 muons , may be used as a luminometer in very high luminosity lhc runs .
@xmath2 and @xmath3 production in high energy @xmath0 and @xmath1 collisions have clean signatures through their leptonic decay modes , @xmath182 and @xmath183 , and so may be considered as potential luminosity monitors @xcite .
a vital ingredient is the accuracy to which the cross sections for @xmath2 and @xmath3 production can be theoretically calculated .
the cross sections depend on parton distributions , especially the quark densities , in a kinematic region where they are believed to be reliably known .
recent determinations of the @xmath2 and @xmath3 cross sections can be found in refs .
@xcite . here
the situation has improved , and next - to - next - to - leading order ( nnlo ) predictions have been made pair production and for @xmath2 bosons produced via @xmath43-quark decays .
these can contribute about 1% of the total signal . ] .
the most up - to - date values are reproduced in fig .
4 @xcite and figs .
5,6 @xcite . to estimate the accuracy with which the @xmath2 and @xmath3 cross sections are known , we start with fig .
4 was obtained using parton distributions found in lo , nlo and nnlo global analyses of the same data set @xcite .
it is relevant to summarize the contents of these plots .
the predictions labelled lo , nlo and nnlo can be schematically written as follows [ eq : f16 ] _ lo & = & f_lo f_lo + _ nlo & = & f_nlo f_nlo + _ nnlo & = & f_nnlo f_nnlo where the label on @xmath184 indicates the order to which the @xmath185-function is evaluated .
the parton distributions @xmath186 obtained from the new global lo , nlo and nnlo analyses @xcite correspond to @xmath187 0.1253 , 0.1175 and 0.1161 respectively .
the nlo and nnlo coefficient functions @xmath188 are known @xcite . however , although the relevant deep inelastic coefficient functions are also known , there is only partial information on the nnlo splitting functions .
recently van neerven and vogt @xcite have constructed compact analytic expressions for the nnlo splitting functions which represent the fastest and the slowest possible evolution that is consistent with the existing partial information .
these expressions have been used to perform global parton analyses at nnlo @xcite .
the uncertainty in the nnlo predictions of @xmath189 and @xmath190 due to the residual ambiguity in the splitting functions is shown by the width of the nnlo bands in fig .
this amounts to about @xmath191 uncertainty at the lhc energy , and less at the tevatron . for completeness
, we note that the dashed lines in fig .
4 correspond to the quasi - nlo prediction [ eq : f17 ] _
nlo^ = f_nlo f_nlo , which was the best that could be done before the work of refs .
@xcite .
we see that the lo @xmath192 nlo @xmath192 nnlo convergence of the predictions for @xmath193 is good .
the jump from @xmath194 to @xmath195 is mainly due to the well - known , large , @xmath196 @xmath197-enhanced drell - yan k - factor correction , arising from soft - gluon emission .
the nlo and nnlo cross sections are much closer and , if this was the end of the story , @xmath2 and @xmath3 production can clearly be predicted with sufficient accuracy .
however let us turn to figs . 5 and 6 ,
each of which combine results presented in refs . @xcite and @xcite .
the solid squares and triangles show the additional uncertainty in the predictions for @xmath189 and @xmath190 which arise from changing the input information in the global parton analyses .
the two major uncertainties appear to be due to the value of @xmath184 and to using different parton densities labelled by q@xmath198 and q@xmath199 .
the plots show the change in @xmath189 and @xmath190 which is caused by changing the value of @xmath200 by @xmath201 respectively .
the change in @xmath189 and @xmath190 at the lhc energy is enhanced as compared to that at the tevatron , since dglap evolution is more rapid at the smaller @xmath202 values , @xmath203 , probed at the higher energy . however , with our present knowledge of @xmath184 , the uncertainty @xmath204 is too conservative , and @xmath205 is a more realistic uncertainty in @xmath193 from this source at the lhc energy .
the normalisation of the input data used in the global parton analyses is another source of uncertainty in @xmath193 .
the hera experiments provide almost all of the data used in the global analyses in the relevant small @xmath202 domain .
the quoted normalisation uncertainties of the measurements of the proton structure function @xmath206 from the h1 and zeus experiments vary with @xmath105 , but a mean value of @xmath207 is appropriate .
the q@xmath198 and q@xmath199 parton sets correspond to separate global fits in which the hera data have been renormalised by @xmath207 respectively . for @xmath2 and @xmath3 production , of @xmath208 origin
, we naively would expect this to translate into a @xmath209 variation in @xmath193 , but the effect of dglap evolution up to @xmath210 is to suppress the difference in the predictions . in summary , allowing for all the above uncertainties , we conclude that the cross sections of @xmath2 and @xmath3 production are known to @xmath211 at the lhc energy , and to @xmath212 at the tevatron .
a major contributer to this error is the uncertainty in the overall normalisation of the h1 and zeus measurements of @xmath206 .
the normalisation may be made more precise by experiments in run ii at the tevatron .
in some circumstances it is sufficient to know the parton - parton luminosity , and not the proton - proton luminosity , see , for example , @xcite . of course
if the proton - proton luminosity is known then the parton - parton luminosities can be calculated from the parton distributions determined in the global parton analyses .
however in this case we rely on the normalisation of experiments at previous accelerators which yielded data that were used in the global analyses
. thus it may be better to monitor the parton - parton luminosities directly in terms of a subprocess which can be predicted theoretically to high precision .
the best example is inclusive @xmath2 ( or @xmath3 ) boson production , which is predicted up to two - loops , that is to nnlo @xcite .
the accurate observation of @xmath2 ( or @xmath3 ) production at the lhc may therefore be used to determine the quark - antiquark luminosity luminosity in a given @xmath213 bin may be obtained by observing the number of @xmath2 events in that bin and dividing by the @xmath214 cross section . however at nlo
we include @xmath215 etc .
, so the only possibility is to use the new @xmath2 data to determine the @xmath208 luminosity within a global parton analysis .
to do the same at nnlo we would require the nnlo expression for @xmath216 , where @xmath217 is the rapidity of the decay lepton . ] .
then the gluon flux , for example , may be determined from the global parton analysis which already is made to describe the measured @xmath2 ( or @xmath3 ) cross sections .
one advantage of this technique is that , for most lhc applications such as the search for new heavy particles , we need to consider a smaller interval of dglap evolution than has been the practice hitherto . other ways to constrain the gluon - gluon luminosity are to study @xmath218 production or the production of two large @xmath22 prompt photons .
the leading order subprocess is @xmath219 .
however at lhc energies the @xmath220 box diagram gives an important contribution .
the two diagrams are shown in fig .
the relative contributions are shown in fig .
8 as a function of the transverse momentum of the photon pair @xmath221 for the case when each photon has transverse momentum @xmath222 and rapidity @xmath223 .
we see that in this kinematic domain the @xmath224 subprocess gives a major contribution due to the higher @xmath225 luminosity .
however there is a strong possibility of contamination by the subprocess @xmath226 , unless severe photon isolation cuts are imposed , see , for example , ref .
@xcite .
so far we have considered conventional parton distributions @xmath227 integrated over the parton transverse momentum @xmath228 up to the factorization scale @xmath15 .
however many reactions are described by unintegrated distributions @xmath229 which depend on both @xmath228 and the longitudinal momentum fraction @xmath202 carried by the parton . in principle , unintegrated distributions are necessary for the description of all processes which are not totally inclusive . in these cases , instead of the conventional qcd factorization
, we have @xmath228 factorization @xcite .
indeed , for some processes , after the specific integration over @xmath228 , the conventional hard qcd factorization may be destroyed .
the unintegrated quark - quark luminosity may be determined , for example , by observing the @xmath228 distribution of @xmath2 ( or @xmath3 ) bosons or drell - yan pairs . at leading order ,
the @xmath228-dependence of cross sections is given by the convolution [ eq:2e ] ^(2 ) ( _ 1 + _ 2 - _ t ) f_q ( x_1 , p_1 ^ 2 , ^2 ) f_|q ( x_2 , p_2 ^ 2 , ^2 ) .
at large @xmath230 , to leading @xmath231 order , the dominant contribution comes from either the domain @xmath232 or the domain @xmath233 .
it is natural to choose the factorization scale @xmath234 .
then the @xmath228 distribution takes the simple form
@xcite [ eq:2f ] , since at leading order [ eq:2 g ] ^k_t^2 f_a ( x , p^2 , k_t^2 ) = xa ( x , k_t^2 ) . moreover diffractive processes ( with one or more rapidity gaps ) are described , in general , by skewed parton distributions .
an example is the double - diffractive exclusive higgs boson production , described by the feynman diagram shown in fig .
this process is described by skewed gluon distributions with @xmath235 and @xmath236 @xcite .
sometimes even the skewed parton flux can be monitored directly via another process with similar kinematics .
for example , we may monitor the effective pomeron - pomeron luminosity for higgs production of fig .
9 by measuring double- diffractive dijet production in the region in which the transverse energy of the jets ( @xmath150 ) is of about half that of the higgs mass @xmath237 @xcite .
fortunately most of the physically relevant processes are described by skewed distributions with @xmath238 . in these cases the skewed distributions
may be reliably reconstructed from the known conventional parton distributions @xcite .
we have studied the theoretical accuracy of the main three proposals for determining the luminosity of the lhc @xmath0 collider namely using forward elastic data , lepton - pair production and @xmath2 or @xmath3 boson production .
the desired goal of a measurement to within @xmath239 seems theoretically attainable .
we focused on potential shortcomings of each method .
first , we demonstrated that the @xmath43 dependence of the elastic cross section is well under control and , in fact , it turns out that it can be safely approximated by a simple exponential in the region @xmath240 . for lepton - pair production we evaluated the corrections to the cross section arising from the strong interactions between the protons .
we showed that in the relevant kinematic domain , with small transverse momentum of the produced lepton - pair , these effects are negligible .
hence a pure qed calculation of the cross section will give sufficient accuracy .
the cross sections for @xmath2 and @xmath3 production can now be predicted to nnlo , which at first sight would seem to provide an lhc luminometer with @xmath191 accuracy , see fig .
however the uncertainties in the input to the global parton analyses mean that the error could , conservatively , be as large as @xmath241
. the uncertainty may be reduced if we work in terms of quark - antiquark luminosity , which is relevant in some future applications of the lhc .
finally , we note that luminosity determinations based on the measurement of the forward elastic cross section and ( most probably ) on two - photon @xmath171 production can only be made in low luminosity runs , and require dedicated detectors and triggers . on the other hand ,
the measurement of @xmath2 or @xmath3 and two - photon @xmath24 production may be performed at high luminosity with the central detector and with standard triggers .
we thank v.s .
fadin , m.a .
kimber , v. nomokonov , a. shamov , a.g .
shuvaev , w.j . stirling and s. tapprogge for useful discussions .
vak thanks the leverhulme trust for a fellowship .
mgr thanks the royal society and pparc for support , and also the russian fund for fundamental research ( 98 - 02 - 17629 ) .
this work was also supported by the eu framework tmr programme , contract fmrx - ct98 - 0194 ( dg 12-miht ) .
s. catani , m. ciafaloni and f. hautmann , phys . lett .
* b242 * ( 1990 ) 97 ; nucl . phys . *
b366 * ( 1991 ) 135 ; + s. catani and f. hautmann , nucl . phys .
* b427 * ( 1994 ) 475 ; + j.c .
collins and r.k .
ellis , nucl .
b360 * ( 1991 ) 3 . | we study the theoretical accuracy of various methods that have been proposed to measure the luminosity of the lhc @xmath0 collider , as well as for run ii of the tevatron @xmath1 collider .
in particular we consider methods based on ( i ) the total and forward elastic data , ( ii ) lepton - pair production and ( iii ) @xmath2 and @xmath3 production . plus 2 mm minus 2 mm 23.0 cm 17.0 cm -1.0 in -42pt ippp/00/01 + dtp/00/58 + 16 october 2000 + * luminosity monitors at the lhc * v.a .
khoze@xmath4 , a.d .
martin@xmath4 , r. orava@xmath5 and m.g .
ryskin@xmath6 + @xmath4 department of physics and institute for particle physics phenomenology , university of durham , durham , dh1 3le + @xmath5 department of physics , university of helsinki , and helsinki institute of physics , finland + @xmath7 petersburg nuclear physics institute , gatchina , st .
petersburg , 188300 , russia |
a core - collapse supernova ( ccsn ) occurs when a star having an initial mass @xmath6 ends its life in a catastrophic explosion .
observationally , ccsne are divided into three groups based on their observed spectra : sne ii show large amounts of hydrogen , sne ib exhibit helium but little or no hydrogen , and sne ic do not show significant amounts of hydrogen or helium ( for a review , see * ? ? ?
sne ic are heterogeneous .
their luminosity , ejected mass , and kinetic energy span over an order of magnitude , from the subluminous sn 2004aw to the overluminous sn 1998bw @xcite .
the light - curve shapes of different events are also quite diverse . a subclass of sne ic whose spectra are characterized by broad lines ( type ic - bl ; prototype sn 1998bw ) is the only one for which clear evidence of an association with gamma - ray bursts ( grbs ) exists ( grb - sne ; see woosley & bloom 2006 for a review ) .
superluminous sne ( slsne ) of type ic are even more powerful ( * ? ? ?
* and references therein ) , but these probably result from a different physical mechanism . while sne ic are common in the center of high - metallicity galaxies @xcite , slsne - i and broad - lined grb - sne tend to be found in dwarf hosts ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , giving untargeted sky surveys an advantage over targeted surveys in detecting these types of cosmic explosions .
the palomar transient factory ( ptf ; law et al .
2009 ; rau et al .
2009 ) is a wide - field untargeted sky survey which explores the transient optical sky .
it uses the ptf cfh12k camera mounted on the palomar 48-inch telescope ( p48 ) .
ptf s short observing cadence and real - time capability ( e.g. , * ? ? ?
* ) enables the discovery and study of sne at early stages of the explosion . in this _
letter _ we report the discovery and study of ptf12gzk , a peculiar sn ic in a dwarf star - forming galaxy located at redshift @xmath7 ( distance 57.8mpc , distance modulus 33.8mag , assuming h@xmath8 = 71kms@xmath3mpc@xmath3 ) .
ptf12gzk was discovered on 2012 july 24.3 ( ut dates are used herein ) at @xmath9(j2000 ) = 22@[email protected]@xmath12 and @xmath13(j2000 ) = @xmath14 , in the dwarf galaxy sdss j221241.53 + 003042.7 ( within the sdss stripe 82 footprint ) , as part of a coordinated ptf - jvla survey .
the discovery magnitude was 20.66 in the @xmath15 band - band magnitudes . ] , and it was not detected down to mag 21.6 ( @xmath16 ) in previous ptf images obtained on july 19 @xcite . shortly after discovery , we initiated an extensive follow - up campaign , including our _ hubble space telescope ( hst ) _ target - of - opportunity ( too ) program for stis ultraviolet ( uv ) spectroscopy of a stripped - envelope sn ( cycle 19 , go-12530 ; pi filippenko ) and _ swift _ x - ray and uv photometry ( cycle 8 , pid 8110099 ; pi kasliwal ) , the results of which are presented herein ( see figures 1 and 2 for photometry and spectroscopy , respectively ) .
we also triggered radio and millimeter observations using the jvla ( program 12a-363 ; pi horesh ) and the combined array for research in millimeter - wave astronomy ( program 12a - c0945 ; pi horesh ) ; see horesh et al .
( in preparation ) .
optical photometry of ptf12gzk was obtained using multiple telescopes ( table 1 ) .
all data were calibrated with respect to the sdss catalog .
light curves of ptf12gzk are shown in figure 1 .
.ptf12gzk photometry .
data were obtained using the p48+ptf camera , the palomar 60-inch telescope + grb camera , the fairchild camera on the 40-inch telescope at mt .
laguna observatory ( mlo ) , fs01 on the 2-m faulkes telescope south ( fts ) , the io : o camera on the liverpool telescope ( lt ) , the pi camera on the wise 1-m telescope , the weizmann institute kraar 16-inch telescope , and the 0.76-m katzman automatic imaging telescope ( kait ; filippenko et al .
the full set of photometric data is available in the electronic version of this paper , as well as from wiserep , http://www.weizmann.ac.il/astrophysics/wiserep/ ; yaron & gal - yam ( 2012 ) . [
cols="^,^,^,^,^",options="header " , ] [ fig : intlines ] the first spectrum was obtained on july 27 , revealing the object to be a peculiar sn i with a blue continuum , reminiscent of the blue continuum observed in the early - time spectra of sn 2006aj @xcite .
initially it was difficult to classify the spectrum securely , but later optical spectra resemble those of sne ic , with an absence of prominent he i lines @xcite .
the august 1 spectrum was taken with the floyds spectrograph mounted on the ftn , a low - dispersion spectrograph ( @xmath17 ) with a single , fixed grating and a cross - dispersing prism , placing the first- and second - order spectra onto the ccd with a single - exposure wavelength coverage of 320010,000 .
while the identical twin floyds spectrographs on ftn and fts will ultimately be robotically operated , the present spectrum was taken during hardware commissioning under human control .
classification with the help of snid @xcite of the august 9 lick 3-m / kast spectrum suggests that the best fit is to the peculiar sn ic sn 2004aw near peak brightness @xcite .
a good match is also obtained for ptf10qts , a type ic - bl sn , at @xmath18days after explosion .
all spectra show prominent , broad absorption lines of ca ii , si ii , fe ii , and o i , with synow fits yielding maximum velocities of @xmath19kms@xmath3 for the july 27 spectrum , and @xmath20kms@xmath3 for the august 9 spectrum .
see figure 3 and 5 for further discussion .
the absence of the 2.1@xmath21 m he i feature in the ir tspec spectrum taken on august 7 ( figure 3 , bottom - right panel ) suggests that he is not abundant in the envelope @xcite .
while a strong absorption line is present at 1@xmath21 m , which could be identified with he i 1.0830@xmath21 m , in this case it is likely contaminated by other elements such as c , mg , s , and ca ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
the first _ hst _
uv spectrum , obtained by the space telescope imaging spectrometer with the near - uv multi - anode microchannel array ( stis / mama ) , was taken between 20:06 august 6 and 03:35 august 7 , with a total exposure time of 11,278s .
we also obtained optical spectra with the stis / ccd , using grisms centered at 4300 and 7500 .
a mosaic of all three _ hst _ spectra is given in figure 3 , with a focus on the uv spectrum in the bottom - right panel .
the flux deficit with respect to blackbody spectra seen in the stis / mama uv spectrum is characteristic of all sne i @xcite and indicates strong line blanketing , evidence for a highly mixed outer envelope devoid of hydrogen .
full analysis of the _ hst _ uv spectra will be presented by ben - ami et al .
( in prep . )
an image and a spectrum of the host galaxy obtained by the sdss prior to explosion are shown in figure 2 .
the sn exploded very close to the center of the host galaxy ( offset @xmath22 n and @xmath23 e ) .
we downloaded the host - galaxy spectrum from the sdss 9th data release ( dr9 ; * ? ? ?
* ) . after correcting for the host - galaxy redshift , we measure integrated emission - line fluxes using standard procedures via _ splot _ in iraf , and follow perez - montero & diaz ( 2003 ) to compute statistical errors , h@xmath24 , h@xmath9 , [ o iii ] @xmath254959,[o iii ] @xmath255007 , and [ n ii ] @xmath256584 emission - line fluxes are @xmath26 , @xmath27 , @xmath28 , @xmath29 , @xmath30 , and @xmath31ergs@xmath3@xmath32 before extinction correction , respectively .
errors include statistical measurement uncertainties ] . using the sdss petrosian magnitudes , correcting for galactic and host - galaxy extinction ( see below ) , and applying @xmath33-corrections via _ kcorrect _
( v4.2 , blanton & roweis 2007 ) at the host - galaxy redshift , we derive an absolute magnitude of @xmath34mag .
this is an extremely underluminous dwarf galaxy , well below the luminosities of the large and small magellanic clouds .
it is one of the least luminous host galaxies of ccsne discovered by ptf , and it is not a common host galaxy for a sn ic , even considering those found via untargeted surveys @xcite , except for sne ic - bl . in addition , the clear detections of numerous bright emission lines allow us to apply standard diagnostics of the star - formation rate ( sfr ) , extinction , and metallicity to this galaxy . using the observed intensity ratio of h@xmath9/h@xmath24 , and assuming the case b recombination value of 2.86 and the cardelli et al .
( 1989 ) extinction law with @xmath35 , we estimate a reddening of @xmath36mag .
the values of the intensity ratios [ n ii]/h@xmath9 and [ o iii]/h@xmath24 indicate that the prominent emission lines are due to recent vigorous star formation rather than to an active galactic nucleus @xcite .
furthermore , we do not detect any absorption features nor a balmer break that may indicate the presence of an older stellar population . from the measured rest - frame equivalent width of h@xmath24 , @xmath37
, we derive an age of @xmath38myr for the young stellar population using the relations of levesque et al .
( 2010 , 3.4 and table 4 ) and the measured metallicity of the host ( for @xmath39 ; see below ) . from the measured integrated h@xmath9 emission - line flux , corrected for extinction , we obtain @xmath40ergs@xmath3 , which translates to sfr(h@xmath9 ) @xmath41m@xmath4yr@xmath3 using the conversion from kennicutt ( 1998 ) .
since the sdss spectrum was obtained through a @xmath42 diameter fiber encompassing an area of @xmath43kpc@xmath44 at the host - galaxy redshift , we measure a normalized sfr of @xmath45m@xmath4yr@xmath3kpc@xmath46 .
since the petrosian radius ( from dr9 ) is @xmath47 , the fraction of global host - galaxy light entering the fiber , which is also at the sn position , is @xmath48 . for computing the oxygen abundance ( modjaz et al .
2011 , and references therein ) , we correct the detected emission - line fluxes of [ o iii ] , [ n ii ] , h@xmath24 , and h@xmath9 for reddening , and employ the scales of pettini & pagel ( 2004 , pp04-o3n2 ) and of kewley & dopita ( 2002 , kd02 ) to obtain values of @xmath49 and @xmath50 , respectively .
we conclude that the metallicity of the host galaxy is 0.20.3z@xmath4 , having used the solar oxygen abundance of @xmath51 @xcite .
the oxygen abundance of the ptf12gzk host is well below that of the hosts of normal sne ic found via untargeted surveys as presented by modjaz et al .
( 2011 ; mean @xmath52 ) and sanders et al .
( 2012 , @xmath53 ) .
indeed , it is much closer to that of the hosts of sne ic - bl and grb - sne @xcite , and of slsne - i @xcite .
ptf12gzk is a luminous sn ic , at the high end of the sn ic luminosity distribution @xcite .
it exhibits a slow rise of 18 days to its peak @xmath15-band magnitude , with @xmath54 peaking @xmath55 days earlier .
this is a large gap relative to other sn i , though similar to sn 2004aw @xcite ; it is caused by metal - line absorption from heavy elements in the outer layers of the ejecta , as is evident from spectra taken after august 1 .
a least - squares fit to a @xmath56 behavior of our well - sampled early photometry places the explosion date between 10 and 40hr prior to our discovery at the 95% confidence level .
we can not rule out the popular quadratic fireball model , but higher values of @xmath1 are possible as well for larger areas in the fit parameter space ( figure 1 , bottom panels ) .
spectroscopically , ptf12gzk exhibits high expansion velocities , @xmath2kms@xmath3 ( si ii absorption velocity ) .
other sne ic with similar velocities are broad - lined sne ic ( figure 4 ) , some of which are associated with grbs ( woosley & bloom 2006 , and references within ) , while no such association was determined for ptf12gzk ( see also sn 2009bb ; soderberg et al .
most similar is sn 2003lw , a sn associated with a grb @xcite .
a possible explanation is a burst misaligned with our line of sight , or a failed grb .
such a scenario is further supported by the host - galaxy characteristics , resembling those of a broad - lined sn ic host galaxy .
we know of no typical sn ic exploding in a host with similar luminosity and oxygen abundance .
the observed relatively narrow lines give a dispersion of @xmath57 , compared to @xmath58 in the case of broad - lined sne ic , and may suggest a nonspherical explosion geometry @xcite , or that the ejecta mass is high or has a very steep density gradient ( a discussion on the effects of a steep density gradient on the lc can be found in piro & nakar 2012 ) .
late - time , nebular spectra will probe the geometry of the explosion in more detail . from the si ii line velocity at peak brightness of ptf12gzk ( 15,300kms@xmath3 from the august 12 spectrum ) and
sn 2004aw ( 12,400kms@xmath3 ; deng et al .
. ) , and the rise time of these two sne , we use the following scaling relations ( arnett 1982 ; mazzali et al . 2009 ; see also mazzali et al .
, in prep . ) to estimate the physical properties of ptf12gzk : @xmath59 and @xmath60 , where @xmath61 is the light - curve rise time , @xmath62 is the kinetic energy , and @xmath63 is the opacity .
the derived ejecta mass is 7.5m@xmath4 ( 612m@xmath4 ) , pointing to a large initial progenitor mass of 2535m@xmath4 , though the latter values are highly uncertain @xcite .
we derive a kinetic energy of @xmath64erg [ ( 510 ) @xmath65 erg ] . using @xmath66 and @xmath67
we get an estimated @xmath68ni mass of 0.37m@xmath4 ( scaling the ptf12gzk light curve to that of sn 2003dh ; mazzali et al . 2003 ) . using the @xmath69-band peak magnitude vs. nickel mass relation presented by perets et al . ( 2010 ) , we derive a @xmath68ni mass of 0.35m@xmath4 , in agreement with the results derived from the scaling relations .
these physical properties , as well as the high expansion velocities and the host galaxy , are unlike those of normal sne ic , which typically occur in large hosts and have low ejecta masses , and kinetic energies , ( 1.7m@xmath4 , and @xmath70erg drout et al .
2011 . for nickel mass in type ic sne
see for example taubenberger et al .
2006 , and sauer et al .
2006 ) . instead , they are reminiscent of grb - sne @xcite .
ptf12gzk is a remarkable example of a sn ic in terms of expansion velocities , evolution timescale , the ejected mass , and the kinetic energy released in the explosion .
we conclude that these properties point to the explosion of a massive star deficient in h and he , at the higher - mass end of sn ic progenitors .
this further illustrates the peculiar population of sne ic exploding in dwarf hosts @xcite , as seen also in the case of grb - sne and most slsne - i .
ptf12gzk demonstrates the advantages of using an untargeted sky survey such as ptf with an extensive network of instruments and telescopes in various wavebands to detect and rapidly characterize unusual cases of cosmic explosions .
s.b . is supported by a ramon fellowship from isa .
acknowledges support by grants from the isf , bsf , gif , minerva and the eu fp7/erc .
a.v.f . and his group benefit from financial assistance from gary & cynthia bengier , the richard & rhoda goldman fund , the sylvia & jim katzman foundation , the christopher r. redlich fund , the tabasgo foundation , nsf grants ast-0908886 and ast-1211916 , and nasa/_hst _ grant go-12530 from stsci ( which is operated by the aura , inc .
, under nasa contract nas 05 - 26555 ) .
p.a.m . and e.p .
acknowledge financial support from grants inaf prin 2011 and asi / inaf i/088/06/0 .
m.i . and y.j .
were supported by the creative initiative program of the nrfk .
m.m.k . acknowledges hubble and carnegie - princeton fellowships .
is supported by nsf grant ast-1009571 . e.o.o .
acknowledges the arye dissentshik career development chair and a grant from the israeli most .
ptf is a collaboration of caltech , lcogt , the weizmann institute , lbnl / nersc , oxford , columbia , ipac , and uc berkeley .
the liverpool telescope is operated on the island of la palma by liverpool john moores university in the spanish observatorio del roque de los muchachos support from the uk stfc .
construction of the laiwo camera was supported by the mpia , gif , and the isf .
we are grateful for the assistance of the staff at the various observatories used to obtain data .
this research has made use of the nasa / ipac extragalactic database ( ned ) which is operated by jpl , caltech , under contract with nasa .
ahn , c. p. , alexandroff , r. , allende prieto , c. , et al .
( sdss - iii collaboration ) 2012 , arxiv:1207.7137 anderson , j. p. , habergham , s. m. , james , p. a. , & hamuy , m. 2012 , , 424 , 1372 arcavi , i. , gal - yam , a. , kasliwal , m. m. , et al .
2010 , , 721 , 777 arnett , w. d. 1982 , , 253 , 785 asplund , m. , grevesse , n. , sauval , a. j. , & scott , p. 2009
, , 47 , 481 baldwin , j. a. , phillips , m. m. , & terlevich , r. 1981 , , 93 , 5 ben - ami , s. , gal - yam , a. , filippenko , a. v. , et al .
2012 , the astronomer s telegram , 4297 , 1 blanton , m. r. , & roweis , s. 2007 , , 133 , 734 blondin , s. , & tonry , j. l. 2007 , , 666 , 1024 ( update : 2011 , astrophysics source code library , 7001 ) breeveld , a. a. , landsman , w. , holland , s. t. , roming , p. , kuin , n. p. m. , & page , m. j. 2011 , in gamma ray bursts 2010 , ed
. j. e. mcenery , j. l. racusin , & n. gehrels ( new york : aip , vol . 1358 ) , 373 cardelli , j. a. , clayton , g. c. , & mathis , j. s. 1989 , , 345 , 245 drout , m. r. , soderberg , a. m. , gal - yam , a. , et al .
2011 , , 741 , 97 filippenko , a. v. 1997 , , 35 , 309 filippenko , a. v. , li , w. d. , treffers , r. r. , & modjaz , m. 2001 , in small - telescope astronomy on global scales , ed . w. p. chen , c. lemme , & b. paczyski ( san francisco : asp , vol .
246 ) , 121 gal - yam , a. 2012 , arxiv:1206.2157 gal - yam , a. , kasliwal , m. m. , arcavi , i. , et al .
2011 , , 736 , 159 hachinger , s. , mazzali , p. a. , taubenberger , s. , et al . 2012 , , 422 , 70 hodgkin , s. t. , irwin , m. j. , hewett , p. c. , & warren , s. j. 2009 , , 394 , 675 iwamoto , k. , mazzali , p. a. , nomoto , k. , et al .
1998 , , 395 , 672 jordi , k. , grebel , e. k. , & ammon , k. 2006 , , 460 , 339 kennicutt , r. c. , jr .
1998 , , 498 , 541 kewley , l. j. , & dopita , m. a. 2002 , , 142 , 35 law , n. m. , kulkarni , s. r. , dekany , r. g. , et al .
2009 , , 121 , 1395 leonard , d. c. , filippenko , a. v. , ganeshalingam , m. , et al .
2006 , , 440 , 505 levesque , e. m. , berger , e. , kewley , l. j. , & bagley , m. m. 2010 , , 139 , 694 mazzali , p. a. , & lucy , l. b. 1998 , , 295 , 428 mazzali , p. a. , deng , j. , hamuy , m. , & nomoto , k. 2009 , , 703 , 1624 mazzali , p. a. , deng , j. , pian , e. , et al . 2006a , , 645 , 1323 mazzali , p. a. , deng , j. , nomoto , k. , et al . 2006b ,
, 442 , 1018 mazzali , p. a. , deng , j. , tominaga , n. , et al .
2003 , , 599 , l95 mazzali , p. a. , iwamoto , k. , & nomoto , k. 2000 , , 545 , 407 modjaz , m. , kewley , l. , bloom , j. s. , et al .
2011 , , 731 , l4 modjaz , m. , kewley , l. , kirshner , r. p. , et al .
2008 , , 135 , 1136 ofek , e. o. , laher , r. , law , n. , et al .
2012 , , 124 , 62 panagia , n. 2007 , in supernova 1987a : 20 years after : supernovae and gamma - ray bursters , ed . s. immler & k. weiler ( new york : aip , vol .
937 ) , 236 patat , f. , cappellaro , e. , danziger , j. , et al .
2001 , , 555 , 900 perets , h. b. , gal - yam , a. , mazzali , p. a. , et al .
2010 , , 465 , 322 prez - montero , e. , & daz , a. i. 2003 , , 346 , 105 pettini , m. , & pagel , b. e. j. 2004 , , 348 , l59 piro , a. l. , & nakar , e. 2012 , arxiv:1210.3032 rau , a. , kulkarni , s. r. , law , n. m. , et al . 2009 , , 121 , 1334 sauer , d. n. , mazzali , p. a. , deng , j. , et al . 2006 , , 369 , 1939 sanders , n. e. , soderberg , a. m. , levesque , e. m. , et al . 2012 , arxiv:1206.2643 soderberg , a. m. , chakraborti , s. , pignata , g. , et al . 2010 , , 463 , 513 stanek , k. z. , gnedin , o. y. , beacom , j. f. , et al .
2006 , , 56 , 333 stoll , r. , prieto , j. l. , stanek , k. z. , et al .
2011 , , 730 , 34 taubenberger , s. , pastorello , a. , mazzali , p. a. , et al .
2006 , , 371 , 1459 woosley , s. e. , & bloom , j. s. 2006 , , 44 , 507 yaron , o. , & gal - yam , a. 2012 , , 124 , 668 young , d. r. , smartt , s. j. , valenti , s. , et al .
2010 , , 512 , a70 | we present the discovery and extensive early - time observations of the type ic supernova ( sn ) ptf12gzk .
our light curves show a rise of 0.8mag within 2.5hr .
power - law fits [ @xmath0 to these data constrain the explosion date to within one day .
we can not rule out a quadratic fireball model , but higher values of @xmath1 are possible as well for larger areas in the fit parameter space .
our bolometric light curve and a dense spectral sequence are used to estimate the physical parameters of the exploding star and of the explosion .
we show that the photometric evolution of ptf12gzk is slower than that of most sne ic .
the high ejecta expansion velocities we measure ( @xmath2kms@xmath3 derived from line minima four days after explosion ) are similar to the observed velocities of broad - lined sne ic associated with gamma - ray bursts ( grbs ) rather than to normal sn ic velocities . yet
, this sn does not show the persistent broad lines that are typical of broad - lined sne ic .
the host - galaxy characteristics are also consistent with grb - sn hosts , and not with normal sn ic hosts . by comparison with the spectroscopically similar sn 2004aw , we suggest that the observed properties of ptf12gzk indicate an initial progenitor mass of 2535m@xmath4 and a large [ ( 510 ) @xmath5erg ] kinetic energy , the later being close to the regime of grb - sn properties . |
field - theoretical renormalization group ( rg ) approach proved to be highly efficient when used to evaluate universal parameters characterizing the behavior of various systems near curie point .
it yields high - precision numerical estimates for critical exponents , renormalized coupling constants , universal ratios , etc .
, provided lengthy enough rg expansions are employed and proper resummation of these diverging series is made ( see , .
e. g. refs.@xcite ) .
numerical estimates for critical exponents of the ising , xy , heisenberg and some other models obtained within field - theoretical rg machinery are referred today as canonical numbers@xcite and widely used in course of comparison of the theory with computer and physical experiments including advanced measurements performed in space@xcite .
rg expansions being power series in renormalized quartic coupling constant @xmath6 or in @xmath7 have coefficients that grow factorially with their number @xmath8 . to struggle this divergency the borel transformation is usually employed which turns divergent series into expansions having non - zero radius of convergence .
resummation methods based on borel transformation work very efficiently when original series are alternating and their coefficients demonstrate regular behavior , i. e. being monotonically decreasing functions of @xmath8 for moderate @xmath8 , they monotonically grow up under @xmath9 .
fortunately , rg expansions for the @xmath10-function and `` big '' critical exponents ( @xmath11 , @xmath12 and some others ) are precisely such regular series both in three and two@xcite dimensions .
this is one of the main reasons why field - theoretical rg approach turned out to be so effective numerically in the phase transition problem .
however , there is a critical exponent for which rg series are not so `` friendly '' .
we mean the fisher exponent @xmath0 .
let us look at the perturbative expansion of @xmath0 for the three - dimensional ising model and at the corresponding @xmath2-expansion that are known today in seven - loop@xcite and five - loop@xcite approximations respectively .
they are as follows : @xmath13 @xmath14 the coefficients of above series as seen to be quite irregular , both in sign and modulo .
that is why the resummation of such series by canonical ( pade - borel , conform - borel , etc .
) methods is much less effective than in the case of wilson fixed point location @xmath15 and big critical exponents . as a results ,
one usually prefers to evaluate the exponent @xmath0 via scaling relations instead of dealing with corresponding rg series ( see , e. g. comprehensive review@xcite ) .
in such a situation it is reasonable to address some alternative technique which is able to turn original rg expansions into more appropriate ones .
here we do not mean avoiding of factorial growth of the coefficients since in series ( 1 ) , ( 2 ) they are small or , at least , not too large .
instead , we are looking for a tool which would convert rg expansion for @xmath0 into the series regular in sign along with making higher - order coefficients to monotonically decrease with growing @xmath8 .
below , it will be shown that the pseudo-@xmath2 expansion can play a role of such a tool .
this approach invented by b. nickel many years ago ( see ref . 19 in the paper of le guillou and zinn - justin @xcite ) exploits the idea that wilson fixed point location in three dimensions may be found iteratively by means of introducing fictitious small parameter @xmath3 into linear term of perturbative series for @xmath10-function .
pseudo-@xmath2 expansion proved to be very efficient when used to estimate critical exponents and other universal quantities characterizing critical behavior of three - dimensional systems @xcite . even in two dimensions , where rg series are shorter and more strongly divergent
, it leads to good or satisfactory numerical results @xcite .
as we will see , for the exponent @xmath0 the pseudo-@xmath2 expansion turns out to be highly effective as well : it generates iteration procedures rapidly converging to the asymptotic values that are in good agreement with the numbers extracted from alternative field - theoretical and lattice calculations .
it is worthy to note that ability of pseudo-@xmath2 expansion approach to accelerate rg iterations and to smooth oscillations of numerical estimates as functions of @xmath8 was discovered just after beginning of its application@xcite .
it was observed also that in many cases pseudo-@xmath2 expansions do not require advanced resummation procedures ; as a rule , use of pad approximants or even direct summation are sufficient to lead to proper numerical results . in our case , however , pseudo-@xmath2 expansion demonstrates an extra advantage it turns the series with a structure rather unfavorable from the computational point of view into those quite suitable for numerical estimates .
critical thermodynamics of three - dimensional systems with @xmath17-symmetric vector order parameters is described by euclidean field theory with the hamiltonian : @xmath18 , \ ] ] where bare mass squared @xmath19 is proportional to the deviation from mean field transition temperature .
perturbative expansions for the @xmath10-function and critical exponents of this model were calculated in the six - loop approximation within the massive theory@xcite .
later , rg series for critical exponents were extended up to seven - loop order by murray and nickel in their unpublished work@xcite ; seven - loop terms were reported in the paper of guida and zinn - justin@xcite .
we derive the pseudo-@xmath2 expansion for critical exponent @xmath0 starting from rg series mentioned . to do
this one has to substitute the @xmath3-series for the wilson fixed point location @xmath15 into perturbative expansion for the fisher exponent and reexpand it in @xmath3 .
pseudo-@xmath2 expansion of @xmath15 for general @xmath1 is known up to @xmath5 term ( six - loop order)@xcite . at first glance , with this expansion in hand @xmath3-series for critical exponents
may be found within the same @xmath5 approximation .
it is really so for all critical exponents but the fisher one .
since the first non - zero term in rg expansion of @xmath0 is proportional to @xmath20 the length of @xmath3-series for @xmath15 turns out to be sufficient to find @xmath4 term .
seven - loop contribution in rg expansion of @xmath0 was calculated for concrete values of @xmath1 most interesting from the physical point of view@xcite .
that is why here we present pseudo-@xmath2 expansions of @xmath0 for @xmath21 only , leaving six - loop ( @xmath5 ) series at generic @xmath1 for section iv .
seven - loop @xmath3-series obtained are as follows : @xmath22 @xmath23 @xmath24 @xmath25 series ( 4)-(7 ) are seen to have much more regular structure than original rg expansions .
their coefficients possess the same sign and monotonically decrease with increasing @xmath8 , apart from those of seven - loop ( @xmath4 ) terms .
these coefficients being small are nevertheless some bigger than their six - loop ( @xmath5 ) counterparts signalizing that @xmath3-series remain divergent . despite of this
, expansions ( 4)-(7 ) turn out to be quite suitable for getting numerical estimates .
numerical values of @xmath0 are extracted from the series ( 4)-(7 ) by means of pad approximants [ l / m ] and by direct summation ( ds ) .
pad triangles for ising and heisenberg models are presented in tables i and ii as typical examples .
note that symbol [ l / m ] denotes here pad approximants constructed for @xmath26 , i. e. with insignificant factor @xmath27 having physical value @xmath28 ignored .
as seen from tables i and ii the estimates of @xmath0 given by highest - order near diagonal approximants [ 3/2 ] and [ 2/3 ] are very close to the numbers resulting from resummed 3d rg expansions@xcite ; they differ from each other by 0.001 ( 3 per cent ) or less
. moreover , the convergence of pseudo-@xmath2 expansion estimates to the asymptotic values turns out to be fast what also may be seen from both tables .
similar behavior of estimates is observed in the case of direct summation of the series ( 4)-(7 ) .
this is clearly demonstrated by table iii where pad and ds estimates of @xmath0 as functions of @xmath8 are collected along with the values the field - theoretical and lattice calculations yield .
one can see that for all four values of @xmath1 both iteration schemes lead to the numbers which agree well with other high - precision estimates .
as seen from table iii the deviation of pseudo-@xmath2 expansion estimates from the alternative values is much smaller than characteristic difference between these values themselves .
ds estimates reach their asymptotes monotonically what is a direct consequence of the @xmath3-series structure . on the contrary
, the behavior of pad estimates turns out to be oscillatory , i. e. typical for this and other , more sophicticated resummation procedures ( see , e. g. classical papers @xcite ) . corresponding oscillation , however , are weak what is known to be specific for the pseudo-@xmath2 expansion technique . keeping in mind optimistic results just obtained , the question arises : will numerically favorable structure of pseudo-@xmath2 expansion for fisher exponent demonstrated at @xmath29 persist for larger @xmath1 ?
in other words , whether numerical power of the pseudo-@xmath2 expansion is its generic property or it manifests itself only for moderate @xmath1 ? to answer this questions we are in a position to study @xmath3-series for @xmath0 at arbitrary @xmath1 .
perturbative rg expansion of @xmath0 for general @xmath1 are known today within six - loop approximation@xcite .
this enables us to derive corresponding pseudo-@xmath2 expansion ranging up to @xmath5 term .
straightforward calculation leads to the following @xmath3-series : @xmath30 analyzing this series under various @xmath1 lying between 4 and 64 we find that : \i ) series ( 8) have positive and monotonically decreasing coefficients up to @xmath31 ; \ii ) coefficients of @xmath32 and @xmath5 terms change their signs at @xmath33 and @xmath31 respectively while other coefficients remain positive and monotonically decreasing ; \iii ) up to @xmath34 coefficients of @xmath32 and @xmath5 terms persist to be tiny ( @xmath35 and smaller ) , so these terms do not influence appreciably upon numerical estimates the series ( 8) yields .
hence , the structure favorable for getting numerical estimates is a generic property of the pseudo-@xmath2 expansion for fisher exponent .
comparison of the values of @xmath0 resulting from expansions ( 4)-(8 ) with each other and with their counterparts obtained within other approaches confirms this conclusion .
these values are collected in table iv , along with the numbers given by the @xmath36-expansion@xcite : @xmath37 all the data presented are seen to be in a good agreement at any @xmath1 .
why the pseudo-@xmath2 expansion technique turns out to be so efficient in particular case considered ?
we have an explanation of this fact .
the point is that mechanism of pseudo-@xmath2 expansion is organized in such a way that it suppresses the divergency of rg expansions for critical exponents provided these expansions are alternating .
the mechanism of suppression works well for alternate series because in course of transformation of rg series into pseudo-@xmath2 expansions multiple mutual subtractions ( `` destructive interference '' ) of the terms of original series take place .
if , however , we apply this technique to series with positive coefficients the subtraction is changed by summation ( `` contructive interference '' ) what makes relevant terms in pseudo-@xmath2 expansion larger than that of rg series . this can be demonstrated explicitly .
let the pseudo-@xmath2 expansion for renormalized quartic coupling constant at criticality ( wilson fixed point location ) @xmath15 be : @xmath38 while rg series for some critical exponent @xmath39 have a form : @xmath40 typically , all @xmath41 are positive , i. e. the series ( 11 ) is alternating . to obtain @xmath3-series for @xmath39
, we have to substitute expansion ( 10 ) into ( 11 ) .
it yields : @xmath42 \tau^4 + ... \ ] ] if coefficients of pseudo-@xmath2 expansion ( 10 ) for @xmath15 are positive , what is really the case for several lower - order terms@xcite , coefficients of the series ( 11 ) interfere within ( 12 ) destructively .
it is clearly seen from the structure of series ( 12 ) .
the character of interference , however , depends crucially on signs of coefficients of initial rg expansion .
indeed , if we changed signs of odd terms in ( 11 ) , i. e. made all terms in ( 11 ) positive , destructive interference would turn into constructive what again is cleary seen from ( 12 ) .
what happens in the case of fisher exponent ?
since lower - order terms in rg expansions for @xmath0 have the same sign ( see , e. g. ( 1 ) ) pseudo-@xmath2 expansion machine grows them up . on the contrary , for the higher - order terms this machine works as suppressive because starting from @xmath43 term series ( 1 ) becomes ( looks as ) alternating . as a result , pseudo-@xmath2 expansion technique transforms rg series with small and irregular coefficients into @xmath3-series which possesses larger lower - order coefficients and decreasing with @xmath8 higher - order ones , i. e. demonstrates behavior similar to that of converging series .
fig.1 illustrates such metamorphosis for the ising ( @xmath44 ) and heisenberg ( @xmath45 ) models .
in fact , pseudo-@xmath2 expansion does not generate convergent expansions .
instead , it replaces one diverging series by another , less strongly divergent .
the resulting expansions , however , have much more favorable structure from the numerical point of view . in our case
pseudo-@xmath2 expansions actually do not require resummation , even in its simplest pad form : as seen from tables iii and iv the highest - order ( @xmath4 ) estimates given by direct summation and found by means of pad analysis differ from each other by 3 percents or less .
this difference is much smaller than individual and overall error bars characteristic for data provided by high - precision field - theoretical and lattice calculations .
thus , as was argued earlier@xcite , the pseudo-@xmath2 expansion approach may be considered as a resummation method .
this method , however , is somewhat specific it does not turn divergent series into convergent but makes them very convenient for practical use .
to summarize , we have calculated pseudo-@xmath2 expansions of the fisher critical exponent up to @xmath4 terms for @xmath16 and within six - loop ( @xmath5 ) approximation for general @xmath1 .
these expansions have been found to possess a structure that is rather favorable for getting numerical estimates .
having processed @xmath3-series obtained by means of pad approximants and performed their direct summation we ve obtained numerical estimates of @xmath0 that are as accurate as those extracted from advanced field - theoretical and lattice calculations .
the structure of @xmath3-series for @xmath0 persists to be favorable within the wide range of @xmath1 signaling that it is a generic property of the pseudo-@xmath2 expansion for @xmath0 .
we have found arguments shedding light on the roots of such fortune .
they lie in the general properties of the pseudo-@xmath2 expansion machinery interfering with some peculiarities of the rg expansion for @xmath0 .
c|*6c @xmath47 & 0 & 1 & 2 & 3 & 4 & 5 + 0 & 0.0110 & 0.0212 & 0.0262 & 0.0294 & 0.0307 & 0.0324 + 1 & 0.1531@xmath48 & 0.0312 & 0.0350 & 0.0316 & 0.0256@xmath49 & + 2 & 0.0232 & 0.0339 & 0.0327 & 0.0333 & & + 3 & 0.0467@xmath50 & 0.0322 & 0.0332 & & & + 4 & 0.0258 & 0.0342 & & & & + 5 & 0.0570@xmath51 & & & & & + c|*6c @xmath47 & 0 & 1 & 2 & 3 & 4 & 5 + 0 & 0.0122 & 0.0226 & 0.0277 & 0.0307 & 0.0322 & 0.0337 + 1 & 0.0815@xmath52 & 0.0324 & 0.0355 & 0.0335 & - & + 2 & 0.0265 & 0.0346 & 0.0341 & 0.0347 & & + 3 & 0.0414 & 0.0339 & 0.0344 & & & + 4 & 0.0304 & 0.0354 & & & & + 5 & 0.0432@xmath50 & & & & & + c|cccccc|ccc @xmath8 & 2 & 3 & 4 & 5 & 6 & 7 & 3d rg & @xmath2-exp . &
lc + + pad & 0.0093 & 0.0182@xmath53 & 0.0266 & 0.0300 & 0.0280 & 0.0285 & 0.0284@xcite & 0.0315@xcite & + ds & 0.0093 & 0.0182 & 0.0225 & 0.0254 & 0.0263 & 0.0280 & & & + + pad & 0.0110 & 0.0212@xmath53 & 0.0312 & 0.0344 & 0.0327 & 0.0332 & 0.0335@xcite & 0.0365@xcite & 0.0360@xcite + ds & 0.0110 & 0.0212 & 0.0262 & 0.0294 & 0.0307 & 0.0324 & 0.0335@xcite & & 0.0364@xcite + + pad & 0.0119 & 0.0224@xmath53 & 0.0326 & 0.0356 & 0.0343 & 0.0348 & 0.0354@xcite & 0.0370@xcite & 0.0380@xcite + ds & 0.0119 & 0.0224 & 0.0276 & 0.0308 & 0.0323 & 0.0339 & 0.0349@xcite & & + + pad & 0.0122 & 0.0226@xmath53 & 0.0324 & 0.0351 & 0.0341 & 0.0346 & 0.0355@xcite & 0.0355@xcite & 0.0375@xcite + ds & 0.0122 & 0.0226 & 0.0277 & 0.0307 & 0.0322 & 0.0337 & 0.0350@xcite & & + * 7c @xmath1 & pad & ds & 3d rg@xcite & 3d rg@xcite & @xmath2-exp.@xcite & @xmath54-exp .
+ 0 & 0.0285 & 0.0280 & & 0.0284 & 0.0315 & + 1 & 0.0332 & 0.0324 & 0.038 & 0.0335 & 0.0365 & + 2 & 0.0348 & 0.0339 & 0.039 & 0.0354 & 0.0370 & + 3 & 0.0346 & 0.0337 & 0.038 & 0.0355 & 0.0355 & + 4 & 0.0329 & 0.0313 & 0.036 & 0.0350 & 0.033 & 0.0260 + 8 & 0.0260 & 0.0252 & 0.027 & & & 0.0271 + 16 & 0.0164 & 0.0163 & 0.017 & & & 0.0157 + 24 & 0.01173 & 0.01168 & 0.012 & & & 0.01078 + 32 & 0.00915 & 0.00902 & 0.009 & & & 0.00820 + 40 & 0.00761 & 0.00732 & & & & 0.00660 + 48 & 0.00627@xmath53 & 0.00616 & & & & 0.00553 + 64 & 0.00490@xmath53 & 0.00466 & & & & 0.00417 + and @xmath45 under the action of pseudo-@xmath2 expansion machinery . left hystograms depict weight of different terms in rg series at the wilson fixed point , right hystograms show analogous distributions for pseudo-@xmath2 expansions ( 5 ) and ( 7 ) , @xmath8 being an order of a term . ] | critical exponent @xmath0 for three - dimensional systems with @xmath1-vector order parameter is evaluated in the frame of pseudo-@xmath2 expansion approach .
pseudo-@xmath2 expansion ( @xmath3-series ) for @xmath0 found up to @xmath4 term for @xmath1 = 0 , 1 , 2 , 3 and within @xmath5 order for general @xmath1 is shown to have a structure rather favorable for getting numerical estimates .
use of pad approximants and direct summation of @xmath3-series result in iteration procedures rapidly converging to the asymptotic values that are very close to most reliable numerical estimates of @xmath0 known today .
the origin of this fortune is discussed and shown to lie in general properties of the pseudo-@xmath2 expansion machinery interfering with some peculiarities of the renormalization group expansion of @xmath0 . |
some of the recent chemical evolution models of galaxies have successfully included the dust content ( wang 1991 ; lisenfeld & ferrara 1998 ; dwek 1998 ; hirashita 1999a , hereafter h99 ) .
takagi , arimoto , & vanseviius ( 1999 ) modeled the dust extinction in ultraviolet ( uv)optical wavelengths and the reprocessed light in the far - infrared ( fir ) region , presenting the spectral evolution in the uv fir wavelengths .
as to the dust formation , supernovae ( sne ) are known to be one of the most dominant sources ( barlow 1978 ; draine & salpeter 1979 ; dwek & scalo 1980 ) , and sn shocks destroy grains ( mckee et al .
1987 ; jones et al . 1994 ; borkowski & dwek 1995 ) .
thus , dust content is connected with star formation histories . in our previous work ,
h99 , the dust - to - gas mass ratio was expressed as a non - linear function of metallicity , which is also related to star formation histories ( see also lisenfeld & ferrara 1998 ) .
h99 confirmed the suggestion proposed by dwek ( 1998 ) that the accretion process onto preexisting dust grains is efficient in spiral galaxies .
this is also supported by the observations of elemental depletions in dense clouds ( e.g , draine 1990 ) .
since the accretion is effective in cold clouds , the global efficiency of the accretion depends on the fractional mass of the cold gas ( seab 1987 ; mckee 1989 ; draine 1990 ; hirashita 1999b ) .
thus , the efficiency varies on a timescale of the phase transition of the interstellar medium ( ism ; @xmath0 yr ; ikeuchi 1988 ; mckee 1989 ) .
mckee & ostriker ( 1977 ) constructed a standard model for the ism composed of three phases : the hot rarefied gas ( @xmath2 k and @xmath3 @xmath4 ) , the warm gas ( @xmath5 k and @xmath6 @xmath4 ) , and the cold cloud ( @xmath7 k and @xmath8 @xmath4 ) .
one of the simplest models for the mass exchange among the three ism phases is described by ikeuchi & tomita ( 1983 , hereafter it83 ; see also habe , ikeuchi & tanaka 1981 ) .
they included the following three processes : [ 1 ] the sweeping of a warm gas into a cold component , [ 2 ] the evaporation of cold clouds embedded in a hot gas , and [ 3 ] the radiative cooling of a hot gas by collisions with a warm gas .
mass fraction of each component can show an oscillatory behavior , which is known as a limit - cycle ( nicolis & prigogine 1977 ; see also fujimoto & ikeuchi 1984 for another modeling of interstellar gas ) .
the limit - cycle behavior is supported by kamaya & takeuchi ( 1997 , hereafter kt97 ) , who interpreted the various level of the star formation activities in spiral galaxies shown by tomita , tomita , & sait ( 1996 ) using the limit - cycle model . in the limit - cycle model by it83 , mass fraction of each phase
oscillates continuously because of the mass exchange among the three components of the ism ( see also korchagin , ryabtsev , & vorobyov 1994 ) .
the timescale of the phase transition in the model is determined by three timescales of the above physical processes [ 1][3 ] .
actually , a static solution as well as the limit - cycle solution for the mass fractions is possible for some range in the parameter space ( this case is also examined in this paper ) . however
, the oscillatory behavior ( i.e. , the limit - cycle model ) of the fractional masses is supported observationally .
indeed , the observed scatter of the fir - to - optical flux ratios of spiral galaxies ( tomita , tomita , & sait 1996 ) is interpreted through the limit - cycle model in kt97 , who suggested that the fractional mass of the cold phase changes in the range of 0.1 to 0.7 ( or more ) on the timescale of @xmath9 yr and that this leads to the time variation of star formation activities if the schmidt law ( schmidt 1959 ) is applicable .
hirashita ( 1999b ) combined the framework of h99 with a theoretical work on multiphase ism and suggested the time variation of the dust - to - gas ratio by the phase transition .
we aim at constructing a model of the time evolution of dust - to - gas ratio including the effect of the ism phase transition .
the model in this paper enables us to simulate the time variation of the dust - to - gas ratio in a galaxy .
this paper is organized as follows .
first of all , in the next section , a model for the time variation of the dust - to - gas ratio is presented .
next , we show the numerical solution of the model in 3 . in 4 , we discuss the solution and present observational implications . finally , 5
is devoted to summary .
first , we prepare a model for the time variation of the dust - to - gas ratio in a galaxy .
the model is developed by h99 ( see also lisenfeld & ferrara 1998 and dwek 1998 ) in order to account for the relation between metallicity and dust - to - gas ratio of nearby spiral galaxies .
he adopted a simple one - zone model ( i.e. , the galaxy is treated as a closed system and spatial variations of physical quantities within the galaxy is neglected ) .
the model equations describe the changing rate of th mass of gas , metal , and dust : @xmath10 here , @xmath11 is the mass of gas ; @xmath12 and @xmath13 denote the total mass of the metal @xmath14 ( @xmath15 , c , si , mg , fe , etc . ) in both gas and dust phases , and only in the dust phase , respectively . the star formation rate is denoted by @xmath16 ; @xmath17 is the total injection rate of mass from stars ; @xmath18 is the mass fraction of the element @xmath14 ( i.e. , @xmath19 ) ; @xmath20 is the total injection rate of element @xmath14 from stars ; @xmath21 is the mass fraction of the element @xmath14 locked up in dust ( i.e. , @xmath22 ) .
the meanings of the other parameters in the above equations are as follows : @xmath23 represents the dust mass fraction in the injected material ( in other words , the dust condensation efficiency in the ejecta such as stellar winds or sne ) ; @xmath24 refers to the efficiency of dust destruction during a star formation [ @xmath25 corresponds to destruction of only the dust incorporated into the star , and @xmath26 corresponds to a net destruction ( formation ) in the star formation ] ; @xmath27 is the accretion timescale of the element @xmath14 onto preexisting dust grains in molecular clouds ; @xmath28 is the timescale of dust destruction by sn shocks .
we hereafter assume that only dust incorporated into stars is destroyed when the stars form ; i.e. , @xmath25 .
the mass loss from the galaxy is neglected in this paper , since we mainly treat spiral galaxies and expect that outflows are prevented by their deep dark matter potentials ( habe & ikeuchi 1980 ) .
when we consider star burst phase ( habe & ikeuchi 1980 ) or dwarf galaxies ( mac low & ferrara 1999 ) , the outflows may be important for the dust content as well as gas and metal content ( lisenfeld & ferrara 1998 ) .
we adopt the instantaneous recycling approximation ( i.e. , a star with the mass larger than @xmath29 dies instantaneously after its birth , leaving the remnant of mass @xmath30 ) .
once the initial mass function is fixed , we can calculate the returned fraction of the mass from formed stars ( @xmath31 ) and the mass fraction of the element @xmath14 that is newly produced and ejected by stars ( @xmath32 ) . using @xmath31 and @xmath32 , @xmath17 and @xmath20 in equations ( [ basic1])([basic3 ] )
is written as ( see h99 for the details ) @xmath33 after some algebra , equations ( [ basic2 ] ) and ( [ basic3 ] ) becomes @xmath34 { \cal d}_i,\label{basic6}\end{aligned}\ ] ] where @xmath35 ( timescale of gas consumption ; e.g. , roberts 1963 ) , @xmath36 , @xmath37 , and @xmath38 .
the timescales above are estimated in the case of typical spiral galaxies as @xmath3910 gyr ( kennicutt , tamblyn , & congdon 1994 ) , and @xmath40 yr ( draine 1990 ) .
the two timescales , @xmath27 and @xmath28 depend on the phase of the ism where the dust exists ( draine 1990 ; hirashita 1999b ) .
the accretion timescale , @xmath27 , is shorter in the cold phase of the ism than in any other phase , since efficient accretion needs a dense environment .
if we express the growth timescale of a dust grain in the cold component as @xmath41 , the accretion timescale is expressed as @xmath42 , where @xmath43 represents the mass fraction of the cold phase to the total mass of ism ( hirashita 1999b ) .
this is equivalent to @xmath44 where @xmath45 . according to draine ( 1990 ) , @xmath46 yr in the cold gas .
the destruction timescale of the dust @xmath47 is kept constant for simplicity , since we would like to concentrate on the time variation of the dust growth efficiency to examine the result in hirashita ( 1999b ) .
the star formation history should be taken into account in order to model the time dependence of @xmath47 . moreover
, the time variation of the sn rate also contributes to the time variation of the fractional masses the three ism components .
because of such complexities in the time dependence of @xmath47 , we simply treat @xmath47 as constant .
next , we review the model by it83 . this model is used to calculate the time evolution of the filling factors of the three ism phases ( see also 1 ) .
the result is used to calculate the dust formation efficiency through equation ( [ tacc ] ) .
the interstellar medium is assumed to consist of three components ( mckee & ostriker 1977 ) ; the hot rarefied gas ( @xmath2 k , @xmath3 @xmath4 ) , the warm gas ( @xmath5 k , @xmath6 @xmath4 ) , and the cold cloud ( @xmath7 k , @xmath8 @xmath4 ) .
the fractional masses of the three components are @xmath48 , @xmath49 , and @xmath43 , respectively .
a trivial relation is @xmath50 the following three processes are considered ( see it83 for the details ) : [ 1 ] the sweeping of a warm gas into a cold component at the rate of @xmath51 ( @xmath52 yr@xmath53 ) ; [ 2 ] the evaporation of cold clouds embedded in a hot gas at the rate of @xmath54 ( @xmath55@xmath56 yr@xmath53 ) ; [ 3 ] the radiative cooling of a hot gas through collisions with a warm gas at the rate of @xmath57 ( @xmath58@xmath59 yr@xmath53 )
. writing down the rate equations and using equation ( [ trivial ] ) , we obtain @xmath60 where @xmath61 , @xmath62 , and @xmath63 .
the solutions of equations ( [ cold ] ) and ( [ hot ] ) are classified into the following three types ( it83 ) : [ 1 ] @xmath64 ; all the orbits in the @xmath65-plane reduce to the node ( 0 , 1 ) , [ 2 ] @xmath66 and @xmath67 ; all the orbits reduce to a stable focus @xmath68 $ ] , [ 3 ] @xmath66 and @xmath69 ; all the orbits converge on a limit - cycle orbit , where @xmath70 . as a representative parameter sets for [ 1 ] and [ 2 ] ,
we examine @xmath71 and @xmath72 , respectively ( denoted as models 1 and 2 in table 1 , respectively ) .
we especially concentrate on the case [ 3 ] , since we treat spiral galaxies with cold gas and with various levels of star formation activities ( e.g. , tomita et al .
the two properties of spiral galaxies are naturally realized ( kt97 ) if the filling factor of the cold gas shows the limit - cycle behavior and the star formation rate changes according to the schmidt law ( @xmath73 ; @xmath742 ; schmidt 1959 ) . as representative parameter sets of the limit - cycle orbit
, we examine the case of @xmath75 and @xmath76 ( denoted by model 3 and 4 in table 1 , respectively ) .
as explained in 2.3 ( model 3a , 3b , 4a , and 4b ) . ]
the cycle periods of the former case is about three times shorter than that of the latter . in table 1 ,
the adopted parameters are summarized .
the column `` type '' means the behavior of the solutions and named `` runaway , '' `` stationary '' and `` limit - cycle '' for the cases [ 1 ] , [ 2 ] and [ 3 ] , respectively .
we note that all the parameter sets above are examined in it83 ( their figures 1 and 3 ) . here , we summarize the equations and the parameters .
first , the filling factor of the cold gas is calculated by the following set of equations : @xmath77 with parameters @xmath78 and @xmath79 listed in table 1 . according to these equations , the time variation of the cold - gas fractional mass @xmath43 is calculated to obtain the efficiency of the heavy - element accretion onto dust grains ( eq . [ tacc ] ) .
next , the time variation of the dust - to - gas ratio is simultaneously solved .
since we are interested in a timescale much shorter than @xmath80 , which is also interpreted as metal - production timescale through equation ( [ basic5 ] ) , we treat @xmath18 as constants .
then , equation ( [ basic6 ] ) is solved for oxygen abundance ( @xmath15 , because we adopted the oxygen as a tracer element in the previous work , h99 : see also lisenfeld & ferrara 1998 ) : @xmath81 { \cal d}_{\rm o},\label{basic7}\end{aligned}\ ] ] where we express the time in a nondimensional manner by using @xmath82 and we assume @xmath25 ( equivalent to the assumption that only dust incorporated into stars is destroyed in the star formation processes ; 2.1 ) .
we note the relation @xmath83 .
we fix the parameters as @xmath84 yr ( e.g. , kennicutt et al . 1994 ) ,
@xmath85 ( h99 ) , @xmath86 , @xmath87 , @xmath88 ( corresponding to @xmath89 yr ; we note that @xmath90 and @xmath91 are related through eq .
[ tacc ] ) , @xmath92 ( h99 ) , and @xmath93 ( the solar system abundance ) .
the dependence on these parameters above are described in lisenfeld & ferrara ( 1998 ) , h99 , and hirashita ( 1999c ) except for @xmath94 and @xmath80 .
the parameter @xmath94 scales the phase transition timescale , and we examine the cases of @xmath95 and 1000 ( table 1 : @xmath96@xmath97 yr ; it83 ) .
we note that @xmath94 appears in the form of the nondimensional parameter @xmath98 .
we examine only the case of @xmath95 in model 1 and 2 , since the qualitative behavior of the dust - to - gas ratio is the same , irrespective of which value of @xmath98 is adopted .
finally , the total dust - to - gas ratio @xmath99 is calculated from the following equation : @xmath100 where we assume @xmath101 ( the galactic value ; whittet 1992 , p. 52 ) .
we solve the basic equations ( [ basic8])([conversion ] ) for the parameters listed in table 1 .
the initial condition for the filling factors are set according to it83 : @xmath102 except for the model 4 and @xmath103 for model 4 .
the initial dust - to - gas ratio is set as @xmath104 ( galactic value ; whittet 1992 ) for all the models .
models 3 and 4 are examined for two different values of @xmath94 listed in table 1 .
first , the result of model 1 is displayed in figure 1 .
because the cold gas disappears , the dust growth becomes ineffective .
then the dust destruction makes the dust - to - gas ratio converge to @xmath1 in the dust destruction timescale ( @xmath105 yr , i.e. , 10 in the unit of @xmath106 ) .
this model represents the efficient sweeping of warm gas in comparison with the radiative cooling of the hot gas .
thus , the swept gas is first converted into the cold component and then into hot gas ( it83 ) . if the hot gas escapes out of the galaxy owing to its large thermal energy , this model corresponds to the galactic wind model of elliptical galaxies ( it83 ; arimoto & yoshii 1987 ) .
the final value of the dust - to - gas ratio in model 1 is calculated by considering a stationary state , @xmath107 , in equation ( [ basic7 ] ) : @xmath108 because of the constant dust formation from metal injection at the rate of @xmath109 , the dust - to - gas ratio is not exactly 0 even after a long time . since @xmath110 , @xmath111 .
thus , the solution of equation([stationary ] ) is @xmath112 for the adopted parameters @xmath93 , @xmath113 , @xmath86 , @xmath114 , and @xmath92 .
this indeed matches the final value of the dust - to - gas ratio in model 1 .
next , the result of model 2 is presented in figure 2 .
as shown in it83 , the solution for @xmath43 reduces to a stable point @xmath115 .
the stable stationary state is realized at a finite dust - to - gas ratio , which is determined so that the dust formation rate is balanced with the dust destruction rate .
the value is calculated from equation ( [ basic7 ] ) .
putting the adopted parameters , the positive solution is @xmath116 , which matches the stationary value in figure 2 .
finally , the limit - cycle case is examined in figure 3 and 4 .
the four sets of parameters listed in table 1 are examined ( models 3a , 3b , 4a , and 4b ) .
the resulting time evolution of the cold component of ism and the dust - to - gas ratio is presented in figures 3a , 3b , 4a , and 4b for the four cases .
as expected , the dust - to - gas ratio oscillates in response to the fractional mass of the cold component , though the amplitude of the variation largely depends on the parameters . comparing the figures 3a and 4a ( or figures 3b and 4b )
, we see that the amplitude becomes larger if the period of the oscillation is longer .
this is because the dust has enough time to grow if the period is long . also from the comparison between the figures 3a and 3b ( or figures 4a and 4b )
, we find that the smaller @xmath94 makes the amplitude larger .
since small @xmath94 means that a timestep of the phase transition model ( eqs . [ cold ] and [ hot ] ) is long in the real time unit ( @xmath117 corresponds to @xmath118 ) , the dust has enough time to grow in the case of smaller @xmath94 .
the timescale of the dust growth relative to that of the phase transition determines the amplitude .
this point is again discussed in the next section .
in the runaway and stationary solutions ( models 1 and 2 ) , the dust - to - gas ratio is settled into a stationary value determined by equation ( [ stationary ] ) after the dust formation and destruction timescales ( @xmath119 yr ) . in the limit - cycle type solutions ( models 3 and 4 ) , the amplitude of the dust - to - gas oscillation largely depends on the timescale of the phase transition . in this subsection
, we discuss the phase - transition and dust - formation timescales in the limit - cycle model . from the solution of the model equations in the previous section ( figs . 3 and 4 ) , we find that the dust - to - gas ratio oscillates in response to the filling factor of the cold component
. however , the amplitude of the oscillation is highly dependent on the values of the parameters @xmath78 , @xmath79 , and @xmath94 .
in other words , the amplitude is determined by the relation between the timescale of the phase transition ( or the period of the oscillation ) and that of the dust growth timescale . in the following ,
we discuss the relation between the the period of the phase transition ( denoted by @xmath120 ) and the dust growth timescale ( @xmath41 ) . in order for the dust to grow ,
the dust must have enough time for the growth , i.e. , @xmath121 . by using equation ( [ tacc ] ) ,
the condition is equivalent to @xmath122 which reads the following inequality by using @xmath90 : @xmath123 putting @xmath88 and @xmath124 , we obtain @xmath125 for the enough dust growth , which realizes the large amplitude of the time variation of the dust - to - gas ratio .
this means that the timescale of the phase transition should be larger than @xmath126 yr for the sufficient dust growth , since @xmath127 gyr ( kennicutt et al .
1994 ) . here
, we note that the timestep in figures 3 and 4 is @xmath106 in the physical unit and that @xmath96@xmath97 yr . in the longer unit ( @xmath128 yr ; corresponding to figures 3a and 4a ) , the amplitude of the dust - to - gas - ratio oscillation is large . on the contrary , in the shorter unit ( @xmath96 yr ; corresponding to figures 3b and 4b ) ,
the amplitude is much smaller because the timestep is shorter and thus the condition ( [ cond1 ] ) is difficult to be satisfied . moreover , comparing figures 3a and 4a , we see that the amplitude is larger if the period is longer . in figure 4a ,
the maximum of the dust - to - gas ratio is an order of magnitude larger than the minimum .
thus , the dust - to - gas ratio in spiral galaxies can show an order - of - magnitude oscillation .
we should note that the parameter range is reasonable for the spiral galaxies ( ikeuchi 1988 ) .
thus , we have confirmed the interpretation of the scatter of the dust - to - gas ratio in hirashita ( 1999b ) , who interpreted the scatter of the dust - to - gas ratio of the nearby spiral sample as the short - term ( @xmath129@xmath130-yr ) variation of the dust - to - gas ratio with the maximum - to - minimum ratio of more than 4 .
the important point is that the dust - to - gas - ratio oscillation with large amplitude occurs in a reasonable parameter range for spiral galaxies ( model 4a ) .
however , the amplitudes are small for the other models after the stable limit cycle is attained .
thus , some spiral galaxies may not experience the prominent oscillation of the dust - to - gas ratio .
but it is possible for the spiral galaxies to experience the oscillation of the dust - to - gas ratio , which is observed in the scatter of the dust - to - gas ratio of the spiral sample . in the discussions above
, the metallicity was fixed because the timescale of the phase transition is much shorter than that of the chemical enrichment
. however , the time range shown in the figures in this paper ( 250 timesteps ) corresponds to 7.5 gyr when the @xmath95 is adopted .
thus , we need to examine the effect of the chemical evolution here . to include the effect of chemical evolution
, we solve equation ( [ basic5 ] ) to examine the time variation of metallicity .
if @xmath80 is fixed for simplicity , the solution of equation ( [ basic5 ] ) becomes @xmath131 in other words , we simply assume a constant rate of metal enrichment to examine a effect of the chemical evolution qualitatively .
we solve basic equations in 2.3 for the time - variable @xmath132 described by equation ( [ varx ] ) . as a representative case
, we adopt the parameter set identical to model 4a in table1 , since the oscillation behavior of the dust - to - gas ratio is the most prominent ( figure 4a ) . the result is shown in figure 5 .
we see that the qualitative behavior of the dust - to - gas ratio is the same as the results for constant metallicity , except for the gradual increase of dust - to - gas ratio .
this increase is due to the metal production , from which dust is formed .
thus , two behaviors of the temporal variation is coupled : one is the gradual increase of the dust - to - gas ratio on the timescale of chemical enrichment , and the other is the short - term variation of the dust - to - gas ratio owing to the ism phase changes .
we note that the effect of chemical evolution is of significant importance if @xmath133 timescale of the chemical enrichment ) is short . especially for starburst galaxies , @xmath134 yr , which is comparable to @xmath120 ( moorwood 1996 ) . in this case
, the metal enrichment also proceeds from the starburst in @xmath80 , and thus the assumption of the constant metallicity is never satisfied . the same discussion
may also be applied to the initial burst in the formation epoch of galaxies , when @xmath80 is nearly the value of starburst galaxies ( @xmath119 yr ; the dynamical time ) .
we have shown that the dust - to - gas ratio varies on a timescale of the ism phase transition ( @xmath129@xmath130 yr ) in spiral galaxies .
the amplitude of the variation can be an order of magnitude ( figs .
this confirms the previous suggestion in hirashita ( 1999b ) that the dust - to - gas ratio scatter of the nearby spiral galaxies are explained by the time variation .
the result also suggests that the variation is partially responsible for the scatter of the fir - to - optical flux ratio of spiral sample in tomita , tomita , & sait ( 1996 ) , since the dust content as well as the stellar heating ( dust temperature ) is responsible for the far - infrared luminosity ( e.g. , whittet 1992 , p. 170 ) .
we comment on the dust content in dwarf galaxies .
lisenfeld & ferrara ( 1998 ) modeled the time evolution of the dust amount in star - forming dwarf galaxies ( dwarf irregular galaxies and blue compact dwarf galaxies ) . based on their model and considering the dust formation in cold clouds ( dwek 1998 ) , hirashita ( 1999c ) presented that the dust growth in dwarf galaxies is negligible compared with the dust condensation from metal ejected by stars .
this is because the metallicity of dwarf galaxies are much smaller than that of spiral galaxies and the collision between grains and metal atoms are not frequent enough for efficienct growth of the grains .
thus , the oscillation of dust - to - gas ratio through the variation of the dust growth efficiency is difficult in dwarf galaxies . in this case
the scatter of the dust - to - gas ratio of the dwarf sample may be interpreted to reflect the various efficiency of the gas outflow ( lisenfeld & ferrara 1998 ) .
the wind may easily blows out of the dwarf galaxies because of their shallow gravitational potentials ( larson 1974 ; de young & heckman 1994 ) , though the distribution of dark matter largely affects the process of the outflow ( mac low & ferrara 1999 ; ferrara & tolstoy 1999 ) .
we should note that the timescale of the phase transition in dwarf galaxies is expected to be longer because of their longer cooling timescale . since the metal - line cooling is the dominant cooling mechanism for the hot gas ( raymond , cox , & smith 1976 ) , the small metallicity of dwarf galaxies , typically @xmath135 of the spiral galaxies , makes the cooling time an order of magnitude longer ( @xmath136 yr ) .
thus , once the hot component becomes dominant in a dwarf galaxy , the dwarf galaxy must wait for the hot component to cool down for more than @xmath130 yr , during which the growth of dust grains is prevented .
this as well as the mechanism in the previous paragraph makes the dust growth difficult .
indeed , the dust - to - gas ratios in dwarf galaxies are known to be much smaller than those in spiral galaxies ( e.g. , gondhalekar et al .
we have discussed the time evolution of dust - to - gas mass ratio in the context of multi - phase model of interstellar medium in spiral galaxies .
the phase transition of interstellar gas occurs on a timescale of @xmath129@xmath130 yr in spiral galaxies ( it83 ) .
the dust growth rate also varies on that timescale owing to the time variation of the fractional mass of the cold gas through the phase transition ( hirashita 1999b ; 2.1 of this paper ) . in order to examine the response of the dust - to - gas ratio to the phase transition , we modeled the time variation of the dust - to - gas ratio with a model of the phase transition .
we adopt the three phase model for the interstellar gas and the ikeuchi - tomita model for the phase transition . according to the model , three types of solutions
are possible : [ 1 ] all the gas is transformed to a hot gas ; [ 2 ] a stable stationary state of three phases is realized ; [ 3 ] the filling factors of each phase cyclically changes .
for each of the three types of solutions , the dust - to - gas ratio behaves as follows ( 3 ) : [ 1 ] the dust is fully destroyed ( the dust - to - gas ratio becomes 0 ) ; [ 2 ] the dust - to - gas ratio converges to a stationary state ; [ 3 ] the dust - to - gas ratio varies cyclically in response to the phase transition .
when the case [ 3 ] is applicable , the amplitude of the variation of the dust - to - gas ratio is large ( nearly an order of magnitude ) if the dust growth timescale is shorter than the phase transition timescale .
this condition is satisfied within the reasonable parameter range of spiral galaxies ( 4.1 ) .
however , the model needs to be modified for dwarf galaxies , since their small metallicity makes the dust growth rate in clouds small ( hirashita 1999c ) , and their timescale of the phase transition is longer owing to their longer cooling timescale . for starburst galaxies ,
the effect of chemical enrichment is important because of their short metal - enrichment timescale .
we wish to thank the anonymous referee for useful comments that substantially improved this paper .
we are grateful to s. mineshige for his continuous encouragement .
this work is supported by the research fellowship of the japan society for the promotion of science for young scientists .
we have made extensive use of the nasa s astrophysics data system abstract service ( ads ) .
ccccc model no .
& @xmath78 & @xmath79 & @xmath98 & type + 1 & 1.2 & 3.0 & 100 & runaway + 2 & 0.4 & 3.0 & 100 & stationary + 3a & 0.3 & 3.0 & 100 & limit - cycle + 3b & 0.3 & 3.0 & 1000 & limit - cycle + 4a & 0.3 & 0.5 & 100 & limit - cycle + 4b & 0.3 & 0.5 & 1000 & limit - cycle + fig .
1 _ upper panel _ : the time evolution of the the cold - gas mass filling factor ( @xmath137 ) for model 1 .
the initial condition is @xmath102 .
the adopted values of the parameters are @xmath138 , @xmath139 , and @xmath140 .
the timestep is normalized by @xmath141 yr . _ lower panel
_ : the time evolution of the dust - to - gas ratio ( @xmath99 )
. fig . 2 the same as fig . 1 , but for model 2 . the initial condition is @xmath102 .
the adopted values of the parameters are @xmath142 , @xmath139 , and @xmath140 .
the timestep is normalized by @xmath141 yr .
3a the same as fig . 1 ,
but for model 3a .
the initial condition is @xmath102 .
the adopted values of the parameters are @xmath143 , @xmath139 , and @xmath140 .
the timestep is normalized by @xmath141 yr .
the same as fig . 1 , but for model 3b
the initial condition is @xmath102 .
the adopted values of the parameters are @xmath143 , @xmath139 , and @xmath144 .
the timestep is normalized by @xmath145 yr .
4a the same as fig . 1 , but for model 4a .
the initial condition is @xmath103 .
the adopted values of the parameters are @xmath143 , @xmath146 , and @xmath140 .
the timestep is normalized by @xmath141 yr .
4b the same as fig . 1 ,
but for model 4b .
the initial condition is @xmath103 .
the adopted values of the parameters are @xmath143 , @xmath146 , and @xmath144 .
the timestep is normalized by @xmath145 yr . | we discuss the time variation of dust - to - gas mass ratio in spiral galaxies using the multi - phase model of interstellar medium .
the typical timescale of the phase change of an interstellar gas is @xmath0 yr in spiral galaxies . since the phase transition changes the filling factor of the cold gas where the dust growth occurs , the dust growth rate varies on that timescale . in order to examine the response of the dust - to - gas ratio to the phase transition
, we construct a model of the time evolution of the dust - to - gas ratio .
we adopt the three phase model for the interstellar gas and the ikeuchi - tomita model for the mass exchange among the phases . according to the model ,
three types of solutions are possible : [ 1 ] all the gas is transformed to a hot gas ; [ 2 ] a stable stationary state of three phases is realized ; [ 3 ] the filling factor of each phase cyclically changes . for each of the three types of solutions
, the dust - to - gas ratio behaves as follows : [ 1 ] almost all the dust is destroyed ( the dust - to - gas ratio becomes @xmath1 ) ; [ 2 ] the dust - to - gas ratio converges to a stationary state ; [ 3 ] the dust - to - gas ratio varies cyclically in response to the phase transition . in the case of [ 3 ] , the amplitude of the variation of the dust - to - gas ratio is large ( nearly an order of magnitude ) if the dust growth timescale is shorter than the phase transition timescale .
this condition is easily satisfied in spiral galaxies .
however , it is difficult for dwarf galaxies to realize the condition because their small metallicity makes the dust growth timescale long .
@align#1#2.6ex accepted to _ apj _
email : [email protected] |
long dna is naturally found in a dense form in most biological systems due to the presence of compacting agents . in vitro
the most used agents are polyamines such as trivalent spermidine ( spd ) and tetravalent spermine ( spe ) @xcite .
these agents play a key role in maintaining cellular dna in a compact state @xcite , modulate ion channel activities of cells , are essential for normal cell growth @xcite and can effectively be applied in gene delivery and in the field of genetic therapy . under physiologic ionic and ph conditions ,
the polyamines are positively charged and hence dna is their prime target of interaction . the molecular mechanism of polyamine function in dna condensation is presumed to involve neutralization of the negatively charged dna backbone by the positively charged amino groups of spd and spe @xcite .
experimental results and counterion condensation theories indicate that non - specific interactions are predominantly electrostatic between polyamines and dna phosphates @xcite .
thus , the electrostatic shielding of phosphates facilitates close helix - helix surface contacts and , ultimately , dna condensation through the correlation attraction @xcite , the attraction between strongly correlated counterion layers on the adjacent dna surfaces . in the last decade ,
different experiments @xcite have shown evidence of dna redissolution ( i.e. dna unbinding ) with increasing concentration of polyamines and several of their structural analogues .
first , addition of a certain amount of multivalent salt causes the collapse of single dna or the bundling in multi - columnar structures .
upon addition of more salt , the polyelectrolyte redissolves and dna assumes again an unbundled conformation .
there are three experimentally well established features of the redissolution phenomena : i ) a linear relationship is found between the threshold concentration of multivalent ions @xmath0 , when the onset of dna condensation takes place , and the initial dna concentration @xmath1 ( the dna concentration in solution free of multivalent ions ) ; ii ) the decondensation threshold @xmath2 of multivalent ions , when the condensed dna returns back to solution , is almost independent of monovalent salt concentration @xmath3 ; iii ) between the threshold values @xmath0 and @xmath2 the dna fragments show two coexisting liquid crystalline phases . concerning the first item , the onset of dna condensation is usually explained by the correlation attraction between almost neutral structures .
thus , the precipitation induced by trivalent or tetravalent ions is not a consequence of the intrinsic structure and flexibility of dna , but is a common feature of a polyelectrolyte solution .
the threshold value @xmath0 grows with increasing monovalent salt concentration @xmath3 @xcite .
a mono - molecular dna collapse into a neutral toroidal structure occurs in highly dilute solutions of long dna chains @xcite , whereas a multi - molecular aggregation is generally observed in more concentrated dna solutions , regardless of the dna length @xcite .
the second item , the dna redissolution at @xmath2 , is currently under intensive debate in the colloidal community with different , and sometimes conflicting , explanations @xcite .
for instance , in ref . @xcite it is argued that after precipitation the increased binding of polyamines will make the dna hydrophilic enough to dissolve in water .
other experiments @xcite show that the dna is still in a condensed state when polyions are added beyond the threshold concentration @xmath2 , but it is more finely dispersed in the solution . in ref.@xcite
the reentrance is explained by resorting to dna overcharging phenomena , which take place roughly in the middle of the condensation plateau .
dna is claimed to experience negative electrophoresis and move opposite to the conventional direction near the reentrance transition .
however , experiments of raspaud et al @xcite do not confirm this claim .
in ref.@xcite it was suggested that the redissolution is very sensitive to the relationship between the condensation of multivalent counterions on the polyelectrolyte and multivalent counterion
monovalent coion association ( bjerrum pairs ) .
thus , if the chemical potential of the multivalent counterions is low , they more likely adsorb on the dna and overcharge it . on the other hand ,
if the chemical potential is high , the counterion - coion associations are more likely to appear in solution .
a resulting condensation of bjerrum pairs creates consecutive layers of oppositely charged ions around the dna molecule @xcite .
while the first two above mentioned items have been studied in considerable detail , much less attention has been paid to the third item , namely the coexistence of two different liquid - like structures in dna condensates . in a series of experiments , livolant and
colleagues demonstrated that the spermidine and spermine ions are capable of provoking several liquid crystalline forms of fragmented dna @xcite .
similar results were published recently by saminathan et al in ref .
@xcite . in this paper
we investigate the condensation and redissolution of dna on a molecular level by using computer simulations of the primitive - model electrolyte with explicit tetravalent counterions and monovalent salt ions .
we trace back the condensation and redissolution to the distance - dependent effective potential @xmath4 between two parallel dna molecules , where @xmath5 is the radial distance between their two centers . using two - dimensional liquid - state theory for the fluid and lattice sums for the solid phases ,
we calculate the phase diagram for columnar dna assemblies .
a previous account of the results was already published elsewhere @xcite .
the remainder of this paper is organized as follows . in section
ii we describe our model system and give the definition of dna - dna interaction forces .
we calculate the interaction forces for different counterion and salt concentrations in section iii .
section iv is devoted to the free energy perturbation theory for defining the liquid - liquid coexistence densities .
we conclude in section v.
we consider a b - dna molecule which is a double helix with a pitch length @xmath6=34 .
there are @xmath7=20 phosphate charges per one helical turn which makes one elementary charge per each 1.7 of axial rise .
the geometrical shape and charge distribution of dna is modeled through the accurate montoro - abascal model ( mam ) @xcite .
its cross section is illustrated in fig .
[ figure_1 ] .
a pair of dna molecules are placed along the _ xy _ diagonal of the cubic simulation box of volume @xmath8 .
the size of the simulation box @xmath9 corresponds to three full turns of dna .
the box also consists of @xmath10 multivalent ions with valency @xmath11 , @xmath12 monovalent coions and @xmath13 monovalent counterions . here
@xmath14 is the number of added salt ion pairs , @xmath15 is fixed by the dna phosphate charge due to the constraint of global charge neutrality ( @xmath16 in our simulations ) .
all ions are modelled as charged hard spheres .
tetravalent counterions are assumed to represent spermine ions .
though the latter is a polyion in _ vivo _ , a spherical approximation can be used , since the bibliographical data support the idea that it is the charge of counterion , rather than its structural specificities , which is important in dna condensation and redissolution processes @xcite . beside of this
, the fluidity of the ordered dna phase also suggests that spermine binds like an ordinary ion to the dna surface , rather than forming inter - strand cross - links to neighboring dna molecules @xcite . assuming that the ions of the same valency are indistinguishable
, the actual number of mobile ion species in the simulation box is reduced from five ( which are multivalent ions and their coions , positive and negative monovalent salt ions and monovalent counterions that neutralize the dna phosphate charge ) to three : multivalent counterions ( charge @xmath17 and diameter @xmath18 ) and positive and negative small ions ( charge @xmath19 and diameter @xmath20 ) .
periodic boundary conditions in all three directions are applied to reduce finite size effects .
the dna replicas in the _ z _ direction produce infinitely long molecules .
the whole system is held at room temperature @xmath21 and the water is modelled as a continuous dielectric medium with dielectric constant @xmath22 .
a typical snapshot of the simulation is shown in figure [ figure_2 ] .
the interaction potentials between the five sort of particles ( three of them are the mobile ions in solution mentioned above , and two of them , one charged and the other neutral , belong to the dna molecule , see figure [ figure_1 ] ) are a combination of hard core and coulomb potentials : @xmath23 here @xmath24 is the inter - particle distance , @xmath25= @xmath26 for multivalent counterions , @xmath27 for small ions , @xmath28 for phosphate charges and @xmath29 for neutral spheres in the mam ( with @xmath30=0 ) . there is also an excluded volume potential @xmath31 between the dna hard cylinder and the free ions @xmath32
. the basic quantity of interest is the effective force per helical turn @xcite @xmath33 between two dna molecules . here
@xmath34 is the direct coulomb repulsion per helical turn of one dna molecule as exerted from the phosphate groups of the other dna .
it does not depend on salt density and its calculation is straightforward @xcite .
the second term @xmath35 in eq .
( [ eq_1 ] ) corresponds to coulomb interactions between the phosphate charges at positions @xmath36 ( @xmath37 ) and the mobile ions at positions @xmath38 ( @xmath39 , @xmath40 ) , @xmath41 here @xmath42 denotes canonical average over small ion configurations .
the third term @xmath43 in eq .
( [ eq_1 ] ) arises from the entropic contribution of small ions due to their moment transform to the dna surface @xmath44 , @xmath45 here @xmath46 is a surface normal vector pointing outwards the dna core and @xmath47 ( @xmath48 ) is the inhomogeneous microion concentration .
the canonical averages in @xmath35 and @xmath43 are carried over all configurations of the mobile ions .
we have performed extensive grand - canonical molecular dynamics ( gcmd ) simulations , similar to those of ref .
@xcite , for a range of different tetravalent counterion and monovalent salt concentrations .
each simulation is referred by its nominal tetravalent counterion concentration @xmath49 ( salt ion concentration @xmath3 ) defined as a ratio between the total number of ions @xmath10 ( @xmath50 ) in the cell without the dna molecules and the system volume @xmath51 , @xmath52 ( @xmath53 ) .
additional simulations have been carried for these bulk phases in order to calculate the chemical potential @xmath54 of solution .
then in the simulations with dna molecules the number of ions in the simulation cell was automatically adjusted to the specified value of chemical potential @xmath54 .
the ion diameter was chosen to be @xmath18=8 for tetravalent counterions and @xmath20=4 for other free ions in the system .
this parameter defines the closest approach of the ion to the dna surface and has a strong impact on the polyion electrostatics .
the time step @xmath55 of the simulation corresponds to an average ion displacement of @xmath56 per time step such that the reflection of counterions following the collision with the combined surface of dna is calculated with high precision . about @xmath57 md steps
are required on average to reach equilibrium .
the time evolution is then followed for @xmath58 steps to gather statistics to calculate canonical averages of the interaction forces .
the case of a single dna molecule in the presence of spermine ions has been considered in ref .
it was shown there ( see figure 2 in ref .
@xcite ) that the ionic cloud may not only compensate the polyion charge but even exceed it , resulting in an opposite sign of the electrostatic potential at some distances from the dna surface .
the adsorption of bjerrum pairs @xcite onto the dna surface at high tetravalent counterion concentration creates consecutive layers of charges of different sign around the dna molecule .
the onset of a multilayer structure occurs at @xmath59 mm .
addition of monovalent salt shifts this threshold concentration to lower values of @xmath49 , in accordance with experimental observations and two - component manning condensation theory @xcite . for multivalent
counterion concentrations exceeding @xmath59 mm , the total charge in the dna grooves remains constant and only the total ionic charge adsorbed on the strands contributes to the overcharging , similar to our earlier findings @xcite . beside of this , there is a competition between the multivalent and monovalent counterions in binding to the dna surface as @xmath49 increases .
the multivalent ions tend to replace the monovalent counterions on the dna surface .
thus , at higher @xmath49 the main contribution to the formation of charged layers around dna molecules comes from bjerrum association between big counterions and small coions .
such charged layers give rise to an attraction between two parallel dna molecules , as shown in figure [ force_charge ] .
a decrease of the big counterion charge leads to the break - up of the bjerrum counterion - coion pairs and thus destroys the layer formation around the dna molecule .
this ultimately results in the loss of intermolecular attraction , see simulation data for @xmath601,2 and 3 in figure [ force_charge ] .
it should be mentioned that in addition to the intermolecular ( or axis - to - axis ) distance @xmath5 , there are angular variables which define the mutual configuration of two parallel dna molecules corresponding to the orientation of their grooves and strands @xcite . at short surface - to - surface distances between the two dna molecules up to 5 there are strong contributions from particular dna - dna configurations @xcite . for larger separation distance ,
@xmath61 we find no detectable dependence of the interaction forces on the mutual orientations of dna molecules . on the other hand ,
there is experimental evidence that in dna condensates two neighboring molecules never approach each other closer than 5 .
this apparently means that at such small distances a strong repulsion between dna solvation shells exists .
thus in all figures hereafter we show orientationally averaged interaction forces starting from the distance @xmath62 .
the interaction forces are scaled per dna pitch , i.e. per 10 dna base pairs .
the electrostatic @xmath63 and entropic @xmath64 components of total interaction force @xmath65 for tetravalent counterions corresponding to dna overcharging are separately plotted in figure [ figure_4 ] .
it can be seen that the electrostatic force shows oscillations around zero , which are reminiscent of the multilayered structure of charges around a single dna molecule @xcite .
the deep attractive minimum of the total force @xmath65 has an entropic origin , whereas the second minimum at intermolecular distance r=41 has a purely electrostatic origin . in figure [ force_salt ]
we plot the dna - dna interaction force for both the undercharged and overcharged cases at different salt concentrations .
it is evident that in dense salt solutions the attractive minimum becomes weaker . for overcharged dna , @xmath66 mm , the position of the second minimum shifts towards the dna surface
thus , whereas the dna overcharging does not strongly depend on the added salt concentration @xmath3 @xcite , the effective forces do .
the positions of the minimum and the maximum of the force shift towards the dna surface also for higher spermine concentrations .
this trend is shown in figure [ figure_5a ] , where the dna - dna interaction force is plotted for different spermine and salt concentrations at fixed distance @xmath67 .
for low @xmath49 , which in figure [ figure_5a ] corresponds to the area to the left of point a , the dna - dna interaction is overall repulsive . between the points @xmath68 and @xmath69 a first minimum develops in @xmath65 .
as the spermine concentration increases further , the minimum shifts toward the dna surface and the force has a repulsive tail .
this tail indicates the appearance of second maximum , which obviously is followed by second minimum .
the dependence of the dna - dna interaction force on @xmath49 for two fixed intermolecular distances are shown in figure [ force_min_max ] .
five different arrows in this picture point to different values of @xmath49 which characterize the number of attractive minima in the interaction force @xmath65 . for small @xmath70 mm , indicated by the arrow next to _ a _ in figure [ force_min_max ] , the interaction force has no minimum and thus is totally repulsive . for spermine concentration @xmath71 mm , corresponding to the arrow next to _
b _ , the force has a single minimum . for intermediate @xmath72 mm and @xmath73 mm , see the arrows next to _ c _ and _ d _ respectively , there are two minima in the interaction force ( note that a positive force at @xmath74 38 for @xmath49 corresponding to arrow _ d _ indicates the occurrence of second maximum , which obviously if followed by second minimum at larger @xmath5 ) . and
finally , at even higher @xmath75 mm , the first minimum has disappeared , however the second minimum is retained . a full set of distance resolved dna - dna interaction force curves for different tetravalent
counterion densities @xmath49 are presented in figure [ interctionforce ] .
it can be seen that even a small trace of spermine ions
well below the overcharging threshold induces an attraction between the dna molecules , except at very close distances , see curve for @xmath49=0.1 mm .
as the dna molecules get more overcharged , which corresponds to high spermine concentrations , the main minimum narrows and becomes more shallow . at the same time , the width and the height of the maximum increases .
we note that the attractive minimum for undercharged dna pair has mainly a pure electrostatic origin and arises due to charge correlations in the electrolyte .
however , for overcharged dna the main contribution to the force at this minimum is due to the spermine crowding near the dna surface .
the second maximum originates both from electrostatic and entropic forces .
finally , the second minimum emerges from a pure electrostatic effect . in the following
we calculate the total effective pair interaction potential @xmath4 per unit length for a given bulk salt concentration @xmath3 and different spe concentrations @xmath49 . the quantity @xmath4 is obtained by integrating the distance - resolved interaction force averaged over all microion configurations @xcite .
results are shown in figure [ figure_5 ] .
the oscillations in the force imply that the interaction potential also oscillates . with increasing @xmath49
, the first minimum of @xmath4 is getting deeper and is achieving a maximal depth at the overcharging concentration @xmath76 mm . a further increase of @xmath49 again reduces the depth of this minimum .
the position of the minimum , on the other hand , hardly depends on @xmath49 .
the spe - layering around the pair of dna molecules induces a second minimum at larger separations as revealed in the enlarging inset of figure [ figure_5 ] .
this minimum is of electrostatic origin and occurs for @xmath7765 mm . again
the depth of the second minimum first increases and then decreases with increasing @xmath49 . at intermediate spe concentrations ,
we are thus confronted with a double minimum potential which is induced by layering .
it is worth to mention that positions of the second minimum can be related to intermolecular distances between the dna molecules in cholesteric phase induced by polyamines @xcite .
direct measurements @xcite and theoretical investigations @xcite of intermolecular forces demonstrated that the energetics of this cholesteric phase is determined primarily by electrostatic interactions . since the potentials in figure [ figure_5 ] are scaled for one dna pitch length , the interaction strength corresponding to the minimum of curve ( 2 ) for very long dna molecules is sufficiently large to induce condensation .
this implies that dna aggregation can take place well _ below _ overcharging spe concentrations .
the characteristic double - minimum structure of the interaction potential @xmath4 will give rise to an unusual phase behavior .
we have calculated the phase diagram of a columnar dna assembly on the basis of our simulated effective pair interactions .
to do so , we assume that the dna molecules are parallel along a certain length @xmath78 .
this length is an additional parameter which we fix to be @xmath79 .
we comment on the dependence of the phase diagram on @xmath78 in section v. the assembly of parallel dna can be considered as a two - dimensional many - body system interacting via a potential @xmath80 and being characterized by a dna particle number density @xmath81 .
we calculated the free energies of the fluid and solid phases by using different techniques outlined below and perform the traditional maxwell double tangent construction to identify the coexisting densities .
the free energy of dilute _ fluid _ phase is approximated by the two - dimensional perturbation theory @xcite via splitting the total potential into a repulsive and attractive parts , @xmath82 .
the repulsive potential @xmath83 , identical to @xmath4 but truncated and shifted towards zero at the first minimum at @xmath84 , is mapped onto that of effective hard disks of diameter @xmath85
@xcite @xmath86 } dr.\ ] ] here the cross - section diameter for the dna molecule is @xmath87 .
the total helmholtz free energy involves that of a hard disk fluid with effective area fraction @xmath88 and a mean - field correction which we simply model as @xmath89 . for the free energy of a hard - disc fluid ,
analytical expressions are available @xcite .
the free energy of the _ solid _ phase , on the other hand , is calculated as a lattice sum with the assumption of a two - dimensional triangular lattice structure . the lattice constant is directly related to the dna number density @xmath81 .
figure [ figure_8 ] shows the resulting phase diagram with the coexisting dna densities for the wide range of @xmath49 . at low @xmath49
there is a strong first - order gas - crystal phase transition whichs coexistence region is widened for larger @xmath49 due to the increasing attractions . between the two threshold concentrations @xmath90 mm and @xmath91 mm
there is enough attraction to stabilize a liquid phase of high dna density . at higher spe concentrations a second crystal , with a considerable larger lattice constant than that of the high - density solid emerges .
we call this novel phase a mesocrystal since its density is intermediate between that of the fluid and the almost closed - packed solid
. condensation ( see the cross in figure [ figure_8 ] ) and subsequent redissolution ( dot - dashed line in figure [ figure_8 ] ) of dilute dna solution are other implications of phase diagram . as the spermine concentration increases for fixed @xmath92 , which corresponds to a typical dna concentration of 1mg / ml dna , first the gas - liquid coexistence line is encountered .
this is the condensation transition and the system will split into a low density gas and a high density liquid phase . at
much higher @xmath49 the coexistence line is touched again and the system redissolves back into the dilute gas phase .
the corresponding threshold concentrations of the condensation and redissolution are in the range @xmath93 mm and @xmath94 mm and agree well with the experimental observations @xcite .
one issue we want to discuss is the dependence of the phase diagram on the dna length @xmath78 .
since @xmath78 is a prefactor of the effective potential , it plays formally the role of an inverse system temperature .
we have explored the phase behavior for smaller and larger dna segment lengths of @xmath95 and @xmath96 respectively .
firstly , the stability of liquid pocket depends sensitively on @xmath78 ; it disappears completely for small @xmath78 , but extends towards larger @xmath49 for larger @xmath78 .
secondly , the fluid coexistence density shifts to considerably higher values for smaller @xmath78 .
hence , condensation and redissolution is prohibited for small dna - segment lengths .
this is in line with the experiments of ref .
@xcite , where a threshold value of @xmath97 for the minimal length @xmath78 required for condensation is reported . for a triplex dna ( three stranded dna molecule ) the minimal length
is reported to be about 2@xmath6 @xcite .
the difference between the minimal lengths for duplex and triplex dna segments arises from the dna charge density @xcite .
the higher the linear charge density , the stronger the overcharging .
this in turn results in the precipitation of triplex dna at spermine concentrations , where duplex dna does not aggregate @xcite .
the dna fragments considered here have a length less than the persistence length of the dna molecule which is around 500 .
that is why one can safely adopt a rigid rod assumption for the dna molecule and avoid the intramolecular fluctuations ( off the long dna axis ) .
this is also in line with the experiments [ 11 - 16 ] where the dna redissolution measurements were done for dna segments which are smaller than the persistence length .
hence the particles involved are very stiff rods .
we speculate that the height of the energetical barrier between the two minima in the interaction potential @xmath4 could regulate the dna segment lengths in the crystalline structures observed in experiments .
imagine that the solution consists of a mixture of dna segments of different lengths ( but still smaller than the persistence length 500 ) . at lower spermine concentration ,
when the interaction potential has a single minimum , all dna segments will form a bundle with an average inter - particle distance of the order of @xmath98 . however ,
when the energetical barrier develops at intermediate spermine concentrations with a height around @xmath99 per pinch length , only short dna segment could overcome this repulsion .
thus , short segments will fall into first minimum and form a dense hexagonal structure , while longer dna segments will be trapped in the second minimum and form more swollen , fluid - like structure , apparently a cholesteric phase .
the other issue is the orientational entropy of dna molecules and its influence on the interaction potential and free energy of dna solutions .
the evaluation of the contribution to the free energy from the orientational entropy of two nonparallel dna rods ( intermolecular fluctuations ) in md simulations is a tremendous task .
in fact , the dna - dna effective potential must be averaged over all possible orientations for two dna .
unfortunately , an implementation of tilted dna molecules will break the system symmetry in simulations . on the other hand , simulation of 20p long molecular segments for the system parameters invoked here still is a challenge . for the bundle phase , due to the deep attraction ( more than 40 @xmath100 ) between the dna rods of 20p length ,
all other than parallel configurations of dna molecules have little statistical weight .
thus , only a tiny correction to the energy of bundle is expected . for the dilute gas phase the orientational entropy for dna rod has an upper limit of about @xmath101 . at lower dna densities ,
when the average dna - dna distance is of the order or less than the dna length , this entropy should be added to the free energy of liquid phase . on the length - scale of free energy , where both the liquid perturbation and lattice sum produced curves have deep minima about dozen @xmath100 at intermediate dna densities , such a upward shift of liquid free energy at smaller dna densities will have only a slight effect on the phase diagram for 20p long dna fragments . in other words , a corresponding upward shift of the free energy of
dilute dna system will not strongly affect the coexistence spermine concentration values deduced from the double - tangent procedure .
hence , a length of 20p for the dna segment is enough to safely neglect orientational entropy effects in the phase diagram shown in fig.10 . in conclusion ,
we have calculated the influence of tetravalent counterions on the effective interactions and the phase diagram of columnar dna assemblies by primitive - model - type computer simulations and statistical theories .
we find that a small concentration of tetravalent counterions induces dna condensation .
the layering of the strongly coupled tetravalent counterions on the dna strands yields an oscillatory effective interaction potential with a double - minimum structure at intermediate counterion concentrations .
this explains the redissolution transition and triggers a novel stable mesosolid .
our threshold concentrations are in good agreement with experimental data .
this paper is dedicated to lothar schfer on the occasion of his sixtieth birthday .
we acknowledge a partial support of this work by the european networks of excellence softcomp.
40 cohen s s 1998 , _ a guide to polyamines _ ( new - york : oxford university press ) bloomfield v a 1997 it biopolymers * 44 * 269 saminathan m , thomas t , shirahata a , pillai c k s and thomas t j 2002 it nucleic acids research * 30 * 3722 cason a l 2003 _
european j. of human genetics _ * 1 * 8 wilson r w and bloomfield v a 1979 _ biochemistry _ * 18 * 2192 ; wilson r w , rau d c and bloomfiled v a 1980 _ biophys .
j. _ * 30 * 317 deng h , bloomfeld v a , benevides j m and thomas jr g j 2000 _ nucleic acids research _ * 28 * 3379 braunlin w h , anderson c f and record jr .
m t 1986 _ biopolymers _ * 25 * 205 ; braunlin w h and xu q 1992 _ biopolymers _ * 32 * 1703 olvera de la cruz m , belloni l , delsanti m , dalbiez j p , spalla o and drifford m 1995 _ j. chem
. phys . _ * 103 * 5781 raspaud e , chaperon i , leforestier a and livolant f 1999 _ biophys . j. _ * 77 * 1547 ; raspaud e , olvera m , de la cruz m o , sikorav j - l and livolant f 1998 _ biophys
. j. _ * 74 * 381 stilck j f , levin y and arenzon j j 2002 _ j. statistical physics _ * 106 * 287 pelta j , livolant f and sikorav j - l 1996 _ j. biological chemistry _ * 271 * 5656 ; pelta j , durand d , doucet j and livolant f 1996 _ biophys . j. _ * 71 * 48 saminathan m , antony t , shirahata a , sigal l h , thomas t and thomas t j 1999 _ biochemistry _ * 38 * 3821 solis f j and de la cruz m o 2001 _ eur .
e. _ * 4 * 143 solis f j 2002 _ j. chem .
phys . _ * 117 * 9009 de vries r 2001 _ biophys
. j. _ * 80 * 1186 murayama y , sakamaki y and sano m 2003 _ phys .
_ * 90 * 018102 nguyen t t , rouzina i and shklovskii b i 2000 _ j. chem .
phys . _ * 112 * 2562 ; nguyen t t and shklovskii b i 2001 _ j. chem .
phys . _ * 115 * 7298 bloomfield v a 1996 _ curr .
biol . _ * 6 * 334 burak y , ariel g and andelman d 2003 _ biophys
. j. _ * 85 * 2100 arscott p g , li a - z and bloomfield v a 1990 _ biopolymers _ * 29 * 619 solis f j and de la cruz m o 2000 _ j. chem .
* 112 * 2030 trubetskoy v s , wolff j a and budker v g 2003 _ biophys . j. _ * 84 * 1124 .
tanaka m and grosberg a yu 2001 _ j. chem .
phys . _ * 115 * 567 montoro j c g and abascal j l f 1995 _ j. chem .
phys . _ * 103 * 8273 ; 1998 _ j. chem .
phys . _ * 109 * 6200 ; abascal j l f and montoro j c g 2001 _ j. chem .
phys . _ * 114 * 4277 allahyarov e , lwen h and gompper g 2003 _ phys . rev
. e. _ * 68 * 061903 allahyarov e and lwen h 2000 _ phys .
e _ * 62 * 5542 allahyarov e , gompper g and lwen h 2004 _ phys .
e. _ * 69 * 041904 allahyarov e , lwen h and gompper g 2004 _ euro . phys
. letters _ * 68 * 894 manning g s 1978 _ q. rev .
_ * 11 * 179 strey h h , parsegian v a and podgornik r 1999 _ phys . rev
59 * 999 kornyshev a a and leikin s 2000 _ phys .
lett . _ * 84 * 2537 durand d , doucet j and livolant f 1992 _ j. phys
. ii _ * 2 * 1769 podgornik r , strey h h and parsegian v a 1998 _ curr
colloid interface science _ * 3 * 534 allahyarov e , damico i and lwen h 1998 _ phys .
_ * 81 * 1334 .
weeks j d , chandler d and andersen h c 1971 _ j. chem .
phys . _ * 54 * 5237 evans r 1990 _ liquids at interfaces _
( amsterdam : north - holland ) santos a , lopez de haro m and bravo yuste s 1995 _ j. chem .
_ * 103 * 4622 goobes r , cohen o and minsly a 2002 _ nucleic acids research _ * 30 * 2154 deserno m , jimenes - angeles f , holm c and lozada - cassou m 2003 _ cond - mat _ 0104002 .
deserno m , jimenez - angeles f , holm c and lozada - cassou m 2001 _ j. phys .
b _ * 105 * 10983 | the effective dna - dna interaction force is calculated by computer simulations with explicit tetravalent counterions and monovalent salt . for overcharged dna molecules , the interaction force shows a double - minimum structure .
the positions and depths of these minima are regulated by the counterion density in the bulk . using two - dimensional lattice sum and free energy perturbation theories ,
the coexisting phases for dna bundles are calculated .
a dna - condensation and redissolution transition and a stable mesocrystal with an intermediate lattice constant for high counterion concentration are obtained . |
dynamical systems subject to random abrupt changes , such as manufacturing systems , networked control systems , economics and finance _ etc_. , can be modelled adequately by random jump linear systems ( rjlss ) .
rjlss are a particular kind of stochastic switching systems , which consists of a set of linear systems , also called multiple modes , and the switching among them is governed by a random jump process .
a notable class of rjlss is markov jump linear system ( mjls ) in which the underlying random jump process is a finite state markov chain ( or a finite state markov process ) .
many important results related to stability , control , and applications of such systems have been investigated in the literature , for instance in @xcite , @xcite , @xcite , @xcite , @xcite , @xcite etc .
almost all the works related to mjlss assume that the underlying random jump process is time - homogeneous / time - inhomogeneous markov , which is a restrictive assumption . in this article
, we deal with a class of rjlss in which the evolutions of the random jump process depends on the state variable , and are referred to as _ state - dependent jump linear systems _
( sdjls ) . in the following ,
we list some motivations for sdjls modelling of dynamical systems . in the analysis of random breakdown of components , the age , wear , and accumulated stress of a component affect its failure rate , for instance .
thus , it can be assumed that the failure rate of a component is dependent on state of the component at age @xmath0 @xcite , where the state variable may be an amount of wear , stress etc . as an another instance , in @xcite , a state - dependent markov process was utilized to describe the random break down of cylinder lines in a heavy - duty marine diesel engine .
also , we can examine a stock market with situations : up and down , and the transitions between the situations can be dependent on state of the market , where the state variable may be general mood of investors and current economy etc . also , a state - dependent regime switching model was considered in @xcite to model financial time series .
one can find other instances or examples of sdjls modelling in the literature .
the studies of stability and control of sdjlss have been scanty in the literature .
a study of hybrid switching diffusion processes , a kind of continuous - time state - dependent jump non - linear systems with diffusion , was considered in @xcite by treating existence , uniqueness , stability of the solutions etc . for rjlss , a state and control dependent random jump process was considered in @xcite , where the authors used stochastic maximum principle to obtain optimal control for a given objective function .
sdjls modelling of flexible manufacturing system was proposed in @xcite , and dynamic programming is used to obtain an optimal input which minimizes the mentioned cost .
a state - dependent jump diffusion modelling of a production plant was considered in @xcite to obtain an optimal control . in the sequel
, we bring back the attention to the main ingredients of the problem that we address in this article . in this article
, we consider that the sdjls is affected by possibly unbounded stochastic disturbances , and the perfect state information is assumed .
we also deal with constraints on different variables of the system that is inherent in all practical systems .
model predictive control ( mpc ) , also called receding horizon control ( rhc ) , is an effective control algorithm that has a great potential to handle input and/or state constraints , for problems across different disciplines .
mpc is a form of control in which the current input is obtained at each sampling time by solving on - line a finite horizon optimal control problem in the presence of constraints , using the current information and the predicted information over the finite horizon .
normally more than one input is obtained at the current sampling time , however , only the first controller input will be implemented to the plant . at the next sampling time , these actions will be repeated , that is why the mpc is also called the rhc .
one can refer to @xcite , @xcite , @xcite , @xcite etc .
, for classic contributions in the mpc literature . in the context of rjlss , of late , the rhc scheme has been extended to discrete - time mjlss . for discrete - time mjlss ,
the rhc with hard symmetric constraints and bounded uncertainties in system parameters was dealt by @xcite , @xcite , where the constraints and the objective function were posed as sufficient conditions in terms of linear matrix inequalities ( lmis ) to be solved at each sampling time ; for the similar case without constraints , the rhc was addressed by @xcite following the similar approach .
also , for unconstrained state and control input , optimality of the rhc was addressed via dynamic programming @xcite , variational methods @xcite , solving riccati equations @xcite , etc .
one major issue in the presence of unbounded disturbances is that the rhc can not necessarily guarantee the satisfaction of constraints .
for instance , an additive unbounded disturbance eventually drives the state variable outside any bounded limits , no matter how arbitrarily large they may be .
a possible alternative is to consider the satisfaction of constraints in stochastic manner , which allow occasional constraint violations . in this direction , recent approaches @xcite , @xcite , @xcite , and the references therein treat the rhc of discrete - time linear systems with stochastic constraints . for discrete - time mjlss , in case of perfect state availability , a linear quadratic regulator problem with second moment constraints was considered in @xcite , where the entire problem was converted to a set of lmis .
however , to the best of the authors knowledge , the rhc of discrete - time sdjlss with probabilistic constraints has not been examined . in this article , we address a one - step rhc of discrete - time sdjlss with additive gaussian process noise ( its distribution has an unbounded support ) , and probabilistic state constraints under perfect state availability .
we would like to highlight several challenges in our problem set - up .
first , in the presence of additive process noise with unbounded support , it impossible to guarantee hard bounds on the state , and also on the linear state - feedback control .
second , one needs to pre - stabilize the system before addressing the rhc problem .
third , one needs to obtain a tractable representation of the rhc problem in the presence of probabilistic constraints
. our approach along with main contributions in this article can be listed as follows . in our problem set - up , we consider the control to have a linear state - feedback and an offset term @xcite , @xcite , where linear state - feedback gains are computed off - line for pre - stabilization and admissible offset terms are computed on - line to solve the rhc problem . in the presence of unbounded process noise , it is not possible to ensure hard bounds on the state and the control variables that follow state - feedback law , thus we consider the state variable to be probabilistically constrained and the control to be unconstrained , except for the offset term
. using inverse cumulative distribution function , we convert the probabilistic state constraints to deterministic constraints and the overall rhc problem is replaced by a tractable deterministic rhc problem .
to summarize , for sdjls subject to possibly unbounded random disturbances and probabilistic state constraints , our contributions in this article are : * pre - stabilizing the system state by a state - feedback controller in means square sense , * implementing a one - step rhc scheme on - line with probabilistic constraints on the state variable , which are converted to deterministic constraints . for illustration , we apply our approach to a macroeconomic situation .
the article is organized as follows .
section [ sec : mathematical - model ] presents the problem setup .
we present the pre - stabilization of the system by a state feed - back controller in section [ sec : pre - stabilization ] .
we convert the probabilistic constraints to suitable deterministic constraints in section [ sec : prob_constraints ] .
we give a one - step rhc scheme with probabilistic constraints in section [ sec : one - step - rhc - with ] .
section [ sec : examples ] presents an illustrative example followed by conclusions in section [ sec : conclusions ] . finally , we give majority of the proofs in the appendix to improve readability . _
notation : _ let @xmath1 denotes the @xmath2-dimensional real euclidean space and @xmath3 the set of non - negative integers . for a matrix @xmath4 , @xmath5 denotes the transpose , @xmath6 @xmath7 the minimum ( maximum ) eigenvalue and @xmath8 the trace of @xmath4 . the standard vector norm in @xmath1 is denoted by @xmath9 the corresponding induced norm of a matrix @xmath4 by @xmath10 . given a matrix @xmath11 , @xmath12 ( or @xmath13 ) denotes that the matrix @xmath11 is positive definite ( or negative definite ) . given two matrices @xmath11 and @xmath14 , @xmath15 ( or @xmath16 )
denotes the element wise inequalities .
symmetric terms in block matrices are denoted by @xmath17 a matrix product @xmath18 denotes @xmath19 the identity matrix of dimension @xmath20 is denoted by @xmath21 .
the diagonal matrix formed from its vector arguments is denoted by diag@xmath22 .
the underlying probability space is denoted by @xmath23 where @xmath24 is the space of elementary events , @xmath25 is a @xmath26-algebra , and @xmath27 is the probability measure . the mathematical expectation of a random variable @xmath28 is denoted by @xmath29 $ ] .
further notation will be introduced when required .
consider a discrete - time sdjls : @xmath30 with the state @xmath31 , the control input @xmath32 , the system mode @xmath33 .
let @xmath34 @xmath35 be the system matrices of appropriate dimensions that are assumed to be known , and without loss of generality we assume that @xmath36 are identity matrices .
perfect state information and the mode availability is assumed at each time @xmath37 .
we set @xmath38 the @xmath26-algebra generated by @xmath39 . regarding the different stochastic processes ,
the following assumptions are made : + * assumption 1 .
* let @xmath40 be an @xmath41dimensional standard gaussian distributed random vector i.e. @xmath42 , independent of @xmath43 .
+ * assumption 2 .
* let the mode @xmath44 be a finite - state random jump process described by @xmath45 where @xmath46 is the transition probability from a state @xmath47 to a state @xmath48 , and @xmath49 .
here we further assume that @xmath50 , @xmath51 .
+ if we observe the sdjls ( [ sys : djls_noise ] ) , the actual state consists of two parts : the state @xmath52 and the mode @xmath33 .
if the mode @xmath33 is fixed for all @xmath37 , then the sdjls ( [ sys : djls_noise ] ) is equivalent to the standard linear time - invariant system .
thus the role of the mode @xmath33 is important in the sdjlss , and in general in any switching systems . with this background
, we observe the mode dependence through out the article : in designing state - feedback gains , in handling probabilistic constraints and in weighting matrices in the objective function . the sets @xmath53 and @xmath54 in ( [ eq : p_tran ] )
are described explicitly as follows .
we consider the set @xmath53 as a polyhedral intersection of finite constraints on the state variable and @xmath54 as @xmath55 .
so , in this article , the states of the system that belong to @xmath53 are specifically denoted by @xmath56 where @xmath57 and @xmath58 .
we can regard @xmath59 as a desirable set , and ideally we want the system state to be constrained in @xmath59 .
however , the state variable being affected by gaussian noise that is unbounded in distribution , it is impossible to constrain the state to the set @xmath59 at all times .
thus , we consider probabilistic state constraints , and in section [ sec : prob_constraints ] , we describe in detail the type of probabilistic state constraints that we deal in this article . in our rhc problem
, we address the minimization of the following one - step objective function @xmath60 \label{obj : func}\end{aligned}\ ] ] subject to probabilistic state constraints and the control input that follows an affine state - feedback law . here
@xmath61 , @xmath62 @xmath63 are assumed to be known for @xmath64 in this formulation , the terminal cost depend on the matrices @xmath65 and the state of the system at time @xmath66 .
the choice of one - step prediction in ( [ obj : func ] ) is explained in detail in section [ sec : one - step - rhc - with ] in perspective of the system and the constraints .
we parametrize the control input as ( inspired by @xcite , @xcite ) @xmath67 where feed - back gains @xmath68 for @xmath69 , will be designed to pre - stabilize the system in mean square sense that will be given the next section .
an on - line minimization of a @xmath70 in ( [ obj : func ] ) subject to constraints will be carried out by an appropriate @xmath71 .
+ notice that the state - feedback control input ( [ eq : stabilization_input ] ) is not bounded a priori .
the reason is as follows . from ( [ sys : djls_noise ] ) , the state @xmath52 in ( [ eq : stabilization_input ] ) is a function of gaussian noise .
since gaussian distribution has an unbounded support , we can not assume the a priori boundedness of @xmath72 ( [ eq : stabilization_input ] ) .
thus , we just consider @xmath73 to be bounded to obtain a tractable solution of minimization of @xmath70 in ( [ obj : func ] ) subject to constraints .
thus , we give an explicit assumption on @xmath73 in the following .
+ * assumption 3 . *
[ assume : input_compact ] we assume that @xmath71 belong to a compact set @xmath74 .
+ [ remark : input_measurable]note that @xmath71 is obtained by solving @xmath75 on - line that will be introduced in section [ sec : one - step - rhc - with ] .
it will be observed that the computation of @xmath71 relies on the available information @xmath76 , for each @xmath37 .
thus @xmath73 can be regarded as @xmath38-measurable random variable taking values in @xmath77 . + substituting ( [ eq : stabilization_input ] ) into ( [ sys : djls_noise ] )
, we obtain the closed loop system as @xmath78 with @xmath79 in this article , we consider the linear state - feedback control input ( [ eq : stabilization_input ] ) for pre - stabilization , where we could not ensure hard bounds on the control because of the stochastic unbounded disturbance . to incorporate bounded controls in the class of causal feedback policies
, one alternative could be employing an affine disturbance feedback control with non - linear saturation of noise terms @xcite , @xcite . in this approach ,
the disturbance implicitly depends on the output , hence on the state in a causal manner .
however , in this case the system under consideration needs to be internally stable . in the next section ,
we synthesize the state - feedback gains @xmath80 in ( [ eq : stabilization_input ] ) , which pre - stabilize ( [ sys : stabilized_input ] ) in bounding the state - trajectories in mean square sense .
in this section , we pre - stabilize the system ( [ sys : stabilized_input ] ) with state - feedback gains @xmath80 by solving sufficient lmis off - line . in general , in the presence of unbounded disturbances , it is difficult to guarantee or establish stability .
it is impossible to expect asymptotic stability to origin .
however , we synthesize a state - feedback controller that bound the second moment of the state - trajectories .
the results of this section rely on a stochastic version of lyapunov s second method @xcite , @xcite by the choice of mode - dependent lyapunov function @xmath81 with @xmath82 . in @xcite
, the authors studied the stochastic stability of discrete - time mjlss subject to uncertainties in system parameters , and we extend these results to examine the stochastic stability of sdjlss ( [ sys : stabilized_input ] ) , where the underlying random jump process is state - dependent , which is the theme of our approach .
we begin with the following definitions .
+ let @xmath83 be a linear subspace of real matrices .
set @xmath84 . also , let @xmath85 be a linear subspace of real symmetric matrices
. set @xmath86 .
+ we give a proposition to pre - stabilize the system ( [ sys : stabilized_input ] ) in the following .
[ theorem : msqs ] if there exist matrices @xmath87 , @xmath88 , with @xmath89 , @xmath90 such that @xmath91 and @xmath92 then the trajectories of ( [ sys : stabilized_input ] ) satisfy the estimates : @xmath93\leq\beta_{1}\vert x_{0}\vert^{2}q^{k}+\beta_{2},\;\forall\ ; k\in \mathbb{n}_{\ge 0},\ ] ] _ for some @xmath94 , @xmath95 and @xmath96 .
_ given in appendix a. in the sequel , we provide a proposition to synthesize gains @xmath97 for @xmath98 [ theorem : stabilization_ss ] if there exist matrices @xmath99 and @xmath100 , @xmath90 , such that @xmath101 _ and _ @xmath102 _ where _ @xmath103 _ then the trajectories of the system _ ( [ sys : stabilized_input ] ) _ satisfy _ ( [ eq : msqs ] ) , _ and @xmath104 is given by
_ @xmath105 briefly , we explain the steps to arrive at the result .
consider the schur complement of ( [ lmi1 ] ) , ( [ lmi2 ] ) , and let @xmath106 . by the congruent transformation of schur complement of ( [ lmi1 ] ) , ( [ lmi2 ] ) by diag@xmath107 , and applying a change of variable @xmath108 one obtains ( [ lmi : stabiization1 ] ) , ( [ lmi : stabiization2 ] ) . hence ,
if ( [ lmi : stabiization1 ] ) , ( [ lmi : stabiization2 ] ) are feasible and using proposition [ theorem : msqs ] , it follows that the trajectories of ( [ sys : stabilized_input ] ) satisfy the estimates ( [ eq : msqs ] ) .
hence the proof is complete .
though the conditions to obtain the state - feedback gains @xmath109 , for @xmath110 , in proposition [ theorem : stabilization_ss ] seems computationally heavy , we perform these computations off - line , and only once .
once we obtain the state - feedback gains @xmath109 , we use them as a lookup table , where depending on the mode @xmath33 at each time @xmath111 , the corresponding value of the state - feedback gain is used in updating the state in ( [ sys : stabilized_input ] ) .
in this section , we consider the probabilistic state constraints that we address in this article , and replace them by deterministic constraints .
we mentioned in section [ sec : mathematical - model ] , that it is impossible to constrain the state variable to the set @xmath112 which is polyhedron @xmath113 , where @xmath57 and @xmath58 .
rather , let us consider probabilistic constraint @xmath114 which portrays the probabilistic scenario of staying in the polyhedron . here
@xmath115 represents the level of constraint satisfaction , which needs to be high enough ( @xmath116 ) .
however , without loss of generality , we assume that @xmath115 may take any values in the interval @xmath117 $ ] . in the sequel , we try to obtain deterministic constraints that imply the probabilistic constraint ( [ constraint : multivariate ] ) .
we say it as converting probabilistic constraint to deterministic ones .
a sufficient condition to satisfy ( [ constraint : multivariate ] ) is given by @xcite , @xmath118 now , we present a sufficient deterministic condition to satisfy ( [ constraint : multivariate_final ] ) via approximation of inscribed ellipsoidal constraints @xcite , @xcite . [
theorem : multivariate]let @xmath119 , where @xmath120 denotes chi - square inverse cumulative distribution function with a given probability @xmath115 and @xmath2 degrees of freedom .
if @xmath121 then @xmath122 , where @xmath123 and @xmath124 are @xmath125 rows of @xmath126 and @xmath127 respectively . the proof is given on the similar lines of @xcite , @xcite and outlined in appendix b. note that ( [ constraint : chi_square ] ) is an over conservative condition , because the deterministic condition ( [ constraint : chi_square ] ) implies ( [ constraint : multivariate_final ] ) , which finally implies ( [ constraint : multivariate ] ) .
even for this special case , we could not obtain an equivalent representation of ( [ constraint : multivariate_final ] ) , which is non - convex in general , where it is hard to find its feasible region @xcite .
so , alternatively , we propose the individual probabilistic constraints of type @xmath128 where @xmath129 and @xmath130 represent the @xmath125 row of @xmath126 and @xmath127 respectively . with given @xmath115 ,
the constraints ( [ constraint : univariate ] ) offer satisfaction of each individual constraint of the polyhedron probabilistically , but with more constraint violations @xcite than ( [ constraint : multivariate ] ) that will also be observed in section [ sec : examples ] .
however , we consider the individual probabilistic constraints ( [ constraint : univariate ] ) , because they are simpler to handle and in general convex @xcite , @xcite .
similar to the above treatment , a sufficient condition to satisfy ( [ constraint : univariate ] ) is given by @xmath131 by ( [ sys : stabilized_input ] ) , one obtains , @xmath132 where @xmath133 and @xmath134 are the cumulative distribution and inverse cumulative distribution of the random variable @xmath135 respectively .
the probabilistic constraints ( [ constraint : univariate ] ) result in handling with uni - variate gaussian random variables @xmath135 when converting to deterministic constraints ( [ constraint : det_univariate ] ) , which is straightforward . in this case , given @xmath115 , @xmath136 can easily be obtained .
observe that the conditions ( [ constraint : univariate_final ] ) and ( [ constraint : det_univariate ] ) are equivalent , which imply ( [ constraint : univariate ] ) .
the probabilistic constraints ( [ constraint : multivariate ] ) and ( [ constraint : univariate ] ) are two different ways of treating the constraints @xmath113 in a probabilistic fashion .
we consider the probabilistic constraints ( [ constraint : univariate_final ] ) because of the simplicity and low conservatism involved .
in this section , we provide a one - step rhc problem with the objective function ( [ obj : func ] ) subject to the probabilistic constraints ( [ constraint : univariate_final ] ) and the state - feedback control input ( [ eq : stabilization_input ] ) . at each @xmath37
, we consider the following one - step rhc problem @xmath137\nonumber \\ & \text{s.t.}\ ; ( \ref{sys : djls_noise } ) , ( \ref{eq : p_tran}),\nonumber\\ & \quad\;\ : \text{pr}\big\{g_{j}x_{k+1}\le h_{j}|\mathcal{f}_k\big\}\ge\xi,\ , 1\le j\le r,\,\xi\in[0,1],\label{eq : prob_constraint}\\ & \quad\;\ : u_{k}=k_{\theta_{k}}x_{k}+\nu_{k},\\ & \quad\;\:\nu_{k}\in\mathbb{u}.\label{eq : input_con}\end{aligned}\ ] ] in @xmath138 , we choose the prediction horizon to be one because of the probabilistic constraints ( [ eq : prob_constraint ] ) and the system ( [ sys : stabilized_input ] ) . in section
[ sec : prob_constraints ] , we obtained a deterministic equivalence of the constraints ( [ eq : prob_constraint ] ) in terms of the state variable and the mode at time @xmath139 that are known .
in general , the larger prediction horizon result in better performance and more computational burden depending on the system @xcite .
the choice of the prediction horizon depends on the performance requirements and computational capability .
suppose , if we consider a multi - step prediction , the probabilistic constraints in @xmath138 look like @xmath140 . by proceeding with the similar approach of section [ sec : prob_constraints ] , we can obtain an equality similar to ( [ constraint : remark_univar ] ) that contain additional unknown random variables @xmath141 , where it is not possible to obtain its deterministic equivalence .
thus , we choose a one - step prediction horizon to obtain a deterministic equivalence of the probabilistic constraints ( [ eq : prob_constraint ] ) for tractability of @xmath142 . at this point , we give a brief review of some works in the literature of rjlss with multi - step prediction . in case of perfect state availability , without process noise ,
a multi - step prediction horizon was considered for discrete - time mjlss with hard symmetric input and state constraints @xcite , @xcite , where an upper bound of the objective function was minimized and the overall problem with the constraints was converted to sufficient lmis .
the approach was based on @xcite , where the state variable was restriced to an invariant ellipsoid in the constrained space and the techniques of lmis and lyapunov method were utilized .
however , it brings a computational burden because of the additional lmis , and more importantly it reduces the feasibility region drastically depending on the invariant ellipsoid . to avoid the sufficiency
, the authors in @xcite directly considered ellipsoidal constraints ( in terms of second moment constraints ) for discrete - time mjlss with multi - step prediction , where the constraints were replaced by a set of lmis .
it can be possible that the @xmath138 become infeasible when the constraints ( [ eq : prob_constraint ] ) are tight with a given admissible input . to resolve this issue ,
the constraints can be relaxed by an additional slack variable @xmath143 as @xcite , @xmath144 where @xmath145 denotes a column vector of appropriate size , which contain all ones .
thus from ( [ constraint : det_univariate ] ) , @xmath146 the addition of a variable @xmath147 can be compensated by adding a variable @xmath148 to the objective function in @xmath138 . in particular , @xmath149 should be chosen as a very high value , which act as a penalty to discourage the use of slack variable @xcite . thus @xmath138 will be converted to @xmath150+\alpha\rho_{k}\nonumber \\ & \text{s.t.}\ ; ( \ref{sys : djls_noise } ) , ( \ref{eq : p_tran}),\nonumber\\ & \quad\;\ : g_{j}\big(\tilde{a}_{\theta_{k}}x_{k}+b_{\theta_{k}}\nu_{k}\big)\le h_{j}-f_{g_{j}w_{k}}^{-1}(\xi)+\mathbf{1}\rho_{k},\ , 1\le j\le r,\,\xi\in[0,1],\label{constraint : slack_rhc2}\\ & \quad\;\ : u_{k}=k_{\theta_{k}}x_{k}+\nu_{k},\\ & \quad\;\:\nu_{k}\in\mathbb{u } , \\ & \quad\;\:\rho_{k}\ge0.\label{eq : epsilon}\end{aligned}\ ] ] notice that due to ( [ constraint : slack_rhc2 ] ) , ( [ eq : epsilon ] ) , @xmath151 is always feasible .
the main task is to minimize the objective function in @xmath151 by a proper choice of @xmath152 to accomplish this task , we present @xmath151 in a tractable fashion in the following . in order to write the @xmath151 in terms of tractable terms , consider the term @xmath153 $ ] . from ( [ sys : stabilized_input ] ) , @xmath154 is a random vector with @xmath155=\tilde{a}_{\theta_{k}}x_{k}+b_{\theta_{k}}\nu_{k},\ ] ] and @xmath156=\mathbb{e}\left[w_{k}w_{k}^{t}\right]=\mathbb{i}_{n}.\end{aligned}\ ] ] and now , let us consider the following lemma which is a classic result , [ lem : lemma1 ] let @xmath157 be a random vector with mean @xmath158 and covariance matrix @xmath159 , and let @xmath160 be a given matrix of suitable dimensions .
then @xmath161=m^{t}km+\mathrm{tr}(kr).\label{eq : lemm1}\ ] ] the result ( [ eq : lemm1 ] ) and the law of total probability gives that @xmath162=\left(\tilde{a}_{\theta_{k}}x_{k}+b_{\theta_{k}}\nu_{k}\right)^{t}\pi_{\theta_{k}}\star+\text{tr}\left(\pi_{\theta_{k}}\right),\end{aligned } \label{eq : exp_noise}\ ] ] where @xmath163 by ( [ eq : exp_noise ] ) , the objective function in @xmath151 can be expressed as @xmath164 thus , the problem @xmath151 is equivalent to @xmath165,\label{constraint : feasibility}\\ & \quad\;\ : u_{k}=k_{\theta_{k}}x_{k}+\nu_{k},\\ & \quad\;\:\nu_{k}\in\mathbb{u},\\ & \quad\;\:\rho_{k}\ge 0 \label{eq : eps_1}.\end{aligned}\ ] ] we denote @xmath166 as an optimal solution of @xmath167 . to solve @xmath167 ,
which is a quadratic programming ( qp ) problem , we use the default qp solver in matlab .
when we solve @xmath167 , if @xmath147 is zero , for all @xmath168 , then we have solved original @xmath138 , and the solution of @xmath167 would be equivalent to the solution of @xmath138 . if @xmath147 is positive , then we conclude that @xmath138 was infeasible , and the solution of @xmath167 would approximate the solution of @xmath138 .
this approximation depends on the magnitude of @xmath147 , which need to be lower for better approximation .
in this section , we illustrate the proposed approach on a dynamics of macroeconomic system . we consider a macroeconomic system based on samuelson s multiplier accelerator model that has been considered by several authors @xcite , @xcite , @xcite , @xcite , @xcite etc . , to illustrate the respective results on discrete - time mjlss . in these works ,
basically , the discrete - time mjls modelling of the economic system was considered to describe the relation between the government expenditure ( control input ) and the national income ( system state ) with different economic situations ( modes ) and the switching among the modes was modelled as a homogeneous markov chain ; an interested reader can refer to the above references for the detailed mathematical model .
in all these references , the authors considered that the transitions among the economic situations follow a time homogeneous markov chain .
a more realistic scenario is to consider that the transitions among the economic situations depend on the state of the economic system .
an economic system subject to fluctuations with state - dependent economic situation can be given by the sdjls ( [ sys : djls_noise ] ) , where @xmath169^t\in \mathbb{r}^2 $ ] with @xmath170^{t}$ ] .
here @xmath171 denotes the national income , @xmath172 denotes the one unit time delayed version of @xmath171 , @xmath173 refers to the government expenditure .
let @xmath174 , which represents the fluctuations in the national income .
let @xmath175 represent modes of the economic system under three situations : @xmath176 ( normal ) , @xmath177 ( boom ) , @xmath178 ( slump ) .
the parameters of the mode matrices are : @xmath179,a_{2}=\left[\begin{array}{cc } 0 & 1\\ -4.3 & 4.5 \end{array}\right],a_{3}=\left[\begin{array}{cc } 0 & 1\\ 5.3 & -5.2 \end{array}\right],\ ] ] @xmath180^{t}$ ] .
the state - dependent transitions of the mode @xmath181 are given by ( [ eq : p_tran ] ) , where @xmath182,\mu=\left[\begin{array}{ccc } 0.67 & 0.17 & 0.16\\ 0.30 & 0.47 & 0.23\\ 0.26 & 0.10 & 0.64 \end{array}\right ] , \ ] ] @xmath183,\ , h\equiv h_{k}=\left[\begin{array}{c } -2 - 0.5k\\ 5 + 0.5k \end{array}\right].\ ] ] observe that when the economy ( the state @xmath52 ) satisfies @xmath184 , the transitions among the modes follow @xmath185 , where the transition probabilities to boom are higher , and also the transition probabilities to slump are lower compared to @xmath186 .
thus , we observe the transitions among the economic situations depend on the state of the economy .
consider the state - feedback control input ( [ eq : stabilization_input ] ) with @xmath187 .
for pre - stabilization , we consider proposition [ theorem : stabilization_ss ] that can be satisfied with @xmath188,x_{2}=\left[\begin{array}{cc } 1.9255 & 0\\ 0 & 0.7628 \end{array}\right],x_{3}=\left[\begin{array}{cc } 1.1393 & 0\\ 0 & 0.1044 \end{array}\right],\ ] ] @xmath189,y_{2}=\left[\begin{array}{cc } 8.2797 & -3.4325\\ \end{array}\right],y_{3}=\left[\begin{array}{cc } -6.0384 & 0.5429 \\ \end{array}\right].\ ] ] thus the system is pre - stabilized in the sense of ( [ eq : msqs ] ) , and the state - feedback gains are given by @xmath190,k_{2}=\left[\begin{array}{cc } 4.3 & -4.5\\ \end{array}\right],k_{3}=\left[\begin{array}{cc } -5.3 & 5.2 \\ \end{array}\right].\ ] ] we assume the probabilistic state constraints ( [ constraint : univariate_final ] ) with @xmath191 as a pre - specified monitory target policy at each @xmath139 .
it means that the income ( the state ) is required to meet the target ( the range between the red lines given in figure [ fig : ex_statemc ] and figure [ fig : eco_income_compare ] with a probability @xmath191 , which denotes the level of constraint satisfaction .
we consider the following parameters for the one - step rhc problem @xmath167 : @xmath192 @xmath193 @xmath194 @xmath195 @xmath196 @xmath197 @xmath198 @xmath199 @xmath200 and @xmath201 using the parameters , we solve @xmath167 , at each time @xmath139 . with these parameters
we obtain @xmath202 as zero for all @xmath139 , which imply that the original problem @xmath138 was feasible and the solution of @xmath167 is equivalent to the solution of @xmath138 .
we consider the planning for 5 years of duration , where each unit time represents the period of three months .
a sample mode ( @xmath181 ) evolution is given in figure [ fig : ex_mode ] , which shows a sample evolution of an economic situation ( normal , bloom or slump ) .
a corresponding optimal cost @xmath166 is shown in figure [ fig : ex_input ] .
suppose , if the values of @xmath202 are not zero and larger ( @xmath203 ) to make the constraints ( [ constraint : slack_final ] ) feasible , then this would make the values of @xmath204 , because all the remaining terms of the objective function in @xmath75 are positive . ] ] to observe the probabilistic state constraints ( [ constraint : univariate_final ] ) qualitatively , we performed monte carlo simulations for 50 runs and the incomes @xmath205 are shown in figure [ fig : ex_statemc ] .
one can observe occasional constraint violations , because we consider the constraint satisfaction probabilistically . to compare the probabilistic constraints ( [ constraint : univariate_final ] ) , ( [ constraint : multivariate_final ] )
, we also solved @xmath75 with the joint probabilistic constraints ( [ constraint : multivariate_final ] ) and the incomes @xmath205 are shown in figure [ fig : ex_state_joint ] that are obtained via monte carlo simulations . from figure
[ fig : ex_statemc ] and [ fig : ex_state_joint ] , one can observe more constraint violations with individual probabilistic constraints ( [ constraint : univariate_final ] ) than ( [ constraint : multivariate_final ] ) , which is stated in section [ sec : prob_constraints ] . in figure
[ fig : ex_state_joint ] , we obtain @xmath206 , for some @xmath139 , which is due to smaller feasibility region of ( [ constraint : chi_square ] ) compared to ( [ constraint : det_univariate ] ) that can be observed by comparing the right hand sides of ( [ constraint : chi_square ] ) and ( [ constraint : det_univariate ] ) .
.5 .5 to compare the probabilistic state constraints ( [ constraint : univariate_final ] ) with different values of @xmath115 qualitatively , we perform monte carlo simulations of 50 runs for values of @xmath115 ( 0.95 , 0.5 , 0.3 ) separately by keeping the remaining parameters same as above , and the incomes are shown in figure [ fig : eco_income_compare ] .
observe that the larger the level of constraint satisfaction @xmath115 , the better the incomes meet the required target .
for this experiment , we obtained @xmath147 as zero , for all @xmath139 , when @xmath115 is 0.5 and 0.3 ; @xmath207 for some @xmath139 , when @xmath115 is 0.95 .
it implies that the higher the level of constraint satisfaction @xmath115 , the more the chance of becoming the original @xmath142 infeasible .
this can also be observed from ( [ constraint : det_univariate ] ) , where the larger values of @xmath115 makes the right hand side of ( [ constraint : det_univariate ] ) smaller , thus reducing its feasibility region .
, width=340,height=226 ]
we considered a receding horizon control of discrete - time state - dependent jump linear systems subject to additive stochastic unbounded disturbance with probabilistic state constraints .
we used an affine state - feedback control , and synthesized feedback gains that guarantee the mean square boundedness of the system by solving a linear matrix inequality problem off - line .
we obtained sufficient deterministic conditions to satisfy probabilistic constraints by utilizing inverse cumulative distribution function , and consequently converted the overall receding horizon problem as a tractable deterministic optimization problem .
although , it is difficult to guarantee the recursive feasibility in the case of stochastic unbounded disturbance , we attempted to resolve this issue with an addition of slack variable to the obtained constraints with a penalty in the objective function .
we performed simulations a macroeconomic system to verify the proposed methodology .
[ sec : appendix ]
consider the given system @xmath208 where @xmath181 is defined in ( [ eq : p_tran ] ) and @xmath79 from remark [ remark : input_measurable ] , @xmath209 is a stochastic process with the property that for each @xmath37 , the vector @xmath71 is @xmath210-measurable and @xmath211\leq\delta<\infty$ ] .
+ one knows that there exist matrices @xmath212 and @xmath213 , with @xmath89 , for @xmath90 such that ( [ lmi1 ] ) and ( [ lmi2 ] ) are verified .
consider @xmath214 .
let @xmath215 , @xmath216 , and @xmath217 .
one has @xmath218-v(x_{k},\theta_{k } ) \\ & = x_{k}^{t}\left(\tilde{a}_{\theta_{k}}^{t}\upsilon_{(\theta_{k},x_{k})}\tilde{a}_{\theta_{k}}-p_{\theta_{k}}\right)x_{k}+2x_{k}^{t}\tilde{a}_{\theta_{k}}^{t}\upsilon_{(\theta_{k},x_{k})}\mathbb{e}\left[w_{k}|{\mathcal{f}_{k}}\right ] + 2\nu_{k}^{t}b_{\theta_{k}}^{t}\upsilon_{(\theta_{k},x_{k})}\tilde{a}_{\theta_{k}}x_{k}\\ & \quad + 2\nu_{k}^{t}b_{\theta_{k}}^{t}\upsilon_{(\theta_{k},x_{k})}\mathbb{e}\left[w_{k}|{\mathcal{f}_{k}}\right]+\nu_{k}^{t}b_{\theta_{k}}^{t}\upsilon_{(\theta_{k},x_{k})}b_{\theta_{k}}\nu_{k}+\mathbb{e}\left[w_{k}^{t}\upsilon_{(\theta_{k},x_{k})}w_{k}|{\mathcal{f}_{k}}\right],\\ & \leq-\mu\vert x_{k}\vert^{2}+2x_{k}^{t}\tilde{a}_{\theta_{k}}^{t}\upsilon_{(\theta_{k},x_{k})}\mathbb{e}\left[w_{k}|{\mathcal{f}_{k}}\right ] + 2\nu_{k}^{t}b_{\theta_{k}}^{t}\upsilon_{(\theta_{k},x_{k})}\tilde{a}_{\theta_{k}}x_{k}+2\nu_{k}^{t}b_{\theta_{k}}^{t}\upsilon_{(\theta_{k},x_{k})}\mathbb{e}\left[w_{k}|{\mathcal{f}_{k}}\right]\\ & \quad + \alpha_{1}\vert\nu_{k}\vert^{2}+\alpha_{2}\mathbb{e}\left[w_{k}^{t}w_{k}|{\mathcal{f}_{k}}\right],\end{aligned}\ ] ] where @xmath219 @xmath220 and @xmath221 . because the random vector @xmath222 is independent of @xmath210 we obtain @xmath218-v(x_{k},\theta_{k})\\ & \leq-\mu\vert x_{k}\vert^{2}+2x_{k}^{t}\tilde{a}_{\theta_{k}}^{t}\upsilon_{(\theta_{k},x_{k})}\mathbb{e}\left[w_{k}\right]+2\nu_{k}^{t}b_{\theta_{k}}^{t}\upsilon_{(\theta_{k},x_{k})}\tilde{a}_{\theta_{k}}x_{k}+2\nu_{k}^{t}b_{\theta_{k}}^{t}\upsilon_{(\theta_{k},x_{k})}\mathbb{e}\left[w_{k}\right]\\ & \quad + \alpha_{1}\vert\nu_{k}\vert^{2}+\alpha_{2}\mathbb{e}\left[w_{k}^{t}w_{k}\right],\\ & \leq-\mu\vert x_k\vert^{2}+2\vert\nu_{k}\vert\vert b_{\theta_{k}}\vert\vert\upsilon_{(\theta_{k},x_{k})}\vert\vert\tilde{a}_{\theta_{k}}\vert\vert x_{k}\vert+\alpha_{1}\vert\nu_{k}\vert^{2}+n\alpha_{2}.\end{aligned}\ ] ] from the young s inequality ,
@xmath223 @xmath224 @xmath225 ; note that @xmath226 one has @xmath227 where @xmath228 .
this yields @xmath229-v(x_{k},\theta_{k } ) & \leq-(\mu-\kappa)\vert x_{k}\vert^{2}+(\alpha_{1}+\kappa^{-1}\beta^{2})\vert\nu_{k}\vert^{2}+n\alpha_{2}.\end{aligned}\ ] ] take @xmath230 .
we have @xmath229 & \leq(1-c_{2}^{-1}(\mu-\kappa))v(x_{k},\theta_{k})+(\alpha_{1}+\kappa^{-1}\beta^{2})\vert\nu_{k}\vert^{2}+n\alpha_{2},\label{equa7}\end{aligned}\ ] ] where @xmath231 , with the constraint @xmath232 .
let @xmath233 .
taking expectation on both sides of ( [ equa7 ] ) , we get @xmath234\leq & q\mathbb{e}\left[v(x_{k},\theta_{k})\right]+(\alpha_{1}+\kappa^{-1}\beta^{2})\mathbb{e}\left[\vert\nu_{k}\vert^{2}\right]+n\alpha_{2}.\end{aligned}\ ] ] we obtain recursively @xmath235 & \leq q^{k}\mathbb{e}\left[v(x_{0},\theta_{0})\right]+(\alpha_{1}+\kappa^{-1}\beta^{2})\sum\nolimits _ { p=1}^{k-1}q^{k - p-1}\mathbb{e}\left[\vert\nu_{p}\vert^{2}\right ] + n\alpha_{2}\sum\nolimits _ { p=1}^{k-1}q^{k - p-1},\\ & \leq q^{k}\mathbb{e}\left[v(x_{0},\theta_{0})\right]+\frac{n\alpha_{2}+\delta(\alpha_{1}+\kappa^{-1}\beta^{2})}{1-q}.\end{aligned}\ ] ] finally , we obtain @xmath236 & \leq\frac{c_{2}}{c_{1}}q^{k}\|x_{0}\|^{2}+\frac{n\alpha_{2}+\delta(\alpha_{1}+\kappa^{-1}\beta^{2})}{c_{1}(1-q)},\end{aligned}\ ] ] where @xmath237 .
hence the proof is complete.@xmath238
consider the given probabilistic constraint ( [ constraint : multivariate ] ) , @xmath239 let @xmath240 then ( [ constraint : prob_intheproof ] ) is equivalent to since @xmath245 is a polyhedron , it is difficult to obtain the closed form of the above integral .
so , we consider an inscribed ellipsoidal approximation of @xmath245 @xcite , @xcite .
it is reasonable since the level curves of multivariate gaussian distribution are ellipsoids .
consider an ellipsoid ( since the covariance of @xmath222 is identity matrix ) , where @xmath253 denotes the chi - square cumulative distribution function with @xmath2 degrees of freedom .
the above inequality can be satisfied with @xmath254 value of @xmath255 by utilizing the maximization of a liner functional over an ellipsoidal set , @xmath256 can be ensured by @xcite,@xcite , | in this article , we consider a receding horizon control of discrete - time state - dependent jump linear systems , particular kind of stochastic switching systems , subject to possibly unbounded random disturbances and probabilistic state constraints . due to a nature of the dynamical system and the constraints
, we consider a one - step receding horizon .
using inverse cumulative distribution function , we convert the probabilistic state constraints to deterministic constraints , and obtain a tractable deterministic receding horizon control problem .
we consider the receding control law to have a linear state - feedback and an admissible offset term .
we ensure mean square boundedness of the state variable via solving linear matrix inequalities off - line , and solve the receding horizon control problem on - line with control offset terms .
we illustrate the overall approach applied on a macroeconomic system . |
stochastic theories of motion are a well - established approach to model physical trajectories of single particles embedded in a thermal environment .
the ergodic motions in particular are an essential prerequisite of statistical mechanics , as they exhibit the following convergence .
let , for a given trajectory , the occupation time @xmath0 designate the total amount of time the particle has spent within a domain @xmath1 up to time @xmath2 .
then @xmath3 @xmath4 is the steady - state probability for the occupation of @xmath1 .
for example , for the overdamped motion of a brownian particle immersed in water at temperature @xmath5 and subject to a static , confining potential @xmath6 , the boltzmann - gibbs canonical law yields @xmath7}$ ] ; @xmath8 is boltzmann s constant . in ensemble theories ,
the focus rests on ensemble laws such as @xmath4 .
but in practice , the measurement time @xmath2 is finite , so it is natural to investigate also the fluctuations @xmath9 knowing their precise distribution is relevant , e.g. , when tracking single particles by confocal microscopy or when studying reaction - diffusion dynamics @xcite .
a submanifold of the complete phase space is considered in the problem of phase persistence @xcite .
the time a laser - cooled atom resides in the `` dark '' low - momentum state @xcite determines the cooling efficiency . in a broader context , fluctuation theorems @xcite
describe deviations from equilibrium and refine our understanding of thermodynamic laws . as a rule of thumb , occupation time fluctuations are particularly large in the presence of annealed @xcite or quenched @xcite disorder .
the essential questions in all cases are : how large are the fluctuations , how do they evolve with time , and how are they determined by the precise microscopic dynamics on the one hand and the universal equilibrium laws @xmath4 on the other ?
we study here the site occupation times @xmath10 of a confined continuous - time random walk ( ctrw ) on a lattice , @xmath11 .
ctrws and related models are a standard theoretical approach to describe dynamics where trapping mechanisms induce a variety of remarkable motion patterns ; examples are the charge carrier transport in amorphous semiconductors @xcite , model glasses @xcite , subrecoil laser cooling @xcite , atomic transport in optical lattices @xcite and diffusion in biological cells @xcite , to name a few @xcite .
we focus on a regime where the ergodic convergence holds and the classical boltzmann - gibbs equilibrium and ergodicity ideas apply .
remarkably , the equilibrium probability @xmath12 even controls the distribution of finite - time fluctuations . but close to the non - ergodic regime , these fluctuations can be unexpectedly large , and their distribution exhibits a non - trivial , dual behavior .
firstly , the lvy - gauss central limit theorem suggests a standard scaling approach , which yields the non - gaussian lvy - stable laws .
secondly , it turns out that we have to explicitly consider the large fluctuations , which deviate from the central limit theorem . to study this aspect , we propose a nonstandard scaling procedure .
it leads to a concept which may appear surprising in a probabilistic model : the infinite density .
this density is not normalizable , despite being a limit of a properly normalized probability law .
while such objects play a key role in the mathematical field of infinite ergodic theory @xcite , their usage is less common in physical models , such as subdiffusion on intermittent maps @xcite , diffusion in logarithmic potentials @xcite and recently strong anomalous diffusion @xcite .
our goal is to showcase the scope and applicability of infinite densities and familiarize the reader with its peculiar properties .
we integrate it in the statistical mechanics framework by relating all our findings to the system s equilibrium statistics @xmath12 . finally , we extend our discussion to include the statistics of time - averaged observables .
this exposes the ubiquity of the interplay between dual scaling , stable laws and infinite densities in ergodic ctrws .
we start by introducing the ctrw in sec .
[ sec.model ] and briefly review the concept of equilibrium in the ctrw context . the basic nature of the time scaling duality is explained in sec .
[ sec.duality ] . after a general introduction to the calculation of occupation time statistics and their long - time asymptotic approximation in sec .
[ sec.otf ] , we discuss in detail the emergent stable law of bulk fluctuations in sec .
[ sec.otf:bulk ] , and the infinite density of large fluctuations in sec .
[ sec.otf:ldev ] .
section [ sec.moments ] tackles the question of how to apply these dual limiting laws when it comes to ensemble averaging . in sec .
[ sec.taobs ] , we consider fluctuations of a time average of an arbitrary observable .
to conclude , we summarize our findings and discuss briefly further directions of research , in particular models and applications beyond ctrw , in sec .
[ sec.outlook ] .
we consider a ctrw evolving along a one - dimensional lattice . to each accessible lattice point @xmath11
we assign a probability @xmath13 to make a jump to the right neighboring site ; @xmath14 for jumps to the left .
the jumps are instantaneous , but in between jumps , the particle resides on the lattice site for a random trapping or waiting time .
all jumps and waiting times are assumed to be statistically independent .
the key quantity here is the common probability density function ( pdf ) @xmath15 of waiting times @xmath16 , which reflects the disorder and heterogeneity of the medium . according to ctrw theory @xcite
, the long - time qualities of the motion are related to the width of @xmath15 ; more precisely , to the finiteness of its moments @xmath17 .
motivated by the complex systems collected in the introduction , we consider here distributions with power law tails , i.e. for large @xmath16 @xmath18 with the coefficient @xmath19 , bearing the unit @xmath20 .
some experimental setups permit the fine - tuning of the tail exponent @xmath21 via physical parameters such as temperature @xcite or particle size @xcite .
if the tail exponent is from the range @xmath22 , the mean waiting time @xmath23 diverges .
this regime has been investigated thoroughly in the past , as it gives rise to a wealth of anomalous phenomena @xcite .
we focus on @xmath24 , where waiting times do possess a characteristic relaxation scale , namely the mean @xmath25 .
consequently , as shown below , a confined ctrw is ergodic in the sense of eq . . of a bounded , thermal ctrw in comparison with ensemble occupation probabilities @xmath12
the waiting time distribution is as in eq .
, with @xmath26 and @xmath27 ( arbitrary units ) .
the statistics of three different paths are shown ( symbols ) , each for one value of @xmath24 ( see key ) .
the @xmath12 are boltzmann s equilibrium law , eq . , for the confining harmonic potential @xmath28 .
the occupation time fractions @xmath29 roughly fall onto ensemble laws @xmath12 .
but especially for the smaller @xmath21 , the deviations are huge , considering the long simulation time @xmath30 . ] in this context , @xmath12 is the probability to find one random walker , picked from a large ensemble of non - interacting random walkers , on position @xmath31 after a large relaxation time @xmath32 .
but while the statement is mathematically concise and correct , it may conceal physical reality .
figure [ fig.average ] compares occupation time fractions @xmath29 with ensemble probabilities @xmath12 for a ctrw particle which is bound by a harmonic potential @xmath33 and coupled to a heat bath at temperature @xmath5 . in this case , the boltzmann - gibbs canonical ensemble laws imply a gaussian form of @xmath12 we discuss thermal systems in more detail below . in fig .
[ fig.average ] , we observe a basic agreement with eq . on average , but
deviations are significant at the huge , yet finite process time @xmath34 .
the cause for the scatter are the waiting time statistics , which , with @xmath24 , are so broad that @xmath35 .
we note that kac s theorem @xcite in chaos theory @xcite suggests that hamiltonian systems should be studied under the premise @xmath25 , while @xmath36 might indeed be infinite .
hence , we conduct below a detailed analysis of the occupation time fluctuations . under confinement , the probability to find the random walker at a site @xmath31 converges with time @xmath2 to the @xmath2-independent value @xmath12 .
in this stationary state , the detailed balance condition holds , @xmath37 the existence of a stationary limiting state and detailed balance conditions are a premise for all ctrw dynamics studied in this work .
we may , in particular , choose to consider thermal ctrws .
then , the hopping probabilities satisfy @xcite @xmath38}$ ] , @xmath11 . the aim of a thermal model is to describe systems where the particle is confined by an external potential @xmath39 and in contact with a heat bath at constant temperature @xmath5
. detailed balance then implies boltzmann s canonical distribution @xmath40 equations - define a mapping between the sets @xmath41 , @xmath42 and @xmath43 , thus relating microscopic ctrw dynamics to the boltzmann - gibbs equilibrium concept .
recipes for the implementation in numerical simulations and a careful account on the effects of the lattice discretization can be found in ref .
@xcite , sec . via .
let @xmath10 be the occupation time at fixed @xmath31 up to time @xmath2 since the start of the random walk at time @xmath44 .
we denote with @xmath45 the pdf of fluctuations @xmath46 as in eq . .
its exact form is complicated and depends in particular on initial conditions , the full waiting time pdf @xmath15 and the hopping probabilities @xmath47 . but close to equilibrium , we expect to find a scaling limit form in terms of long - time properties of the process , such as the tail and the equilibrium statistics @xmath12 .
surprisingly , the occupation times of ctrw feature not one , but two such asymptotic scaling forms .
we derive below a bulk ( `` b '' ) scaling limit @xmath48 figure [ fig.otf.pdf ] demonstrates that these scaling laws do indeed coexist .
the scaling functions @xmath49 and @xmath50 are derived and discussed in secs .
[ sec.otf:bulk ] and [ sec.otf:ldev ] , respectively .
but before going deeply into the analytical study , we want to develop an understanding for the character of this duality .
the total occupation time is a sum of independent waiting times with the broad statistics . for the bulk , that is
, the large majority of process realizations , the number of waiting times entering this sum is large , as its average is @xmath51 . by virtue of the generalized central limit theorem @xcite , one can thus indeed anticipate a scaling limit of the form of eq . .
we note that the bulk relation also has a common shorthand notation : @xmath52 .
relations of this form appear in abundance in the physical and other sciences ( with various @xmath49 and @xmath21 ) .
it implies a strong statement on the asymptotic statistics : let a threshold @xmath53 increase faster than @xmath54 ; then the probability to observe @xmath55 goes to zero with @xmath2 . in this sense ,
the scope of the approximation @xmath49 widens , and deviations from bulk behavior become rare . hence , in many other physical contexts , a scaling relation of this type fully describes all asymptotic aspects of the respective process .
but for the ctrw problem at hand , the bulk analysis misses essential features of the distribution .
the broad waiting time pdf implies that the rare , large occupation times are relevant events .
since we can in principle have @xmath56 , the largest occupation times can be of the order of the measurement time @xmath2 itself .
this is why we divide @xmath57 by @xmath2 in the large fluctuations scaling limit , eq . .
but notice that the latter has a nonstandard , counter - intuitive form .
the involved scaling function @xmath50 is not properly normalized : integrating eq .
over all @xmath57 gives @xmath58 on the left - hand side , but on the right - hand side we get the seemingly time - dependent value @xmath59 .
in fact , we will find in sec .
[ sec.otf:ldev ] that @xmath50 has an infinite norm .
hence , it can not be treated as a conventional pdf and we call it an infinite density .
it is a useful and meaningful object , and contains exactly the statistical information that escapes the bulk scaling analysis . in the setting of fig .
[ fig.average ] . in order to study the occupation of the lattice site @xmath60
, one can introduce an occupation observable .
it alternates between two states , namely the particle is `` on @xmath31 '' or `` not on @xmath31 . ''
the site occupation time @xmath10 is then an accumulation of those waiting times that were spent `` on @xmath31 . ''
the stretches of time spent in the `` not on @xmath31''-state are essentially first - return times of the ctrw . ]
let @xmath61 denote the pdf of the occupation time @xmath10 and define its double laplace transform @xmath62 . to calculate it ,
we employ a method developed for the non - ergodic ctrw regime , i.e. @xmath22 @xcite .
it exploits the two - state renewal nature of the occupation process : the site @xmath31 alternates between being occupied or empty , see fig .
[ fig.occupation ] . to these two states
we can associate sojourn time distributions .
the time of continuous occupation of @xmath31 is distributed as @xmath15 .
the time of continuous absence from @xmath31 is essentially a first - return time : it measures how long it takes a ctrw particle to return to a site @xmath31 once it jumped off .
we denote its pdf with @xmath63 ( subscript : `` not on @xmath31 '' ) .
associated laplace transforms are @xmath64 and @xmath65 respectively . from these two distributions
one can infer the statistics of the occupation time on @xmath31 , see eq .
( a13 ) of ref .
@xcite : @xmath66+(s+u ) \left[1-{\widehat{\psi}}_{{\text{n}x}}(s)\right]}{s(s+u)\left[1-{\widehat{\psi } } ( s+u){\widehat{\psi}}_{{\text{n}x}}(s)\right ] } . \label{ot}\end{aligned}\ ] ] we are interested in the large time statistics of occupation times , @xmath67 . in laplace space , this corresponds to small @xmath68 . for broad - tailed waiting time pdfs as in eq . with @xmath24 ,
the laplace space asymptotics read @xmath69 } \label{psiasymptotic } .\end{aligned}\ ] ] as demonstrated in ref .
@xcite , one can use subordination arguments in the first - return time analysis to derive the respective small-@xmath70 expansion for the time of absence from @xmath31 : @xmath71 } , \label{psinxasymptotic}\end{aligned}\ ] ] where @xmath72 is the mean first - return time of an ordinary random walk ( rw ) . here , `` ordinary '' means that the walker is subject to the same hopping probabilities @xmath47 , but the waiting times on each lattice site are fixed to unity .
the random walks studied in this work are recurrent ( i.e. the ultimate return to any @xmath31 is certain ) , since they take place under the influence of a binding field @xmath6 .
they also obey the detailed balance condition , eq . .
under these circumstances , we have the simple relation @xcite @xmath73 in order to obtain an approximation for the occupation time pdf in the limit of large times , we now insert the expansions in eqs . and into the exact eq . ; we drop , separately in the numerator and the denominator , anything beyond the two lowest order terms in @xmath70 and @xmath74 , respectively .
this gives @xmath75 } { s+u{p_{x}}-c\left[{p_{x}}(s+u)^{\alpha}+(1-{p_{x}})s^{\alpha}\right ] } \label{ot : pdf : complete : trafo}\ ] ] for small , comparable values of @xmath76 , @xmath70 .
the ratio @xmath77 quantifies the relative width of the waiting time pdf . by fixing @xmath70
while expanding at @xmath78 , we further get @xmath79}$ ] .
we thus check that the pdf is properly normalized , its mean is indeed the ergodic limit , @xmath80 , and its variance is finite , @xmath81 ^ 2 \right\rangle}<\infty$ ] .
the latter is natural , since occupation times are bounded , @xmath56 .
we can now turn our attention to the fluctuations of the occupation times , @xmath46 . according to their definition , their pdf @xmath45 , is related through @xmath82 .
hence , @xmath83^{\alpha-1}+(1-{p_{x}})(s+ik{p_{x}})^ { \alpha-1}\right\ } } { s - c\left\{{p_{x}}[s - ik(1-{p_{x}})]^{\alpha}+(1-{p_{x}})(s+ik{p_{x}})^{\alpha } \right\ } } .\end{aligned}\ ] ] we would like to elaborate some more on the nature of this approximation .
we have derived it under the premise that we are dealing with long measurement times @xmath2 , hence we took @xmath70 to be small .
occupation times @xmath10 typically grow with time , so we assumed @xmath84 is small and `` comparable '' to @xmath70 . to be precise , the asymptotic approximation in eq .
is meaningful when @xmath70 is small and of the same order as @xmath85 , or @xmath84 , or anything between .
this corresponds exactly to the scaling limits in eqs . and , where the fluctuations @xmath46 are assumed to scale with @xmath54 or @xmath2 , respectively .
we thus find our heuristic arguments in sec .
[ sec.duality ] being compatible with eq . .
this equation encodes the complete information on the asymptotic distribution of occupation times , but it lives in the abstract fourier - laplace space .
additional efforts are required to extract any applicable results .
therefore , we split the problem into two aspects , each focusing on one specific time scaling .
this greatly simplifies and clarifies the picture .
in order to find the precise bulk scaling function @xmath49 , we continue from eq . .
the fourier - laplace space analogue to the scaling limit is to let @xmath70 become small , while fixing the ratio @xmath86 .
equation by this simplifies and becomes @xmath87 \right\}^{-1}\end{aligned}\ ] ] the neglected higher order terms are of the order @xmath88}$ ] . to continue
, we write @xmath89 , \ ] ] choosing principal values for the exponentiation and using @xmath90 . with this
, the bulk statistics can be cast into the more compact form @xmath91 } , \end{aligned}\ ] ] with parameters [ otf : pdf : bulk : asymptotic ] @xmath92\right\}^{1/\alpha } \\
\beta&= \frac{(1-{p_{x}})^\alpha{p_{x}}-(1-{p_{x}}){p_{x}}^\alpha } { ( 1-{p_{x}})^\alpha{p_{x}}+(1-{p_{x}}){p_{x}}^\alpha } .\end{aligned}\ ] ] fourier - laplace inversion of eq .
yields the sought - after pdf @xmath93 .
it is now evident that we can write it in the scaling form @xmath94 thus , at fixed @xmath2 , the scale parameter @xmath95 measures the bulk pdf s width . for any fixed waiting time parameters
, @xmath95 has a maximum at @xmath96 and vanishes if @xmath12 tends to either @xmath44 or @xmath58 .
moreover , the characteristic function @xmath97\right\ } , \end{aligned}\ ] ] defines the @xmath21-stable law @xcite @xmath98 . following the intuitive reasoning along the lines of the generalized limit theorem ( see sec . [ sec.duality ] ) , the basic @xmath21-stable form of the scaling function was to be expected
nonetheless , some of its specific features are remarkable .
the skewness parameter @xmath99 indicates an asymmetry ; see the sample plots in fig .
[ fig.otf.pdf ] .
this contrasts the familiar gaussian fluctuations , which appear , e.g. in the ordinary random walk .
also , in other physical systems where lvy - stable statistics play a role , one encounters typically rather @xmath100 or @xmath101 .
the asymmetry is controlled through @xmath21 and @xmath12 .
a symmetric distribution is obtained when @xmath96 , or in the limit @xmath102 , which generally recovers gaussianity .
another surprising feature of eq .
is its independence of any @xmath103 with @xmath104 .
naturally , at large times , the average occupation time becomes @xmath80 by virtue of ergodicity .
but even more , the complete pdf of fluctuations @xmath46 is controlled by the single steady - state probability @xmath105 , in conjunction with the waiting time characteristics @xmath21 and @xmath106 .
one thing appears to be odd though . because of @xmath56 , the fluctuations have natural bounds , @xmath107 but the support of the stable law @xmath49 is the complete real line .
more seriously , @xmath49 takes over the broad tails , i.e. @xmath108 for large @xmath109 @xcite .
hence , high - order moments diverge .
this includes even the variance , @xmath110 ^ 2 \right\rangle}_{\mathrm{b}}=t^{2/\alpha}\int_{-\infty}^\infty y^2g_{\mathrm{b}}(y)\,dy=\infty$ ] , easily the most common quantitative measure for fluctuations in experimental or numeric data .
evidently , in the case of ctrw , by focusing on the bulk scaling , we are missing vital statistical information . to complete the picture , we need to explicitly consider the rare large fluctuations .
the nonstandard scaling limit can be obtained as follows . according to eq . , the largest occupation time fluctuations @xmath46 are of the order of the measurement time @xmath2 itself .
we therefore restart from eq .
, let again @xmath70 be small , but now fix the ratio @xmath111 .
this gives @xmath112^{\alpha } + ( 1-{p_{x}})\left[1+{p_{x}}\frac{ik}{s}\right]^{\alpha } \\ & \;\ ; -{p_{x}}\left[1-(1-{p_{x}})\frac{ik}{s}\right]^{\alpha-1 } -(1-{p_{x}})\left[1+{p_{x}}\frac{ik}{s}\right]^{\alpha-1 } \bigg\ } \\ & = \frac{1}{s } + c{p_{x}}\;{\widehat{r}}_\alpha\big((1-{p_{x}})k;s\big)+c(1-{p_{x}})\ ; { \widehat{r}}_\alpha\big({-{p_{x}}}k;s\big ) .
\end{aligned } \label{otf : pdf : ldev : trafo}\end{aligned}\ ] ] in the first step , we dropped all terms which are of @xmath113}$ ] . in the second
, we made some simple rearrangements and introduced the auxiliary function @xmath114 ( s - ikq)^{\alpha-2 } .
\label{otf : pdf : ldev : rdef}\end{aligned}\ ] ] we show below that the fourier - laplace inversion of @xmath115 , and hence of eq . yields the scaling function @xmath50 appearing in eq . .
but before we start , some technical remarks are in place . in eq .
, the term @xmath116 is , strictly speaking , the leading order term in this small-@xmath70 expansion .
it becomes @xmath117 upon fourier - laplace inversion , i.e. a dirac @xmath118-distribution .
this does not come as a surprise .
we are now rescaling the statistics linearly with time , which is faster than the bulk scaling @xmath52 , @xmath119 .
hence , we `` squeeze '' the bulk realizations onto the origin . the higher order terms in eq .
contain exactly the information on those fluctuations that deviate from the bulk behavior .
hence , we suppress the @xmath117-term in the oncoming calculations .
however , notice that @xmath117 is by itself a pdf .
removing it from the equation naturally causes some issues with normalization for the remaining parts .
therefore , we should not interpret the @xmath50 literally as a pdf .
mathematically , we should rather treat it as a generalized function .
when calculating an average with it as @xmath120 , then we make it act on a test function @xmath121 .
not all test functions need to be `` suitable '' , i.e. integrable , with respect to the infinite density . in sec .
[ sec.moments ] , we elaborate more on this issue .
we now invert the @xmath122 as defined in eq . .
we proceed in two steps : first , we go from laplace space to time , @xmath123 ; second , from fourier space to fluctuations , @xmath124 .
we write the associated transformed functions as @xmath125 and @xmath126 . if we start from the second line of eq .
, then the first step , the laplace inversion , @xmath123 , becomes @xmath127 we can put this in a more elegant form , which makes explicit the time scaling nature of the function : @xmath128 where @xmath129 .
\label{otf : pdf : ldev : scalingfunctionf}\end{aligned}\ ] ] the remaining task is the fourier inversion @xmath124 , or respectively , @xmath130 . for this , we translate powers of @xmath131 into derivatives @xmath132 and the exponential @xmath133 into @xmath134 .
we thus get the inverse fourier - transform of @xmath135 , @xmath136\,d\omega\right\ } { \nonumber\\}&=\frac{1}{\alpha(\alpha-1 ) } \left\ { \frac{\partial}{\partial z } \left[{\mathbf{1}\!\left\{0<z\leq 1\right\ } } z^{-(\alpha-1)}\right ] \right.{\nonumber\\}&\hspace{12mm}\left .
+ \frac{\partial^2}{\partial z^2}\left [ { \mathbf{1}\!\left\{0<z\leq 1\right\ } } ( 1-z)\,z^{-(\alpha-1 ) } \right]\right\ } .\label{otf : pdf : ldev : scalingfunction : differentiate}\end{aligned}\ ] ] the indicator function @xmath137 is @xmath58 when the condition in the argument is fulfilled , and @xmath44 otherwise .
hence , it is a discontinuous step function , and its differentiation gives rise to several @xmath118-distributions in the expression . for reasons that we elaborated on above , we drop the contributions from the boundary @xmath138 . in app .
[ appsec.otf:pdf:ldev ] , we show that the other peaks at @xmath139 have zero contribution .
performing the differentiations in eq
. therefore gives @xmath140 tracing back the definitions in eqs .
, and , we finally arrive at [ otf : pdf : ldev : asymptotic ] @xmath141 this scaling function is defined within the appropriate bounds and correctly predicts the behavior of the large occupation times , see fig . [ fig.otf.pdf ] . again
, the expression depends solely on @xmath21 , @xmath106 , and the single equilibrium probability @xmath12 , and it is symmetric only if @xmath96 . due to the fast divergence @xmath142 at the origin , @xmath50 is not normalizable .
this is why we call it an infinite density .
this distinguished property should not be a cause of concern .
of course , at any finite time @xmath2 , the exact pdf @xmath143 is perfectly normalized .
its central part is best approximated by the stable law @xmath49 , eq . , while large fluctuations are best studied in terms of @xmath50 , eq . . in particular , the pole of the infinite density is completely congruent with the tails of the @xmath21-stable bulk distribution @xcite : eqs .
, , , and consistently yield @xmath144 [ otf : pdf : powerlaw ] the power law @xmath145 is responsible for both types of unphysical divergences in the asymptotics : the divergent moments of the stable laws and the divergent norm of the infinite density .
but by patching the two asymptotic expressions along the common power law overlap we obtain a complete , consistent statistical long - time description , bare of any such dubious divergences .
when calculating an average of the form @xmath146 , we are confronted with the question of how to apply the above asymptotic results . is the average obtained from the lvy - stable bulk statistics @xmath49 , eq . ?
or is it sensitive to the large fluctuations , so that we need the infinite density @xmath50 , eq . ?
or may we even have to consider the complete , yet implicit solution @xmath147 , eq . ?
in general , we find that @xmath49 and @xmath50 complement one another and yield the entire information on the long - time asymptotics .
this becomes most evident in the study of absolute @xmath148th order moments , @xmath149 .
we show in app .
[ appsec.otf:moments ] that , with the exception of @xmath150 , the moments fall into two distinct classes , which are exclusively integrable with respect to either the stable law or the infinite density : [ otf : moments ] @xmath151 where @xmath152 } , { \nonumber\\}{m_q^>}&= \frac{c\alpha\left [ { p_{x}}(1-{p_{x}})^q + ( 1-{p_{x}}){{p_{x}}}^q \right]}{|\gamma(1-\alpha)| } \int^1_0 z^q\,{{\mathcal{i}}\left(z\right ) } \,dz { \nonumber\\}&= \frac{cq\left [ { p_{x}}(1-{p_{x}})^q + ( 1-{p_{x}}){{p_{x}}}^q \right]}{|\gamma(1-\alpha)|(q-\alpha)(q-\alpha+1 ) } .
\label{otf : moments : coeff } \end{aligned}\ ] ] we stress that the coefficients @xmath153 and @xmath154 are finite in their respective parameter domains .
the stable laws @xmath98 do possess finite moments of order @xmath155 .
the expression @xmath156 is perfectly integrable when @xmath157 .
we can thus use , in this context , the infinite density @xmath50 as if it was an ordinary pdf despite the pole at the origin .
we add that the borderline case @xmath158 is obtained by carefully accounting for contributions from both the bulk and large fluctuations , see app .
[ appsec.otf:moments ] . according to eq . , the different time scalings of bulk versus large fluctuations manifest in a dual scaling of moments .
the same duality has been discussed with respect to strong anomalous diffusion @xcite .
in particular , we have @xmath110 ^ 2 \right\rangle}\sim{\left\langle [ { \delta t_{x}\left(t\right)}]^2 \right\rangle}_{\mathrm{l}}\simeq t^{3-\alpha}$ ] .
even the variance usually not perceived as a high - order moment does probe the statistics of the large fluctuations .
its increase with time is faster than anticipated from the bulk scaling ( @xmath159 ^ 2\sim t^{2/\alpha}$ ] ) .
in this section , we show that the connection between dual scaling laws and stable and infinite densities is not specific to occupation time statistics , but extends to a larger class of observables .
assume an observable @xmath160 takes on the value @xmath161 when the random walker sits on site @xmath31 .
then , by virtue of ergodicity , eq . , the time average @xmath162 in the limit @xmath163 . at finite time @xmath2
, we define the fluctuations of the time average as @xmath164 we denote the respective pdf by @xmath165 . for reasons which become apparent below ,
we restrict ourselves to observables and systems where the equilibrium average latexmath:[${\left\langle{|{\delta{\mathcal{o}}}|^\alpha}\right\rangle_{\!\mathrm{eq } } } = \sum_x is finite .
the general time average is defined through eq . in terms of the occupation time fractions @xmath29 .
conversely , the occupation time fraction of a specific region @xmath1 of the lattice is proportional to the time average of a particular observable .
namely , we can define an occupation observable @xmath160 through an indicator function @xmath137 : [ taobs : otf ] @xmath167 the time average of an occupation observable can be written as @xmath168 and likewise @xmath169 in the simplest case , @xmath1 would only consist of a single lattice point @xmath31 ( see , e.g. , fig .
[ fig.occupation ] , where @xmath60 ) .
we then return to the scenario of the previous sections . in this sense ,
the single - site occupation time fractions @xmath29 are special cases of the general time average @xmath170 in eq . .
we are hence now extending the discussion to a broader class of observables .
the detailed calculus is similar to the previous and can be found in app .
[ appsec.taobs ] .
we focus on the implied generalizations for the dual time scaling , the bulk and the large fluctuation laws
. the fluctuations of a general time average @xmath171 should , according to definition , have essentially the same time scaling behavior as the fluctuations of the occupation time fractions @xmath172 .
that is , for bulk fluctuations we expect that @xmath173 , while the large fluctuations should be of constant order .
indeed , we find the bulk behavior @xmath174 equations and are to be viewed as special cases . this basic nature of the dual time scaling is thus universal for all time averages , including occupation times .
what remains to be investigated is the exact form of the scaling functions @xmath175 and @xmath176 and their sensitivity to the choice of the observable @xmath160 .
the bulk scaling limit for large @xmath2 reads [ taobs : pdf : bulk : asymptotic ] @xmath177 where now @xmath178^{1/\alpha } , \end{aligned}\ ] ] and the @xmath21-stable law @xmath98 is as in eq .
, only with @xmath179 defined as @xmath180 we have introduced here the @xmath21th order absolute moments of equilibrium fluctuations [ taobs : eqmom ] @xmath181 the constrained sums @xmath182 run over all @xmath31 where @xmath183 , respectively .
the main observation here is that the basic @xmath21-stable character of bulk fluctuations is universal for all observables .
still , the details are observable - specific , in particular the skewness parameter @xmath179 .
if the observable @xmath160 in equilibrium symmetrically fluctuates around its mean , then @xmath100 , so the fluctuations of the time average are also symmetric . the converse , however
, does not hold .
one can think of a situation where @xmath160 is asymmetric in equilibrium , but in such a way that we happen to have @xmath184 , so that the time average is symmetric
. thus , there seems to be no simple one - to - one correspondence between the symmetry of an observable in equilibrium and the symmetry of the respective time average that is , with respect to the bulk statistics .
the scaling function in the large fluctuations limit is given through [ taobs : pdf : ldev : asymptotic ] @xmath185 introducing the infinite densities @xmath186 .
\label{taobs : pdf : ldev : asymptotic : scalingfct } \end{aligned}\ ] ] the non - integrable pole @xmath187 at the origin consistently overlaps with the broad tails of the @xmath21-stable bulk pdf in eq . .
therefore , this feature is again universal for all observables .
but the specifics of the observable do determine the shape of the complete infinite density .
we stress that the constrained moments that enter eq . explicitly depend on @xmath188 . because of this ,
the analysis of the large fluctuations does reveal the symmetry and other characteristics of the observable @xmath160 .
in contrast to the mere bulk statistics , the pdf @xmath176 is symmetric if and only if the @xmath160 fluctuates symmetrically around its mean in the equilibrium state . as an example , we study the time - averaged position , @xmath189 for a thermal ctrw as defined in sec .
[ sec.model ] .
we consider a constant force @xmath190 as a bias , which drives the particle against a reflective wall at @xmath191 . for convenience ,
we define a parameter @xmath192 that quantifies the competition between the localizing bias @xmath193 and the diffusive spreading at finite temperature @xmath5 . on our lattice
we define the discrete potential landscape @xmath194 which is linear in @xmath31 for all @xmath195 .
thermal detailed balance , eq .
, then implies that @xmath196 hence , all steps are taken preferably to the left with probability @xmath197 ; only at the origin , they are reflected to the right .
the boltzmann - gibbs equilibrium pdf , eq .
, for these dynamics reads @xmath198 with the partition function @xmath199^{-1}$ ] .
the equilibrium position is given by @xmath200 for the statistics of the time average , we further need the absolute moments of eq .
: [ tax : eqmom : constrained ] @xmath201 for all @xmath202 .
we used the notation @xmath203 to designate the smallest integer @xmath204 .
similarly , we can find , for all @xmath205 , @xmath206 here , @xmath207 is the largest integer @xmath208 .
of course we implicitly obtain @xmath209 equations and contain all the equilibrium statistics that we need to calculate the asymptotic distribution of the time - averaged position @xmath210 .
its ensemble mean falls onto the equilibrium mean , @xmath211 .
the bulk pdf of the fluctuations @xmath212 is given through eq .
, where we insert the equilibrium quantities in eqs . and .
plots are provided in fig .
[ fig.tax.pdf ] , left panel , for different values of the bias strength @xmath213 .
when the bias is stronger ( @xmath213 larger ) , then the pdf of the fluctuations gets more peaked ( @xmath95 decreases ) and more symmetric ( @xmath214 decreases ) .
the large fluctuations scaling limit is obtained by using eqs . and in eq . .
the respective plots can be found in the right panel of fig .
[ fig.tax.pdf ] .
in contrast to the universal lvy - stable bulk statistics , the structure of the emergent infinite density does reveal two defining properties of the underlying ctrw .
first , the reflective boundary condition causes the fluctuations to be bounded from the left , @xmath215 .
second , the infinite density exhibits discontinuous steps at points where @xmath216 becomes an integer , an obvious signature of the discrete lattice structure .
the effect is most pronounced for a strong bias , since then the ctrw samples only a small number of lattice sites .
within the framework of ctrw , we studied bounded , ergodic dynamics close to the non - ergodic phase .
in this regime , equilibrium statistical mechanics , in particular the boltzmann - gibbs laws , dictate the values of occupation times and time averages in the limit of infinite measurement times .
but at finite times , the fluctuations of these quantities are significant and are to be analyzed on the footing of dual scaling laws , such as eqs . - or - .
bulk fluctuations are to be distinguished from the rare , large fluctuations .
the two emergent asymptotic scaling laws complement one another .
thus , the behavior of physical observables , like the @xmath148th order moments considered in eqs . , are to be derived from bulk or large fluctuations , or both .
the bulk fluctuations are asymptotically distributed according to asymmetric lvy - stable laws .
this qualitative feature is universal for occupation times and general time averages alike , see eq . or .
only a single parameter , the skewness @xmath179 , depends to some extent on the binding potential and the observable of interest .
the rare large fluctuations also exhibit universality , in so far as they are generally described in terms of infinite densities , see eq . or .
but the precise analytic form of the infinite density is much more sensitive to the details of the system and the particular observable under study .
for instance , the natural bounds of the occupation times or time averages are only probed by the large fluctuations ; also the discrete / continuous structure of the phase space is only reflected in the infinite density behavior .
our equations connect the emergent lvy - stable laws and infinite densities with the occupation probability @xmath12 of an equilibrium ensemble of independent random walkers . in a thermal system
, we can thus relate the boltzmann - gibbs equilibrium statistics with the finite - time fluctuations around the ergodic limit . in the case of occupation times
, this relationship is remarkably simple . for a given lattice site @xmath31 , the single equilibrium occupation probability @xmath12 controls the asymptotic distribution of both bulk and large fluctuations . beyond that
, only two dynamical parameters determine the distribution : the tail exponent @xmath21 and the relative width @xmath77 of the waiting time distribution .
hence , the equilibrium laws of statistical mechanics have a strong connection with asymmetric lvy - stable laws and infinite densities .
we believe that similar relationships can be found in a vast variety of systems beyond ctrw .
they derive from the broad sojourn time distributions with power law tails as in eq . .
this characteristic also appears in stochastic @xcite and deterministic @xcite models of complex motion , and has been measured experimentally @xcite . while the detailed asymptotic statistics of these systems might be complicated
, our dual scaling approach may prove to be an effective analytical tool .
experimental studies of such systems may reveal the connection between model parameters ( @xmath21 , @xmath106 , @xmath12 ) and other measurable quantities ( temperature , particle size , confinement characteristics , etc . ) .
an ambitious , yet desirable long - term goal is to embed the limit laws found here into large deviations theory @xcite .
furthermore , aging effects @xcite should be taken into consideration , if the ctrw - like process has been initiated long before we start to measure occupation times or time averages .
we expect aging to be relevant in particular for the infinite density of large fluctuations .
the regime @xmath217 is an alternative direction to extend our studies .
we believe that , in this case , an infinite density still provides the large fluctuation statistics , due to the divergence of moments @xmath218 with @xmath219 . but how rare these large fluctuations are , and how their infinite density looks like or connects to the presumably gaussian bulk distribution , are outstanding questions .
we acknowledge funding from the israel science foundation .
the considerations that we expressed at the beginning of sec . [ sec.otf:ldev ] have led us to exercise care when treating the outer boundaries @xmath139 in eq . , and respectively @xmath221 in eq . .
generally , derivatives of the step function @xmath222 lead to peaks of the form @xmath223 .
but for the ctrws we studied in this work , such contributions are not present , as we show now .
we abbreviate the expression @xmath224 appearing in eq . with @xmath225 and
write down the distributional identity @xmath226\right\ } & = { \mathbf{1}\!\left\{z\leq1\right\ } } f'(z ) -\delta(z-1 ) f(z ) .\end{aligned}\ ] ] in principle , @xmath225 could be any sufficiently well - behaved conventional function , which in particular does not have a pole at @xmath139 . similarly , we can write @xmath227\right\}= { \nonumber\\}&\hspace{5mm}= { \mathbf{1}\!\left\{z\leq1\right\ } } f''(z ) -2\delta(z-1 ) f'(z ) + \delta'(z-1 ) f(z ) { \nonumber\\}&\hspace{5mm}= { \mathbf{1}\!\left\{z\leq1\right\ } } f''(z ) -\delta(z-1 ) f'(z ) + \delta(z-1 ) f(z)\frac{\partial}{\partial z } .\end{aligned}\ ] ] [ app : differentiate ] the derivative on the right - hand side acts upon any subjected test function . with respect to eq .
, the above relation is useful for evaluating the expression containing @xmath228 . with the identities
, we can continue eq . as @xmath229z^{-\alpha-1}\big\ } { \nonumber\\}&\quad -\frac{\delta(z-1)}{\alpha(\alpha-1 ) } \big\ { z^{-(\alpha-1)}-\left[\alpha-1+(2-\alpha)z\right]z^{-\alpha}\big\ } { \nonumber\\}&\quad + \frac{\delta(z-1)}{\alpha(\alpha-1 ) } \big\ { ( 1-z)z^{-(\alpha-1 ) } \big\}\frac{\partial}{\partial z } { \nonumber\\}&={\mathbf{1}\!\left\{0<z\leq1\right\ } } z^{-1-\alpha } \left(1-\frac{\alpha-1}{\alpha}z\right ) .\end{aligned}\ ] ] we observe that the @xmath118-peaks at the boundary @xmath139 indeed have zero contribution , and eq .
holds true .
we know that at large times @xmath2 , the central part of the pdf @xmath45 is well approximated by @xmath93 , while the outer parts are close to @xmath230 . inside two intermediate overlap regions ,
both approximations feature the power law behavior given in eq . .
we now introduce a positive , monotonic function @xmath231 , such that @xmath231 and @xmath232 are bound to lie inside the positive and negative power law regions , respectively , i.e. @xmath233 notice that the distance between the boundaries broadens as time increases .
therefore , we can always find a function @xmath231 that remains inside this region , at least beyond some finite threshold time .
we even have some freedom in defining its time dependence : @xmath234 is fine , or @xmath235 , or anything in between .
we make use of this fact below . with the function
@xmath231 we can divide the value of an absolute moment into two contributions : @xmath236 on the last line , we used the asymptotic expressions as in eqs . and .
this is allowed , since the boundary @xmath231 asserts that both integrals operate on regions where the respective approximation becomes more and more accurate with time .
we now treat the three classes of moments separately .
we choose @xmath238 , with some @xmath239 , and continue eq .
: @xmath240 } { \nonumber\\}&\sim\quad { m_q^<}\,t^{q/\alpha } \qquad(0<q<\alpha)\end{aligned}\ ] ] when going from the first to the second line , we used the fact that @xmath241 is @xmath21-stable , and therefore the related integral converges with time to the finite moment @xmath153 of order @xmath155 . in the third line
, the bulk contribution dominates at large times , since we have the relations @xmath242 we choose @xmath244 , with some @xmath245 , and continue eq . :
@xmath246 } & \quad&+\quad t^{-(\alpha-1)+q } \int_{-{p_{x}}}^{1-{p_{x } } } the integral on the second line gives the value of @xmath154 as in eq . .
we stress that the latter is finite : the infinite density @xmath247 diverges at the origin like @xmath248 , but the factor @xmath249 guarantees convergence for any @xmath157 . finally , due to relation , the large fluctuations with time
dependence @xmath250 dominate higher order moments @xmath157 .
we set again @xmath251 , @xmath252 . for @xmath150 , both bulk and large fluctuations
contribute to moments in the same way .
this can be seen as follows .
@xmath253}\quad -\quad t\,\frac{c\alpha}{|\gamma(1-\alpha)| } \left [ ( 1-{p_{x}})^\alpha{p_{x}}\ln\left(t^{1/\alpha-1}\,c_2\right ) + ( 1-{p_{x}}){p_{x}}^\alpha\ln\left(t^{1/\alpha-1}\,c_2\right)\right ] { \nonumber\\}&\sim\quad \frac{c(\alpha-1)}{|\gamma(1-\alpha)|}\left[(1-{p_{x}})^\alpha{p_{x}}+(1-{p_{x } } ) { p_{x}}^\alpha\right ] \;t\ln t\end{aligned}\ ] ] since @xmath254 behaves at the origin as @xmath255 , the large fluctuations integral on the first line diverges logarithmically as @xmath163 .
this is why in this case , we had to consider explicitly the form of the power law pole , eq . .
terms of the order of @xmath2 , were neglected on the last line .
this includes in particular the contributions from the very center of the stable density ( @xmath256 ) and the extremities of the infinite density ( @xmath257 and @xmath258 ) and the @xmath259-dependence of function @xmath231 .
hence , in a sense , only the power law overlap between the lvy - stable and the infinite density determines the asymptotic form of the moment @xmath150 .
we outline here the derivation of the long - time asymptotics of the pdf @xmath165 for the fluctuations @xmath171 of the time average of a general observable @xmath160 .
our starting point is the joint pdf @xmath260 for measuring the @xmath261 at time @xmath2 , while having made exactly @xmath262 random walk steps up to that point .
we can write down its fourier - laplace representation as @xmath263}\delta[n(t)=n ] \right\rangle}\,dt .\end{aligned}\ ] ] the average @xmath264 is to be interpreted here as an average over all realizations of the ctrw process . according to ref .
@xcite , sec .
5.3 , one can write @xmath265^{n^{{\text{rw}}}_x(n ) } \bigg\rangle .
\label{app : taobs : pdfn}\end{aligned}\ ] ] this should be read as follows .
the average with respect to the random , independent and identically distributed waiting times has already been carried out .
the statistics of waiting times is embedded via the laplace transform @xmath266 .
the remaining average is to be taken with respect to two aspects of the trajectories of an analogue _ ordinary _ random walk ( rw ) , i.e. where the time between steps is fixed to unity .
first , we have to average over all terminal positions @xmath267 after @xmath268 steps .
second , we average over the possible sets of visitation numbers @xmath269 , i.e. the number of times an ordinary random walker has visits each of the sites @xmath31 of the lattice within @xmath268 steps .
we want to approximate eq . in the limit of large times and ,
consequently , large values of @xmath270 .
this translates to small @xmath70 , @xmath76 . on the one hand , for the broad - tailed waiting times as in eq .
, this means we expand @xmath271 on the other hand , it is natural to assume that we need to concentrate on large @xmath268 , so we require the statistics of the ordinary random walk after many steps . since we confine our discussion to ergodic random walks , it is easy to say what happens when @xmath268 becomes infinite .
the position @xmath267 is distributed according to the @xmath268-independent equilibrium distribution @xmath12 , while the visitation numbers converge as @xmath272 . at finite @xmath268 , deviations from these ergodic limits appear .
however , for ordinary random walks , these kind of fluctuations do typically not exhibit broad - tailed statistics .
we rather assume that their contribution in eq .
is of the order @xmath273}$ ] .
they are then negligible as compared to the fluctuations caused by the broad - tailed waiting times . recall that in the specific case of the site occupation times @xmath46 , sec .
[ sec.otf ] , we gave a more rigorous argument for this assumption in terms of a first - passage time analysis .
mean _ of the first - passage time of the ordinary random walk plays a role in the asymptotics of occupation time fluctuations . for this or similar arguments to hold , it is obviously crucial that @xmath274 .
we thus now approximate eq . by using eq .
, by writing the average with respect to @xmath267 as an equilibrium average and by replacing the visitation numbers @xmath275 by their ergodic limit values @xmath276 .
this yields @xmath277^{n{p_{x } } } { \nonumber\\}&\quad= \left [ { \left\langle \tau \right\rangle}-a{\left\langle{(s+u{\mathcal{o}})^{\alpha-1}}\right\rangle_{\!\mathrm{eq } } } \right ] \times { \nonumber\\}&\quad\quad \exp\!\left\ { n\sum_x{p_{x}}\ln\left[1-{\left\langle \tau \right\rangle}(s+u{o}_x)+a(s+u{o}_x)^\alpha\right ] \right\ } .
\end{aligned}\ ] ] at this point , we marginalize out the number of steps . since we are operating in a limit where @xmath268 is assumed large , we find it appropriate to treat @xmath268 as continuous and integrate , that is @xmath278 } { \nonumber\\}&\sim \frac{1-c{\left\langle{(s+u{\mathcal{o}})^{\alpha-1}}\right\rangle_{\!\mathrm{eq}}}}{s+u{\left\langle{{\mathcal{o}}}\right\rangle_{\!\mathrm{eq}}}-c{\left\langle{(s+u{\mathcal{o}})^ { \alpha}}\right\rangle_{\!\mathrm{eq } } } } .
\label{app : taobs : pdf : completel}\end{aligned}\ ] ] in the last step , we used @xmath279}$ ] for small @xmath31 .
equation is the generalization of eq . .
the latter can be recovered by considering the occupation observable in eq . .
the pdf of the fluctuations @xmath171 are most suitably discussed in fourier - laplace space . by shifting by @xmath280 ,
we get @xmath281 } \right\rangle}\,dt= { \widehat{f}}(-ik;s+ik{\left\langle{{\mathcal{o}}}\right\rangle_{\!\mathrm{eq } } } ) { \nonumber\\}&\sim \frac{1-c{\left\langle{(s - ik{\delta{\mathcal{o}}})^{\alpha-1}}\right\rangle_{\!\mathrm{eq}}}}{s - c{\left\langle{(s - ik{\delta{\mathcal{o}}})^{\alpha}}\right\rangle_{\!\mathrm{eq } } } } , \label{app : taobsf : pdf : completel}\end{aligned}\ ] ] generalizing eq . .
by methods which are completely analogous to those used in sec .
[ sec.otf:bulk ] one can compute the limit where @xmath70 becomes small while @xmath86 is fixed .
this gives the lvy @xmath21-stable bulk statistics of the fluctuations of time averages , eqs .
the limit where @xmath70 is small and @xmath111 fixed is obtained through the procedure discussed in sec .
[ sec.otf:ldev ] and [ appsec.otf:pdf:ldev ] .
it yields the infinite density of large fluctuations , eq . .
n. agmon , chem .
. lett . * 497 * , 184 ( 2010 ) .
a. m. berezhkovskii , v. zaloj , and n. agmon , phys .
e * 57 * , 3937 ( 1998 ) .
o. bnichou , m. coppey , j. klafter , m. moreau , and g. oshanin , j. phys .
a * 36 * 7225 ( 2003 ) . c. godrche and j. m. luck , j. stat
. phys . * 104 * , 489 ( 2001 ) .
f. bardou , j .-
bouchaud , o. emile , a. aspect , and c. cohen - tannoudji , phys .
lett . * 72 * , 203 ( 1994 ) . c. jarzynski , annu .
rev . condens .
matter phys .
* 2 * , 329 ( 2011 ) thermodynamics at the nanoscale u. seifert , rep . prog . phys . * 75 * , 126001 ( 2012 ) . c. monthus , and j .-
bouchaud , j. phys .
a * 29 * , 3847 ( 1996 )
. g. bel , and e. barkai , phys .
lett . * 94 * , 240602 ( 2005 ) .
g. bel , and e. barkai , phys .
e * 73 * , 016125 ( 2006 ) s. n. majumdar , and a. comtet , phys .
lett . * 89 * , 060601 ( 2002 ) .
s. burov , and e. barkai , phys .
lett . * 98 * , 250601 ( 2007 ) h. scher and e. w. montroll , phys .
b * 12 * , 2455 ( 1975 ) .
e. lutz , phys .
lett . * 93 * , 190602 ( 2004 ) . j .-
jeon , v. tejedor , s. burov , e. barkai , c. selhuber - unkel , k. berg - srensen , l. oddershede , and r. metzler , phys .
lett . * 106 * , 048103 ( 2011 ) .
a. v. weigel , b. simon , m. m. tamkun , and d. krapf , pnas * 108 * , 6438 ( 2011 ) .
single - molecule tracking q. xu , l. feng , r. sha , n. c. seeman , and p. m. chaikin , phys .
* 106 * , 228102 ( 2011 ) .
i. y. wong , m. l. gardel , d. r. reichman , e. r. weeks , m. t. valentine , a. r. bausch , and d. a. weitz , phys .
lett . * 92 * , 178101 ( 2004 ) .
networks p. c. holz , a. dechant , and e. lutz , europhys .
* 109 * , 23001 ( 2015 ) .
e. lutz and f. renzoni , nature phys .
* 9 * , 615 ( 2013 ) .
a. rebenshtok , s. denisov , p. hnggi , and e. barkai , phys .
* 112 * , 110601 ( 2014 ) , limit theorem _ ibid .
e * 90 * , 062135 ( 2014 ) .
j. h. p. schulz , e. barkai , and r. metzler , phys .
x * 4 * , 011028 ( 2014 ) .
a. rebenshtok , and e. barkai , j. stat . phys . * 133 * , 565 ( 2008 ) .
s. burov , r. metzler , and e. barkai , pnas * 107 * , 13228 ( 2010 ) .
m. kac , _ probability and related topics in physical sciences _ ( interscience , new york , 1958 ) .
we note that eq .
is derived under the assumption that at time @xmath282 , the random walker is not residing on @xmath31 . however , in the large time limit , the choice of this initial condition has no effect ( see ref .
@xcite ) .
the authors of ref . @xcite use a definition of the first - return time @xmath283 which slightly differs from ours : it includes the time spent at the site @xmath31 before jumping off .
hence , for us , @xmath284 .
g. samorodnitsky and m. s. taqqu , _ stable non - gaussian random processes : stochastic models with infinite variance _
( chapman & hall , 1994 ) .
p. castiglione , a. mazzino , p. muratore - ginanneschi , and a. vulpiani , physica d ( amsterdam ) * 134 * , 75 ( 1999 ) .
h. touchette , phys . rep .
* 478 * , 1 ( 2009 ) . | we study occupation time statistics in ergodic continuous - time random walks . under thermal detailed balance conditions , the average occupation time is given by the boltzmann - gibbs canonical law .
but close to the non - ergodic phase , the finite - time fluctuations around this mean are large and nontrivial .
they exhibit dual time scaling and distribution laws : the infinite density of large fluctuations complements the lvy - stable density of bulk fluctuations .
neither of the two should be interpreted as a stand - alone limiting law , as each has its own deficiency : the infinite density has an infinite norm ( despite particle conservation ) , while the stable distribution has an infinite variance ( although occupation times are bounded ) .
these unphysical divergences are remedied by consistent use and interpretation of both formulas .
interestingly , while the system s canonical equilibrium laws naturally determine the mean occupation time of the ergodic motion , they also control the infinite and lvy - stable densities of fluctuations .
the duality of stable and infinite densities is in fact ubiquitous for these dynamics , as it concerns the time averages of general physical observables . |
it was pointed out in ref . that a finite hall response for a model with gaussian disorder requires particle - hole asymmetry in the normal state dos .
we remind the reader , that for a quadratic dr , the dos is we will need the real space green s function @xmath115 to zeroth order in @xmath40 .
the limit @xmath116 is taken before any other limit , and thus formally @xmath117 is assumed .
it is determined by the fermi surface and thus , for any dispersion relation , @xmath118 furthermore , we will need @xmath119 to first order in @xmath40 .
it turns out that the zeroth order contribution is determined by the microsopic details at the uv , but the first order contribution is determined by the fermi surface .
@xmath120\right \rbrace \label{app : eq : realspacexigf}\ ] ] in eqs . and
, @xmath47 and @xmath121 denote bessel functions of the first kind .
the function @xmath48 represents an off - shell contribution and therefore depends on microscopic details of the model . in the case of a quadratic dispersion we find @xmath49 , where @xmath50 is the zeroth bessel function of the second kind . for a general dr , we can write @xmath122
. the struve function @xmath123 stems from the vicinity of the fermi surface while @xmath124 , which vanishes for large distances , stems from other parts of the spectrum . for @xmath125 microscopic details
become unimportant , this is reflected in the asymptotic relation @xmath126 .
we first present the derivation of @xmath115 for a general dispersion relation .
it will be useful to use the notation @xmath63 , @xmath127 and denote the lower ( upper ) band edge by @xmath128 ( @xmath129 ) .
we will omit the difference between @xmath130 and @xmath131 which is of order @xmath132 .
similarly we find @xmath134 we see that for @xmath115 [ @xmath119 ] even [ odd ] powers of @xmath131 are determined by the residue at the fermi surface . on the other hand ,
the odd [ even ] terms in @xmath131 are determined by off - shell contributions and therefore dependent on microscopic details .
for example it depends on the details of the cut - off , which is @xmath136 for quadratic dr . for this concrete case , we present the real space green s function up to @xmath137 . all calculations in this supplemental material
are performed in matsubara green s function technique .
it should be understood , that @xmath143 ( @xmath144 ) are fermionic ( bosonic ) matsubara frequencies .
we introduce the notation @xmath145 and restrict ourselves to the case @xmath146 .
the case @xmath147 immediately follows , since anomalous green s functions are always adjacent to the current vertices . throughout the calculation we employ the following identity @xmath148 we now investigate diagram ( 2a ) .
@xmath149\notag\\ & = & - \frac{e^2 t}{2 ( 2\pi \nu_0)^4 ( \omega \tau)^2 } \sum_{\epsilon , \epsilon_1 , \epsilon_2 } \int_{{\textbf{\textit p}}_1 , { \textbf{\textit p}}_2 , { \textbf{\textit k } } } { { \textbf{\textit p}}_1 \wedge { \textbf{\textit p}}_2}{(\delta_{\epsilon_1 , \epsilon + \omega } - \delta_{\epsilon_1 , \epsilon})(\delta_{\epsilon_2 , \epsilon + \omega } - \delta_{\epsilon_2 , \epsilon } ) } \notag \\ & & \times { \text{tr}}\left [ g(\epsilon_1 , { \textbf{\textit p}}_1)\tau_z g ( \epsilon + \omega , { \textbf{\textit k } } ) \tau_z g ( \epsilon_2 , { \textbf{\textit p}}_2 ) \tau_z g(\epsilon , { \textbf{\textit k } } ) \tau_z\right ] \end{aligned}\ ] ] we perform the trace over numerators of the green s function , cf .
eq . , using the antisymmetrization under @xmath150 .
we obtain @xmath151[\mathcal{g}(\epsilon , { \textbf{\textit p}}_2)-\mathcal{g}(\epsilon + \omega , { \textbf{\textit p}}_2 ) ] \notag \\ & = & - \frac{e^2 t \delta_0 ^ 2}{(2\pi \nu_0)^4 \tau^2 p_f^2 } \sum_\epsilon ( 2\epsilon + \omega ) \left
\lbrace \int_{{\textbf{\textit p } } } p^2 \mathcal g(\epsilon , { \textbf{\textit p}})\mathcal g(\epsilon+\omega , { \textbf{\textit p } } ) \right \rbrace^2 \left \lbrace ( \xi \mathcal g)(\epsilon+\omega , { \textbf{\textit r}}= 0)-(\xi \mathcal g)(\epsilon , { \textbf{\textit r}}= 0 ) \right \rbrace .
\label{eq : q2a}\end{aligned}\ ] ] for the evaluation of the current correlator of diagrams ( 2e)-(2h ) it is useful to investigate the self - energy at born level , first
. it is momentum independent @xmath152 and @xmath153.\end{aligned}\ ] ] in particular , the self - energy contains a uv divergent contribution .
however , @xmath154 is determined by @xmath155 , which is finite .
\notag \\ & & - { \text{tr}}\left [ g(\epsilon + \omega , { \textbf{\textit p } } ' ) \sigma(\epsilon + \omega ) g(\epsilon + \omega , { \textbf{\textit p } } ' ) \tau_z g(\epsilon + \omega , { \textbf{\textit p } } ) g(\epsilon , { \textbf{\textit p } } ) \tau_z g(\epsilon , { \textbf{\textit p } } ' ) \right ] \big \rbrace \notag \\ & & - [ \omega \rightarrow ( -\omega ) ] \notag \\ & = & \frac{e^2 t}{2 ( 2\pi \nu_0)^3 \tau \omega^2 } \sum_\epsilon \int_{{\textbf{\textit p } } , { \textbf{\textit p } } ' } { { \textbf{\textit p } } \wedge { \textbf{\textit p } } ' } { \text{tr}}\big \lbrace \tau_z \left [ g(\epsilon , { \textbf{\textit p } } ) - g(\epsilon + \omega , { \textbf{\textit p } } ) \right ] \tau_z \notag \\ & & \times [ g(\epsilon , { \textbf{\textit p } } ' ) \sigma(\epsilon + \omega ) g(\epsilon + \omega , { \textbf{\textit p } } ' ) -g(\epsilon + \omega , { \textbf{\textit p } } ' ) \sigma(\epsilon + \omega ) g(\epsilon , { \textbf{\textit p } } ' ) ] \big \rbrace \notag \\ & & - [ \omega \rightarrow ( -\omega ) ] .\end{aligned}\ ] ] we use @xmath157 take the trace over nambu space and obtain @xmath158 \mathcal g(\epsilon , { \textbf{\textit p } } ' ) \mathcal g(\epsilon + \omega , { \textbf{\textit p } } ' ) \notag \\ & & - [ \omega \rightarrow ( -\omega ) ] \notag \\ & = & \frac{e^2 t \delta_0 ^ 2}{(2\pi \nu_0)^4 \tau^2 \omega p_f^2 } \sum_\epsilon ( \xi \mathcal g ) ( \epsilon + \omega , { \textbf{\textit r } } = 0 ) \int_{{\textbf{\textit p } } , { \textbf{\textit p } } ' } ( { { \textbf{\textit p } } \wedge { \textbf{\textit p}}'})^2 \left [ \mathcal g(\epsilon , { \textbf{\textit p } } ) - \mathcal g(\epsilon + \omega , { \textbf{\textit p } } ) \right ] \mathcal g(\epsilon , { \textbf{\textit p } } ' ) \mathcal g(\epsilon + \omega , { \textbf{\textit p } } ' ) \notag \\ & & - [ \omega \rightarrow ( -\omega ) ] \notag \\ & = & \frac{e^2 t \delta_0 ^ 2}{2 ( 2\pi \nu_0)^4 \tau^2 p_f^2 } \sum_\epsilon ( 2\epsilon + \omega ) \left \lbrace \int_{{\textbf{\textit p } } } p^2 \mathcal g(\epsilon , { \textbf{\textit p}})\mathcal g(\epsilon+\omega , { \textbf{\textit p } } ) \right \rbrace^2 \left \lbrace ( \xi \mathcal g)(\epsilon+\omega , { \textbf{\textit r}}= 0)-(\xi \mathcal g)(\epsilon , { \textbf{\textit r}}= 0 ) \right \rbrace .
\label{eq : q2eh}\end{aligned}\ ] ] we see , that the functional form of contributions from diagrams ( 2a ) , eq . , and ( 2e)-(2h ) , eq . , is the same .
we use @xmath159 , \notag \\ ( \xi \mathcal g)(\epsilon+\omega ,
{ \textbf{\textit r}}= 0)-(\xi \mathcal g)(\epsilon , { \textbf{\textit r}}= 0 ) & = & - \pi \nu_0 ' \left [ \sqrt{(\epsilon + \omega)^2 + \delta_0 ^ 2 } - \sqrt{\epsilon^2 + \delta_0 ^ 2}\right ] , \notag\end{aligned}\ ] ] and finally obtain for the current current correlator in the non - crossing approximation @xmath160 for the case of parabolic dispersion , @xmath161 and thus the contribution of non - crossing diagrams vanishes .
this can be checked explicitly , since @xmath162 \notag \\ & = & - \frac{\nu_0}{2 } \left [ \ln[\mathcal e_f^2 + ( \epsilon + \omega)^2 + \delta_0 ^ 2]-\ln[\mathcal e_f^2 + \epsilon^2 + \delta_0 ^ 2]\right ] .\end{aligned}\ ] ] as the summation over @xmath163 is dominated by @xmath164 , we find that , for the model with parabolic dispersion , the non - crossed diagrams are a factor of @xmath40 smaller than the crossed diagrams and thus negligible .
we now turn our attention to the crossed diagrams ( 2b)-(2d ) . for diagram ( 2b ) we find @xmath165 \notag \\ & = & - \frac{e^2}{4 ( 2\pi \nu_0)^4 ( \omega \tau)^2 } t \sum_{\epsilon , \lbrace \epsilon_i \rbrace } { ( \delta_{\epsilon_1 , \epsilon + \omega } - \delta_{\epsilon_1 , \epsilon})(\delta_{\epsilon_2 , \epsilon + \omega } - \delta_{\epsilon_2 , \epsilon } ) } \left ( \delta_{\epsilon_3 , \epsilon + \omega}\delta_{\epsilon_4 , \epsilon}-\delta_{\epsilon_4 , \epsilon + \omega}\delta_{\epsilon_3 , \epsilon}\right ) \notag \\ & & \int_{\lbrace { \textbf{\textit p}}_i \rbrace}(2 \pi)^2 \delta ( { \textbf{\textit p}}_1 + { \textbf{\textit p}}_2 - { \textbf{\textit p}}_3 - { \textbf{\textit p}}_4 ) { \textbf{\textit p}}_1 \wedge { \textbf{\textit p}}_2 { \text{tr}}[g(\epsilon_1 , { \textbf{\textit p}}_1 ) \tau_zg(\epsilon_3 , { \textbf{\textit p}}_3 ) \tau_zg(\epsilon_2 , { \textbf{\textit p}}_2 ) \tau_zg(\epsilon_4 , { \textbf{\textit p}}_4 ) \tau_z ] , \notag \\\end{aligned}\ ] ] while for the sum of diagrams ( 2c ) and ( 2d ) we obtain @xmath166 \notag \\ & & + { \text{tr}}\left [ g(\epsilon , { \textbf{\textit p}}_1 ) g(\epsilon + \omega , { \textbf{\textit p}}_1 ) \tau_z g(\epsilon + \omega , { \textbf{\textit p}}_2)g(\epsilon , { \textbf{\textit p}}_2 ) \tau_z g(\epsilon , { \textbf{\textit p}}_4 ) \tau_z g(\epsilon , { \textbf{\textit p}}_3 ) \tau_z \right ] \big \rbrace \notag \\ & = & - \frac{e^2}{2 ( 2 \pi \nu_0)^4 ( \omega \tau)^2 } t \sum_{\epsilon , \lbrace \epsilon_i \rbrace } { ( \delta_{\epsilon_1 , \epsilon + \omega } - \delta_{\epsilon_1 , \epsilon})(\delta_{\epsilon_2 , \epsilon + \omega } - \delta_{\epsilon_2 , \epsilon } ) } \left ( \delta_{\epsilon_3 , \epsilon + \omega}\delta_{\epsilon_4 , \epsilon+\omega}-\delta_{\epsilon_3 , \epsilon}\delta_{\epsilon_4 , \epsilon}\right ) \notag \\ & & \int_{\lbrace { \textbf{\textit p}}_i \rbrace}(2 \pi)^2 \delta ( { \textbf{\textit p}}_1 - { \textbf{\textit p}}_2 - { \textbf{\textit p}}_3 + { \textbf{\textit p}}_4 ) { \textbf{\textit p}}_1 \wedge { \textbf{\textit p}}_2 { \text{tr}}[g(\epsilon_1 , { \textbf{\textit p}}_1 ) \tau_zg(\epsilon_3 , { \textbf{\textit p}}_3)\tau_zg(\epsilon_4 , { \textbf{\textit p}}_4 ) \tau_zg(\epsilon_2 , { \textbf{\textit p}}_2 ) \tau_z ] . \notag \\\end{aligned}\ ] ] we evaluate the traces and find for @xmath167- and @xmath168diagrams here , we introduced the integral@xmath170 \notag \\ & - \frac{\pi \beta_2}{8 } \left [ \frac{\epsilon_4}{\prod_{i = 1,3,4 } \sqrt{\epsilon_i^2 + \delta_0 ^ 2}}+\frac{\epsilon_4}{\prod_{i = 2,3,4 } \sqrt{\epsilon_i^2 + \delta_0 ^ 2}}-\frac{\epsilon_1+\epsilon_2}{\prod_{i = 1,2,3 } \sqrt{\epsilon_i^2 + \delta_0 ^ 2}}\right ] \label{eq : crossedintegral}\end{aligned}\ ] ] the two constants @xmath171 are defined by the following integrals only the last term of the square brackets in eq
. enters the current - current correlator of x- and @xmath173 diagrams .
this is because the previous terms are @xmath174 or @xmath175 independent , which in turn follows from @xmath176 , see also ref . , eq .
( 56 ) . with the help of these expressions ,
we find the following formula for the current - current correlator @xmath177 the constant @xmath178/{8}$ ] is @xmath65 for the case of quadratic dispersion and generally expected to be of order unity .
this concludes the derivation of eq . of the main text . | we calculate the temperature dependent anomalous ac hall conductance @xmath0 for a two - dimensional chiral @xmath1-wave superconductor .
this quantity determines the polar kerr effect , as it was observed in sr@xmath2ruo@xmath3 [ j. xia _ et al .
_ , phys .
rev .
lett . * 97 * , 167002 ( 2006 ) ] .
we concentrate on a single band model with arbitrary isotropic dispersion relation subjected to rare , weak impurities treated in the born approximation .
as we explicitly show by detailed computation , previously omitted contributions to extrinsic part of an anomalous hall response , physically originating from diffractive skew scattering on quantum impurity complexes , appear to the leading order in impurity concentration . by direct comparison with published results from the literature we demonstrate the relevance of our findings for the interpretation of the kerr effect measurements in superconductors .
= 1 * _ introduction . _
* unconventional superconductivity remains a very active field of condensed matter research .
notably , the chiral @xmath1-wave superconductor is a particularly spectacular state of matter . not only it demonstrates the extraordinary effects of electronic correlations , but it also displays exciting topological features , such as majorana zero modes bound to half quantum vortices . in a chiral @xmath1-wave superconductor , the electrons which constitute the cooper pairs rotate around each other with magnetic quantum number @xmath4 .
clearly , such a state breaks time - reversal symmetry ( trs ) and by pauli s exclusion principle , the cooper pair wave function ought to be symmetric in spin or band indices of a given material . to present date
, the chiral @xmath1-wave superconducting phase has not yet been unambiguously observed experimentally in solids .
nonetheless , there is wide consensus in the community , that strontium ruthenate ( sr@xmath2ruo@xmath3 ) constitutes a promising candidate material.@xcite experimental evidence for triplet - pairing in sr@xmath2ruo@xmath3 relies on the knight shift@xcite and neutron scattering@xcite while a peculiar phase sensitivity of the josephson effect @xcite is believed to reveal the odd parity of the order parameter .
furthermore , the observation of half quantum vortices in magnetometry@xcite indicates spin triplet @xmath1-wave superconductivity .
the spontaneous breaking of trs was first observed in the muon spin - relaxation@xcite and later in the polar kerr effect ( pke).@xcite in this paper , we concentrate on the latter probe .
a nonzero kerr angle @xmath5\ ] ] in a layered material ( such as sr@xmath2ruo@xmath3 ) with interlayer distance @xmath6 and complex index of refraction @xmath7 relies on a finite , 2d , optical anomalous hall conductivity @xmath8/2 $ ] , with @xmath9 the ac frequency .
superconductor with weak impurities and , for concreteness , a quadratic dispersion relation ( ac frequency @xmath9 , superconducting pairing amplitude @xmath10 , elastic scattering time @xmath11 and @xmath12 $ ] ) .
the solid red ( blue dashed ) curve represents the imaginary ( real ) part of @xmath13 .
inset : real space illustration of quantum mechanical probabilities for processes contributing to @xmath13 and corresponding to diagrams ( 2a)-(2h ) from fig .
[ fig : pertdiagssecond ] . while generally all of those diagrams contribute , in the specific case of a parabolic band , the response stems from the processes ( 2b)-(2d ) , only .
these contributions rely on diffractive scattering from quantum impurity complexes ( yellow ellipses ) with spatial extension @xmath14 comparable to the fermi wavelength @xmath15 . ]
theories of the anomalous hall effect@xcite ( ahe ) are most often developed on the basis of either the semiclassical boltzmann equation@xcite or the kubo - streda@xcite diagrammatic formalism . while both approaches are equally justified and
should yield the same results,@xcite the semiclassical approach seems to be more intuitive while diagrams appear to be more systematic . in the boltzmann treatment ,
the ahe is attributed to the addition of the following effects .
first , the intrinsic or anomalous velocity contribution , which relies on the berry curvature of the bands .
second , the extrinsic contributions , which stem from ( a ) asymmetric skew scattering from impurities , and ( b ) the side jump , a lateral displacement of semiclassical trajectories near scattering centers .
these contributions are automatically accounted for in the diagrammatic treatment of the problem .
most recently , the importance of diagrams with two crossed impurity lines@xcite was uncovered .
@xcite physically , these diagrams represent diffractive skew scattering from quantum impurity complexes.@xcite it is important to emphasize , that for a disorder potential with gaussian distribution , diagrams with two crossed impurity lines are of the same order as diagrams within the noncrossing approximation .
theoretically , the ac ahe in the context of chiral @xmath1-wave superconductors has been studied in refs . for clean single band models .
however , @xmath16 for such models,@xcite a result that can be understood as a consequence of galilean invariance.@xcite therefore , the observed finite kerr effect was considered within clean multiband models@xcite and single band models with impurities.@xcite notwithstanding the significant theoretical interested , to the best of our knowledge the effect of diffractive skew scattering from quantum impurity complexes has been disregarded in the literature , so far
. it will therefore be the subject of the present paper .
we concentrate on a single band model for a chiral @xmath1-wave superconductor and treat weak impurities perturbatively and in the gaussian ( i.e. born ) approximation . in this case , the contribution to the zeroth and first order in the impurity concentration vanishes .
we will show that diffractive skew scattering , represented by crossed diagrams ( 2b-2d ) of fig .
[ fig : pertdiagssecond ] , contributes to the same order as diagrams in the noncrossing approximation , ( 2a ) and ( 2e-2o ) in fig .
[ fig : pertdiagssecond ] . * _ model and assumptions . _ * we employ the following 2d mean - field bogoliubov - de gennes hamiltonian @xmath17 to describe the single band chiral @xmath1-wave superconductor under consideration . here
, @xmath10 is the mean - field superconducting amplitude , @xmath18 is the fermi momentum and @xmath19 determines the chirality of the superconductor .
pauli matrices in nambu space are denoted by @xmath20 .
the dispersion relation ( dr ) @xmath21 is assumed to be isotropic @xmath22 . while we derive and present all results for a generic dr
, we will additionally discuss our findings for a parabolic band @xmath23 .
we remind the reader , that sr@xmath2ruo@xmath3 is a layered material and that the conduction mainly takes place in the ru - o planes . the model hamiltonian should be a good description of the cylindrical @xmath24-sheet in sr@xmath2ruo@xmath3 .
@xcite in addition to eq .
our model contains point - like impurities of strength @xmath25 and density @xmath26 that we treat in the born approximation .
then , the disorder potential @xmath27 , which enters the hamiltonian as @xmath28 follows to have a gaussian white noise distribution @xmath29 [ eq : model ] in our notation , @xmath30 is the density of states ( dos ) at the fermi level and @xmath11 the elastic scattering time , both taken in the normal phase .
we consider a superconductor in the bcs limit in a degenerate electron gas with rare impurities .
these assumptions correspond to the following hierarchy of energy scales : @xmath31 here , @xmath32 is the temperature , @xmath9 the ac frequency , @xmath33 the fermi velocity and we set boltzmann s and planck s constants as well as the speed of light to unity @xmath34 .
our calculations are perturbative in impurity concentration , with the leading contributions being of second order .
furthermore , we keep only terms up to zeroth order in the small parameter @xmath35}/{\mathcal e_f } \ll 1 $ ] . *
_ calculation .
_ * since all diagrams to zeroth and first order in impurity concentration vanish,@xcite we concentrate on second order contributions , see fig .
[ fig : pertdiagssecond ] .
to second order in impurity concentration for the model defined by eqs . .
diagrams ( 2a ) and ( 2e)-(2h ) were presented in ref .
[ ] .
diagrams ( 2i)-(2o ) are zero .
diagrams ( 2b)-(2d ) are the diffractive contributions which are the major focus in this work .
diagram @xmath36 : mercedes star " diagram@xcite occuring for a model with non - gaussian disorder . ]
we are interested in the response of the @xmath1-wave superconductor to a vector potential , which slowly varies on the length scale of the coherence length .
such a slow vector potential does not enter the momentum dependent order parameter .
the physical reason is that slow electromagnetic fields can not resolve the relative momentum of the electrons forming the cooper pair . technically , this is a consequence of @xmath37 gauge invariance , keeping in mind that the order parameter field transforms as a bilinear of two creation operators .
the current vertex is thus ( electron charge @xmath38 ) @xmath39 for a generic dr , the last equation is valid to leading order in @xmath40 while it is exact for a parabolic band .
we now outline the calculation of the ac hall response , more details can be found in ref .
[ ] .
we use the matsubara green s functions @xmath41^{-1 } = n_{\textbf{\textit p}}(\epsilon_n ) \mathcal g(\epsilon_n , { \textbf{\textit p}}),\end{aligned}\ ] ] with fermionic frequency and momentum @xmath42 and @xmath43 , \\ & & \hskip-.5 cm \mathcal g(\epsilon_n , { \textbf{\textit p } } ) = \frac{1}{\epsilon_n^2 + \xi_{{\textbf{\textit p}}}^2 + ( { \textbf{\textit p}}\delta_0 /p_f)^2 } .\end{aligned}\ ] ] [ eq : cleangf ] we also need real space expressions for the green s function and for @xmath44 to order @xmath45
@xmath46 in eq .
, @xmath47 denotes the zeroth bessel function of the first kind .
the dimensionless function @xmath48 represents an off - shell contribution and therefore depends on microscopic details of the model . in the case of a quadratic dispersion we find @xmath49 , where @xmath50 is the zeroth bessel function of the second kind .
these expressions are valid for length scales @xmath51 , i.e. the regime of length scales which of relevance for nonvanishing diagrams ( 2a)-(2h ) , see e.g. eq . .
the transverse current - current correlator @xmath52 is evaluated at finite photon frequency @xmath53 .
we first evaluate diagrams ( 2i-2o ) of fig .
[ fig : pertdiagssecond ] . in view of the antisymmetrization
@xmath54/2 $ ] it is readily seen that diagrams ( 2i-2o ) identically vanish after angular momentum integration .
we next concentrate on the other diagrams in the noncrossing approximation .
diagram ( 2a ) contributes @xmath55 we expanded the density of states near the fermi surface as @xmath56 and introduced the dimensionless constant @xmath57 as well as the function @xmath58 \notag \\ & \times \left [ \frac{1}{\sqrt{(\epsilon_n+ \omega_l)^2 + \delta_0 ^ 2 } } -\frac{1}{\sqrt{\epsilon_n^2 + \delta_0 ^ 2 } } \right ] ^2 \bigg \rbrace.\end{aligned}\ ] ] similarly , the evaluation of diagrams ( 2e-2h ) yields @xmath59 we now turn our attention to the crossed diagrams ( 2b-2d ) . their contribution is @xmath60 with @xmath61 \notag \\ & \times \left [ \frac{1}{\sqrt{(\epsilon_n + \omega_l)^2 + \delta_0 ^ 2}}-\frac{1}{\sqrt{\epsilon_n ^2 + \delta_0 ^ 2}}\right ] ^2\bigg \rbrace.\end{aligned}\ ] ] in eq .
we introduced a nonuniversal constant @xmath62 ^ 3 \;f(\rho ) \notag \\ & + 3 \int_0^\infty d\rho \;\partial_{\rho}^2 j_0(\rho)\ ; \partial_{\rho } [ j_0(\rho)]^2 \;f(\rho ) \big \rbrace .
\label{eq : beta}\end{aligned}\ ] ] the integration variable @xmath63 denotes the distance between the two impurities of diagrams ( 2b)-(2d ) . for the general dispersion relation we expect @xmath64 , while for the specific case of a parabolic band we find @xmath65 . also , note that the integral is determined by lengthscales @xmath66 , i.e. by length scales much smaller then the coherence length .
we conclude this section with a comment on the role of the particle - hole ( ph ) transformation , i.e. of interchanging electronic creation and annihilation operators . as explained in ref .
[ ] , for our model this transformation is equivalent to mapping @xmath67 and @xmath68 .
it was also shown there , that ph symmetry in the normal phase , i.e. @xmath69 , combined with the fact @xmath70 which follows from the time reversal operation , implies @xmath71 for our model with gaussian disorder .
furthermore , a generic dr , treated in the linearized approximation , @xmath72 , is ph symmetric upon a redefintion of momenta . by consequence , since the contributions presented in eqs . and stem from the fermi surface
, they will vanish whenever @xmath73 .
similarly , if the system has ph invariance , @xmath74 and thus @xmath75 and @xmath76 , eqs . and , will be zero in accordance with the general arguments exposed in this section .
* _ results .
_ * evaluating matsubara sums in expressions for @xmath77 and @xmath78 followed by an analytical continuation @xmath79 one finds @xmath80 with dimensionless functions @xmath81 } - i\frac{1}{2 } \tanh\left ( \frac{\delta_0}{2 t}\right ) \left [ \sqrt{\frac{z}{z-2 } } + \sqrt { \frac{z}{z+2 } } \right ] \notag \\ & + 2 \int_1^\infty \frac{dx } { 2\pi } \tanh\left ( \frac{x \delta_0}{2t}\right ) \sqrt{x^2 -1 } \left [ \frac{1}{(2x + z)[1 - ( x+ z)^2]}-\frac{1}{(2x - z)[1 - ( x- z)^2]}\right ] , \\
h(z ) & = \int_1^\infty dx \frac{\tanh\left ( \frac{x \delta_0}{2t}\right ) } { \pi\sqrt{x^2 - 1 } } \left [ \frac{3x + 2z}{1-(x+z)^2}- \frac{3x - 2z}{1-(x - z)^2}\right ] - \frac{\tanh\left ( \frac{\delta_0}{2 t}\right ) } { 2 } \left [ \frac{z - 3}{\sqrt{2z - z^2 } } + \frac{z + 3}{\sqrt{-2z - z^2}}\right ] .\end{aligned}\ ] ] in this notation , the contribution of noncrossing diagrams ( 2a ) , ( 2e-2h ) to the hall response is @xmath82 the contribution of crossed diagrams ( 2b-2c ) to the hall response , i.e. the major result of this work , is @xmath83 [ eq : finalresult ] the total hall response is the sum of eqs . and .
while above two contributions have very different functional dependence on temperature and on the ac frequency their asymptotic behavior is close . indeed , for @xmath84 the limiting cases of functions @xmath85 and @xmath86 with @xmath87 are @xmath88 * _ discussion . _ * first , we would like to dwell on the physical meaning of the diagrammatic calculation . in the inset of fig .
[ fig : hallplot ] , the quantum mechanical probability for connecting source and drain , @xmath89 , is depicted .
amplitudes @xmath90 ( their complex conjugate @xmath91 ) are represented in red ( blue ) .
also notice the `` anomalous '' propagation with two opposite arrows on a single line .
it represents reflection off the condensate and , as a consequence of averaging over the fermi surface , occurs in the vicinity of the current vertex.@xcite the proportionality @xmath92 immediately follows .
concerning the diffractive , crossed diagrams , eq .
, recall that those involve a prefactor @xmath75 which is determined by the function @xmath48 and thus stems from virtual ( off - shell ) processes . as a consequence , by means of heisenberg s incertainty principle , @xmath93
is determined by impurities residing about one fermi wavelength @xmath15 from each other , i.e. from impurity complexes represented by yellow areas in fig .
[ fig : hallplot ] .
interestingly , those impurity complexes act similarly to strong impurities .
indeed , diagrams ( 2b-2d ) have the same functional form , eq . , as the mercedes star " diagram@xcite @xmath36 in fig . [
fig : pertdiagssecond ] , which involves the third moment in the distribution of @xmath27 .
however , in contrast to diagram @xmath36 , we repeat that @xmath48 and thus @xmath93 vanish for a strictly ph symmetric model , in accordance with general arguments@xcite reviewed above . on the basis of these considerations , the relative importance of the diffractive contribution , eq . , as compared to the previously known result , eq . , is apparent : @xmath94 in a model for which the dos is nearly constant @xmath95 and consequently the diffractive contribution can be parametrically enhanced as compared to other impurity - induced processes computed within ladder approximation . in particular , in the case of a parabolic band , the hall response is finite , see fig .
[ fig : hallplot ] , as compared to the vanishing result that one obtains from noncrossing diagrams.@xcite * _ summary .
_ * the microscopic origin and quantitative understanding of the kerr effect in trs - broken state of unconventional superconductors remains as a topic of ongoing debate and active research .
existing calculations in various models and initial assumptions yield very different results concerning the functional dependence of anomalous ac hall conductance on essential parameters such as frequency , electronic mean free path , and temperature . in particular , in the experimentally relevant frequency range , @xmath96 , two - band model calculations in the clean limit result in the quadratic decay of @xmath13 with inverse frequency @xmath97 .
@xcite in contrast , impurity based calculations performed either in the model of non - gaussian disorder or without particle - hole symmetry predict @xmath98 scaling from skew - scattering , see eq . and refs .
[ ] .
however , each of these two extrinsic mechanisms implies different dependence on impurity concentration as @xmath26 and @xmath99 respectively . the most recent full @xmath32-matrix analysis @xcite uncovered that skew scattering of low - energy quasiparticle on strong impurities results in anomalous hall response being linearly proportional to @xmath100 and falling off linearly with inverse frequency .
the kinetic approach of ref .
[ ] is however limited to low frequencies , @xmath101 . furthermore
, since quasiparticle density decreases exponentially fast with temperature , this mechanisms dominates kerr rotation only in the immediate vicinity of the critical temperature @xmath102 .
apart from its purpose in the context of @xmath1-wave superconductors , our work can be seen as a proof of principle for the importance of diffractive skew scattering ( crossed diagrams ) beyond the context of the dc ahe . in particular , our study of the ac ahe evokes a similar investigation for time reversal symmetry breaking superconductors with other unconventional order parameter symmetries .
this is also motivated by recent kerr measurements in high-@xmath103 cuprates yba@xmath2cu@xmath104o@xmath105
@xcite and la@xmath106ba@xmath107cuo@xmath3 , @xcite and heavy fermion superconductors upt@xmath104 @xcite and uru@xmath2si@xmath2 @xcite , that already triggered new theories .
@xcite we will devote a separate publication to this topic .
more generally , it can be expected that diffractive skew scattering plays a substantial role for a plethora of other anomalous physical observables ( e.g. the spin hall effect @xcite ) and thus constitutes a focus for future research . * _ acknowledgements . _
* we thank a. andreev , i. dmitriev , p. ioselevich , m. khodas , r. lutchyn , p. nagornykh , p. ostrovsky , d. pesin , j. sauls and v. yakovenko for useful discussions , and acknowledge hospitality by the department of physics at university of michigan ( e.j.k . ) .
this work was financially supported in part by nsf grants no .
dmr-1606517 and eccs-1560732 .
support for this research at the university of wisconsin - madison was provided by the office of the vice chancellor for research and graduate education with funding from the wisconsin alumni research foundation .
34ifxundefined [ 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 * * , ( ) y. liu and z .- q .
mao , physica c : superconductivity and its applications * 514 * , 339 - 353 ( 2015 ) . c. kallin and j. berlinsky , rep .
. phys .
* 79 * , 054502 ( 2016 ) . @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) phys .
rev .
lett . * 117 * , 046601 ( 2016 ) .
phys .
rev .
b * 94 * , 041403(r ) ( 2016 ) . @noop * * , ( ) @noop * * , ( ) , @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) in @noop _ _ , vol .
( , ) p. @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) supplemental materials to this paper . j. xia , e. schemm , g. deutscher , s. a. kivelson , d. a. bonn , w. n. hardy , r. liang , w. siemons , g. koster , m. m. fejer , and a. kapitulnik , phys .
rev .
lett .
* 100 * , 127002 ( 2008 ) .
h. karapetyan , jing xia , m. hucker , g. d. gu , j. m. tranquada , m. m. fejer , and a. kapitulnik , phys .
rev .
lett . * 112 * , 047003 ( 2014 ) .
e. r. schemm , w. j. gannon , c. m. wishne , w. p. halperin , a. kapitulnik , science * 345 * , 190 ( 2014 ) .
e. r. schemm , r. e. baumbach , p. h. tobash , f. ronning , e. d. bauer , and a. kapitulnik , phys .
rev .
b * 91 * , 140506(r ) ( 2015 ) .
y. wang , a. chubukov , and r. nandkishore phys .
rev .
b * 90 * , 205130 ( 2014 ) .
@noop ( ) * supplemental materials to * * `` kerr effect from diffractive skew scattering in chiral @xmath108 superconductors '' * + elio j. knig and alex levchenko |
seeking tractable solutions of nonlinear evolution equations has been the focus of intense study for the past several decades ( see , e.g. , @xcite-@xcite ) .
while many strategies have been employed , such as the hirota bilinear method @xcite , homogeneous balance method @xcite , trigonometric method @xcite , the hyperbolic method @xcite , the jacobi elliptic function method @xcite and iterative method @xcite to name a few , perturbative techniques have also been used to extract multi - order exact periodic solutions based on lam@xmath1 equation and jacobi elliptic functions @xcite . in this note
we propose a generalized class of lam@xmath1 equation comprising an auxiliary function @xmath2 to consider correlations between it and the perurbatively reduced nonlinear evolution equation for different orders of the perturbative parameter .
a generalized lam@xmath1 equation is expressed as @xmath3y=0,\ ] ] where @xmath4 stands for an eigenvalue , @xmath5 is a positive integer and we restrict the auxiliary function @xmath2 to obey an elliptic equation @xmath6 in ( [ proposedode ] ) and ( [ elip ] ) , @xmath7 are real constants .
it is well known that equation ( [ elip ] ) admits of several categories of solutions for different values of @xmath7 all expressible in terms of modulus @xmath8 of the jacobi elliptic functions . for later use
we summarize a few relevant ones in table [ table1 ] .
.@xmath2 is provided along with the values of the parameters @xmath9 .
the complementary modulus is denoted by @xmath10 @xcite . [
cols="^,^,^,^ " , ] [ table1 ] it should be mentioned that while we would attend to the enlisted set of @xmath2 in table [ table1 ] to widen the scope of enquiry , some particular cases of @xmath2 , namely @xmath11 and @xmath12 , have been studied in the literature @xcite to arrive at solutions of the relevant pdes by applying the perturbation method on the latter . however , as is evident from table [ table1 ] there do remain other variants of @xmath2 which open up different cases of new solutions not only for the already considered pdes but other types as well by perturbatively reducing them to an ode form so as to match with ( [ proposedode ] ) .
we have selected a physically meaningful model of a generalized shallow water wave equation @xcite which is amenable to a perturbative treatment by looking for a travelling wave variable @xmath13 with @xmath14 where @xmath15 is the wave number and @xmath16 is the wave speed .
we discuss the procedure of obtaining the solutions in the next section . in the following section we also take up applicability of our scheme to other nonlinear evolution equations namely , the korteweg de vries ( kdv ) equation , modified kdv equation , boussinesq equation , klein - gordon equation and modified benjamin - bona - mahony ( mbbm ) equation with a view to tracking down new hitherto unexplored multi - order solutions .
the main point of our analysis is to demonstrate that all such solutions can be written down in a generic form in terms of the function @xmath0 . to make equation ( [ proposedode ] ) accessible in terms of known lam@xmath1 functions
, we recast it in terms of a new variable @xmath17 by applying the transformation @xmath18 . using ( [ elip ] )
this gives @xmath19\frac{dy}{d\eta } -\frac{\nu+(n+1)(n+2)\eta}{4\eta(\eta+\frac{1}{a_{1}})(\eta+\mu)}y=0,\ ] ] where the parameters @xmath20 and @xmath21 are defined by @xmath22 and @xmath23 .
equation ( [ lam ] ) is readily solvable for the special cases of @xmath24 and @xmath25 for different choices of @xmath20 and @xmath21 subject to certain relation between them . the solutions expressible by the corresponding lam@xmath1 functions are @xcite @xmath26.\nonumber \end{aligned}\ ] ] @xmath27.\nonumber \end{aligned}\ ] ] @xmath28.\nonumber \end{aligned}\ ] ] @xmath29.\nonumber \end{aligned}\ ] ] henceforth we will be guided by the solutions ( [ fun2 ] ) - ( [ fun1 ] ) along with an appropriate @xmath0 from table 1 for the various multi - order cases that follow from a nonlinear pde .
we turn attention to a generalized class of shallow water wave ( gsww ) equation given by @xcite @xmath30 where @xmath31 under boussinesq approximation the derivation of ( [ gsww ] ) results from the classical study of water waves .
more interestingly , there also follows various classical and non - classical reductions from gsww such as the kdv and bbm equations .
investigations of painlev@xmath1 tests reveal complete integrability for specific values of @xmath32 and @xmath33 namely @xmath34 or @xmath35 @xcite .
recently we extended @xcite the jacobian elliptic function method to classify new exact travelling wave solutions expressible in terms of quasi - periodic elliptic integral function and doubly - periodic jacobian elliptic functions .
for the travelling wave solutions @xmath36 , equation ( [ gsww ] ) can be reduced in terms of the variable @xmath37 to the ode form @xmath38 where @xmath39 , @xmath40 and @xmath41 is an integrating constant .
to tackle ( [ rgsww ] ) perturbatively we set @xmath42 where @xmath43 is a small parameter and @xmath44 represent various multi - order solutions like the zeroth - order , first - order , second - order solutions etc . of equation ( [ rgsww ] ) .
accordingly we can write @xmath45 as @xmath46 where @xmath47 substituting the series ( [ pert ] ) in equation ( [ rgsww ] ) we obtain for each power of @xmath48 the corresponding equations @xmath49 and so on .
we now proceed to solve the above chain of equations by first expanding @xmath50 namely , @xmath51 where @xmath52 s are constants and then comparing highest order linear and nonlinear terms in ( [ reduode ] ) . in this way
we obtain @xmath53 by making use of ( [ elip ] ) .
thus ( [ z - ord ] ) gets reduced to the quadratic form @xmath54 with @xmath50 given by ( [ z - ord1 ] ) we are led to the following system of coupled equations : @xmath55 & = & 0\ , , \label{syst1-b}\\ 4(a_{2}+a_{1}a_{3})a_{2}+p(a_{1}^{2}+2a_{0}a_{2})+qa_{2 } & = & 0\ , , \label{syst1-c } \\
a_{1}[a_{1}a_{2}+pa_{2 } ] & = & 0 \ , , \label{syst1-d}\\ a_{2}[6a_{1}a_{2}+pa_{2 } ] & = & 0.\ , \label{syst1-e } \end{aligned}\ ] ] it is readily seen that we generate the following set of consistent solutions @xmath56 $ ] , @xmath57 and @xmath58 along with the constraint @xmath59-q^{2}+4pc_{1}=0.\ ] ] from ( [ cons ] ) we obtain wave speed as @xmath60^{-1}\ ] ] where the two signs signal the two directions .
note that in order to have @xmath16 real we can always adjust the integration constant @xmath41 to keep @xmath61 .
further for the finiteness of @xmath16 we require @xmath62 . from ( [ reduode ] ) and ( [ z - ord1 ] )
we get for the first integral @xmath63-\frac{6a_{1}a_{2}}{p}f^{2}\ ] ] which in turn gives the zeroth - order solution @xmath64\xi-\frac{6a_{1}a_{2}}{p}\int f^{2}~d\xi.\ ] ] as is evident , @xmath65 depends upon the choice of the auxiliary function @xmath0 . in table
[ table2 ] we furnish the various forms for @xmath65 corresponding to different elliptic functions of table [ table1 ] . ,
width=302,height=226 ] vs constant of integration @xmath66 for @xmath67 , width=226,height=151 ] [ table2 ] a sample graph of the zeroth - order solution @xmath65 for @xmath68 is depicted in figure [ fig1 ] and a plot of wave - speed for variation of the constant @xmath41 is shown in figure [ fig2 ] .
knowing @xmath50 from ( [ v0 ] ) , equation ( [ f - ord - eq ] ) reads @xmath69v_{1}=0,\ ] ] which matches with ( [ proposedode ] ) for @xmath70 .
this gives @xmath71 implying that the solution of ( [ r - f - ord - eq ] ) can be written as @xmath72 where @xmath73 is an arbitrary constant .
integration of ( [ v1 ] ) with the use of ( [ elip ] ) immediately provides @xmath74 which serves as a first order approximation to equation ( [ gsww ] ) .
figure [ fig3 ] gives a plot of @xmath75 for @xmath68 .
, width=302,height=226 ] using ( [ v0 ] ) and ( [ v1 ] ) the second - order equation ( [ s - ord - eq ] ) takes the form @xmath76v_{2}=-pc_{2}^{2}f^{2}(1+a_{1}f^{2})(a_{3}+a_{2}f^{2}).\ ] ] we observe that the coefficients of ( [ s - ord ] ) are polynomials in @xmath0 .
this prompts us to take a polynomial ansatz for @xmath77 namely @xmath78 substitution of the above into ( [ s - ord ] ) we get @xmath79 on equating the highest power of @xmath0 from both sides .
thus ( [ v2-series ] ) is reduced to the biquadratic form @xmath80 putting ( [ ans ] ) into ( [ s - ord ] ) and equating the coefficients of @xmath81 to zero , we have the following system of coupled equations , @xmath82 solving for @xmath83 we obtain @xmath84 , width=302,height=226 ] hence the solution of equation ( [ s - ord ] ) can be written as @xmath85\ ] ] using ( [ elip ] ) we can integrate ( [ v2 ] ) to get @xmath86 @xmath87 is plotted for @xmath68 in figure [ fig4 ] .
in the previous section , we considered a nonlinear second - order ode containing a functional parameter @xmath0 and obtained specific solutions in terms of @xmath0 .
it was shown that the use of jacobi elliptic function method in a perturbative way yielded new multi - order exact solutions that depended upon @xmath0 in a generic way . in this section
we show that our approach can be profitably applied to other nonlinear equations resulting in more general types of solutions that have not been reported to the best of our knowledge .
a complete list of our results are presented in table [ table3 ] .
* @xmath88 kdv equation * + the concerned equation is of the form @xmath89 where @xmath32 and @xmath33 are real parameters .
let us choose @xmath90 for which @xmath91 . substituting these values into the zeroth - order , first - order and second - order exact solutions listed in table 3 , we find @xmath92.\nonumber \end{aligned}\ ] ]
these solutions have been obtained in @xcite for @xmath93 where the authors used the notations @xmath94 and @xmath8 in places of @xmath8 and @xmath15 respectively and their scale is @xmath95 .
the kdv equation has also been studied in @xcite and their results can be extracted from ours for @xmath68 corresponding to @xmath96 .
+ + + * @xmath97 mkdv equation * + @xmath98 our general result listed in table [ table3 ] contains , as special cases , the following results reported in earlier works .
* @xmath68 ( ref @xcite ) + @xmath99\nonumber \end{aligned}\ ] ] * @xmath90 ( ref @xcite ) + @xmath100\nonumber \end{aligned}\ ] ] * @xmath101 boussinesq equation * + @xmath102 from the general solution given in terms of @xmath0 ( see table [ table3 ] ) one can deduce the ones obtained in @xcite as special case for @xmath103 .
+ + + * @xmath104 klein - gordon equation * + @xmath105 the above equation was considered in @xcite .
their results correspond to the case @xmath106 and we do indeed recover their solutions from table 3 as given below : @xmath107\nonumber \end{aligned}\ ] ] * @xmath108 mbbm equation * + @xmath109 for @xmath68 the multi - order solutions to ( [ mbbm ] ) read from table 3 @xmath110.\nonumber \end{aligned}\ ] ] these results were obtained in ref @xcite .
we have reported exact multi - order periodic solutions for a generalized shallow water wave equation based on various types of jacobi elliptic functions in a perturbative framework .
the presence of an auxiliary function , which makes reference to any particular jacobi function explicit , can be well adjusted to connect with other classes of pde such as the kdv , modified kdv , boussinesq , klein - gordon and modified benjamin - bona - mahony equation .
the general character of our results encompasses those special cases which have earlier been studied in the literature .
99 cavaglia a. , fring a. and bagchi b. 2011 _ j. phys . a : math . and theor .
_ * 44 * 325201 .
hietarinta j. introduction to the hirota bilinear method _
( arxiv : solv - int/9708006 ) _ and references therein .
1996 _ _ phys .
a__**216 * * 67 .
1996 _ phys .
a _ * 224 * 77 .
malfliet w and hereman w 1996 _ phys .
_ * 54 * 563 , 569 .
parkes e.j . , duffy b.r . and abbott p.c .
2002 _ phys .
a _ * 295 * 280 .
daftardar - gejji v and jafari h 2006 _ j. math .
appl . _ * 316 * 753 .
bhalekar s and daftardar - gejji v 2008 _ appl .
_ * 203 * 778 .
liu s. , fu z. , liu s. and wang z. 2004 _ chaos soliton fractals _ * 19 * 795 . fu z. , yuan n. , mao j. and liu s. 2009 _ phys . lett .
a _ * 374 * 214 .
guan - ting l. 2009 _ appl .
math . and computation _ * 212 * 312
. whittaker e.t . and watson g.n .
1952 _ modern analysis _
( cambridge university press ) .
hietarinta j 1990 partially integrable evolution equations in physics _
( nato asi series c : mathematical and physical sciences vol .
310 ) _ ed r conte and n boccara ( dordrecht : kluwer ) 459 .
clarkson p.a . and
mansfield e.l .
1994 _ nonlinearity _ * 7 * 975 .
chen h. and zhang h. 2003 _ chaos soliton fractals _ * 15 * 585 .
bagchi b. , das s. and ganguly a. 2010 _ physica scrypta _ * 82 * 025003 . | we seek multi - order exact solutions of a generalized shallow water wave equation along with those corresponding to a class of nonlinear systems described by the kdv , modified kdv , boussinesq , klein - gordon and modified benjamin - bona - mahony equation .
we employ a modified version of a generalized lame equation and subject it to a perturbative treatment identifying the solutions order by order in terms of jacobi elliptic functions .
our multi - order exact solutions are new and hold the key feature that they are expressible in terms of an auxiliary function @xmath0 in a generic way . for appropriate choices of @xmath0
we recover the previous results reported in the literature .
+ keywords : * _ travelling waves ; lam@xmath1 equation ; lam@xmath1 function ; jacobi elliptic function ; elliptic integral function . _ * |
recent experiments on the production of @xmath13 mesons in the reaction @xmath14 , where @xmath15 is a hadronic state with strangeness @xmath16 and baryon number @xmath17 and @xmath18 is a nucleon , have led to the prediction for the ratio of the cross sections for the reactions @xmath19 and @xmath20 @xmath21 the estimate has been obtained for proton beam energies @xmath22 and @xmath23 giving @xmath24 and @xmath25 , respectively , to fit the experimental data on the differential cross section for the reaction @xmath14 @xcite .
the parameter @xmath26 has been used as an input parameter .
this has been justified by the fact that the scattering of the proton by the deuteron can be treated in the _ impulse _ approximation @xcite , since there is no indication for any collective target behaviour @xcite ( see also @xcite ) . since the main contribution to the cross sections for the reactions @xmath27 and @xmath20 comes from the channels @xmath28 and @xmath29 @xcite , the ratio eq.([label1.1 ] )
can be rewritten as follows @xmath30 can such a ratio be valid near threshold of the final state @xmath5 ?
@xcite .
a theoretical explanation of the ratio @xmath31 near threshold of the final @xmath32 state has been proposed in @xcite and @xcite . as has been pointed out in @xcite
, the value @xmath33 can be mainly due to the dominant contribution of the @xmath34meson exchange in addition to the contribution of the resonances @xmath35 and @xmath36 with the quantum numbers @xmath37 @xcite , described within the breit
wigner approach @xcite . in @xcite
the cross sections for the reactions @xmath38 have been calculated within the resonance model of @xmath39 scattering for energies of the incident proton far from threshold .
within such an approach the ratio @xmath26 can be obtained assuming the validity of the cross sections in the vicinity of threshold @xcite .
recently @xcite the strangeness production in @xmath4 collisions @xmath0 and @xmath40 near threshold of the final states has been analysed by using chiral lagrangians with linear realization of chiral symmetry and non derivative meson
baryon couplings @xcite@xcite .
the matrix elements of the transitions @xmath41 and @xmath42 have been calculated in the one meson exchange approximation @xcite . for the calculation of the amplitudes of the reactions @xmath0 and @xmath43 there has been taken into account the @xmath4 rescattering in the initial state , i.e. the initial state interaction @xcite , and @xmath44 and @xmath45 final state interaction @xcite .
as has been shown in @xcite the same analysis of the strangeness production has turned out to be very useful for the correct description of the reactions @xmath46 of @xmath47 scattering at threshold , where @xmath48 , @xmath49 and @xmath50 . in this paper
we apply such an analysis of the strangeness production to the calculation of the cross sections for the reactions @xmath38 near threshold of the final state @xmath5 and the ratio @xmath26 .
the paper is organised as follows . in section 2
we calculate the cross section for the reaction @xmath0 .
the obtained result agrees with the experimental data @xcite with an accuracy better @xmath51 .
we show that the contribution of the vector @xmath34 and @xmath52meson exchanges to the cross section for the reaction @xmath0 makes up about @xmath53 only . in section 3
we calculate the cross section for the reaction @xmath54 .
we calculate the cross sections for the reactions with the @xmath3 pair in the state with isospin @xmath55 and isospin @xmath56 and demonstrate that @xmath57 .
we find that the contribution of the vector @xmath34 and @xmath52meson exchanges to the cross section for the reaction @xmath58 with the @xmath3 pair in the isospin
triplet state is about @xmath59 . in turn , the vector
meson exchanges do not influence on the value of the cross section for the reaction @xmath60 with the @xmath3 pair in the isospin singlet state . in section 4
we calculate the ratio @xmath26 of the cross sections for the reactions @xmath61 and @xmath0 .
we get @xmath62 .
this agrees with the enhancement of the @xmath3 interaction with respect to the @xmath4 interaction in the reactions of the strangeness production observed by the anke collaboration at cosy and the theoretical prediction in @xcite and @xcite .
we show that such an enhancement is caused by the contribution of @xmath3 interaction in the isospin
singlet state @xmath63 , which is stronger than the @xmath3 interaction in the isospin triplet state @xmath64 . in section 5
we analyse the contribution of the resonances @xmath35 and @xmath36 , treating them as elementary particles @xcite@xcite .
we show that the contribution of these resonances to the amplitudes of the reactions @xmath38 can be neglected with respect to the contribution of the ground state baryon octet coupled to the octets of the pseudoscalar and scalar mesons . in section 6
we calculate the cross section for the reaction @xmath7 , caused by the pure @xmath65 transition with pseudoscalar and scalar meson exchanges .
near threshold the theoretical cross section agrees well with the experimental data . in section 7
we calculate the cross sections for the reactions @xmath66 and @xmath9 .
we show that the theoretical cross section for the reaction @xmath9 agrees well with the experimental data for the energy excess @xmath67 @xcite and the results obtained within @xmath68 chiral dynamics with coupled channels @xcite .
we show that for the correct description of the cross sections for the reactions @xmath69 and @xmath9 the contribution of the resonances @xmath35 and @xmath36 as well as @xmath70 is very important due to the cancellation of the non resonance contribution
. the dominant role of resonances @xmath35 and @xmath36 in the amplitudes of the reactions @xmath71 and @xmath9 has been pointed out by kaiser _
_ @xcite . in the conclusion
we discuss the obtained results . in appendix
a we analyse the contribution of the final
state interaction to the amplitudes of the reactions @xmath72 .
following @xcite and @xcite we define the amplitude of the transition @xmath74 in the tree approximation with one meson exchanges .
a complete set of feynman diagrams , describing the transition @xmath74 near threshold of the final @xmath75 state , is depicted in fig.1 and fig.2 . the calculation of these diagrams we carry out using chiral lagrangian with non derivative meson baryon couplings invariant under linear transformations of chiral @xmath76 symmetry @xcite@xcite . since masses of scalar partners of pseudoscalar mesons are not well
defined , we calculate the amplitudes in the limit of infinite masses of scalar mesons . according to @xcite , such a limit corresponds to the calculation of the amplitudes with chiral lagrangian invariant under non linear transformations of chiral @xmath76 symmetry@xcite .
of course , such an equivalence has been proved only in the tree
approximation @xcite@xcite . for the derivation of the analytical expressions of these diagrams we use the meson baryon interactions ( see eqs.(d.4 ) , ( d.7 ) and ( d.9 ) of ref.@xcite and also @xcite ) and the wave functions of the initial and the final states in the particle number representation : @xmath77 where creation ( annihilation ) operators of baryons and the @xmath78meson obey standard relativistic covariant anti commutation and commutation relations @xcite . near threshold of the final state @xmath44 the feynman diagrams in fig.1 and fig.2 can be described by the effective local lagrangian of the transition @xmath0 @xcite : @xmath79[\bar{p}(x)i\gamma^5 p(x)]\,k^{+\dagger}(x)\nonumber\\ & + & b^{pp}_{p\lambda^0k^+}[\bar{\lambda}^0(x)i\gamma^5 p(x)][\bar{p}(x)p(x)]\,k^{+\dagger}(x),\end{aligned}\ ] ] where the coefficients @xmath80 and @xmath81 are equal to @xmath82 where @xmath83 is the coupling constant of @xmath84 interactions @xcite , @xmath85 is the axial vector coupling constant , renormalized by strong low
energy interactions , @xmath86 is the pcac constant and @xmath87 is the gell
mann parameter @xcite , @xmath88 is the kinetic energy of the relative motion of the protons at threshold in the center of mass frame , calculated for @xmath89 , @xmath90 and @xmath91 .
we have used also @xmath92 , @xmath93 and @xmath94 @xcite . by a fierz transformation
we get @xmath95[\bar{p}(x)i\gamma^5 p(x ) ] } & = & - \frac{1}{4}\ , [ \bar{\lambda}^0(x)i\gamma^5 p^c(x)][\bar{p^c}(x)p(x)]\nonumber\\ & & + \frac{1}{4}\,i\ , [ \bar{\lambda}^0(x)\gamma_{\mu } p^c(x)][\bar{p^c}(x)\gamma^{\mu } \gamma^5 p(x ) ] + \ldots , \nonumber\\ { [ \bar{\lambda}^0(x)i\gamma^5p(x)][\bar{p}(x)p(x ) ] } & = & - \frac{1}{4}\ , [ \bar{\lambda}^0(x)i\gamma^5 p^c(x)][\bar{p^c}(x)p(x)]\nonumber\\ & & - \frac{1}{4}\,i\ , [ \bar{\lambda}^0(x)\gamma_{\mu } p^c(x)][\bar{p^c}(x)\gamma^{\mu } \gamma^5 p(x ) ] + \ldots\,.\end{aligned}\ ] ] hence , the effective lagrangian of the transition @xmath96 is equal to @xmath97[\bar{p^c}(x)p(x)]\,k^{+\dagger}(x)\nonumber\\ & & + \,\frac{1}{4}\ , c^{(pp)_{{^3}{\rm p}_1}}_{(p\lambda^0)_{{^3}{\rm s}_1}k^+}\ , [ \bar{\lambda}^0(x)i\gamma^{\mu } p^c(x)][\bar{p^c}(x)\gamma_{\mu}\gamma^5 p(x)]\,k^{+\dagger}(x),\end{aligned}\ ] ] where the first and the second terms describe the production of the @xmath98 pair in the @xmath99 and @xmath100 state by the @xmath4 pair in the @xmath101 and @xmath102 state , respectively .
the effective coupling constants @xmath103 and @xmath104 are equal to @xmath105 the total cross section for the reaction @xmath106 is given by @xmath107 where @xmath108 is the excess energy for the relative energy of the @xmath4 pair in the center of mass frame @xmath109 .
it is measured in mev .
the function @xmath110 is related to the account for the final
state interaction in the @xmath98 and @xmath111 channels defined in appendix a ( see also @xcite ) .
for @xmath112 the contribution of the coulomb repulsion in the @xmath113 pair makes up about @xmath114 .
therefore for @xmath112 we carry out the calculation of @xmath110 at the neglect of the coulomb repulsion .
the amplitudes @xmath115 and @xmath116 describe the @xmath4 rescattering in the @xmath117 and @xmath102 state at threshold of the final @xmath44 state , respectively .
the detailed procedure for the calculation of the amplitudes @xmath115 and @xmath118 is expounded in @xcite .
following this procedure one gets @xmath119\big\}^{-1}\big| = 0.15,\nonumber\\ |f_{(pp)_{{^3}{\rm p}_1}}(p_0)| & = & \big|\big\{1 - \frac { c_{(pp)_{{^3}{\rm p}_1}}(p_0)}{12\pi^2}\ , \frac{p^3_0}{e_n}\,\big[{\ell n}\big(\frac{e_n + p_0}{e_n - p_0}\big ) -\,i\,\pi\big]\big\}^{-1}\big| = 0.24,\end{aligned}\ ] ] where the effective coupling constants @xmath120 and @xmath121 are defined by the @xmath122 and @xmath123meson exchanges and the results obtained in @xcite we can estimate the values of the phase shift @xmath124 and the inelasticity @xmath125 of @xmath4 scattering in the @xmath101 state .
we get @xmath126 and @xmath127
. this does not contradict the said analysis of the experimental data @xcite : @xmath128 and @xmath129 . ] .
for the relative momentum of the @xmath4 pair @xmath130 at threshold of the reaction @xmath106 , the effective coupling constants @xmath131 and @xmath121 amount to @xcite @xmath132 the cross section for the reaction @xmath0 is equal to @xmath133 where @xmath134 is calculated at @xmath135 ( see appendix a ) .
the theoretical value ( [ label2.10 ] ) agrees with the experimental data @xmath136 @xcite within an accuracy better than @xmath137 .
now we take into account the contribution of the @xmath138 and @xmath52mesons .
the effective lagrangian of the @xmath106 transition , caused by the vector
meson exchanges , is equal to , @xmath139 , @xmath140 , @xmath141 and @xmath142 . ]
@xmath143[\bar{p}(x ) \gamma^{\mu}p(x)]\,k^{+\dagger}(x).\end{aligned}\ ] ] by a fierz transformation we reduce the effective interaction to the form @xmath144[\bar{p}(x ) \gamma^{\mu}p(x ) ] \to -\,[\bar{\lambda}^0(x)i\gamma^5 p^c(x)][\bar{p^c}(x)p(x)]\nonumber\\ & & \hspace{2.15in}-\,\frac{1}{2}\ , [ \bar{\lambda}^0(x)i\gamma_{\mu}p^c(x)][\bar{p^c}(x ) \gamma^{\mu}\gamma^5 p(x ) ] + \ldots\,.\end{aligned}\ ] ] the contributions of the @xmath138 and @xmath52meson exchanges to the coefficients @xmath103 and @xmath104 are equal to @xmath145 for the numerical calculation we set @xmath146 @xcite . the contributions of the @xmath138 and @xmath52meson exchanges to the effective coupling constants @xmath120 and @xmath121 amount to @xmath147 this gives @xmath148 and @xmath149 .
the cross section for the reaction @xmath0 with the @xmath138 and @xmath52meson exchanges is @xmath150 thus , the @xmath138 and @xmath52meson exchanges decrease the value of the cross section for the reaction @xmath151 by about @xmath53 and lead to the agreement with the experimental data @xmath136 @xcite with an accuracy about @xmath51 .
the @xmath3 pair in the reaction @xmath54 can interact both in the isospin
triplet @xmath64 state and isospin singlet @xmath63 state . according to the _ generalised pauli exclusion principle _
, the @xmath3 pair with isospin @xmath55 can be in the @xmath101 and @xmath102 state with the @xmath50 pair in the @xmath99 and @xmath100 state , respectively , whereas the @xmath3 pair in the state with isospin @xmath153 can be in the @xmath154 state only with the @xmath50 pair in the @xmath155 state .
a complete set of feynman diagrams for the @xmath156 transition near threshold of the final state is depicted in fig.3 and fig.4 .
the wave functions of the initial and final states we take in the form @xmath157 the matrix element of the @xmath54 transition , defined by the feynman diagrams in fig.3 and fig.4 , is described by the effective local lagrangian of the transition @xmath54 : @xmath158[\bar{n}(x)i\gamma^5 p(x ) ] + b^{pn}_{n\lambda^0k^+}[\bar{\lambda}^0(x)p(x)][\bar{n}(x)i\gamma^5 n(x ) ] \nonumber\\ & + & c^{pn}_{n\lambda^0k^+}[\bar{\lambda}^0(x)i\gamma^5 n(x)][\bar{n}(x ) p(x ) ] + d^{pn}_{n\lambda^0k^+}[\bar{\lambda}^0(x)i\gamma^5 p(x)][\bar{n}(x ) n(x ) ] ) , \nonumber\\ & & \times\,k^{+\dagger}(x),\end{aligned}\ ] ] where the effective coupling constants @xmath159 , @xmath160 , @xmath161 and @xmath162 are equal to @xmath163 since at low energies the @xmath50 pair can be produced in the @xmath99 and @xmath100 state only , we have to extract these interactions from the effective lagrangian eq.([label3.2 ] ) by a fierz transformation .
we get @xmath164[\bar{n}(x)i\gamma^5 p(x ) ] } = -\,\frac{1}{4}\ , [ \bar{\lambda}^0(x)i\gamma^5 n^c(x)][\bar{n^c}(x)p(x)]\nonumber\\ \hspace{-0.3 in } & & + \,\frac{1}{4}\,[\bar{\lambda}^0(x)i\gamma_{\mu}n^c(x ) ] [ \bar{n^c}(x)\gamma^{\mu}\gamma^5 p(x ) ] -\,\frac{1}{8}\ , [ \bar{\lambda}^0(x)i\sigma_{\mu\nu } n^c(x)][\bar{n^c}(x)\sigma^{\mu\nu}\gamma^5 p(x ) ] + \ldots , \nonumber\\ \hspace{-0.3 in } & & { [ \bar{\lambda}^0(x)p(x)][\bar{n}(x)i\gamma^5 n(x ) ] } = -\,\frac{1}{4}\ , [ \bar{\lambda}^0(x)i\gamma^5 n^c(x)][\bar{n^c}(x)p(x)]\nonumber\\ \hspace{-0.3 in } & & + \,\frac{1}{4}\,[\bar{\lambda}^0(x)i\gamma_{\mu}n^c(x ) ] [ \bar{n^c}(x)\gamma^{\mu}\gamma^5 p(x ) ] + \,\frac{1}{8}\ , [ \bar{\lambda}^0(x)i\sigma_{\mu\nu } n^c(x)][\bar{n^c}(x)\sigma^{\mu\nu}\gamma^5 p(x ) ] + \ldots , \nonumber\\ \hspace{-0.3 in } & & { [ \bar{\lambda}^0(x)i\gamma^5 n(x)][\bar{n}(x)p(x ) ] } = -\,\frac{1}{4}\ , [ \bar{\lambda}^0(x)i\gamma^5 n^c(x)][\bar{n^c}(x)p(x)]\nonumber\\ \hspace{-0.3 in } & & -\,\frac{1}{4}\,[\bar{\lambda}^0(x)i\gamma_{\mu}n^c(x ) ] [ \bar{n^c}(x)\gamma^{\mu}\gamma^5 p(x ) ] -\,\frac{1}{8}\ , [ \bar{\lambda}^0(x)i\sigma_{\mu\nu } n^c(x)][\bar{n^c}(x)\sigma^{\mu\nu}\gamma^5 p(x ) ] + \ldots , \nonumber\\ \hspace{-0.3 in } & & { [ \bar{\lambda}^0(x)i\gamma^5 p(x)][\bar{n}(x ) n(x ) ] } = -\,\frac{1}{4}\ , [ \bar{\lambda}^0(x)i\gamma^5 n^c(x)][\bar{n^c}(x)p(x)]\nonumber\\ \hspace{-0.3 in } & & -\,\frac{1}{4}\,[\bar{\lambda}^0(x)i\gamma_{\mu}n^c(x ) ] [ \bar{n^c}(x)\gamma^{\mu}\gamma^5 p(x ) ] + \,\frac{1}{8}\ , [ \bar{\lambda}^0(x)i\sigma_{\mu\nu } n^c(x)][\bar{n^c}(x)\sigma^{\mu\nu}\gamma^5 p(x ) ] + \ldots.\end{aligned}\ ] ] hence , the effective lagrangian of the @xmath165 transition is equal to @xmath166[\bar{n^c}(x)p(x)]\,k^{+\dagger}(x)\nonumber\\ & & + \,\frac{1}{2}\ , c^{(pn)_{{^3}{\rm p}_1}}_{(n\lambda^0)_{{^3}{\rm s}_1}k^+}\ , [ \bar{\lambda}^0(x)i\gamma_{\mu } n^c(x)][\bar{n^c}(x)\gamma^{\mu}\gamma^5 p(x ) ] \,k^{+\dagger}(x)\nonumber\\ & & - \,\frac{1}{4}\ , c^{(pn)_{{^1}{\rm p}_1}}_{(n\lambda^0)_{{^3}{\rm s}_1}k^+}\ , [ \bar{\lambda}^0(x)i\sigma_{\mu\nu } n^c(x)][\bar{n^c}(x)\sigma^{\mu\nu}\gamma^5 p(x ) ] \,k^{+\dagger}(x),\end{aligned}\ ] ] where the first and the second terms describe the production of the @xmath50 pair in the @xmath99 and @xmath100 state by the @xmath3 pair in the isospin triplet and the @xmath101 and @xmath102 states , respectively , the third term corresponds to the interaction of the @xmath3 pair in the isospin singlet and the @xmath154 state . at threshold for the @xmath3 pair in the @xmath167 state with isospin @xmath56 the @xmath50 pair produces itself in the @xmath100 state .
the effective coupling constants are equal to @xmath168 the total cross section for the reaction @xmath169 is given by @xmath170 where the cross sections @xmath171 and @xmath172 are defined by @xmath173 the function @xmath174 takes into account the final state interaction in the @xmath50 channel ( see appendix a ) .
the amplitude @xmath175 describes the rescattering of the @xmath3 pair in the isospin singlet and @xmath167 state .
since the amplitudes of the @xmath3 rescattering in the @xmath101 and @xmath102 state , @xmath176 and @xmath177 , are equal to the amplitudes of the @xmath4 rescattering , the value of the cross section for the reaction @xmath58 is equal to @xmath178 where @xmath179 for @xmath180 ( see appendix a ) . for the cross sections of the reactions @xmath181 and @xmath73 , defined by eqs.([label2.10 ] ) and ( [ label3.9 ] ) , we get the ratio @xmath182 agreeing well with isospin invariance of strong @xmath183 interactions . according to @xcite ,
the result of the calculation of the amplitude @xmath175 is @xmath184\big\}^{-1}\big| = 0.49,\end{aligned}\ ] ] where the coupling constant @xmath185 is defined by the @xmath186 and @xmath123meson exchanges and is equal to @xmath187 @xcite .
the cross section for the reaction @xmath60 amounts to @xmath188 for the total cross section for the reaction @xmath54 we get @xmath189 where we have denoted @xmath190 now we can take into account the contribution of the vector meson exchanges .
unlike the reaction @xmath0 the contribution of the @xmath138 and @xmath52meson exchanges is cancelled and there are only the contributions of the charged @xmath34meson exchange .
the effective lagrangian , caused by the charged @xmath34meson exchanges , is equal to @xmath191[\bar{n}(x ) \gamma^{\mu}p(x)]\,k^{+\dagger}(x).\end{aligned}\ ] ] by a fierz transformation @xmath192[\bar{n}(x)\gamma^{\mu } p(x ) ] } & = & -\,\frac{1}{4}\ , [ \bar{\lambda}^0(x)i\gamma^5 n^c(x)][\bar{n^c}(x)p(x)]\nonumber\\ & & -\,\frac{1}{4}\,[\bar{\lambda}^0(x)i\gamma_{\mu}n^c(x ) ] [ \bar{n^c}(x)\gamma^{\mu}\gamma^5 p(x ) ] + \ldots\end{aligned}\ ] ] we obtain the contributions of the charged @xmath34meson exchanges to the effective coupling constants @xmath193 , @xmath194 and @xmath195 @xmath196 the contributions of the charged @xmath34meson exchanges to the effective coupling constants @xmath197 and @xmath198 amount to @xmath199 this gives @xmath200 and @xmath201 .
the cross section for the reaction @xmath58 with the contribution of the vector
meson exchanges is @xmath202 this gives the following value of the ratio @xmath203 : @xmath204 which is in agreement with isospin invariance of strong @xmath183 interactions .
the vector
meson exchanges lead to the decrease of the cross section for the reaction @xmath205 by about @xmath59 .
thus , the total cross section for the reaction @xmath169 is equal to @xmath206 hence , due to the dominant contribution of the @xmath3 interaction in the isospin singlet state , the vector
meson exchanges have practically no influence on the cross section for the reaction @xmath169 .
the decrease of the total cross section for the reaction @xmath29 , caused by the vector
meson exchanges , is about @xmath207 .
using the results obtained above we can calculate the ratio of the cross sections for the reactions @xmath54 and @xmath106 . for the theoretical cross sections , calculated with the account for vector meson exchanges eq.([label3.12 ] ) and eq.([label2.15 ] ) and without vector meson exchanges eq.([label3.12 ] ) and eq.([label2.10 ] ) , we get @xmath208 the obtained result confirms an enhancement of the @xmath3 interaction with respect to the @xmath4 interaction near observed by the anke collaboration at cosy @xcite . in our analysis
such an enhancement is due to the contribution of the @xmath3 interaction in the isospin singlet state @xmath209 .
as has been pointed out by wilkin @xcite , an important contribution to the cross sections for the reactions @xmath72 should come from the exchange with the @xmath35 resonance @xcite .
this is the @xmath210 resonance with the quantum numbers @xmath211 @xcite .
the inclusion of the @xmath35resonance exchange demands the knowledge of its interactions with the octets of the ground baryons and the pseudoscalar mesons . since the @xmath35 resonance is the member of octet @xcite , the effective lagrangian of its interactions with the ground state baryons and the pseudoscalar mesons reads @xmath212\,\pi^0(x)\nonumber\\ & + & \frac{1}{\sqrt{3}}\,(3 - 4\alpha_{d1})\,g_{\pi nn_1}\,[\bar{p}(x)p_1(x ) + \bar{n}(x)n_1(x)]\,\eta(x)\nonumber\\ & + & \frac{1}{\sqrt{3}}\,(3 - 2\alpha_{d1})\ , g_{\pi nn_1}\,\bar{\lambda}^0(x)p_1(x)k^-(x ) + \ldots + { \rm h.c.},\end{aligned}\ ] ] where @xmath213 are the local interpolating field of the resonance @xmath35 , which we treat as an elementary particle @xcite@xcite as well as the @xmath214 resonance @xcite . due to the ambiguity of the parameters of the @xmath35 resonance
the numerical value of the contribution of this resonance is rather ambiguous @xcite .
below we use the following parameters : the mass @xmath215 , the width @xmath216 and the branching ratios @xmath217 and @xmath218 .
we leave @xmath219 of the total width for other decay channels such as @xmath220 and so on @xcite .
for such a choice of the parameters the coupling constants are equal to @xmath221 and @xmath222 . for the reaction @xmath0 the @xmath35 resonance gives the contribution only to the effective coupling constant @xmath80 .
we get @xmath223 = -\,0.22\times 10^{-6}\,{\rm mev}^{-3}.\end{aligned}\ ] ] we can also take into account the contribution of the @xmath36 resonance ( the @xmath224 resonance ) with the quantum numbers @xmath211 , which is the octet partner of the resonances @xmath225 and @xmath70 @xcite .
the coupling constants of the @xmath36 resonance with the ground state baryons and the pseudoscalar mesons are equal to @xmath226 and @xmath227 .
they are calculated for @xmath228 , @xmath229 and @xmath230 and @xmath231 @xcite .
the @xmath36 resonance as well as the @xmath35 resonance gives the contribution to the effective coupling constant @xmath80 .
the result is @xmath232 = 0.08\times 10^{-6}\,{\rm mev}^{-3}.\end{aligned}\ ] ] hence , the total contribution of the resonances @xmath35 and @xmath36 is about @xmath233 .
the values of the parameters of the resonances @xmath35 and @xmath36 are rather ambiguous . nevertheless , as we show in section 7 our choice of the parameters of the resonances @xmath35 and @xmath36 fits well the experimental data on the amplitude and the cross section for the reaction @xmath234 near threshold of the @xmath235 state .
we would like to accentuate that a cancellation of contributions of the resonances @xmath35 and @xmath36 occurs only in the lagrangian approach , where the resonances are treated as elementary particles @xcite@xcite .
we can not claim that such a cancellation between the contributions of these resonances should be within the approaches developed by wilkin _
@xcite or by sibirtsev _
@xcite , where the @xmath35 and @xmath36 resonances were described within the relativistic generalisation of the breit wigner approach @xcite .
thus , treating the the resonances @xmath35 and @xmath36 as elementary particles we have shown that the total contribution of the resonances @xmath35 and @xmath36 to the matrix elements of the @xmath236 transitions near threshold of the final state is less important in comparison with the contribution of the ground state baryon octet coupled to the octets of the pseudoscalar and scalar mesons .
in this section we show that the same procedure , which we have used for the analysis of the strangeness production in @xmath183 collisions near threshold , can be applied well to the description of the reaction @xmath237 . near threshold
the reaction @xmath7 is defined by a pure @xmath65 transition @xcite@xcite .
the effective lagrangian responsible for the @xmath238 transition , calculated in the analogy with the effective lagrangians ( [ label2.2 ] ) and ( [ label3.5 ] ) , is @xmath239 [ \bar{p^c}(x)i\gamma^5p(x)]\pi^0(x).\end{aligned}\ ] ] the effective coupling constant @xmath240 is equal to @xmath241 = \nonumber\\ & = & 6.04\times 10^{-6}\,{\rm mev}^{-3},\end{aligned}\ ] ] where @xmath242 is a relative threshold momentum of the @xmath4 pair in the initial state . for the cross section for the reaction
@xmath7 we obtain the following expression @xmath243 where @xmath244 eq.([label2.8 ] ) .
the excess energy @xmath245 is related to the kinetic energy of the proton in the laboratory frame as @xmath246 .
it is measured in @xmath247 . the function @xmath248 is defined by the phase volume of the @xmath249 state .
it is given by @xmath250 the wave function @xmath251 , describing the final
state interaction in the @xmath4 channel accounting for the coulomb repulsion , is determined by @xcite ( see also @xcite ) : @xmath252 where we have denoted @xcite @xmath253,\end{aligned}\ ] ] where @xmath254 is euler s constant , @xmath255 with @xmath256 is the fine structure constant , @xmath257 is the phase shift of low energy elastic pp scattering in terms of the s wave scattering length @xmath258 and the effective range @xmath259 @xcite . at @xmath260 and @xmath261 the wave function @xmath251 tends to unity @xmath262 ( see @xcite and appendix a ) .
the theoretical cross section for the reaction @xmath263 together with experimental data are depicted in fig.5 . in table 1
we adduce numerical values of the theoretical and experimental cross section for the reaction @xmath263 . , defined by the @xmath65 transition .
the experimental data are taken from @xcite , @xmath264 is a kinetic energy of the incident proton in the laboratory frame . ]
.the theoretical and experimental values of the cross section for the reaction @xmath265 , caused by the @xmath266 transition .
the experimental data are taken from @xcite .
the cross section is measured in @xmath267 .
[ cols= " < , > , > , > , > , > , > , > , > " , ] the theoretical cross section agrees well with the experimental data .
this implies that the analysis of the strangeness production near threshold of @xmath183 collisions given above works also well for the description of the pion production in the @xmath4 collisions .
the cross sections for other channels @xmath268 @xcite can be calculated in a similar way .
within chiral lagrangian approach the reaction @xmath270 has been investigated in @xcite ( see also @xcite ) . due to isospin invariance
the amplitude of the reaction @xmath271 is related to the amplitude of the reaction @xmath269 as @xmath272 the relative threshold momentum of the @xmath273 pair is equal to @xmath274 . at threshold
we define the amplitude of the reaction @xmath275 @xmath276 and the cross section @xcite @xmath277 here @xmath278 is a relative momentum of the @xmath279 pair , measured in @xmath280 , and @xmath281 is equal to @xmath282 for the amplitude @xmath283 we obtain the following expression @xmath284 = \nonumber\\ \hspace{-0.3in}&&= ( 0.109 - 0.190_{\sigma^0 } + 0.073_{n_1 } - 0.047_{n_2 } + 0.005_{\sigma_2 } ) \,{\rm fm } = -\,0.050\,{\rm fm},\end{aligned}\ ] ] where @xmath285 .
we remind that the resonance @xmath70 is the octet partner of the @xmath36resonance @xcite .
one can show that the contribution of the resonance @xmath286 , the octet partner of the resonance @xmath35 @xcite , is negligible small . for the constant @xmath281 we get the value @xmath287 due to isospin invariance the constant @xmath288 of the reaction @xmath271 is equal to @xmath289 near threshold of the reaction @xmath290 the constant @xmath288 defines the cross section @xcite @xmath291 where @xmath292 is a @xmath293meson momentum in the laboratory frame and
@xmath294 is the relative momentum of the @xmath295 pair . with the @xmath296 and @xmath235 pairs in the s wave state .
the experimental points are taken from @xcite , @xmath292 is a momentum of the incident @xmath293meson in the laboratory frame . ]
the theoretical value @xmath297 agrees well with the experimental data @xmath298 @xcite .
the cross section for the reaction @xmath271 eq.([label7.8 ] ) agrees also well with the theoretical results obtained in @xcite within @xmath68 chiral dynamics with coupled channels . in the laboratory frame the threshold momentum of
the @xmath293meson is equal to @xmath299 . in the vicinity of threshold
the momentum @xmath294 of the @xmath295 pair depends on the laboratory momentum @xmath292 of the @xmath293meson as follows @xmath300 it is measured in @xmath280 . for the cross section for the reaction @xmath271 as a function of @xmath292
we get @xmath301 for @xmath302 the theoretical cross section @xmath303 agrees well with the experimental one @xmath304 @xcite .
one can see that unlike the reactions @xmath38 the contribution of the resonances @xmath35 and @xmath36 is very important for the correct description of the reactions @xmath305 and @xmath9 near threshold of the final states @xcite .
indeed , the neglect of the resonance contributions leads to the increase of the cross section for the reaction @xmath306 by a factor 3 .
this disagrees with the experimental data @xcite .
the importance of the contributions of the resonances @xmath35 and @xmath36 is caused by the substantial cancellation of the non resonance contributions described by the first two terms in r.h.s .
of eq.([label7.5 ] ) . in fig.6
we show that the region of the applicability of our theoretical cross section is restricted from above by the laboratory momentum @xmath307 .
this corresponds to @xmath308 . in terms of the energy excess
this region is defined by @xmath309 , where @xmath310 and @xmath311 is the reduced mass of the @xmath295 pair .
it is obvious that the curve in fig.6 is a branch of hyperbola .
using chiral lagrangians with linear realisation of chiral @xmath76 symmetry and non derivative meson
baryon couplings we have calculated the cross sections for the reactions @xmath312 near threshold of the final @xmath5 state .
we have taken into account the contributions of the rescattering in the initial @xmath183 state and the final state interactions in the @xmath313 channels .
we have shown that the ratio of the cross sections is equal to @xmath62 .
this agrees with an enhancement of the strangeness production in the @xmath3 interaction with respect to the strangeness production in the @xmath4 interaction observed by the anke collaboration at cosy @xcite .
we have explained such an enhancement by the contribution of the @xmath3 interaction in the isospin singlet state , which is much stronger than the @xmath3 interaction in the isospin
triplet state .
we have shown that the contribution of the vector
meson exchanges is at the level of a few percent .
hence , the vector
meson exchanges can not be fully responsible for the dynamics of strong low
energy @xmath183 interactions near threshold of the reaction @xmath72 .
we have also shown that the contribution of the resonances @xmath35 and @xmath36 , treated as elementary particles , to the matrix elements of the @xmath38 transitions does not exceed @xmath233
. however , such an assertion should not be valid within wilkin s approach @xcite , where the main contributions to the amplitudes of the reactions @xmath72 come from the resonances @xmath35 and @xmath36 , described by the breit wigner shapes .
in addition to vector meson exchanges and resonances @xmath35 and @xmath36 one can estimate the contributions of the scalar meson @xmath314 and the tensor meson @xmath315 @xcite .
following @xcite and @xcite one finds that the contribution of the scalar meson @xmath314 makes up about @xmath316 @xcite , whereas the contribution of the tensor meson @xmath315 is about @xmath317 for the coupling constants @xmath318 and @xmath319 @xcite .
since the theoretical accuracy of our approach is not better than @xmath114 , one can drop the contributions of vector , tensor , exotic scalar mesons and resonances @xmath35 and @xmath36 with respect to the contributions of ground state baryon octet coupled to octets of pseudoscalar mesons and their chiral scalar partners , which are taken in the infinite mass limit . for the confirmation of our results , obtained for the strangeness production near threshold of @xmath183 interactions
, we have calculated the cross sections for the reactions @xmath7 , @xmath320 and @xmath321 .
the theoretical results agree well with the experimental data .
unlike the strangeness production in the @xmath183 interactions the contribution of resonances @xmath35 and @xmath36 to the amplitudes of the reactions @xmath322 has turned out to be very important .
this is caused by the substantial cancellation of the non resonance contributions , defined by interactions of ground state baryon octet with octets of pseudoscalar and scalar mesons .
we are grateful to vladimir koptev for calling our attention to the problem , investigated in this paper , and numerous and fruitful discussions stimulating this work .
one of the authors ( a. ivanov ) is grateful to colin wilkin , alexander sibirtsev , alexander kudryavtsev , vadim baru and all participants of the seminar at institut fr kernphysik at forschungszentrum in jlich for fruitful discussions and constructive critical comments .
a. ivanov thanks prof .
strher for warm hospitality extended to him during his stay at institut fr kernphysik at forschungszentrum in jlich .
the assistance of yurii valdau during the stay of a. ivanov at institut fr kernphysik is greatly appreciated .
we are grateful to igor strakovsky for the discussions and comments on the said analysis of the experimental data on @xmath4 scattering @xcite .
the amplitude of the reaction @xmath38 , where @xmath313 pair in the s wave state , is related to the matrix element of the @xmath324matrix as follows @xmath325 where @xmath326 is a 4dimensional volume defined by @xmath327 and the @xmath324matrix obeys the unitary condition @xcite @xmath328 the amplitude @xmath329 , defined by eq.([labela.1 ] ) , describes the total amplitude of the reaction @xmath236 accounting for the all kinds of interactions . in order to separate the contributions of the final
state interaction we propose to follow @xcite ( see also @xcite and @xcite ) .
first we define the amplitude of the reaction @xmath38 , which is calculated by the diagrams depicted in fig.1fig.4 : @xmath330 where @xmath331 is the wave function of the final state with free particles @xcite .
the account for the final
state interaction in the @xmath313 channel with the @xmath313 pair in the s wave state can be given by means of the wave function @xmath332 defined by @xcite @xmath333 where @xmath334 is a relative moment of the @xmath313 pair , @xmath335 and @xmath336 are the phase shift and the s wave scattering length of @xmath313 scattering in the s wave state .
as has been shown in @xcite ( see also @xcite ) such a wave function ( [ labela.4 ] ) can be obtained by summing up an infinite series of bubble diagrams in the @xmath313 channel .
the wave function @xmath331 is normalised by @xcite @xmath337 where @xmath338 and @xmath339 with @xmath340 are momenta of particles in the initial and final states , @xmath341 is the @xmath13meson energy , @xmath342 and @xmath343 are the energy and polarisation of the particle @xmath15 . due to ( [ labela.5 ] ) the wave function @xmath332 is normalised as follows @xmath344 the factor @xmath345 can be rewritten in terms of the cross sections for @xmath313 scattering in the s wave state @xmath346 where the cross sections @xmath347 and @xmath348 are equal to @xmath349 the factor @xmath345 can be interpreted as a probability to find a non interacting system @xmath32 with a relative momentum @xmath334 of the @xmath313 pair @xcite .
for @xmath350 , when the particles @xmath18 and @xmath351 are separated by an infinite relative distance @xmath352 at @xmath353 , the probability to find non
interacting particles in the state @xmath354 is equal to unity , i.e. @xmath355 . for any finite @xmath334 it is less than unity , i.e. @xmath356 @xcite .
the amplitude of the reaction @xmath357 accounting for the final state interaction in the @xmath313 channel is defined by @xcite @xmath358 the same result can be obtained by summing up an infinite series of bubble diagrams in the @xmath313 channel @xcite ( see also @xcite ) .
hence the amplitude of the reaction @xmath72 taking into account the final
state interaction in the @xmath313 channel is @xmath359 where the amplitude @xmath360 is defined by the feynman diagrams in fig.1fig.4 . at @xmath350
we get @xmath361 .
the reduction of the amplitude of the final state @xmath362 interaction to unity at @xmath350 is clear due to the short range character of strong interactions . indeed , at @xmath350 ,
corresponding to infinitely large relative distance of the baryons in the @xmath313 pair , the baryons do not influence each other , since the wave functions of them do not overlap .
the contribution of the coulomb interaction can be taken into account according to the prescription @xcite and balewski _ et al .
this gives @xmath363 here @xmath364 is the gamow factor @xcite @xmath365 where @xmath256 is the fine structure constant , @xmath366 is a relative momentum of the @xmath367 pair , @xmath368 is the inverse relative velocity of the @xmath369 pair , @xmath370 for @xmath371 and @xmath372 for @xmath373 . in the case of the coulomb interaction
the wave function @xmath374 has the normalisation ( [ labela.6 ] ) with the factor @xmath375 equal to @xmath376 it defines the probability to find a non interacting system @xmath5 at relative momenta @xmath334 and @xmath366 .
the definition ( [ labela.14 ] ) is correct in the vicinity of threshold , where for the calculation of the amplitude @xmath379 of the reaction @xmath38 ( or the constant @xmath378 ) one can neglect the dependence on the momenta of the particle in the final state @xmath5 .
the function @xmath380 is determined by @xmath381 where the function @xmath382 describes the final state interaction in the @xmath313 channel @xcite .
it is normalised to unity at @xmath383 and equal to @xcite @xmath384 where @xmath385 and @xmath386 are the s wave scattering length and effective range of @xmath313 scattering @xcite .
the gamow factor depends on the variables @xmath387 and @xmath388 as follows @xmath389 the function @xmath390 is equal to to unity for @xmath391 and @xmath392 . for the reactions @xmath393 and @xmath156
we get @xmath394 where we have denoted @xmath395 and @xmath396 for @xmath135 the contribution of the coulomb repulsion in the @xmath113 pair to function @xmath397 makes up about @xmath114 . at the neglect of the contribution of the coulomb repulsion @xmath398 at @xmath180 . in the reaction @xmath7 the account for the final
state interaction in the @xmath4 channel has been carried out within the scheme expounded above .
as we have shown the theoretical cross sections for the reactions @xmath393 and @xmath399 agree well with the experimental data .
9 m. bscher _
et al . _ ,
j. a * 22 * , 301 ( 2004 ) .
t. e. o. ericson and w. weise , in _ pions and nuclei _ , clarendon press , oxford , 1988 .
e. chiavassa _
et al . _ ,
b * 337 * , 192 ( 1994 ) .
k. tsushima _
et al . _ ,
c * 59 * , 369 ( 1999 ) .
v. koptev ( private communication ) , 2005 . g. fldt and c. wilkin , eur . phys
. j. a * 24 * , 431 ( 2005 ) , nucl
th/0411019 ; z. phys .
a * 357 * , 241 ( 1997 ) .
s. eidelman _
et al . _ ( the particle data group ) , phys .
b * 595 * , 1 ( 2004 ) .
c. wilkin ( private communication ) , jlich 2005 .
a. sibirtsev ( private communication ) , jlich 2005 .
et al . _ ,
j. a * 9 * , 425 ( 2000 ) ; eur .
j. a * 12 * , 341 ( 2001 ) . v. de alfaro , s. fubini , g. furlan , and c. rossetti , in _ currents in hadron physics _ , north holland publishing co. , amsterdam @xmath400 london , american elsevier publishing co. , inc .
, new york , 1973 .
s. weinberg , phys .
. lett . * 18 * , 188 ( 1967 ) .
w. bardeen and b. w. lee , phys . rev . * 177 * , 2389 ( 1969 ) . s. gasiorowicz and d. a. geffen , rev .
phys . * 41 * , 531 ( 1969 ) .
n. kaiser , eur .
j. a * 5 * , 105 ( 1999 ) . c. hanhart and k. nakayama , phys .
b * 454 * , 176 ( 1999 ) .
j. t. balewski _
et al . _ ,
eur . phys .
j. a * 2 * , 99 ( 1998 ) and references therein .
( see @xcite eq.(4 ) ) .
a. n. ivanov _
et al . _ , eur .
j. a * 23 * , 79 ( 2005 ) ; nucl th/0406053 .
j. t. balewski _
et al . _ ,
b * 420 * , 211 ( 1998 ) .
c.h . lee , d.p .
min , and m. rho , phys .
b * 326 * , 14 ( 1994 ) ; c.h .
lee , g. e. brown , and m. rho , phys .
b * 335 * , 266 ( 1994 ) ; c.h .
lee , g. e. brown , d.p .
min , and m. rho , nucl .
a * 585 * , 401 ( 1995 ) ; c.h .
lee , d.p .
min , and m. rho , nucl .
a * 602 * , 334 ( 1996 ) .
a. n. ivanov , m. nagy , and n. i. troitskaya , phys .
* c59 * , 451 ( 1999 ) and references therein . t. e. o. ericson and a. n. ivanov , phys .
b * 634 * , 39 ( 2006 ) .
a. n. ivanov _
et al . _ ,
j. a * 21 * , 11 ( 2004 ) ; nucl th/0310081 .
a. n. ivanov _
et al . _ ,
phys . rev .
a * 71 * , 052508 ( 2005 ) .
a. n. ivanov _
et al . _ ,
j. a * 25 * , 329 ( 2005 ) , nucl
a. n. ivanov _
_ , _ recent theoretical studies on hadronic atoms _ , invited talk at international conference on exotic atoms exa05 at stefan meyer institute of austrian academie of sciences , vienna , austria , 21 - 25 february 2005 , nucl
j. j. jones _
* 26 * , 860 ( 1971 ) .
n. kaiser , t. waas , and w. weise , nucl .
a * 612 * , 297 ( 1997 ) .
m. m. nagels _ et al .
_ _ compilation of coupling constants and low energy parameters _ , nucl .
b * 147 * , 189 ( 1979 ) ; o. dumbrajs _
_ , _ compilation of coupling constants and low energy parameters _ , nucl .
b * 216 * , 277 ( 1983 ) .
et al . _ , eur .
j. c * 21 * , 473 ( 2001 ) .
r. a. arndt , i. i. strakovsky , and r. l. workman , phys .
c * 62 * , 034005 ( 2000 ) , nucl
th/0004039 ; http://gwdac.phys.gwu.edu i. i. strakovsky ( private communication ) , 2005 .
( see @xcite p.872 ) ( see @xcite p.158 ) h. o. meyer _
et al . _ ,
a * 539 * , 633 ( 1992 ) .
h. machner and j. haidenbauer , j. phys .
part . phys . * 25 * , r231 ( 1999 ) .
v. bernard , n. kaiser , and ulf.g .
meiner , eur .
phys . a * 4 * , 259 ( 1999 ) . c. hanhart and n. kaiser , phys .
c * 66 * , 054005 ( 2002 ) and references therein .
a. n. ivanov , n. i. troitskaya , h. oberhummer , and m. faber , eur .
j. a * 8 * , 223 ( 2000 ) .
k. b. mather and p. swan , in _ nuclear scattering _ , cambridge at the university press , 1958 , p.214 .
v. cirigliano , g. ecker , h. neufeld , a. pich , jhep * 0306 * , 012 ( 2003 ) .
b. renner , phys .
b * 33 * , 599 ( 1970 ) .
y. oh and t.s . h. lee , phys .
c * 69 * , 025201 ( 2004 ) .
s. s. schweber , in _ an introduction to relativistic field theory _ ,
row , peterson and co@xmath400 evanston , ill . ,
elmsford , new york , 1961 .
( see also @xcite p.241 ) . | the cross sections for the reactions @xmath0 and @xmath1 are calculated near threshold of the final states .
the theoretical ratio of the cross sections @xmath2 shows the enhancement of the @xmath3 interaction with respect to the @xmath4 interaction near threshold of the strangeness production @xmath5 .
such an enhancement is caused by the contribution of the @xmath6 interaction in the isospin
singlet state , which is stronger than the @xmath3 interaction in the isospin triplet state . for the confirmation of this result
we calculate the cross sections for the reactions @xmath7 , @xmath8 and @xmath9 near threshold of the final states .
the theoretical cross sections agree well with the experimental data . _
@xmath10atominstitut der sterreichischen universitten , technische universitt wien , wiedner hauptstrasse 8 - 10 , a-1040 wien , sterreich + @xmath11stefan meyer institut fr subatomare physik , sterreichische akademie der wissenschaften , boltzmanngasse 3 , a-1090 , wien , sterreich + @xmath12state polytechnic university , polytechnicheskaya 29 , + 195251 , st petersburg , russian federation _
pacs : 11.10.ef , 12.39 .
fe , 13.75.jz , 14.20.jn |
type ia supernovae ( sne ia ) are thought to be thermonuclear disruptions of white dwarf ( wd ) stars @xcite , but the details remain uncertain .
one possibility for the progenitors is the single degenerate model in which main sequence stars or post main sequence red giants transfer mass to a wd through roche lobe overflow or a common envelope and the wd grows close to the chandrasekhar mass @xmath6 . due to compression , the thermonuclear runaway starts near the center leading to the explosion of the wd and a rapidly expanding envelope with a mass close to @xmath6 .
a second possibility is the double degenerate model in which a pair of wds merge and lead to an explosion . in most such cases ,
the resulting mass of the rapidly expanding envelope will be different from @xmath6 .
residual material from these mergers surrounding the explosions will get swept up by the ejecta forming dense , shell - like structures .
although the explosion of single wds seems to be favored for the majority of objects , we may expect mergers to contribute to the sne ia population ( see [ obs ] ) .
the possibility of different progenitor channels , the population of which may vary with redshift , may pose a challenge for the use of sne ia in cosmological studies that rely on a single parameterization , such as a light curve width to peak luminosity relation , lwr , to reduce the intrinsic scatter in the peak magnitudes and render them standard candles @xcite .
to first order , the lwr relation can be understood as a result of different amounts of @xmath7ni produced during the explosion @xcite .
there may be some spread and an offset in lwr introduced by one of the channels if the masses of the envelope differ from @xmath6 , and/or the density structures differ .
this can lead to a systematic shift of lwr with redshift if the evolutionary time scales of the progenitor systems differ . even if the different progenitor scenarios obey the same lwr
, differences in the color could introduce systematic errors in cosmological studies because sne ia are known to suffer to some degree from reddening in their respective host galaxies which has to be taken into account . to correct for this , the maximum light color excess ( usually @xmath8 ) and
an average reddening law are used to determine the amount of absorption .
sne ia that are intrinsically redder as compared to the average local sample will thus be over - corrected in this fashion to a higher luminosity .
similarly to the two distinct progenitor channels , qualitative variations in the explosion physics may lead to various classes of sne ia even within the single degenerate scenarios .
standard explosion models include delayed detonations ( dd ) and deflagrations . in these scenarios , burning during the deflagration phase
leads to an unbound wd . in dd models ,
the deflagration turns into a detonation in an expanding envelope . because the density structure of the wd declines monotonically with radius , the resulting density structure in the expanding envelope also smoothly declines with mass and radius .
a variation of dd models are the pulsating delayed detonation models ( pdd ; @xcite ) . in these models ,
the total energy production during the deflagration phase is , by construction , lower and insufficient to unbind the wd .
this results in large amplitude pulsations . because the fall - back time increases with distance , the inner regions contract and
leave a shell behind at larger distances . due to infall driven compression , a detonation
is triggered , the material behind the burning front is accelerated , and this expanding material runs into the low - velocity material left at larger distances .
similar to the merger scenario , a shell - like structure is formed with very similar light curve and spectroscopic properties , but with a total mass close to @xmath6 @xcite .
these two groups , consisting of dd and deflagration models such as w7 @xcite , which lack shells , and the models with shells ( mergers and pdds ) , can be differentiated by their predictions for the photospheric evolution and maximum light colors @xcite . for the former group ,
the photospheric velocities , @xmath9 , smoothly decline with time and the models show a blue color at maximum light , @xmath10 ; in the latter group , @xmath9 shows a plateau in the evolution as the photosphere recedes through the shell .
these models are intrinsically redder and slightly over - luminous because of the lower expansion rate in the inner region . as shown in @xcite the color , length and velocity of the plateau
are correlated with the mass of the shell , and this potentially allows the two groups to be distinguished even for similar brightnesses .
indeed , there is a growing sample of sne ia showing photospheric velocity plateaus ( e.g. 1990n ; @xcite ; 1991 t , 1999aa ; @xcite ; 1999ee ; @xcite ; 2000cx ; @xcite ; see also @xcite ) .
many of these sne ia have been reported as having a red color @xmath11 at maximum , but this is typically attributed to reddening along the line of sight .
alternatively , this sample may suggest the contribution of events with shell - like density structures in the observed population .
these events may be understood in terms of mergers or pdds ; however , the inhomogeneities and incompleteness of individual data sets in the literature preclude definite conclusions . to address this problem and others , we started the texas supernovae search ( tss ; quimby et al . in prep . ) with the goal of providing a homogeneous set of quality data for several supernovae beginning well before maximum light .
in this paper , we present our observations of sn 2005hj and analysis of the data . in [ obs ]
we describe the discovery and give the details for both the photometric and spectroscopic follow - up . in [ models ] we discuss generic properties of explosion models and suggest a secondary parameter to separate models with and without shells , and analyze the peculiarities of sn 2005hj .
conclusions and discussion are presented in [ conclusions ] .
sn 2005hj was discovered on october 26.13 ut in the field of abell 194 as part of the tss .
the tss uses the wide field ( @xmath12 ) 0.45 m rotse - iiib telescope @xcite at the mcdonald observatory in texas to scan nearby galaxy clusters nightly for transients with a modified version of the psf - matched image subtraction code from the supernova cosmology project .
sn 2005hj was found at an unfiltered magnitude ( calibrated against the usno - b1.0 r2 ) of @xmath13 and is located at @xmath14 , @xmath15 .
the foreground reddening at this location is @xmath16 @xcite .
examination of rotse - iiib images from oct .
20 and oct .
22 shows the sn was detected prior to discovery , but not significantly well to pass the search pipeline s automatic cuts .
figure [ lc ] shows the rotse - iiib light curve for sn 2005hj through 40 days after maximum light . to construct the light curve
, we co - added images taken on a given night ( usually 6 ) excluding any frames of significantly lower quality due to passing clouds or wind sheer , and then subtracted the reference image convolved to the same psf .
magnitudes were determined by fitting the local psf ( derived from the co - added nightly images ) to the location of the sn on the subtracted frame using custom software and the daophot psf - fitting routines ( @xcite ported to idl by @xcite ) . the unfiltered ccd response of rotse - iiib has an approximate full width of @xmath17 centered in the @xmath18-band around 6000 .
because we do have some sensitivity in the blue and since the @xmath11 colors of sne ia typically grow @xmath19 mag redder in the 30 days after maximum @xcite , there is a blue deficit at later times that causes our unfiltered magnitudes to decline more rapidly than the true @xmath18-band fading .
note that @xmath20 colors of sne ia are close to zero at maximum light .
we therefore limit the light curve fitting to data taken before 10 days after maximum ( determined through several iterations of the fit ) , during which the color evolution is minimal . the best fit @xmath18-band template from @xcite is also shown in figure [ lc ] .
the date of maximum light determined from the fit is nov .
1.6 with a formal error of 0.7 days ( note the template phases are relative to the @xmath21-band maximum ) .
the best fit stretch factor @xcite for the light curve width is @xmath22 .
the preliminary measurement of the observed @xmath11 color at @xmath23 maximum from the carnegie supernova project is @xmath24 after removal of the host light but before any extinction or @xmath25-corrections are applied ( m. m. phillips , private communication ) . near real - time photometric analysis combined with target of opportunity ( too ) time on the neighboring 9.2 m hobby - eberly telescope ( het ) allowed us to obtain optical spectra just 4 hours after the discovery images were taken and every few days over the next 6 weeks .
these observations are detailed in table [ spec ] .
the instrumental response is such that very little second order light is expected blue of 8900 even with the gg385 blocking filter .
the data were reduced in the optimal manner using iraf and custom idl scripts .
the wavelength scale was calibrated against cd and ne lamps and its accuracy was verified by comparing night sky lines to the spectral atlas of @xcite . because the het pupil size varies for different tracks , absolute flux calibration can not reliably be achieved ; however , we used the standard stars of @xcite and @xcite , which were observed using the same setups , to achieve _ relative _ spectrophotometric calibration and to remove strong telluric features .
the redshift of the host galaxy was derived from narrow emission lines around 7000 ( observed ) , which we attribute to h-@xmath26 , [ ] , and [ ] in the host galaxy .
we combined all the spectra and simultaneously fit these lines with gaussians to determine the line centers .
the line redshifts are best fit by @xmath27 , and we adopt this value for the sn .
this gives sn 2005hj an absolute peak magnitude of @xmath1 in our unfiltered band pass ( assuming h@xmath28=71 km s@xmath3 mpc@xmath3 , @xmath29 , and @xmath30 ) , and places the host well behind abell 194 ( @xmath31 ; @xcite ) .
the brightness and broad light curve shape suggest that sn 2005hj is a slightly over - luminous sn ia .
the unfiltered rotse - iiib reference image shows that the host for sn 2005hj is relatively bright ( @xmath32 ) and compact , and is therefore likely a significant contaminant to our spectra .
thus , we have to subtract the galaxy contribution ( see fig . [ spec0 ] ) . lacking an observed spectrum for the host galaxy excluding the sn light , we constrained the galaxy sed using archival sloan digital sky survey ( sdss ) @xmath33 observations and obtained a template galaxy spectrum ( n. drory 2005 , private communication ) .
the relative amounts of sn and galaxy light in the spectral apertures will vary not only with the changing sn brightness , but also with the seeing , slit width and positioning .
also plotted in figure [ spec0 ] is a spectrum of sn 1999aa ( blue curve ) constructed via a linear interpolation of the @xmath34 day and @xmath35 day spectra presented by @xcite . noting the similarity of the spectral features of sn 1999aa and sn 2005hj
, we assume that we can model our observed spectra as a linear combination of our galaxy template and the sn 1999aa spectra interpolated to the same phase as the sn 2005hj observations .
we perform a least squares fit to determine the relative contributions of each component .
the red line in figure [ spec0 ] shows the derived contribution of galaxy light in the @xmath36 day spectrum . aside from a few small differences ( most noticeably in the @xmath374481 triplet ) ,
some of which may be explained by calibration errors , the combined sn 1999aa + host spectrum ( purple curve ) is a good fit .
the over all fit is improved if we interpolate the sn 1999aa spectra to @xmath38 days instead of @xmath36 , especially in the 5400 to 6500 range , which could imply a @xmath391 day error in the date of maximum light or different time scales for the spectral evolutions of the two sne .
we repeated this process for all the sn 2005hj spectra , each time using the same galaxy template and the sn1999aa spectra ( interpolated to the appropriate phase ) as reference to determine the relative amount of galaxy light .
in general , the galaxy template added to the sn 1999aa spectra does an excellent job of reproducing the observed sn 2005hj spectra .
the galaxy light typically dominates the flux red of 7000 .
figure [ spec ] shows the spectral evolution of sn 2005hj recorded by the het between days @xmath36 and @xmath40 with the derived galaxy contribution subtracted .
overall , sn 2005hj shows spectra with lines dominated by intermediate mass and iron group elements as is typical for sne ia .
while the lines show normal expansion velocities , the absorption components are more narrow and , for the early phases , weaker than typically observed , as exemplified by the @xmath376355 line ( see fig .
[ spec1 ] ) .
sn 2005hj also shows an atypical velocity evolution of these features over time .
line minima are useful diagnostic indicators of the ejecta structure as they give the abundances and velocities of the material .
the actual measurement of the velocity at the minimum of the line profile is complicated by the presence of the continuum , other blended lines , and some uncertainty in the true line profile shape .
detailed modeling is required to accurately sort out all the components and how they relate to the photospheric layer to reveal the velocity distribution of the ejecta larger than those measured from weak lines . however , for shell models the steep density gradients cause even strong lines to from very close ( in radius ) to the actual photosphere . ] .
such models have shown that the absorption minima approximate the photospheric expansion velocities to within about 1000 km s@xmath3 at maximum light @xcite .
thus , simple line fitting can lead to a rough description of the ejecta velocities , and allows a useful comparison to discriminate between different models . at late times
the photosphere will recede below the si rich layer and so the velocities derived from the @xmath376355 line will become increasingly discrepant with the photospheric velocity . for deflagration and classical detonation models , this departure will begin to set in 1 - 2 weeks after maximum light @xcite .
the strength of the @xmath376355 line and its persistence from at least 2 weeks before to 4 weeks after maximum light make it a valuable tool for probing the ejecta .
its evolution with time is shown in figure [ siii ] for the case of sn 2005hj . to determine the velocity , we smooth the spectra by fourier transform filtering , divide by the estimated continuum , and then select the lowest point using spline interpolation over a selected range . the continuum is represented by a medium order ( 6th - 7th ) polynomial fit to regions of the spectra that are not strongly affected by lines . to smooth the spectra
, we use a fourier transform to convert the data into a power spectrum , and then multiply this by a filter to remove high frequency variations .
we then apply a reverse ft to the filtered power spectrum to recover the smoothed spectrum .
the filter has the functional form @xmath41 & \rm{otherwise } \end{array } \right.\ ] ] the filter cutoff frequency , @xmath42 , and attenuation scale , @xmath43 , were determined as follows : 1 ) the spectra were converted into a power spectrum , @xmath44 , via fourier transform ; 2 ) the slope of @xmath45 is fit over the noise dominated high frequencies and interpolated through the low frequencies to determine the noise spectrum ; 3 ) @xmath42 is taken as the frequency at which @xmath45 drops to within three times the dispersion about the noise spectrum ; 4 ) @xmath43 is chosen such that the slope of @xmath46 $ ] is twice the noise spectrum slope ( i.e .. @xmath47 is the frequency above which noise is clearly the dominate component ) . for this analysis ,
only the spectral bins with signal to noise above 25 were considered ( note the peak throughput for het / lrs is near the @xmath376355 line ) . for consistency , we adopt a single filter for all our analysis , choosing the results from our nosiest data , @xmath48 @xmath3 and @xmath49 @xmath3 , which removes noise in the data but also some real information related to `` sharp '' features in the spectra such as the narrow core to the @xmath376355 absorption in the day @xmath50 spectrum . using the relativistic doppler formula and the @xmath51-weighted @xmath376355 rest velocity in the host galaxy frame , we convert the wavelengths of the line profile minima into expansion velocities . for each spectrum
we conducted 250,000 monte carlo simulations in which normally distributed noise based on the statistical flux errors was added to the data and the ft smoothed minimum was found .
the peak of the distribution and the interval containing 68% of the simulation results were used to calculate the velocity of the minimum and its error , respectively .
we also measured the relative shift in the region lines over all epochs and found the scatter to be 80 km s@xmath3 , which we add in quadrature to the individual errors .
the results are given in table [ linedata ] and plotted in figure [ linevel ] .
we find that the data points are at 10,600 @xmath52 150 km s@xmath3 between maximum light and @xmath53 days , somewhat faster prior to maximum , and significantly slower on day @xmath54 . by day @xmath40
, the @xmath376355 absorption has all but completely disappeared . from maximum light through day @xmath50
, the @xmath376355 line profile shows little change in both depth and width in addition to maintaining a constant absorption minimum velocity .
of specific relevance is the blue wing of the absorption profile ; this section of the line is formed by the material at the greatest distance from the photosphere and at the highest velocities , and as such it should be the first to vanish as the photosphere recedes . the consistency of this blue wing from maximum light through day @xmath50 suggests the photosphere falls within the si enriched layers for at least this period . by day @xmath53
the blue wing has shifted significantly to the red , while the red wing remains constant except for the effects of an blend around 6250 .
other features begin to appear or strengthen at this phase as well .
this behavior could be a signal that the si layers are becoming detached from the photosphere by day @xmath53 .
the day @xmath54 spectra show a double minimum at the location of the @xmath376355 feature ( see figure [ day25 ] ) .
telluric absorption is weak in this wavelength range , and the line profile is clearly seen in each of the three individual exposures , which support the reality of this feature .
a possible explanation for this feature is contamination from the host that is not removed by the template subtraction ; however , galaxy spectra do not typically exhibit features in this range that could cause such interference , and even if such were the case , we would expect to see similar behavior in the @xmath40 day spectra .
a second possibility is contamination from lines . using the spectral analysis tool synow @xcite , and the example of sn 1994d as a starting place @xcite
, we find that while likely produces the absorption dips @xmath55 away on either side of the @xmath376355 line , it is unlikely responsible for the double minimum .
the third possibility , which we favor , is that this double minimum simply appears because we are resolving the @xmath376355 doublet .
this result implies that the seen in the @xmath54 day spectra is confined to a very narrow region of velocity space ( @xmath56 km s@xmath3 ) .
if accurate , the true minimum of the @xmath376355 doublet would be about 100 - 200 km s@xmath3 faster than indicated in figure [ linevel ] and table [ linedata ] , but still significantly below the plateau velocity
. the emergence of this thin layer may also be responsible for the appearance of the narrow core in the @xmath50 day spectrum as well as the apparent double minimum to the @xmath53 day data .
some remnant of the blue component to the doublet may persist to the @xmath40 day spectrum .
figure [ day25 ] also shows the spectra of several other sne ia taken around 25 days after maximum light . while the distinctly double minimum appears unique to sn 2005hj
, the width and depth of the feature is roughly consistent with the others .
sn 2005hj clearly belongs to the low velocity gradient ( lvg ) group in the classification scheme of @xcite , but moreover the velocity derivative from maximum light through day @xmath53 , @xmath57 km s@xmath3 day@xmath3 , is consistent with no change is the average daily rate of decrease in the expansion velocity from maximum light through the last available spectrum before the @xmath376355 line disappears ; therefore , including the day @xmath54 spectrum , sn 2005hj formally has @xmath58 km s@xmath3 day@xmath3 , but with a @xmath59 per degree of freedom of 3.2 ] . from the line profile evolution ( table [ linedata ] , figures [ siii ] and [ linevel ] ) we can deduce a plateau phase starting at @xmath60 days which lasts no more than 30 days . noting the change in the @xmath376355 line profile in the @xmath53 day spectrum , we conservatively mark the end of the plateau phase as day @xmath61 days , which gives the plateau phase a total duration of @xmath62 days .
the @xmath376355 velocity evolution derived from the minima of ft smoothed spectra of several selected sne ia is plotted in figure [ linevels ] .
the velocity plateau of sn 2005hj is similar to that of other over - luminous sne ia such as sn 1999aa @xcite and sn 2000cx @xcite , but it is distinct from normal sne ia such as sn 1994d @xcite and sn 1992a @xcite that do not show a plateau phase from the sn 1999aa curve because all spectral features in these data seem to be frequency shifted including the telluric features . ] .
there is general agreement that sne ia result from some process involving the combustion of a degenerate c / o white dwarf @xcite . within this general picture ,
two classes of models are most likely .
the first is an explosion of a c / o - wd with a mass close to the chandrasekhar limit ( @xmath6 ) that accretes matter through roche - lobe overflow from an evolved companion star @xcite . in this case , the explosion is triggered by compressional heating near the wd center .
alternatively , the sn could be an explosion of a rotating configuration formed from the merging of two low - mass wds , after the loss of angular momentum @xcite .
candidate progenitor systems have been observed for both scenarios : wd binary systems with the correct period to merge in an appropriate time scale with an appropriate total mass @xcite ; and supersoft x - ray sources @xcite showing accretion onto the wd from an evolved companion .
there are still open questions about the details of both the merging and accretion processes ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
. from the observed spectral and light curve properties , the first scenario appears to be the most likely candidate for the majority of normal sne ia . in particular , delayed detonation ( dd ) models @xcite have been found to reproduce the majority of the observed optical / infrared light curves ( lc ) and spectra of sne ia reasonably well @xcite . in the dd scenario
, a slow deflagration front turns into a detonation .
the wd pre - expands during the deflagration and undergoes complete burning during the detonation phase .
similarly , the classical deflagration models w7 @xcite show similar behavior to dds but only by neglecting instabilities due to the deflagration fronts @xcite . for recent reviews
see @xcite .
despite the success of classical dd and w7 models , both lack the basic features seen in sn 2005hj .
neither predicts a long plateau in velocity ; they instead show a smooth decline of the photospheric velocity6355 line is an imperfect tracer of the photospheric velocity as mentioned in
[ spec_char ] , the observed sne ia population typically exhibits a 1000 - 3000 km s@xmath3 decrease in the measured line velocities between 1 week before maximum light to 2 weeks after@xcite , and the deflagration and classical delayed detonation models employed to explain these events have shown a correspondingly large decrease in photospheric velocities over the same period @xcite .
these models are inconsistent with the corresponding @xmath63 km s@xmath3 shift measured for sn 2005hj . ] as a function of time ( fig .
[ modelvel ] ) .
this happens because in expanding envelopes the photosphere recedes in mass and , because of the homologous expansion , in velocity as well .
this behavior results from the smoothly declining density structure of the wd and the fact that variations in the specific energy production are small .
in contrast , shell - like density structures will produce velocity plateaus in a natural way because the photosphere remains in the shell for some time as shown by @xcite . to form a shell - like structure
requires interaction of rapidly expanding material with a surrounding envelope .
various mechanisms have been suggested to supply this surrounding matter : the pulsating delayed detonation scenario @xcite , mergers or tamped detonation models .
shells may also form by the interaction of an exploding single wd within the progenitor system @xcite .
we analyzed the observations of sn 2005hj based on detailed , spherical models for supernovae published in the literature .
the models are based on detailed calculations for the explosion , light curve and spectra .
the models considered include delayed detonations , deflagrations , pulsating delayed detonations and tamped detonation / merger scenarios . in figure
[ modelvel ] , we show the photospheric velocities as a function of time for these models along with the branch - normal sne ia to illustrate the formation of a plateau in the models that naturally form a shell . note for lower shell masses , this `` plateau '' is more accurately described as a period of slowly declining velocities . in classical delayed detonation models and for normal - bright sne
ia , si is present over a wide range in mass , spanning about @xmath64 to @xmath65 , which corresponds to velocities from about 8,000 to 9,000 km s@xmath3 to more than 20,000 km s@xmath3 .
the si layer is thick ( in the mass frame ) because explosive oxygen burning occurs over a wide range of temperatures .
the density gradient is smooth and si is mostly in , so initially the velocity derived from the minimum of the @xmath376355 line smoothly declines with the receding photosphere governed by the geometrical dilution of the expanding envelope .
eventually , the photosphere begins to recede below the si layer at which point the evolution of the @xmath376355 line profile will show the following behavior : 1 ) the optical depth of the highest velocity material at the largest radii will begin to decline below 1 and as such the blue wing of the line profile will start to drift toward the red ; 2 ) as the optical depth decreases , the strength of the line as measured from the line depth will decrease ; 3 ) the line minimum may continue to slow , but it will grow increasingly discrepant with the photospheric velocity .
this phase typically begins 1 - 2 weeks after maximum light for normal sne ia and is heralded by the appearance of lines .
while this behavior is commensurate with observations of normal - bright sne ia such as sn 1994d , this behavior is not consistent with the observations of sn 2005hj .
the @xmath376355 line seen in sn 2005hj is narrow , and during the plateau phase the wings do not change , the depth does not change , and the velocity of the minimum does not change to within the errors .
the data require a narrow distribution of in velocity space , and we suggest this may be explained by an interaction that compresses the si rich layers as predicted by merger and pulsating delayed detonation models .
the shell models are also consistent with the velocity drop seen after the plateau because a significant amount of si is located below the shell @xcite . in fig .
[ modelprop ] , we show general properties of these models .
as discussed in the papers above , to first order , the observational signatures of the shell depend on the mass of the shell @xmath4 .
almost the entire wd is burned , and momentum conservation determines the amount of high velocity matter that can pile up in the expanding shell . with increasing shell mass , more material of the sn envelope
is slowed down . as a consequence ,
the velocity of the shell @xmath66 decreases with @xmath4 . because it will take longer for the photosphere to recede into the lower velocity matter , the time until the beginning of the plateau phase , @xmath67 , increases with @xmath4 .
the optical depth increases with @xmath4 , duration of the plateau , @xmath68 , also increases , the temperature gradient becomes steeper , and the photosphere becomes cooler ( i.e. @xmath11 increases ) with increasing @xmath4 @xcite .
the duration of the plateau , @xmath68 , is defined by the velocity spread @xmath69 around @xmath70 with @xmath71 km s@xmath3 , which puts the end of the plateau phase safely into the parts of a rapidly declining @xmath9 .
we choose a larger value than in the observations to avoid ambivalences due to discreetness , which , in some of the models , is of the order of @xmath72 km s@xmath3 . by increasing @xmath73 from 200 to 500 km
s@xmath3 the nominal duration is increased by @xmath74 day
. however , we also note that the actual width depends on the velocity spread in the shell ( see [ conclusions ] )
. given the model predictions , we can use different observational indicators to test which @xmath4 is consistent with sn 2005hj ( fig .
[ modelprop ] ) .
all three parameters , @xmath66 , @xmath75 , and @xmath67 suggest @xmath76 , with the allowed ranges specifically bracketed by 0.15 - 0.6 , 0.1 - 0.25 , and 0.1 - 0.25 @xmath77 for the plateau length , shell velocity , and plateau onset , respectively , taking the observed errors into account .
the comparison between the @xmath11 color as a function of @xmath66 , @xmath75 , or @xmath67 , however , shows only marginal consistency between the observations and the models if we assume only foreground redding by the galaxy .
we note , that the intrinsic @xmath11 color of the models is uncertain by about 0.05 to @xmath78 at maximum light .
the two best - fitting models , pdd3 and det2env2 , show a peak brightness , @xmath79 , of @xmath80 and @xmath81 , respectively , with an uncertainty of @xmath82 @xcite vs. a typical dd model with @xmath83 @xcite ,
i.e. they are brighter by about 20 % mostly due lower escape probablity of @xmath84-rays that results when the @xmath7ni layers are slowed down because of the interaction @xcite .
we have presented photometric and spectroscopic data for sn 2005hj , a slightly over - luminous type ia .
the most striking feature is an apparent plateau in the expansion velocity evolution , which we derive from the @xmath376355 line .
the velocities remain at about @xmath85 km s@xmath3 for about 3 weeks starting slightly before maximum light , and this plateau is bracketed by preceding and succeeding decelerations .
we find that si is confined to a relatively narrow velocity region .
analysis of the detailed observations in concert with published models suggest there may be some physical distinction between sn 2005hj and other normal - bright sne ia that may systematically affect their use as distance indicators if not properly taken into account .
the models considered include delayed detonations , deflagrations , pulsating delayed detonations and tamped detonation / merger scenarios . in order to explain the narrow @xmath376355 line and its plateau in velocity ,
we suggest an early interaction that forms a dense shell as predicted by merger and pdd models . the spectral and photometric peculiarities are consistent with respect to the velocity , duration , and onset of the plateau , and marginally consistent with the maximum light color , for models that have shells of about @xmath86 .
as indicated by earlier works @xcite , the mass of the interacting shell has been found to be the parameter that dominates the details of these observational signatures independent of how this shell may form .
the tight predicted relation between each of @xmath66 , @xmath75 , and @xmath67 may provide a stable means to separate sn 2005hj like events from regular branch - normal sne ia .
although the agreement between the shell models and the observations is good , the predictions are not necessarily unique and other possibilities may exist . for example , we have not considered 3-d models such as the detonation from a failed deflagration scenario recently examined by @xcite . for sn 2005hj
then the agreement of the plateau velocity and its duration to that predicted by shell models may simply be a fluke , and in such case this concordance should then not hold for other sne with similar @xmath376355 evolution .
given the data and models considered , we suggest either pdds or merger events are responsible for sn 2005hj , and this implies the existence of two different progenitor channels . it is important to understand how these two progenitor channels , which may occur in relatively varying fractions as a function of redshift , will impact studies using sne ia as distance indicators .
@xcite estimate that 20% of sne ia in their sample are either 1991t - like or 1999aa - like .
these sne show spectral features and a velocity plateau similar to sn 2005hj .
@xcite found 5 1999aa - like events in the @xcite sample out of 20 total sne ia that were observed early enough to show 1999aa - like spectral features , and one that was 1991t - like ; however , in the pre - loss sample they do not classify any of the 7 sne ia with early spectra as 1999aa - like .
these nearby samples are constructed from targeted galaxy searches that have different selection biases than the deep cosmological surveys , but we will assume a uniform 2005hj - like rate of 25% for all sne ia .
sne ia that appear spectroscopically similar to sn 2005hj in a single epoch could none the less arise from different progenitors , and the mass of the low - density envelope around pdds or mergers may effect their peak magnitudes and/or light curve shapes , but we will further assume that all such events deviate uniformly from the lwr of branch normal sne ia .
@xcite calculated the relation between peak @xmath23 band magnitudes , @xmath87 , and the fading between maximum light and + 20 days , @xmath88 , for a variety of theoretical models and found that shell models produced @xmath88 that were 0.2 to 0.3 mag smaller than for ( in vacuum ) delayed detonations reaching the same peak magnitude . therefore using the same lwr for shell models will result
in corrected peak magnitudes systematically offset by 0.1 to 0.2 mag .
also , the observed peak magnitudes of sne ia are usually corrected for absorption along the line of sight using the observed @xmath11 color at maximum light and a reddening law . for events that are intrinsically red
, this will increase the estimated peak magnitude above its already over - luminous intrinsic value .
cosmological studies may therefore need to remove or at least separately handle sn 2005hj - like events to avoid systematic errors in distance . as a case for
the importance of separating different progenitors , let us consider sn 1999ee . very similar to sn 2005hj
, sn 1999ee shows a plateau with @xmath89 km sec@xmath3 , a duration of @xmath90 days , and an onset at day @xmath91 relative to maximum ( @xcite ; see fig .
[ linevels ] ) .
the @xmath11 color of sn 1999ee was also quite red at maximum light ; @xmath92 after correction for galactic extinction @xcite . based on the standard brightness decline relation and the corresponding colors ,
@xcite derived reddening in the host galaxy of @xmath93 which implies an absolute brightness of @xmath94 similar to snls-03d3bb which @xcite attributed to a super - chandrasekhar mass wd . taking into account the spectroscopic information about the velocity plateau , its length and onset
, we attribute a portion of the red color to the intrinsic properties of the supernova .
we find that the duration of the velocity plateau , its onset and size are consistent with a shell mass of @xmath95 which suggests an intrinsic color @xmath11 of @xmath96 ( see fig .
[ modelprop ] ) .
this reduces the reddening in the host galaxy to @xmath97 and the absolute brightness @xmath87 to @xmath98 , which compares favorably to the model predictions of @xmath99 and @xmath81 for pdd3 and detenv2 , respectively , within the model uncertainties .
note that there is an interstellar sodium line in the spectra that implies some degree of reddening within the host .
there are some apparent spectral differences when compared to sn 2005hj , namely sn 1999ee has a slightly broader blue wing in and stronger absorption around 4900 .
this may either hint toward different explosions scenarios ( i.e. pulsations versus mergers ) , or different viewing angles of asymmetric envelopes .
this brings us to the limitation of our studies . except for the color ,
sn 2005hj fits remarkably well with the merger and pdd model predictions but , still , it is a single event and the good agreement may be coincidental .
we need a large , well - observed sample of similar objects to test and verify or falsify the models and to determine the shell mass distribution .
moreover , 3d effects have been neglected . in reality
, we must expect some dispersion . though pulsating delayed detonation models may be expected to be rather spherical
, mergers may be asymmetric with directionally dependent luminosities and colors .
in fact , both classes may be realized in nature .
as mentioned above , the duration of the plateau , @xmath68 , is defined by the velocity spread around @xmath70 .
the physical width of the shell depends , to first order , on the distance at which the interaction occurs and the density distribution of the interacting expanding media and shell during the hydrodynamical phase of the interaction @xcite .
for obvious reasons , asymmetries of the shell will increase the velocity gradient seen over the shell .
the observations of sn 2005hj indicate a very flat plateau that , in principle , may further constrain the properties of the shell . for sn 2005hj , this may already indicate a rather spherical shell and hint toward the pdd scenario or mergers with an intermediate disk of very large scale heights .
however , additional information needs to be taken into account such as detailed line profiles and statistical properties to break this degeneracy between mergers and pdds . as a next step ,
detailed models for the explosion , light curves and spectra tailored toward individual events need to be constructed .
whereas the mean velocity of the shell for a given mass is dictated by momentum conservation , the thickness of the shell is limited by the distance of the shell material , the distance sound can travel during the interaction , and the specific density profile within the shell . with increasing distance of the shell , the relative size ( and corresponding velocity spread ) becomes smaller because the sound speed remains about the same .
the intrinsic color will be sensitive to the optical depth of the shell , which is governed by the magnitude of the density jump and thus depends on the distance of the interacting shell from the wd @xcite . the blue @xmath11 color for sn 2005hj may hint of a need to modify the distance and structure of the shell .
precise analysis of such `` non - stable '' features requires detailed model fitting beyond the scope of this paper . in the recent past ,
both the scenarios leading to shell - like structures have been discounted .
pdd models have been dismissed because 3d deflagration models showed that the wd becomes unbound and thus pulsations would not occur @xcite .
however , it has recently been shown that this solution depends mainly on the ignition conditions , namely the number and locations of ignition points leading to single or multiple bubble solutions , and mixture of bubble solutions leading to raleigh - taylor instabilities . as a result ,
solutions with fewer bubbles are likely to result in a reduced amount of burning , thus only slightly unbinding the wd and increasing the possibility of pdds @xcite .
similarly , the merging scenario has been dismissed because the wd may undergo burning during the merger and result in an accretion induced collapse @xcite , and also on the basis of the long merging time scale .
however both of these results depend sensitively on the initial conditions , and new pathways to the actual merging may effect the results @xcite . in light of our results ,
the predicted death of both of these scenarios may be premature , and further studies are needed .
we would like to thank the staff of the hobby - eberly telescope and mcdonald observatory for their support and the rotse collaboration .
we give specific thanks to j. caldwell , s. odewahn , v. riley , b. roman , s. rostopchin , m. shetrone , e. terrazas , and m. villarreal for their skilled observations with the het , and to f. castro , p. mondol , and m. sellers for their efforts in screening potential sn candidates .
this work made use of the suspect on - line database of sne spectra ( http://bruford.nhn.ou.edu/$\sim$suspect/index1.html ) .
this research is supported , in part , by nasa grant nag 5 - 7937 ( ph ) and nsf grants ast0307312 ( ph ) and ast0406740 ( rq & jcw ) .
cccccc oct 26.30 & 53669.30 & -6 & 2x600 & gg385 & 2.0 + oct 27.20 & 53670.20 & -5 & 4x600 & og590 & 1.5 + oct 27.28 & 53670.28 & -5 & 2x600 & gg385 & 1.5 + nov 1.27 & 53675.27 & 0 & 2x600 & gg385 & 1.5 + nov 1.29 & 53675.29 & 0 & 2x550 & og515 & 1.5 + nov 3.27 & 53677.27 & 2 & 2x480 & gg385 & 1.5 + nov 4.26 & 53678.26 & 3 & 2x600 & og515 & 1.5 + nov 6.17 & 53680.17 & 5 & 2x600 & gg385 & 1.5 + nov 6.18 & 53680.18 & 5 & 2x600 & og515 & 1.5 + nov 11.25 & 53685.25 & 10 & 3x600 & og515 & 1.5 + nov 19.22 & 53693.22 & 18 & 4x600 & og515 & 1.5 + nov 26.20 & 53700.20 & 25 & 2x600 & og515 & 1.5 + dec 5.18 & 53709.18 & 34 & 3x600 & og515 & 2.0 + rrrrr @xmath36 & 10820 & 140 & 0.38 & 140 + @xmath38 & 10800 & 110 & 0.32 & 120 + 0 & 10640 & 90 & 0.52 & 110 + 2 & 10440 & 100 & 0.60 & 110 + 3 & 10640 & 90 & 0.57 & 110 + 5 & 10680 & 80 & 0.57 & 110 + 10 & 10530 & 100 & 0.60 & 100 + 18 & 10550 & 120 & 0.48 & 90 + 25 & 9850 & 90 & 0.25 & 60 + 34 & & & & + | het optical spectra covering the evolution from about 6 days before to about 5 weeks after maximum light and the rotse - iiib unfiltered light curve of the `` branch - normal '' type ia supernova sn 2005hj are presented .
the host galaxy shows region lines at redshift of @xmath0 , which puts the peak unfiltered absolute magnitude at a somewhat over - luminous @xmath1 .
the spectra show weak and narrow lines , and for a period of at least 10 days beginning around maximum light these profiles do not change in width or depth and they indicate a constant expansion velocity of @xmath2 km s@xmath3 .
our observations indicate that si is confined to a relatively narrow velocity region .
we analyzed the observations based on detailed radiation dynamical models in the literature .
the models considered include delayed detonations , deflagrations , pulsating delayed detonations , and tamped detonation / merger scenarios .
whereas the first two classes of models have been used to explain the majority of sne ia , they do not predict a long velocity plateau in the minimum with an unvarying line profile . pulsating
delayed detonations and merger scenarios form shell - like density structures with properties mostly related to the mass of the shell , @xmath4 , and we discuss how these models may explain the observed line evolution ; however , these models are based on spherical calculations and other possibilities may exist .
sn 2005hj is consistent with respect to the onset , duration , and velocity of the plateau , the peak luminosity and , within the uncertainties , with the intrinsic colors for models with @xmath5 .
our analysis suggests a distinct class of events hidden within the branch - normal sne ia . if the predicted relations between observables are confirmed , they may provide a way to separate these two groups .
we discuss the implications of two distinct progenitor classes on cosmological studies employing sne ia , including possible differences in the peak luminosity to light curve width relation . |
the correlation between stellar mass and gas - phase metallicity in galaxies ( the mass - metallicity , or mz relation ) is sensitive to the processes that regulate their growth .
this relation is shaped by gas outflow and accretion rates , the enrichment of these gas flows , and the star formation efficiency of galaxies @xcite .
it is expected , then , that observations of the mz relation towards low masses and high redshifts will constrain the assembly of galaxies and the metal enrichment of the intergalactic medium ( igm ) . to date , a complete picture of metallicity evolution has not emerged . while studies focusing on higher mass galaxies at @xmath16 have shown lower metallicities than are seen locally @xcite , evolution below @xmath17 remains unconstrained . in @xcite , we began to probe masses just below @xmath18 at intermediate redshifts ; nevertheless measuring metallicities at higher redshifts ( @xmath19 ) has been challenging because of the requirement for infrared spectroscopy of large , faint samples of galaxies .
although lower masses and metallicities have been reached for some strongly lensed galaxies ( e. g. @xcite ) , the uncertainties of these measurements are large . in this letter , we present the mz relation at @xmath20 , derived by stacking 83 galaxies with observations covering the @xmath2 metallicity diagnostic , ( [ ] @xmath3 + [ ] @xmath4)/h@xmath5 @xcite . our sample , which is drawn from _ hubble space telescope _ wide field camera 3 ( wfc3 ; @xcite ) grism observations , reaches @xmath21 at @xmath22 .
by focusing on measurements from stacked spectra , we avoid bias introduced by the requirement that h@xmath5 be detected in each spectrum .
we adopt a @xcite initial mass function ( imf ) and take @xmath23 , @xmath24 and @xmath25 km s@xmath26 . when reporting measurements of doublet lines ( i.e. [ ] @xmath4 ) , we use `` [ ] '' or `` [ ] '' to refer to both lines .
we use data from the wfc3 infrared spectroscopic parallel ( wisp ) survey , as described by @xcite , @xcite and @xcite .
the wisp survey uses both the g102 ( @xmath27 , @xmath28 ) and g141 ( @xmath29 , @xmath30 ) grisms , enabling measurements of @xmath2 from @xmath22 . as the wisp survey is ongoing ,
we restrict our analysis to 29 fields that have emission - line catalogs and supporting optical imaging , either from wfc3/uvis ( 24 fields ) or from the large format camera @xcite on the 200-inch hale telescope at palomar observatory ( @xmath31 and @xmath32 bands ; 5 fields ) . five - band galaxy spectral energy distributions ( seds ; including _
spitzer_/irac 3.6 photometry for most ) are obtained from photometric catalogs described in @xcite and @xcite . using the wisp emission line catalogs
@xcite , we identify 128 galaxies with secure redshifts measured from multiple lines . objects with contamination from overlapping spectra were removed , leaving 74 galaxies . of these ,
we exclude five galaxies which have inadequate mass constraints : three fall in the uvis chip gap , one is detected in only one broad - band filter , and one is blended with another object in ground - based @xmath31 and @xmath32 images .
these cuts leave 69 galaxies from wisp .
we supplement this sample with publicly released grism data covering the ultra deep field ( udf ) ; since the udf is observed only in g141 , @xmath2 is covered only for the 25 galaxies between @xmath33 . of these , we exclude three galaxies because of contamination , two because their spectra do not contain emission lines , and one with incomplete imaging near the edge of the udf .
we construct seds for the remaining 19 galaxies using the nine bands of imaging in the hst extreme deep field ( @xcite ) .
photometric catalogs were created by using sextractor @xcite in dual image mode , with f160w as the detection image .
we use mag_auto to estimate the total flux .
errors are increased to account for correlated noise in drizzled data @xcite . among the 88 galaxies in the combined sample , we identify and remove five agn candidates ( [ agn ] ) .
the mean redshift for the remaining 83 galaxies is @xmath34 .
we measure emission line fluxes from the individual spectra by simultaneously fitting gaussian profiles to the [ ] , [ ] , and h@xmath5lines , as well as h@xmath35 and [ ] when they are covered . contribution to h@xmath35from unresolved [ ] @xmath36 is taken to be negligible , consistent with our followup spectroscopy ( masters et al . , in prep ) .
each region of the spectrum ( i.e. [ ] , [ ] + h@xmath5 , h@xmath35 + [ ] ) is also fit with a linear continuum . because the dispersion in the grism spectra diverges from the adopted solution at the red and blue ends , we allow for shifts of up to @xmath37 between the three widely separated groups of emission lines .
we take the redshift from the fit to [ ] and h@xmath5 .
additionally , since the [ ] @xmath38 doublet is marginally resolved , we fit both lines assuming a 3:1 flux ratio
. finally , the emission lines should be spectrally unresolved with widths set by the size of the galaxy in the dispersion direction . since the dispersion is two times higher in the g102 spectra , we require lines measured in this grism to have widths a factor of two narrower ( in ) than those of the redder lines .
stellar masses are derived with the fast software @xcite , fitting bc03 @xcite template spectra to our broadband photometry .
we take a grid of stellar population parameters that include : exponentially declining star formation histories with e - folding times ranging from 40 myr to 10 gyr ; ages ranging from 50 myrs to the age of the universe , and @xmath39 for a @xcite extinction curve .
additionally , we use a @xcite imf with metallicities of 0.004 , 0.008 , and 0.02 ( @xmath40 ) .
we removed the emission line contribution to our broadband photometry , estimating the corrections by integrating our continuum subtracted spectra ( see [ stacking ] ) under the bandpasses that cover the line emission . in figure [ agn_diagnostics_fig ]
, we use the mass excitation ( mex ) diagram to identify agn that could contaminate our stacked spectra .
this diagnostic uses mass as a proxy for the [ ] @xmath41/h@xmath35 ratio , which is not measured for our sample .
the solid orange lines show the agn threshold from @xcite ; objects above and to the right are classified as agn , and objects between the solid curves at @xmath42 are star - forming / agn composites
. however , the use of the mex diagnostic is compromised by the evolution of the mz relation . at fixed metallicity ( or [ ] /h@xmath5 ) , galaxies at @xmath43 are 1 - 2 dex more massive than @xmath44 galaxies @xcite .
therefore , we expect that the star - forming locus in figure 1 shifts to the right at higher redshifts .
the dashed curve illustrates the possible evolution , by shifting the juneau et al . curves a conservative 1 dex in stellar mass . taking this modified threshold
, we find five agn candidates .
these objects are excluded from further analysis .
we acknowledge that the more aggressive agn selection would slightly modify our mz relation above @xmath45 , increasing the metallicity derived in [ met ] by 0.03 dex at @xmath46 .
nevertheless , this shift falls within our uncertainties , and does not affect the conclusions drawn in 4 and 5 . for the galaxies that are undetected in h@xmath5 , we can not rule out an agn for any given object .
however , given the distribution of detections and limits in figure [ agn_diagnostics_fig ] , it is unlikely that a sizable fraction of these galaxies have very high [ ] /h@xmath5 ratios .
we also note that none of the udf galaxies are identified as agn from x - ray counterparts in the @xcite catalog .
therefore , we conclude that stacked spectra will be minimally affected by this contamination .
overall , the low agn content in our sample is consistent with the _ low mass _ , grism - selected objects reported by @xcite . in order to stack spectra
, we follow a procedure similar to @xcite .
first , we subtract a model continuum for each galaxy , made by masking the emission lines and smoothing with a 30 pixel boxcar .
each spectrum is then normalized by its [ ] flux , so that all galaxies contribute equally . finally , the spectra are de - redshifted and combined by taking the median .
the relative line fluxes are determined by fitting gaussian profiles .
the measurement follows the same procedure described in
[ em_line_meas ] , except the widths of the emission lines are allowed to differ from one another because the relative contribution of the g102 and g141 data varies with rest - wavelength .
error spectra are generated by bootstrap resampling the spectra in each mass bin to generate 500 artificial stacks , and taking the rms at each wavelength .
errors on the relative line fluxes are calculated by propagating the uncertainties on the fit parameters . our sample of 83 galaxies
is divided into four mass quartiles with @xmath47 , @xmath48 , @xmath49 , @xmath50 .
the stacked spectra are shown in figure [ spec_fig ] .
lcccccccccc 8.2 & 21 & 1.82 & 110 & @xmath51 & @xmath52 & @xmath53 & @xmath54 & @xmath55 & 7 & + 8.8 & 22 & 1.73 & 47 & @xmath56 & @xmath57 & @xmath58 & @xmath59 & @xmath60 & 9 & 8.17 + 9.2 & 21 & 1.77 & 36 & @xmath61 & @xmath62 & @xmath63 & @xmath64 & @xmath65 & 13 & 8.33 + 9.8 & 19 & 1.74 & 29 & @xmath66 & @xmath67 & @xmath68 & @xmath69 & @xmath70 & 21 & 8.60 + 8.2 & 21 & 1.82 & 110 & @xmath71 & @xmath72 & @xmath73 & @xmath74 & @xmath75 & 3 & + 8.8 & 22 & 1.73 & 47 & @xmath76 & @xmath77 & @xmath78 & @xmath79 & @xmath80 & 4 & 8.22 + 9.2 & 21 & 1.77 & 36 & @xmath81 & @xmath82 & @xmath83 & @xmath84 & @xmath85 & 6 & 8.39 + 9.8 & 19 & 1.74 & 29 & @xmath86 & @xmath87 & @xmath88 & @xmath89 & @xmath90 & 9 & 8.67 [ results_table ] a few steps must be taken to infer metallicities from our stacked spectra .
first , since [ ] and [ ] are widely separated in wavelength , a dust correction is required .
we estimate extinction by measuring the balmer decrement for each of the five galaxies in our @xmath91 subsample where both h@xmath35 and h@xmath5 are detected ( at @xmath92 significance or greater ) .
consistent with our findings in @xcite , we measure an average h@xmath35/h@xmath5 @xmath93 , with an rms of 1.1
. we also stack all of the 28 galaxies with @xmath94 ; here we find h@xmath35/h@xmath5 @xmath95 , consistent with no dust . in the analysis that follows , we calculate line ratios assuming both no dust , and h@xmath35/h@xmath5 @xmath96 ( with a @xcite extinction curve ) . in the latter case
, the scatter on the balmer decrement is included in the metallicity errors .
second , the flux measured from the h@xmath5 emission line can be artificially reduced by stellar absorption , which is estimated to have an equivalent width ( ew ) of 3 - 4 in similar grism - selected galaxies @xcite .
since our stacked spectra only represent relative line fluxes , we estimate the h@xmath5 ew from the the mean [ ] ew of the individual stacked galaxies and [ ] /h@xmath5 ratio for that stack .
the values for the rest - frame h@xmath5 ew range from 30 to 110 ( see table [ results_table ] ) .
the relative h@xmath5fluxes are increased by @xmath97 to account for this absorption .
finally , @xmath2 is double - valued ; a single measurement can correspond to either a high or low metallicity .
determining whether a galaxy lies on the high or low metallicity branch can be challenging @xcite .
nevertheless , under the reasonable assumption that metallicity decreases towards lower masses , an average trend is expected : as mass decreases , @xmath2 should increase to log@xmath2 @xmath98 , before turning over and decreasing .
for the first time at high redshift , this trend is observed in figure [ branch_tests ] .
we conclude that the @xmath99 bin marks the turn - around , " with the two higher mass bins falling on the upper branch of @xmath2 and the lowest mass bin implying lower - branch metallicities .
the middle panel of figure [ branch_tests ] supports this conclusion ; as in the local universe , the lowest metallicity galaxies have more highly ionized gas , as indicated by higher o32 ( log [ ] / [ ] ) .
we have tested that the results from stacking are robust to scatter between the upper and lower - branch bins .
taking a simulated mz relation with the same slope as we have measured , we adopt an intrinsic metallicity scatter of 0.1 dex ( as at @xmath100 ; @xcite ) and 0.2 dex of scatter in stellar mass due to our typical measurement error .
we find that fewer than 10% of the galaxies in the bins adjacent to the turn - around should be attributed to the opposite @xmath2 branch , and less than 1% of galaxies in our high - mass bin will fall on the lower branch .
this exercise shows that metallicities may be overestimated by 0.03 - 0.04 dex in the @xmath101 bin , and underestimated by the same amount in our lowest mass bin .
we do not apply a correction for this effect , since its amplitude is smaller than our uncertainties .
finally , as detailed in @xcite , different metallicity calibrations of strong - line ratios yield systematically different abundances . in table
[ results_table ] , metallicities are reported on both the kobulnicky & kewley ( 2004 ; kk04 ) and @xcite scale .
in the local universe , metallicity , stellar mass , and sfr form a plane called the fundamental metallicity relation ( fmr ; @xcite ) , in the sense that galaxies with high specific sfrs have lower metallicities . as a result , it is important to consider selection effects when interpreting the mz relation .
for example , uv - continuum or emission - line selected samples may be biased towards higher sfrs and lower metallicities than mass - selected samples .
therefore , we use h@xmath5 luminosity ( measured analogously to the h@xmath5 ew , using the mean [ ] luminosity and the [ ] /h@xmath5 ratio ) to estimate the sfr for each stack .
the h@xmath5 luminosity is corrected for dust and stellar absorption , and multiplied by 2.86 to obtain the h@xmath35 luminosity .
we then calculate the sfr from the relation in @xcite , converted to a @xcite imf and @xmath102 ) .
however , in order to compare to previous measurements of the star - forming main sequence , we report sfrs for solar metallicity . ] . in table [ results_table ] we list the average sfr in each mass bin .
figure [ mz_fig ] shows that our dust - corrected sfrs lie very close to an extrapolation of the @xmath14 star - forming main sequence from @xcite .
this comparison shows that our grism - selected sample is not likely biased towards high specific sfrs .
remarkably , @xcite report that the fmr does not evolve with redshift ( at least out to @xmath103 ) .
further studies confirm this claim for smaller samples of low mass galaxies @xcite .
using our stack - averaged measurements , we can also test for evolution . in table 1
we list the oxygen abundance , predicted from the fmr in @xcite .
this quantitiy shows excellent agreement with the metallicities calculated on the @xcite calibration ( on which the fmr is defined ) .
we conclude that our sample shows no evidence for evolution of the fmr .
figure [ mz_fig ] shows the mz relation derived from our stacked spectra , for both the dust corrected and uncorrected measurements .
it is clear that the dust correction represents only a small adjustment .
excluding our lowest mass bin , our ( dust corrected ) metallicities are 0.33 , 0.17 , and 0.10 dex lower than the sdss @xmath100 measurement at @xmath104 , and @xmath105 @xcite , suggesting more significant evolution in low mass galaxies . a similar trend has been observed in the intermediate - redshift mz relation ( also shown in figure [ mz_fig ] ; @xcite ) .
in fact , in @xcite we interpreted this evolution as evidence for downsizing , where higher mass galaxies evolve onto the local mz relation at earlier times .
while the uncertainty on the evolution measured by the present data remains large , it is consistent with downsizing .
stacking larger samples of grism spectra can help to reduce the uncertainties , but , at the same time , robust measurements of the local mz relation are needed for @xmath106 . comparison to local dwarf galaxies remains difficult , as most have metallicities derived directly from electron temperatures ( which differ systematically from strong - line metallicities ) , or suffer from uncertainties regarding the @xmath2 branch @xcite .
we also compare our mz relation to the measurement from @xmath13 galaxies reported by @xcite .
these metallicities were measured from the [ ] @xmath41/ h@xmath35 ratio in stacked spectra ( using the @xcite calibration ) , and are converted to kk04 metallicities using @xcite .
figure [ mz_fig ] shows that our mz relation agrees with the results from erb et al .
the somewhat higher metallicities in our data are consistent with the lower redshift ( @xmath107 ) of our sample .
the broad agreement between these measurements suggests that the systematic offsets between metallicity calibrations ( which we removed ) are not drastically different at low and high redshifts . finally ,
while some low mass lensed galaxies have metallicity measurements at these redshifts ( e.g. @xcite ) , large scatter and many weak upper limits preclude meaningful comparisons . finally , we compare our data to theoretical models for the mz relation .
following our implementation in @xcite , we show two models calculated from the prescriptions in @xcite .
figure [ mz_fig ] shows that a model regulated by momentum - driven winds predicts a shallower mz slope than an energy - driven wind model .
while we showed in @xcite that the intermediate redshift mz relation is reproduced by momentum - driven winds , the higher redshift relation suggests evolution , as it more closely follows the energy - driven wind model .
it is unlikely that a different choice of metallicity calibration would flatten our observed mz relation to bring it closer in line with momentum - driven winds ; among the widely used strong - line metallicity calibrations , only @xcite and @xcite yield mz relations with significantly shallower slopes compared to a kk04 mz relation .
however , these shallow slopes may be inaccurate , owing to a saturation of the [ ] @xmath108 diagnostic at high metallicities @xcite .
nevertheless , the metallicity errors remain large , so while our data favor energy - driven winds , we can not rule out the shallower slope associated with momentum - driven winds .
we have presented the first measurement of the mz relation from a large sample of low mass , high redshift galaxies .
although emission - line selected , these galaxies are not extreme in their sfrs or metallicities .
rather , our results show excellent agreement with extrapolations of the star - forming main sequence and mz relation at similar redshifts .
we conclude that the low - mass , high - redshift mz relation suggests a downsizing evolution and a preference for feedback from energy - driven winds .
we acknowledge dawn erb , kate whitaker and danielle berg for helpful discussions .
this research was supported by an appointment to the nasa postdoctoral program at the goddard space flight center , administered by oak ridge associated universities through a contract with nasa .
ah also acknowledges support from hst go 11696 , 12284 , and 12568 . | we present the first robust measurement of the high redshift mass - metallicity ( mz ) relation at @xmath0 , obtained by stacking spectra of 83 emission - line galaxies with secure redshifts between @xmath1 . for these redshifts , infrared grism spectroscopy with the _ hubble space telescope _
wide field camera 3 is sensitive to the @xmath2 metallicity diagnostic : ( [ ] @xmath3 + [ ] @xmath4)/h@xmath5 . using spectra
stacked in four mass quartiles , we find a mz relation that declines significantly with decreasing mass , extending from 12+log(o / h ) @xmath6 at @xmath7 m@xmath8 , to 12+log(o / h)@xmath9 at @xmath10 . after correcting for systematic offsets between metallicity indicators , we compare our mz relation to measurements from the stacked spectra of galaxies with @xmath11 and @xmath12 . within the statistical uncertainties , our mz relation agrees with the @xmath13 result , particularly since our somewhat higher metallicities ( by around 0.1 dex ) are qualitatively consistent with the lower mean redshift ( @xmath14 ) of our sample . for the masses probed by our data , the mz relation shows a steep slope which is suggestive of feedback from energy - driven winds , and a cosmological downsizing evolution where high mass galaxies reach the local mz relation at earlier times .
in addition , we show that our sample falls on an extrapolation of the star - forming main sequence ( the sfr - m@xmath15 relation ) at this redshift .
this result indicates that grism emission - line selected samples do not have preferentially high sfrs .
finally , we report no evidence for evolution of the mass - metallicity - sfr plane ; our stack - averaged measurements show excellent agreement with the local relation . |
radio frequency identification ( rfid ) is a promising technology for the proliferation of internet of things ( iot ) applications , and it can be used to detect and identify the items in the proximity@xcite . due to their cost
effective , durable , and energy efficient operation @xcite , rfid technology has been used in wide range of applications such as asset management @xcite , access control@xcite , public safety@xcite , localization@xcite , and tracking @xcite . among these , enabling high accuracy localization for massively deployed iot devices carries critical importance for a diverse set of iot applications @xcite .
localization using radio frequency ( rf ) signals has been actively researched in the literature over the past decades@xcite .
outdoor localization is mostly handled with global positioning system ( gps ) technology whereas indoor localization requires alternative approaches since gps needs a line - of - sight connection between user equipment and satellites .
moreover , massive deployment of iot devices necessitates energy and cost efficient localization methods for prolonged durations .
the rfid technology hence becomes a promising alternative for cost - effective , energy efficient indoor identification and localization for massively deployed iot . an ultra high frequency ( uhf ) rfid communication is fundamentally different from the conventional rf communication since it has two distinct links : the forward ( power - up ) and the reverse ( backscatter ) link .
the forward link powers the passive rfid tags and the reverse link carries the information of tags .
ability to power - up tags in the forward link enables _ battery - less _ operation of rfid tags@xcite , which is a major advantage of rfid systems for low - power iot applications . in general , there are two configurations for uhf rfid systems : 1 ) monostatic configuration , and 2 ) bistatic configuration . in the monostatic configuration ,
a single reader antenna transmits the continuous wave , which powers up the passive tag , and subsequently receives the backscattered information signal from the tag . in the bistatic configuration the transmission and reception
are handled by different reader antennas as shown in fig .
[ fig : iotframework ] .
these antennas might be co - located ( i.e. , at same location , closely spaced ) or dislocated ( at separate locations ) .
a particular challenge with both configuration is that complex , directional , and three dimensional rfid propagation models need to be explicitly taken into account to accurately characterize the real - world forward / backward propagation channels . in this paper
, we use sophisticated and realistic 3d path - loss and radiation models to study fundamental lower bounds on the localization accuracy of received signal strength ( rss ) based uhf rfid localization systems for both monostatic and bistatic configurations .
the main contributions of this work are as follows : 1 ) cramer - rao lower bound ( crlb ) on the localization accuracy are derived in closed - form considering an enhanced rss model , using the _ directional _ and _ 3d _ radiation pattern from uhf rfid reader antennas , and the concept of _ localization coverage _ ; 2 ) tag and reader sensitivity is incorporated into analytic derivations both for monostatic and bistatic scenarios , to derive _ localizability _ and _ localization coverage _ metrics ; 3 ) extensive computer simulations are carried out to compare the localization accuracy of the maximum likelihood ( ml ) technique with the crlbs , considering directional radiation patterns and using different configurations for rfid reader antennas .
our analysis and simulation results show that for certain scenarios , using bistatic antenna configuration as in fig .
[ fig : iotframework ] may increase the average localization coverage by @xmath0 when compared to monostatic rfid configuration .
another important parameter in the antenna configurations is the elevation angle @xmath1 . especially with lower transmit powers
, it affects the localization coverage and accuracy .
corner placement of antennas for @xmath2 with @xmath3 mw gives @xmath4 localization coverage , while @xmath5 and @xmath6 results in @xmath7 .
our results for the specific rfid configuration show that it is possible to locate a tag within @xmath8 meter error with a probability of @xmath9 with corner placement of antennas , whereas this probability drastically reduces to @xmath10 when side placement is used for @xmath11 with bistatic configuration .
the rest of this paper is organized as follows .
literature review for rss - based localization in passive uhf rfid systems is provided in section [ sect : litrev ] . in section [
sect : systemmodel ] , the system model is described in detail which involves a 3d radio propagation model for rfid systems .
the concept of localizability is defined , as well as localization coverage percentage in section [ sect : covareas ] .
section [ sect : crlb ] derives the crlbs and the maximum likelihood estimator ( mle ) based on the likelihood function for an rfid tag s location for the considered rfid scenario .
numerical results are provided in section [ sect : nresults ] , and concluding remarks are given in section [ sect : conclusion ] .
.35 .55 [ fig : system ] although there are several studies in the literature that investigate rss - based localization with rfid technology @xcite , fundamental lower bounds on rfid - based wireless localization are relatively unexplored . in @xcite , authors used a mobile robot with rfid reader antennas to generate map of an indoor environment with rfid tags on the walls .
after the mapping phase , the robot may locate itself inside the building based on the closest tag information . in the landmarc localization technique introduced in @xcite ,
reference rfid tags are used for implementing rss - based indoor localization method , where fixed - location reference tags with known locations are used to localize the tags . in @xcite , authors improve landmarc approach to tackle with multipath effects and rf interference .
a probabilistic rfid map - based technique with kalman filtering is used to enhance the location estimation of the rfid tags in @xcite . another approach to localize the rfid tags is studied in @xcite , which uses the phase difference information of backscattered signal of the rfid tags . in @xcite , authors consider a multipath environment to derive the crlbs on the position error of an rfid based wireless localization system .
geometry of the deterministic multipath components and the interfering diffuse multipath components are considered in the backscatter channel model .
typically a simple path - loss model is used for rfid propagation models in the existing literature @xcite , which employs free - space path - loss signal strength model .
these models are not capable of accurately capturing the radiation pattern of rfid reader antennas since they are highly directional .
there are also several experimental studies in the literature related to rss - based uhf rfid localization systems . in @xcite , an experimentation with passive uhf rfid system
is conducted to investigate the relationship between rss and distance .
recently in @xcite , crlb of rss - based localization are derived considering a frequency dependent path - loss propagation model , where the model explicitly depends on the transmit power level and the transmission frequency .
accuracy of several localization techniques are compared to crlb with given path - loss model via simulations and experiments . in @xcite , authors used @xmath12-nearest neighbor ( knn ) algorithm to estimate the location of the target tag from rss information . an experiment involving four antennas and seventy tags is conducted , which resembles to the simulation scenario in our manuscript . it is shown that power control techniques may significantly improve localization accuracy .
effects of multipath propagation and signal scattering are considered in @xcite for passive uhf rfid localization , using mle and linear least square techniques .
a localization algorithm using the differences of rss values from various tags under same conditions is also proposed .
its performance , which is shown to outperform the knn algorithm used in landmarc@xcite . a two - parameter path - loss model for uhf rfid systems is constructed in @xcite , which shows that the rss of rfid systems are slightly more stable than wifi rss values , and this yields more precise location estimates for rfid rss - based localization . in our earlier work@xcite
, we have studied the bounds on rfid localization for monostatic rfid configuration . in this study ,
our additional contributions include : 1 ) use of bistatic antenna configuration and different antenna placement which provides a more generalized framework , 2 ) use of an enhanced rss model with lognormal distributed noise which yields different crlb formulations , 3 ) incorporation of reader antenna and tag sensitivity into theoretical analysis , 4 ) study of _ localization coverage _ for rfid tags , outside of which they can not be localized with a reasonable accuracy , and 5 ) extensive new simulations to study the effects of various parameters and configurations .
in the rest of this paper , we consider the rfid localization scenario as shown in fig . [
fig : system ] .
in particular , fig . [
fig : system](a ) illustrates a monostatic antenna configuration , where the reader antenna is both the transmitter and the receiver . on the other hand , the bistatic antenna configuration is shown in fig .
[ fig : system](b ) , where one antenna transmits the power - up signal for rfid tag , and the other antenna receives the backscattered signal from the tag .
we will consider the more general case of bistatic antenna configuration , and study the monostatic configuration as a special case .
for the considered scenario , let @xmath13 rfid reader antennas be mounted on the walls , located at a height of @xmath14 meters from the ground for the @xmath15th antenna . as shown in fig .
[ fig : system](b ) , rfid reader antennas @xmath15 and @xmath16 ( which are the forward and reverse antennas , respectively ) are tilted by an angle @xmath17 and @xmath18 , respectively , with reference to the azimuth plane .
the goal is to localize an rfid tag , which is located at a distance @xmath19 below a reader antenna @xmath15 .
the total backscattered received power @xmath20 at a bistatic configuration of reader antenna @xmath15 and antenna @xmath16 , which are located at @xmath21 and @xmath22 , respectively while the position of the tag is @xmath23 , is given by @xcite : @xmath24 which can be written in logarithmic scale as @xmath25= 20\log_{10}\big(\tau\mu_{\rm t}\rho_{\rm l}p_{\rm tx}g^2_{\rm t}|h_ih_j\gamma|^2\big)\nonumber\\&+20\log_{10}\big(g_{\rm r}^i\big)+20\log_{10}\big(g_{\rm r}^j\big)\nonumber\\ & + 20\log_{10}\big(l(d_i)\big)+20\log_{10}\big(l(d_j)\big),\label{eq : preceived2}\end{aligned}\ ] ] @xmath26 where @xmath27 is a coefficient that quantifies the specific data encoding modulation details that can be calculated using power density distribution of the tag s signal . according to the epcglobal c1g2 specifications @xcite , any tag in the interrogation zone of the reader
can send back its information by reflecting the incoming continuous wave .
the power transfer efficiency @xmath28 $ ] in quantifies how well the tag is impedance - matched to the antenna .
polarization loss factor @xmath29 captures the loss due to the mismatch between the polarization of a transmitter antenna and a receiver antenna . the effective isotropic radiated power ( eirp ) of the rfid reader antenna
is shown as @xmath30 , while @xmath31 and @xmath32 are the gain of the rfid reader antenna @xmath15 and tag antenna , respectively , and @xmath33 is the channel pathloss defined by : @xmath34 where @xmath35 is the wavelength of the signal , and @xmath36 is the distance between the tag and the reader antenna @xmath15 .
the transmit power limit of rfid reader antennas , which is critical for coverage of the reader , is 2 w in effective radiated power ( erp ) as stated in epc gen2 protocol for uhf rfid systems .
this makes the eirp limit for rfid readers 35.15 dbm , which is larger than the highest transmit power of 3 w that was used in our simulations@xcite .
the forward - link and backscatter - link channels are represented with @xmath37 and @xmath38 .
the parameter @xmath39 in is the differential reflection coefficient of the tag which is a function of the tag antenna gains @xmath32 , the _ radar cross section _ rcs denoted by @xmath40 , and the communication wavelength @xmath35 as follows@xcite : @xmath41 in passive uhf rfid applications , the goal is to maximize rcs , which characterizes the scattered power , while still absorbing sufficient power to operate the chip of the tag . in our study , we have utilized statistical models for rcs and @xmath39 which we obtained from @xcite .
assuming a scenario as illustrated in fig .
[ fig : system ] , and adopting the expression provided by @xcite , a modified _ directional _ gain of a patch antenna for a 3d propagation environment can be expressed as follows : @xmath42 ^ 2~ , \label{eq : patchantennagain2}\end{aligned}\ ] ] where @xmath43 , with @xmath17 and @xmath44 being the elevation and azimuthal angles of the patch antenna @xmath15 , respectively .
the parameter @xmath19 in is the difference between height of the reader antenna and the height of tag .
location of a tag with respect to reader antenna is defined with relative elevation and azimuthal angles , and distance between tag and reader antenna . on the other hand , derivation of crlb
requires translation from polar coordinate system to the cartesian coordinate system .
gain of patch antenna is defined in , which can be represented in the cartesian coordinate system as @xmath45 where @xmath46 is coordinate of the antenna-@xmath15 , and @xmath23 is the location of tag . the distance between tag and antenna-@xmath15 is defined with @xmath47 , while its projection on the @xmath48-plane is given with @xmath49 . as in fig .
[ fig : system](a ) , monostatic rfid is a special case of bistatic rfid configuration , there the transmitter and receiver antenna are identical .
this makes @xmath50 and @xmath51 equal ( @xmath52 ) .
therefore , the received power in dbm at the reader antenna with monostatic configuration simplifies to : @xmath53for @xmath54 .
note that , for monostatic configuration , @xmath55 when @xmath56 , for all @xmath57 .
therefore , monostatic configuration essentially uses a subset of the antenna reader pairs in bistatic configuration during localization . in the rest of this paper , bistatic configuration as defined in will be assumed to capture measurements at all pairwise combinations of antenna readers , including those corresponding to monostatic configurations .
0.33 0.33 0.33 [ fig : t45 ] in this section , we will first introduce the concepts of _ tag antenna sensitivity _ , _ reader antenna sensitivity _ , and _ localization coverage _ of an rfid system , which corresponds to the spatial region in which an rfid tag will be considered _
localizable_. the passive tags do not have an internal power structure to modulate or transmit any signal .
they use the received power to both modulate the signal with the internal chip , and backscatter modulated signal to reader . as one can expect , rfid tags have certain power requirements .
state - of - art tags are able to modulate signals with rss as low as @xmath58 dbm@xcite , and will not be able to detect the received signal at lower power levels .
the rss from @xmath15th reader at tag is defined with @xmath59for @xmath54 .
note that is a subset of and which characterize the rss after round - trip signal propagation , since represents only the forward - link . in either monostatic or bistatic configuration
, an rfid reader must correctly detect the backscattered modulation from the tag , which relies on the reader antenna sensitivity .
therefore , the received power as in [ eq : preceived2 ] and [ eq : preceived2mono ] must be larger than the reader antenna sensitivity in order to be detected .
the _ detection coverage _ of an rfid configuration is defined as detectability of a tag at a certain location with that configuration .
the detectability is assumed deterministic with respect to rss and sensitivity of rfid reader antenna . in this subsection
, we investigate the impact of the sensitivity of the tag and reader antennas on localization performance .
we introduce below several new metrics for characterizing tag / reader sensitivities and localization coverage . * _ definition 1 : _ * the _ coverage _ for a given antenna pair @xmath60 at a given location @xmath61 is captured by a binary deterministic parameter @xmath62 , which is defined as : @xmath63 using and .
due to nonlinearity of antenna propagation model , in order to localize a tag , at least two different rss measurements from that tag at a particular position @xmath61 are needed . on the other hand
, there might be some tags which are detected from only a monostatic antenna or a bistatic antenna pair , and those tags can not be localized due to limited information . note that for monostatic configuration , @xmath64 if @xmath56 .
then we can define the localizability of a tag as follows .
* _ definition 2 : _ * a tag at a given location @xmath61 is _ localizable _ if the following condition is satisfied @xmath65 where @xmath13 is the number of antennas in the system , and @xmath66 is the total number antenna configurations that can detect the tag at location @xmath61 . * _ definition 3 : _ * the _ localization coverage _ of a tag at position @xmath61 is defined with @xmath67 as follows : @xmath68 the rfid tag can be localized at position @xmath61 when @xmath69 , and is not localizable when @xmath70 . *
_ definition 4 : _ * the _ localization coverage percentage _ at a physical area @xmath71 can be formally expressed as follows : @xmath72 note that defines the percentage of localizable area to total area . in fig .
[ fig : t45 ] , results from an example deployed scenario for parameters @xmath73 , @xmath74 meter for @xmath75 , and @xmath76 mw are shown
. maximum achievable rss at any possible tag location is represented in fig .
[ fig : t45](a ) , while monostatic and bistatic localization coverage are shown in fig .
[ fig : t45](b ) and fig .
[ fig : t45](c ) , respectively .
the localization coverage percentage is @xmath77 for monostatic configuration , while it is above @xmath78 for bistatic configuration .
the number of maximum measurements increases from 4 for monostatic configuration to 16 for bistatic configuration .
the crlb is a bound on the variance of any unbiased estimator for an unknown variable , such as the location of an rfid tag , based on a set of observations . in this section ,
we define likelihood function , and derive the crlb on the accuracy of rss - based uhf rfid localization systems as a function of various parameters of interest .
we consider both monostatic and bistatic cases for crlb analysis .
subsequently , the mle for unknown rfid tag location is also defined .
comparison of crlb and mle for monostatic and bistatic configurations in various scenarios will be presented through numerical results in section [ sect : nresults ] . when a tag is localizable , then its exact location can be estimated using the measurements obtained at different antenna pairs .
the probability of an rfid tag being at a certain location can be characterized by its _
_ likelihood function__@xcite .
let @xmath79 $ ] denote the unknown location of the tag , assuming that the received power in log scale at an rfid reader antenna is subject to gaussian noise@xcite .
consider that the observations of received power in from different rfid antennas mounted on the walls are stacked in a vector @xmath80 $ ] .
then , this vector can be modeled as follows @xmath81^t,\omega_{ij}\sim\mathcal{n}(0,\sigma^2 ) , \label{eq : measured}\ ] ] where @xmath82 , @xmath83 , and @xmath84 is a vector of true rss values which has a size of @xmath85 for a bistatic configuration , and a size of @xmath13 for a monostatic configuration .
the additive noise on received power , which is assumed independent and identically distributed ( iid ) , is captured by @xmath86 , corresponding to the measurement at antenna couple @xmath15 and @xmath16 , with @xmath87 denoting the gaussian distribution with mean @xmath88 and variance @xmath89 .
then , for the general case of bistatic antenna configuration , the respective likelihood function for the received power at a location @xmath90 can be written as : @xmath91 where @xmath20 is the value of rss for reader antennas @xmath15 and @xmath16 , and it depends on the unknown tag location @xmath92 as defined in .
for the monostatic configuration , it can be easily shown that simplifies to the following likelihood function : @xmath93 based on the 3d and directional propagation model defined in , the localization coverage parameter @xmath94 defined in , and the likelihood function defined in the crlb on the variance of an unbiased estimator for @xmath90 can be defined as follows .
[ prop : p1 ] the crlb on the root mean square error ( rmse ) of an unbiased position estimator @xmath95 based on the measurements model in and the likelihood function in is given by : @xmath96 where @xmath97 $ ] is the fisher information matrix ( fim ) for @xmath98 , @xmath99 = \begin{bmatrix } \textbf{i}_{11 } & \textbf{i}_{12 } \\
\textbf{i}_{21 } & \textbf{i}_{22 } \end{bmatrix } , \label{eq : fim}\ ] ] whose elements are as derived in - .
_ proof : _ see appendix [ sect : derivation ] .
@xmath100 an example derivation of the crlb for the special case of @xmath2 for all @xmath15 is explained in detail in appendix [ sect : derivationexample ] .
while the crlb gives a lower bound on the localization rmse , an effective estimator is needed to find an rfid tag s location as accurate as possible , ideally with an rmse close to the crlb . in here
, we will define a simple mle estimator for comparison purposes with the crlb . using the likelihood function defined in
, the mle can be formulated as follows @xcite @xmath101 having a closed form solution for the mle in is not mathematically tractable due to the complexity of the directional antenna radiation pattern as captured through - . in particular ,
due to entangled sines and cosines , after equating differentiation of the likelihood function as in to zero , one can not obtain a closed form solution .
thus our problem could be solved with mle grid search , which can be represented as follows @xmath102 in our computer simulations in section [ sect : nresults ] , we consider a densely sampled grid of nearly @xmath103 uniformly spaced points .
the granularity of the grid is set to @xmath104 cm
. then , the mle solution corresponds to the grid position that maximizes the likelihood function in and can be found using exhaustive search . to reduce complexity
, the mle solution is found by a constrained search over the region that is defined by the number of rss measurements and corresponding antennas .
when there are only two rss measurements available , the search is conducted only over the positions where @xmath105 . as it is stated in section [ subsect : coverageareas ] , a grid location with only two rss measurements is still localizable , although the accuracy is relatively limited when compared to locations where more than two rss measurements are available .
based on our numerical results that will be shown in section [ sect : nresults ] , overall localization accuracy is still acceptable .
accuracy of the mle will be compared with the crlb in various scenarios in the next section .
.passive uhf rfid system parameters . [ cols="^,^ " , ]
numerical results are provided to validate analytic derivations with computer simulations and to compare the performance of the mle with the crlb for rfid based iot localization .
the simulation parameters for the passive uhf rfid system is given in table [ table : parameter ] .
as stated in section [ sect : crlb ] , the received power at the rfid reader antenna is subject to lognormal noise .
the noise variance is adopted from the statistical models in @xcite , which were derived from rfid propagation measurements .
our computer simulation considers rfid antennas that are installed in a square shaped room with @xmath106 meters width , and the height of the reader antennas are @xmath107 meters above floor level .
the channel is assumed to be frequency flat slow fading channel in our system .
there are two antenna placement configurations , one is placing the antennas to centers of side walls which is referred as ` side ' , and the second is placing them on the corners of the room which is referred as ` corner ' in figures . the reader uses circularly polarized antennas which have a radiation pattern as defined in , and the tag antennas are assumed to be vertically polarized .
the height of the tag is assumed to be known and @xmath8 meter .
elevation angles of reader antennas are defined as @xmath108 , @xmath109 , and @xmath110 .
elevation angles lower than @xmath108 are not considered due to lack of localization coverage for those angles . in fig .
[ fig : coveragep ] , localization coverage percentage in is illustrated for different elevation angles , antenna placement configurations , and transmit power levels for monostatic ( fig .
[ fig : coveragep](a ) ) and bistatic ( fig .
[ fig : coveragep](b ) ) antenna configuration .
the localization coverage is below @xmath78 for monostatic cases other than side @xmath110 .
the coverage percentage for monostatic configuration increases rapidly with increasing transmit power from @xmath111 on the average for @xmath112 mw to @xmath113 for @xmath114 mw transmit power .
localization coverage for bistatic cases show improvement with increased transmit power as well .
the mean localization coverage percentage for @xmath112 mw is @xmath115 , while increasing transmit power to @xmath114 mw substantially boosts it to @xmath116 .
the elevation angle also plays a critical role in localization coverage of the system . in monostatic and bistatic configurations ,
@xmath6 is superior to other angles for both corner and side placement of antennas . in general
, the coverage is increased with increased elevation angle .
corner placement of the antennas is better in bistatic configuration , whereas in monostatic configuration side placement has larger coverage area in general .
the corner placement of the antennas covers @xmath117 of the area for bistatic configuration on the average for all available transmit powers , whereas side placement enables to localize the tags in @xmath118 of the area .
things are different for monostatic case , where corner placement has @xmath119 coverage , while side placement achieves better performance with @xmath120 .
this is expected since side placement increases the overlap possibility of monostatic antenna coverages with less distances between antennas , whereas corner placement exploits the radiation coverage with increased distances between antennas .
0.49 , @xmath74 meter , for @xmath121 , and @xmath122 mw.,title="fig : " ] [ fig : mle45 m ] 0.49 , @xmath74 meter , for @xmath121 , and @xmath122 mw.,title="fig : " ] [ fig : mle45b ] + 0.49 , @xmath74 meter , for @xmath121 , and @xmath122 mw.,title="fig : " ] [ fig : crlb45 m ] 0.49 , @xmath74 meter , for @xmath121 , and @xmath122 mw.,title="fig : " ] [ fig : crlb45b ] in fig .
[ fig : t45mc ] , average mle and crlb rmse for monostatic and bistatic configurations with @xmath2 , for @xmath76 mw at each possible tag location is given .
the localization coverage for monostatic configuration is just above @xmath123 , while in bistatic configuration it is above @xmath78 as represented in fig .
[ fig : coveragep ] .
monostatic configuration has localization coverage above @xmath78 for only side placement of antennas with @xmath6 , thus they are not represented in median localization rmse results which they do not have . the median localization rmse of crlb and mle are compared in fig .
[ fig : pi4all](a ) , for elevation angle of @xmath124 .
monostatic configuration is not in the results since it does not have a coverage above @xmath78 as in fig .
[ fig : coveragep ] .
median rmse of crlb for side placement of antennas begin with @xmath125 meters at @xmath126 mw and gets as low as @xmath127 meters , while corner placement has lower median error in general from @xmath128 meters at @xmath129 mw to @xmath130 meters at @xmath131 mw . as expected , mle gets closer performance to the crlb as transmit power increases .
median rmse of mle for side placement of antennas begin with @xmath132 meters at @xmath133 mw to @xmath9 meters at @xmath131 mw , while corner placement does better with @xmath134 meters at @xmath129 mw , and @xmath135 meters at @xmath131 mw . in fig .
[ fig : pi4all](b ) , performance of side placement , and in fig .
[ fig : pi4all](c ) , performance of corner placement is are shown . in fig . [
fig : pi4all](b ) , the localization probability of a tag with mle below an error of @xmath8 meter for monostatic configuration with side placement and @xmath131 mw is @xmath136 , while for bistatic configuration with same parameters it gets to @xmath10 .
the cdf values of crlb for those are @xmath137 and @xmath138 , respectively . in fig .
[ fig : pi4all](c ) , the localization probability of a tag with mle below an error of @xmath8 meter for monostatic configuration with corner placement and @xmath131 mw is @xmath139 , while for bistatic configuration with same parameters it gets to @xmath9 .
the cdf values of crlb for those are @xmath136 and @xmath140 , respectively .
the side placement of antennas has better performance with monostatic mle compared to corner placement , while bistatic performance substantially lower .
increasing elevation angle to @xmath5 helps to decrease median localization rmse and improve localization performance .
the median localization rmse of crlb and mle are compared in fig .
[ fig : pi3all](a ) , for elevation angle of @xmath141 . as shown in fig
[ fig : coveragep ] , bistatic configuration is always above @xmath78 in localization coverage . in fig .
[ fig : pi3all](a ) , median rmse of crlb for side placement of antennas begin with @xmath142 meters at @xmath76 mw and gets as low as @xmath143 meters , while corner placement has lower median error in general from @xmath144 meters to @xmath145 meters at @xmath146 mw . similar to @xmath2 , mle converges to crlb as transmit power increases .
median rmse of mle for side placement of antennas begin with @xmath147 meters at @xmath76 mw , which reduces to @xmath148 meters at @xmath131 mw , while corner placement does better with @xmath149 meters and @xmath150 meters , respectively . in fig .
[ fig : pi3all](b ) cdf of localization rmse for side placement is shown for side placement with @xmath5 .
the localization probabilities of a tag below an error of @xmath8 meter for monostatic and bistatic configuration are @xmath151 and @xmath152 , respectively , while their crlb are @xmath153 and @xmath154 , respectively . in fig .
[ fig : pi3all](c ) , the localization probability of a tag with mle below an error of @xmath8 meter for monostatic configuration with corner placement and @xmath131 mw are @xmath155 and @xmath143 , while their crlb are @xmath153 and @xmath156 , respectively .
side placement of antennas increase the performance of monostatic configuration while degrading bistatic configuration performance similar to @xmath2 . in fig .
[ fig : pi2all](a ) , the median localization rmse of crlb and mle are compared for elevation angle of @xmath157 with side and corner placement of antennas .
median rmse of crlb for side placement of antennas begin with @xmath158 meters at @xmath76 mw and gets as low as @xmath159 meters , while corner placement has lower median error in general from @xmath160 meters to @xmath161 meters at @xmath146 mw .
similar to @xmath2 and @xmath5 , mle converges to crlb as transmit power increases .
median rmse of mle for side placement of antennas begin with @xmath162 meters at @xmath76 mw and reduce to @xmath163 meters at @xmath131 mw , while corner placement does better with @xmath164 meters and @xmath151 meters , respectively . in fig .
[ fig : pi2all](b ) cdf of localization rmse for side placement is represented .
the localization probabilities of a tag below an error of @xmath8 meter for monostatic and bistatic configuration are @xmath165 and @xmath166 , respectively .
the crlb for those are @xmath153 and @xmath154 , respectively . in fig .
[ fig : pi2all](c ) , the localization probability of a tag with mle and crlb is shown with respect to localization rmse .
the probability of having an error below @xmath8 meter for monostatic configuration with corner placement and @xmath131 mw are @xmath167 and @xmath168 , while the cdf of crlb for those are @xmath149 and @xmath169 , respectively .
side placement of antennas increase the performance of monostatic configuration slightly while degrading bistatic configuration performance substantially .
in general , configurations with larger elevation angle results better localization coverage and lower localization rmse . in fig .
[ fig : pi4all](a ) , the median localization rmse for @xmath2 has much higher values compared to @xmath5 in fig .
[ fig : pi3all](a ) and @xmath6 in fig .
[ fig : pi2all](a ) , for example , at @xmath76 mw localization rmse is not available for @xmath2 since its localization coverage is all below @xmath78 for either corner and side placement of antennas , while @xmath5 and @xmath6 have acceptable accuracies .
especially @xmath6 has median localization rmse of @xmath170 meters for both side and corner configuration . at all elevation angles , corner placement of antennas has better localization coverage for bistatic configuration at @xmath131 mw .
monostatic configuration does better with side placement of antennas , since in that case the coverage of antennas overlaps in larger areas .
increasing transmit power not only increases the localization coverage , but also reduces the localization error . as a conclusion ,
an elevation angle larger than @xmath5 is crucial for localization coverage and accuracy as well as corner placement of antennas with transmit power at @xmath146 mw which is the eirp limit in epc gen2 protocol of uhf rfid systems .
in this paper , fundamental limits on the iot localization accuracy of a passive uhf rfid tag is studied considering realistic propagation models for reader antennas .
our results show that high accuracy of localization does not only depend on the transmit power , but also depends on the use of right elevation angle and antenna placement and the use of bistatic configuration in localization system . in our simulations
it is shown that among the considered elevation angles , @xmath6 yields the best results for the given deployment scenario , since it maximizes the received power , results in largest localization coverage for iot and minimizes the localization error .
we observed that bistatic localization coverage drops with the use of side placement of antennas , while it increases monostatic localization coverage .
using bistatic configurations improves the probability of localizing the tag with higher accuracies when compared with monostatic configurations .
the best results are achieved with bistatic configuration and side placement of the antennas .
this work was made possible by the national science foundation grant ast1443999 .
the statements made herein are solely the responsibility of the authors .
in this appendix we will show derivation of crlb through obtaining fim .
individual elements of the fim in can be calculated using the likelihood function @xmath171 in as follows@xcite : @xmath172 , \label{eq : fim_element } \end{aligned}\ ] ] where @xmath173 is the @xmath174-th element of the fim for @xmath175 . as in @xcite , using the fim element in can be derived as @xmath176=\frac{1}{\sigma^2}\sum_{i=1}^{n}\sum_{j=1}^{n}\bigg(\frac{\partial\hat{{p } } _ { ij}}{\partial { \rm x}_m}\times\frac{\partial\hat{{p } } _ { ij}}{\partial { \rm x}_n}\bigg).\label{eq : expectfim } \end{aligned}\ ] ] note that is in logarithmic scale .
derivative of each element in received power is calculated separately since it can be written as summation of different functions in logarithmic scale .
partial derivative of can be represented as @xmath177 the ( unknown ) location of the tag ( @xmath90 ) does not affect the parameters @xmath178 of received power , and hence the resulting partial derivative of is then given by @xmath179
in this appendix we will derive the crlb for parameters @xmath73 and @xmath180 for @xmath75 .
the gain function in for those particular values of @xmath17 and @xmath44 becomes @xmath181 first derivative of with respect to @xmath182 , for @xmath183 , is @xmath184 where @xmath185 then for @xmath186 in , @xmath187 and @xmath188 can be solved as @xmath189 the same solution for @xmath190 is given in @xmath191 the path loss function @xmath33 does not change with @xmath1 and @xmath192 , and it only depends on the distance between the reader antenna and the tag . then , the derivative of @xmath33 with respect to @xmath193 and @xmath194 is as follows @xmath195 based on these derivations , using
, the crlb for any location can be calculated with known set @xmath196 and @xmath197 for @xmath198 with given parameters @xmath73 and @xmath180 . in fig .
[ fig : t45mc](c ) and fig .
[ fig : t45mc](d ) , crlb for monostatic and bistatic configurations respectively are calculated for any possible location of tag .
l. yan , y. zhang , l. yang , and h. ning , _ the internet of things : from rfid to the next - generation pervasive networked systems _ , ser .
wireless networks and mobile communications.1em plus 0.5em minus 0.4em taylor & francis , 2008 .
e. welbourne , l. battle , g. cole , k. gould , k. rector , s. raymer , m. balazinska , and g. borriello , `` building the internet of things using rfid : the rfid ecosystem experience , '' _ ieee internet computing _ , vol . 13 , no . 3 , pp .
4855 , may 2009 .
k. akkaya , i. guvenc , r. aygun , n. pala , and a. kadri , `` iot - based occupancy monitoring techniques for energy - efficient smart buildings , '' in _ wireless communications and networking conference workshops ( wcncw ) , 2015 ieee _ , march 2015 , pp . 5863 .
x. jia , q. feng , t. fan , and q. lei , `` rfid technology and its applications in internet of things ( iot ) , '' in _ proc .
consumer electronics , communications and networks ( cecnet ) _ , apr .
2012 , pp . 12821285 .
s. wamba and e. w. ngai , `` importance of the relative advantage of rfid as enabler of asset management in the healthcare : results from a delphi study , '' in _ proc .
hawaii int .
system science ( hicss ) _
2012 , pp . 28792889 .
a. al - ali , f. aloul , n. aji , a. al - zarouni , and n. fakhro , `` mobile rfid tracking system , '' in _ proc .
information and communication technologies : from theory to applications _ , apr .
2008 , pp . 14 .
n. patwari , j. ash , s. kyperountas , a. hero , r. moses , and n. correal , `` locating the nodes : cooperative localization in wireless sensor networks , '' _ ieee sig .
proc . mag .
_ , vol . 22 , no . 4 , pp
5469 , july 2005 .
s. gezici , z. tian , g. giannakis , h. kobayashi , a. molisch , h. poor , and z. sahinoglu , `` localization via ultra - wideband radios : a look at positioning aspects for future sensor networks , '' _ ieee sig .
proc . mag .
_ , vol . 22 , no . 4 , pp .
7084 , july 2005 . i. guvenc , s. gezici , and z. sahinoglu , `` fundamental limits and improved algorithms for linear least - squares wireless position estimation , '' _ wireless comm . and mobile computing _ , vol . 12 , no . 12 , pp . 10371052 , 2012 .
l. geng , m. bugallo , a. athalye , and p. djuric , `` real time indoor tracking of tagged objects with a network of rfid readers , '' in _ proc .
european signal processing conference ( eusipco ) _ , aug .
2012 , pp . 205209 .
m. moreno , m. zamora , j. santa , and a. skarmeta , `` an indoor localization mechanism based on rfid and ir data in ambient intelligent environments , '' in _ proc .
innovative mobile and internet services in ubiquitous computing ( imis ) _ , july 2012 , pp .
805810 .
d. hahnel , w. burgard , d. fox , k. fishkin , and m. philipose , `` mapping and localization with rfid technology , '' in _ proc .
robotics and automation _ , vol . 1 , apr .
2004 , pp . 10151020 vol.1 .
a. bekkali , h. sanson , and m. matsumoto , `` rfid indoor positioning based on probabilistic rfid map and kalman filtering , '' in _ proc .
wireless and mobile computing _ , oct .
2007 , pp . 2121
. b. s. iftler , a. kadri , and i. guvenc , `` experimental performance evaluation of passive uhf rfid systems under interference , '' in _ proc .
on rfid tech . and appl .
, rfid - ta _ , sept .
2015 , pp . 8186 .
x. zheng , h. liu , j. yang , y. chen , r. martin , and x. li , `` a study of localization accuracy using multiple frequencies and powers , '' _ ieee trans .
parallel and distributed syst . _ , vol . 25 , no . 8 , pp . 19551965 , aug .
akre , x. zhang , s. baey , b. kervella , a. fladenmuller , m. zancanaro , and m. fonseca , `` accurate 2-d localization of rfid tags using antenna transmission power control , '' in _ wireless days ( wd ) , 2014 ifip _ , nov 2014 , pp . 16 .
d. lieckfeldt , j. you , and d. timmermann , `` exploiting rf - scatter : human localization with bistatic passive uhf rfid - systems , '' in _ ieee int .
conf . on wireless and mobile computing ,
netw . and comm .
2009 , pp .
179184 .
m. hasani , e .- s .
lohan , l. sydanheimo , and l. ukkonen , `` path - loss model of embroidered passive rfid tag on human body for indoor positioning applications , '' in _ proc .
ieee rfid tech . and
appl . conf .
( rfid - ta ) _ , sept .
2014 , pp . 170174 .
a. bekkali , s. zou , a. kadri , m. crisp , and r. penty , `` performance analysis of passive uhf rfid systems under cascaded fading channels and interference effects , '' _ ieee trans .
wireless commun .
_ , vol . 14 , no . 3 , pp . 14211433 , mar | passive radio - frequency identification ( rfid ) systems carry critical importance for internet of things ( iot ) applications due to their energy harvesting capabilities .
rfid based position estimation , in particular , is expected to facilitate a wide array of location based services for iot applications with low - power requirements . in this paper , considering monostatic and bistatic configurations and 3d antenna radiation pattern , we investigate the accuracy of received signal strength based wireless localization using passive ultra high frequency ( uhf ) rfid systems .
the cramer - rao lower bound ( crlb ) for the localization accuracy is derived , and is compared with the accuracy of maximum likelihood estimators for various rfid antenna configurations .
numerical results show that due to rfid tag / antenna sensitivity , and the directional antenna pattern , the localization accuracy can degrade at _ blind _ locations that remain outside of the rfid reader antennas main beam patterns . in such cases
optimizing elevation angle of antennas are shown to improve localization coverage , while using bistatic configuration improves localization accuracy significantly .
beamforming , bistatic , crlb , iot , localization , maximum likelihood estimation , monostatic , position estimation , public safety , radiation pattern , uhf rfid . |
the optical quality of the beam is a critical issue for intense lasers : a good optical quality is necessary to optimize stimulated amplification in optical amplifiers , to prevent optical surfaces from deteriorating due to hot spots , and to optimize the flux density at focus . for these reasons the optical quality of the beams of intense lasers
must be monitored . to this end
, wavefront analysis has to cope with the presence of high spatial frequencies in the beam profile due to speckle patterns .
commonly used wavefront sensors , whether shack - hartmann @xcite ( sh ) or shearing interferometers @xcite , rely on the assumption of a continuous phase structure . in order to measure high spatial frequencies ,
the sampling of the wavefront has to be fine .
this requires additional optical components and many measurement points , which are manageable assuming large focal plane sensors .
moreover , these concepts require the reconstruction of the wavefront from wavefront gradient measurements .
the reconstruction process is only valid if the wavefront is continuous and can be measured on a domain that is a connected set in the topological sense . to bypass the above limitations , far field wavefront sensing techniques such as phase diversity are an appealing alternative .
phase diversity , i.e.,the recovery of the field from a set of intensity distributions in planes transverse to propagation , is a well - established method .
the vast majority of the work on this technique has concentrated on the estimation of the phase for applications related to imaging , assuming that the modulus is known , as is often reasonable in astronomy at least see in particular @xcite for seminal contributions , @xcite for a review on phase diversity , and @xcite for an application of phase diversityto reach the diffraction limit with strehl ratios as high as 98.7% . yet , with intense lasers , the wave profile may present strong spatial variations @xcite . for this reason , in this application of phase diversityone needs to estimate the complex amplitude
. early work on phase and modulus estimation was performed for the characterization of the hubble space telescope ( hst ) .
roddier @xcite obtained non - binary pupil modulus images using an empirical procedure combining several gerchberg - saxton ( gs ) type algorithms @xcite , while fienup @xcite used a metric minimization approach @xcite to estimate the binary pupil shape , parametrized through the shift of the camera obscuration .
the gs algorithm belongs to a larger class of methods based on successive mathematical projections , studied in @xcite .
although there is a connection between projection - based algorithms and the minimization of a least - square functional of the unknown wavefront @xcite , the use of an explicit metric to be optimized is often preferable to projection - based algorithms for several reasons .
firstly , it allows the introduction of more unknowns ( differential tip - tilts between images , a possibly extended object , etc ) ; secondly , it allows the incorporation of prior knowledge about the statistics of the noise and/or of the sought wavefront ; thirdly , projection based algorithms are often prone to stagnation .
the first work on estimating a wave complex amplitude from phase diversity data with an explicit metric along with experimental results was published by jefferies @xcite , in view of incoherent image reconstruction in a strong perturbation regime .
the authors encountered difficulties in the metric minimization , which may in part be due to the parametrization of the complex amplitude with its polar form rather than its rectangular one .
experimental results in a strong perturbation regime have also been obtained by almoro _
et al . _
@xcite using a wave propagation - based algorithm .
this algorithm , which can be interpreted as successive mathematical projections @xcite , required many measurements recorded at axially - displaced detector planes .
the said technique was adapted for smooth test object wavefronts using a phase diffuser and , for single plane detection , with a spatial light modulator@xcite .
more recently , thurman and fienup @xcite studied , through numerical simulations , the influence of under - sampling in the reconstruction of a complex amplitude from phase diversity data with an explicit metric optimization , in view of estimating the pupil modulus and aberrations of a segmented telescope , but this metric was not derived from the data likelihood .
this paper aims at presenting a likelihood - based complex amplitude retrieval method relying on phase diversity and requiring few images , together with an experimental validation . as in @xcite
, the complex amplitude is described by its rectangular form rather than the polar one to make the optimization more efficient .
the method , named camelotfor _ complex amplitude measurement by a likelihood optimization tool _ , is described in section [ sec - method ] . in section [ sec - manip ]
we validate it experimentally and assess its performance on a laboratory set - up designed to shape the complex amplitude of a laser beam and record images of the focal spot at several longitudinal positions .
in particular , a cross - validation with conventional phase diversityis presented . finally , experimental results are confronted to carefully designed simulations taking into account many error sources in section [ sec : perf_analysis ] . in particular , the impact of photon , detector and quantization noises on the estimation precision is studied .
the schematic diagram for phase diversitymeasurement is presented on fig .
[ mod_dir ] .
an imaging sensor is used to record the intensity distributions . as with any wavefront sensor ,
the laser beam is focused by optics in order to match the size of the beam with that of the sensor . with phase diversity , the sensor area
is installed at the image focal plane of a lens .
note that it is advisable to install a clear aperture at the front focal plane of the optics in order to ensure a correct sampling of the intensity distributions .
the focal length of the optics and the diameter of the aperture are chosen in order to satisfy the shannon criterion with respect to the pixel spacing . in order to retrieve the complex amplitude of an electromagnetic field
, the relationship between the unknowns and the measurements must be described mathematically .
this description is called the image formation model , or direct model .
let @xmath0 denote the complex amplitude in plane @xmath1 .
@xmath0 is decomposed onto a finite orthonormal spatial basis with basis vectors @xmath2}$ ] : @xmath3 the coefficients of this decomposition are stacked into a single column vector , denoted by @xmath4_{j=[1,n_k]}\in { \mathbb{c}}^{n_k}$ ] . in the following , a pixel basis is used without loss of generality .
the field complex amplitude in the plane of the above - mentioned clear aperture , @xmath5 , is supposed to be the unknown .
@xmath5 is called hereafter the estimation plane .
we assume that the phase diversityis performed by measuring intensity distributions in @xmath6 planes , perpendicular to the propagation axis .
@xmath1 ( @xmath7 ) refer to these planes .
the transverse intensity distributions of the field are measured by translating the image sensor along the optical axis around the focal plane .
the measured signal in plane @xmath1 is a two dimensional discrete distribution concatenated formally into a single vector of size @xmath8 denoted by @xmath9 . as the detection of the images
is affected by several noise sources , denoting @xmath10 the noise vector , the direct model reads : @xmath11 where @xmath12 and @xmath13 represents a component - wise product .
the component - wise product of two complex column vectors of size @xmath14 denoted @xmath15^t_{j=[1,n]}$ ] and @xmath16^t_{j=[1,n]}$ ] is defined as the term by term product of their coordinates : @xmath17^t_{j=[1,n]},\ ] ] in eq .
( [ eq : i_k ] ) , the spatial integration of the intensity distribution by the image sensor is not taken into account . in practice
, this assumption will remain justified as long as the spatial sampling rate exceeds the shannon criterion .
each @xmath18 can be expressed as a linear transformation ( a transfer ) of @xmath19 and therefore be described by the product of the propagation matrix @xmath20 by @xmath19 .
unfortunately , the transverse registration of the different measurements is experimentally difficult to obtain with accuracy . in order to take these misalignments into account ,
differential shifts between planes are introduced in the direct model via the dot product of @xmath19 by a differential shift phasor @xmath21 : @xmath22.\ ] ] the @xmath23-th differential shift phasor is decomposed on the zernike tip and tilt polynomials @xmath24 and @xmath25 expressed in the pixel basis @xcite , @xmath26 and @xmath27 being their respective coefficients : @xmath28 the misalignment vector @xmath29 is defined as : @xmath30 . without loss of generality , the first measurement plane ( k=1 ) is chosen as the reference plane , so that @xmath31 .
finally the image formation model is : @xmath32|^2 } + n_k.\ ] ] the propagation from the estimation plane down to the reference plane is simulated numerically using a discrete fourier transform ( dft ) , @xmath33 being the far field of @xmath34 : @xmath35}\ ] ] the propagation between the reference plane ( @xmath36 ) and plane @xmath37 is simulated by a fresnel propagation performed in fourier space : @xmath38}\odot\exp{(i\pi\lambda d_{1k } \nu^2)}]}.\ ] ] where @xmath39 is the distance between plane @xmath40 and plane @xmath23 , @xmath41 is the norm of the spatial frequency vector in discrete fourier space , and idft the inverse of the dft operator . in order to retrieve an estimation of @xmath19 from the set of measurements , @xmath42 ,
the basic idea is to invert the image formation model , i.e.,the direct model . for doing so
, we adopt the following maximum a posteriori ( map ) framework @xcite : the estimated field @xmath43 and misalignment coefficients @xmath44 are the ones that maximize the conditional probability of the field and misalignment coefficients given the measurements , that is the posterior likelihood @xmath45 . according to bayes rule : @xmath46 where @xmath47 , respectively @xmath48 , embody our prior knowledge on @xmath19 , respectively @xmath29 .
the map estimate of the complex field corresponds to the minimum of the negative logarithm of the posterior likelihood : @xmath49 which , under the assumption of gaussian noise , takes the following form : @xmath50 } |^2})^t c_k^{-1 } ( { \text{\boldmath$i_k$}\xspace } -{| { m_k[\psi_0{\odot}{\text{\boldmath$s_k$}\xspace } ] } |^2 } ) -\ln p(\psi_0 ) - \ln p(a ) , \ ] ] where @xmath51 is the covariance matrix of the noise on the pixels recorded in plane @xmath23 ( diagonal if the noise is white ) and @xmath52 and @xmath53 are regularization terms that embody our prior knowledge on their arguments . in the experiment and in the simulations
presented hereafter , we have not witnessed the need for a regularization , so in the following we shall take @xmath54constant , and the map metric of eq .
( [ eq : j ] ) reduces to a maximum - likelihood ( ml ) metric .
this metric is similar to but different from the intensity criterion suggested by fienup in @xcite .
indeed , in our likelihood - based approach , @xmath55 enables us to take into account not only bad pixels ( either saturated or dead ) but also noise statistics . in eq .
( [ eq : argmin ] ) @xmath56 is a non - linear real - valued function of @xmath57 and @xmath58 . in order to perform its minimization a method based on a quasi - newton algorithm , called variable
metric with limited memory and bounds method ( vmlm - b ) @xcite is used .
the vmlm - b method requires the analytical expression of the gradient of the criterion .
the complex gradient of @xmath56 with respect to @xmath59 @xcite , denoted @xmath60 , is defined in eq .
( [ nabla_je_1 ] ) as the complex vector having the partial derivative of @xmath56 with respect to @xmath61 ( respectively @xmath62 ) as its real ( respectively imaginary ) part .
@xmath63}_{2}){\odot } ( \underbrace{{\text{\boldmath$i_k$}\xspace}-|m_k [ \psi{\odot}{\text{\boldmath$s_k$}\xspace}]|^2)}_{1}\right]\right).\ ] ] four factors can be identified in eq .
( [ nabla_je_1 ] ) , which allows a physical interpretation of this somewhat complex expression : 1 .
the computation of the difference between measurements and the direct model ; 2 . weighing this difference by the result of the direct model ; 3 . whitening the noise to take into account noise statistics ; 4 . a reverse propagation that enables the projection of the gradient into the space of the unknowns and takes into account differential tip / tilts .
the minimization of @xmath56 must also be performed with respect to misalignment coefficients .
the following analytical expression has been obtained and implemented : @xmath64){\odot } ( { \text{\boldmath$i_k$}\xspace}-|m_k [ \psi{\odot}{\text{\boldmath$s_k$}\xspace}]|^2)\right]}^*}.\ ] ]
the objective of the experiment is to validate camelotexperimentally on a perturbed laser beam .
the experiment is performed using a low power continuous fibered laser whose complex amplitude is modulated to create spatial perturbations .
the experimental setup is illustrated on fig .
[ fig_expe_setup ] .
the spatial modulation of the laser beam ( phase and modulus ) is controlled by a field control module and conjugated with the clear aperture plane .
the beam going through the clear aperture crosses a beam splitter and is focused on the camelotcamera of fig .
[ mod_dir ] .
the camera is used to record three near focal plane intensity distributions , which are given as inputs to camelot .
the field estimated by the latter will be denoted @xmath65 in the following .
the beam reflected on the beam - splitter is used to record the intensity distribution @xmath66 with an image sensor conjugated with the aperture plane .
camelot s estimation of the complex field is cross - validated in the two following ways : regarding field modulus , @xmath67 is compared with @xmath68 , called measured modulus of the field hereafter , which is computed as the square - root of image @xmath66 : @xmath69 . regarding phase , camelot s estimate
is compared to the result of a straightforward adaptation of a classical phase diversityalgorithm @xcite that will be referred to as conventional phase diversity .
phase diversity is now a well established technique : it has been used successfully for a number of applications @xcite , including very demanding ones such as extreme adaptive optics @xcite , and its performance has been well characterized as a function of numerous factors , such as system miscalibration , image noise and artefacts , and algorithmic limitations @xcite . in classical phase diversity ,
only the pupil phase is sought for , while the pupil transmittance is supposed to be known perfectly .
this is probably the main limitation of this method , regardless of how the transmittance is known , whether by design or by measurement . in the case of a transmittance measurement ,
a dedicated imaging subsystem must be used , hence increasing the complexity of the setup .
misalignment issues ( scaling , rotation of the pupil ) have to be addressed due to the potential mismatch between the true pupil ( which gives rise to the recorded image ) and the pupil that is assumed in the image formation model . in this paper , for conventional phase diversity , the pupil transmittance is set to the measured modulus @xmath68 and the data used are two of the three near focal plane intensity distributions : the focal plane image and the first defocused image . in this section , we present the modulation method used to control the field and how the result of the modulation is measured in the aperture plane .
the modulation of the complex amplitude is obtained using the field control method suggested by bagnoud @xcite : a phase modulator is followed by a focal plane filtering element .
the phase modulation is performed with a phase only slm ( hamamatsu lcos slm x10 - 468 - 01 ) with 800x600 @xmath70 pixels .
the laser source is a @xmath71 continuous laser diode at @xmath72 injected into a @xmath73 core single - mode fiber . at the exit of the fiber ,
the beam is collimated , linearly polarized and then reflected on the slm .
the clear aperture diameter is @xmath74 on the surface of the slm .
it is conjugated with a unit magnification onto the clear aperture plane .
the spatial modulation and filtering are designed to control 15@xmath7515 resolution elements in the clear aperture i.e.,15 cycles per aperture .
the effective result of the control in the clear aperture plane is called the true field .
its modulus is denoted @xmath76 .
the spatial modulation and filtering have been modeled by means of an end - to - end simulation .
the result of the simulation is denoted by @xmath77 .
its modulus , denoted @xmath78 , is presented on the left of fig .
[ fig_control ] .
it shows smooth variations as can be typically observed on an intense pulsed laser @xcite .
it has been truncated in the upper right corner to simulate a strong vignetting effect .
the phase of the field is presented on fig .
[ fig_phase ] .
it is dominated by a zernike vertical coma ( @xmath79 ) of @xmath80 peak - to - valley ( pv ) , typical of a strong misalignment of a parabola for instance . *
1@c ) , center : measured modulus ( @xmath68 ) , right : @xmath81.,title="fig : " ] the modulus of the true field is measured with the aperture plane imaging camera ( see fig.[fig_expe_setup ] ) .
the image is recorded with a high spatial resolution ( 528 pixels in a diameter ) and with a high snr per pixel ( 60 in average ) .
the result of the measurement,@xmath68 , is presented at the center of fig .
[ fig_control ] . in order to quantify the proximity of the simulated modulus @xmath78 with the measured one @xmath68 ,
the following distance metric is defined : @xmath82 to enable their comparison , @xmath78 and @xmath68 have been normalized in flux ( @xmath83 ) . the spatial distribution of the modulus of the difference between @xmath78 and @xmath68 is presented on the right of fig .
[ fig_control ] .
it has been multiplied by a factor 2.5 to present a dynamic comparable to @xmath68 .
the corresponding distance @xmath84 is found to be @xmath85 , whereas in the case of a photon noise limited measurement simulations show that one should rather obtain @xmath86 .
a preliminary fourier - based analysis of the difference between @xmath78 and @xmath68 enables to identify a clear cut separation between spatial frequencies below the cut - off frequency of the spatial modulation ( 15 cycles per aperture ) and higher spatial frequencies . due to the spatial filtering by the control module ,
the low spatial frequencies part of the difference between @xmath78 and @xmath68 can be attributed mostly to model errors between the simple simulation performed to compute @xmath78 and the effective experimental setup of the control module .
for instance in the simulation , the spatial filter is assumed to be centered on the optical axis , the slm illumination is assumed to be perfectly homogeneous , and lenses and optical conjugations are supposed to be perfect .
this part of @xmath84 is evaluated to @xmath87 , that is more than @xmath88 of the total .
the high frequencies part of @xmath84 comes from optical defects on relay optics and noise influence on @xmath68 measurement .
it is evaluated to @xmath89 .
this gives insight on the measurement precision of @xmath68 , which should thus be of the order of @xmath89 .
the intensity distributions in the different planes are recorded by translating the sensor along the optical axis around the focal plane .
the amount of defocus must be large enough to provide significantly different intensity distributions so as to facilitate the estimation process . however , small translation distances are preferred for the sake of experimental easiness . for conventional phase diversitya defocus of @xmath90 pv
is often chosen as it maximizes the difference between the focal plane intensity distribution and the defocused image .
therefore , the amplitude of defocus between the position of each plane is fixed to @xmath90 pv .
considering such a defocus , it appears that no less than three different measurement planes are required with camelot .
the first image is located in the focal plane , the second one is at a distance corresponding to @xmath90 of defocus pv , the third one at a distance of 2@xmath90 pv from the focal plane .
the relation between the pv optical path difference @xmath91 in the aperture plane and the corresponding translation distance between two successive planes @xmath92 is : @xmath93 for the experiment , the focusing optics focal length is @xmath94 and the aperture diameter is @xmath74 .
consequently the translation distances are @xmath95 between the successive recording planes .
the intensity distributions are recorded with a hamamatsu cmos camera ( orca @xmath96 ) .
its characteristics are the following : a pixel size and spacing @xmath97 , a readout noise standard deviation @xmath98rms , a full well capacity of @xmath99 , and a @xmath100 bit digitizer . as explained in section [ sec - statement ]
, the sampling must fulfill the shannon criterion . the focusing optics focal length and the aperture diameter combined with the pixel size lead to a theoretical shannon oversampling factor @xmath101 for @xmath72 .
the total number of photo - electrons must be large enough to prevent the estimation from being noise limited .
simulations of the system that will be detailed in section [ sec : perf_analysis ] demonstrate that @xmath102 total photo - electrons are sufficient . due to the limited full well capacity of the sensor , @xmath103 short exposures images are added to reach this number . for each image ,
background influence has been removed by a subtraction of an offset computed from the average of pixels located on the side of the images ( hence not illuminated ) .
the noise covariance matrix @xmath104 is approximated by a diagonal matrix whose diagonal terms , @xmath105_{jj}$ ] , correspond to the sum of the variances of the photon noise , read out noise and quantization noise : @xmath106_{jj } = [ \sigma^2_{ph , k}]_j + p(\sigma^2_{ron } + q^2/12)\ ] ] where @xmath107 is the quantization step . in practice ,
the readout noise variance map @xmath108 is calibrated beforehand and the photon noise variance @xmath109 is approximated from the image by : @xmath110_j=\max([{\text{\boldmath$i$}\xspace}_k]_j,0)$ ] on pixel @xmath111 @xcite .
the three recorded near - focal intensity distributions are presented at the top of fig .
the full size of one image is @xmath112 pixels . for the figure , a region of interest of @xmath113 pixels , centered on the optical axis ,
is selected . from left to right the focal plane image , the first defocused image and the second one are displayed .
the colorscale is logarithmic .
the three focal plane images are now used in the camelotestimation .
the minimization process is initiated with a homogeneous modulus field , a phase set to zero and differential tip / tilts also set to zero .
the number of estimated points in the estimation plane is @xmath114 .
the current implementation of the algorithm is written in the idl language .
a relevant measurement of the success of our inversion method is the quality of the match between the data @xmath9 and the model that can be computed from the estimated field through eq ( [ eq : mod_direct ] ) .
this latter map is presented on the middle row of fig .
the moduli of the differences between the measurements and the direct model , that is to say estimation residuals , are displayed on the bottom row of the same figure .
these residuals are below @xmath115 of the maximum of the measurements on the three planes .
the estimated shifts are @xmath116 and @xmath117 pixel for x and y directions for @xmath118 and @xmath119 respectively . ) ; center : first defocused images ( @xmath120 ) ; right : second defocused image ( @xmath121 ) .
top : experimental images ; middle : direct model result for @xmath122 ; bottom : residuals , i.e. , @xmath123 .
logarithmic scale , @xmath113 pixels roi centered on the optical axis . ] the modulus of camelot s estimated field , @xmath67 , is presented on the left of fig . [ fig_amp ] .
it has been normalized in flux to enable the comparison with the measured modulus @xmath68 , shown in the center of the figure . for the comparison ,
@xmath68 has been sub - sampled and resized to @xmath124 pixels .
the modulus of the difference between @xmath67 and @xmath68 is represented on the right .
the main spatial structures of @xmath68 are well estimated by the method .
the estimation residuals are below @xmath125 of the maximum of the measured modulus , even in the zones where the flux is low ( top right corner ) .
the distance between the two moduli is @xmath126 only .
this must be compared to the error between @xmath68 and @xmath78 reported in section [ sec : control ] which was found to be @xmath127 .
camelotthus delivers a modulus estimation that is several times closer to @xmath68 than @xmath78 is . ,
@xmath68 and @xmath128 . ]
we now compare the phase estimated by camelotto the result of the conventional phase diversitymethod described in section [ sec - principle ] .
the comparison is presented on fig .
[ fig_phase_c ] .
the phase of @xmath129 , @xmath130 , is presented on the left of the figure , the phase of the conventional phase diversity,@xmath131 , is in the center of the figure , the modulus of their difference is presented on the right . for comparison of these phase maps , their differential piston has been set to 0 . in the zones where the modulus is greater than @xmath132 of the maximum modulus of @xmath133 ,
the maximum of the phase residuals is below @xmath134 .
additionally concerning conventional phase diversity , setting @xmath67 as the pupil transmittance instead of @xmath68 enables a better fit to the data with a 5% smaller criterion at convergence of the minimization , which is yet another indicator of the quality of the modulus @xmath67 estimated by camelot .
camelotand conventional phase diversity algorithms have about the same complexity .
the former converges after less than 300 iterations and requires less than @xmath135 of computation , while the latter takes slightly more iterations and time .
the main difference between the two methods lies in the modeling of the defocus between the measurement planes : in camelotit is performed by a fresnel propagation , hence requiring two additional dfts for each defocused plane .
however , camelotappears here as fast as conventional phase diversity to achieve comparable results .
the convergence time demonstrated here makes camelotsuitable for the measurement and control of slowly varying aberrations such as those induced by thermal expansion of mechanical mounts for instance .
the use of an appropriate regularization metric should contribute to speed up the computation .
recent work on real time conventional phase diversity demonstrated that few tens of hertz are achievable with a dedicated ( but commercial off - the - shelf ) computing architecture @xcite . according to the author of the latter reference
, several hundreds of hertz could even be achieved in a very near future , making phase diversity compatible with the requirements of the measurement and control of atmospheric turbulence effects .
we believe that these conclusions can be generalized to camelotconsidering the similar convergence demonstrated here compared to conventional phase diversity and the the possibilities brought by graphical processing units to speed up fresnel propagation computations @xcite . , @xmath131 and @xmath136 . ]
we now analyze the way these results compare with simulations .
we simulate focal plane images from @xmath129 while taking into account the main disturbances that affect image formation : misalignments , photon and detector noises , limited full well capacity , quantization and miscalibration . for the numerical simulation of the out - of - focus images ,
a random tip / tilt phase is added to the field before propagation to the imaging plane in order to simulate the effect of misalignment .
the standard deviation of the latter corresponds to half a pixel in the focal plane .
then two cases are considered .
the first case , hereafter called perfect detector , assumes a detector with noise but an infinite well capacity and no quantization .
each image is first normalized by its mean number of photo - electrons @xmath137 , then for each pixel a poisson occurrence is computed , and a gaussian white noise occurrence with variance @xmath138 is then added to take into account the detector readout noise .
the second case corresponds to a more realistic detector : for a given value of the desired number of photo - electrons @xmath137 , the corresponding image is computed as the addition of as many `` short - exposures '' as needed in order to take into account the finite well capacity , and each of these short exposures is corrupted with photon noise , read - out noise and a 12 bit quantization noise .
the same number of photo - electrons is attributed to each image : @xmath139 where @xmath140 designates the total number of photo - electrons .
the simulated long exposure focal plane image is presented on the left of fig .
[ fig_psf_dyn ] .
this image is obtained for @xmath141 photo - electrons by adding @xmath142 short exposures .
it can be visually compared with the experimental focal plane image recorded in comparable conditions ( right ) : the similarity between the two images illustrates the relevance of the image simulation . , @xmath142 short exposures ) ,
right : experiment . ]
the modulus estimation error @xmath143 is plotted on fig . [ fig_varepsilon_a ] as a function of @xmath140 for the two detector cases .
ten uncorrelated occurrences are averaged to compute @xmath144 .
the result of the comparison of @xmath67 with @xmath68 obtained from the experiment is also reported on the figure ( abscissa @xmath145 photo - electrons ) . as a function of average photo - electrons per pixel @xmath140 .
black diamonds ( @xmath146 ) : perfect detector , green triangles ( @xmath147 ) : finite well capacity and quantization , red asterisk ( @xmath148 ) : experiment result , dotted line : @xmath149 , dashed line : @xmath150 . ] for the perfect detector case and @xmath140 small compared to @xmath151 , @xmath144 follows a @xmath149 power law . for greater values
, the power law turns to @xmath150 .
this can be explained by the relative weights of the readout and the photon noise .
for @xmath152 , the average flux on illuminated pixels , that can be approximated by @xmath153 where @xmath154 is the average number of illuminated pixels per image plane ( @xmath155 ) , is smaller than the total readout noise contribution for one image plane , that is @xmath156 .
thus , readout noise dominates . for @xmath157 ,
the photon noise contribution becomes predominant and the noise propagation of @xmath144 follows a @xmath150 power law .
it is confirmed here , as stated in section [ practical_implementation ] , that @xmath144 for the experiment is not limited by photon noise .
the analysis of @xmath144 for the case with finite well capacity and quantization shows that it follows the perfect detector case up to about @xmath158 photo - electrons , then starts to follow a @xmath150 power law .
@xmath158 corresponds to the flux necessary to saturate the well capacity of the sensor . above this limit
, images are added to emulate the summation of `` short - exposures '' images .
hence , the noise level in the measurements starts to depend on the number of summed images , that is to say on @xmath140 , with a @xmath150 power law .
the error level obtained from the comparison between the modulus measurement and camelotestimation ( @xmath159 ) is comparable to the error level between camelotestimation and the true modulus in the case of the more realistic detector ( @xmath160 ) : they differ approximately by a factor two .
a significant part of this difference comes from the fact that the measured modulus @xmath68 too is imperfect i.e.,is only an estimate of the true modulus @xmath76 , due to experimental artefacts , notably differential optical defects that affect image formation on the aperture plane imaging setup and noise influence .
this claim is also supported by the fact that the phase estimated by conventional phase diversityfits the measurements better when the pupil transmittance is set to the modulus estimated by camelot , @xmath67 , instead of the measured modulus @xmath68 . the fourier - based analysis mentioned at the end of section [ sec : control ] delivers an estimate of the measurement precision of @xmath68 that is evaluated to @xmath89 . as a final remark
, one can note that fig .
[ fig_varepsilon_a ] can also be useful to evaluate not only the estimation error on the field modulus but the total error on the complex field itself .
indeed , we have calculated that this total error is on average simply twice the error on the modulus , or @xmath161 .
this can be particularly useful for designing a complex field measurement system based on camelot .
in this paper , we have demonstrated the applicability of phase diversityto the measurement of the phase and the amplitude of the field in view of laser beam control .
the resolution of the inverse problem at hand has been tackled through a map / ml approach .
an experimental setup has been designed and implemented to test the ability of the method to measure strongly perturbed fields representative of misaligned power lasers .
the estimated field has been confronted to measured modulus using pupil plane imaging and to phases estimated with classical two - plane phase diversity(using the measured aperture plane modulus ) .
it has been shown that the estimation accuracy is consistent with carefully designed numerical simulations of the experiment , which take into account several error sources such as noises and the influence of quantization .
noise propagation on the field estimation has been studied to underline the capabilities and limitations of the method in terms of photometry .
several improvements to the method are currently considered .
they include the estimation of a flux and of an offset per image , adding a regularization metric in the criterion to be optimized for the reconstruction of the complex field on a finer grid and speeding up the computations .
the work presented in this paper focuses on the case of a monochromatic beam .
the application of camelotto the control of intense lasers with a wider spectrum , as it is the case for femtosecond lasers , must be investigated .
another limitation of the method lies in the computation time . for intense lasers ,
the correction frequency is small ( typically a fraction of hertz ) .
camelotcould therefore be used assuming a reasonable increase of computation speed . for real - time compensation of atmospheric turbulence in a an adaptive optics loop ,
a significant effort is requested to manage control frequencies above several hundred of hertz .
application of the method to imaging systems with complicated pupil transmittance is also of interest . | the control of the optical quality of a laser beam requires a complex amplitude measurement able to deal with strong modulus variations and potentially highly perturbed wavefronts .
the method proposed here consists in an extension of phase diversity to complex amplitude measurements that is effective for highly perturbed beams . named camelotfor complex amplitude measurement by a likelihood optimization tool
, it relies on the acquisition and processing of few images of the beam section taken along the optical path .
the complex amplitude of the beam is retrieved from the images by the minimization of a maximum a posteriori error metric between the images and a model of the beam propagation .
the analytical formalism of the method and its experimental validation are presented .
the modulus of the beam is compared to a measurement of the beam profile , the phase of the beam is compared to a conventional phase diversity estimate .
the precision of the experimental measurements is investigated by numerical simulations .
99 r. b. shack and b. c. plack , `` production and use of a lenticular hartmann screen ( abstract ) , '' j. opt .
. am . *
61 * , 656 ( 1971 ) .
j. primot , `` three - wave lateral shearing interferometer , '' appl . opt . * 32 * , 62426249 ( 1993 ) .
d. l. misell , `` an examination of an iterative method for the solution of the phase problem in optics and electron optics : i. test calculations , '' j. phys .
d : appl .
phys . * 6 * , 22002216 ( 1973 ) .
r. a. gonsalves , `` phase retrieval and diversity in adaptive optics , '' opt .
eng .
* 21 * , 829832 ( 1982 )
. l. m. mugnier , a. blanc , and j. idier , `` phase diversity : a technique for wave - front sensing and for diffraction - limited imaging , '' in `` advances in imaging and electron physics , '' , vol .
141 , p. hawkes , ed .
( elsevier , 2006 ) , chap . 1 , pp .
176 .
j .- f .
sauvage , t. fusco , g. rousset , and c. petit , `` calibration and pre - compensation of non - common path aberrations for extreme adaptive optics , '' j. opt .
soc .
am .
a * 24 * , 23342346 ( 2007 ) .
s. fourmaux , s. payeur , a. alexandrov , c. serbanescu , f. martin , t. ozaki , a. kudryashov , and j. c. kieffer , `` laser beam wavefront correction for ultra high intensities with the 200 tw laser system at the advanced laser light source , '' opt .
express * 16 * , 1198711994 ( 2008 ) . c. roddier and f. roddier ,
`` combined approach to the hubble space telescope wave - front distortion analysis , '' appl . opt . * 32 * , 29923008 ( 1993 ) .
r. w. gerchberg , `` super - resolution through error energy reduction , '' opt .
acta * 21 * , 709720 ( 1974 ) .
j. r. fienup , j. c. marron , t. j. schulz , and j. h. seldin , `` hubble space telescope characterized by using phase - retrieval algorithms , '' appl . opt . * 32 * , 17471767 ( 1993 ) .
j. r. fienup , `` phase - retrieval algorithms for a complicated optical system , '' appl
. opt . * 32 * , 17371746 ( 1993 ) .
h. stark , ed . , _ image recovery : theory and application _
( academic press , 1987 ) .
j. r. fienup , `` phase retrieval algorithms : a comparison , '' appl
. opt . * 21 * , 27582769 ( 1982 ) . s. m. jefferies , m. lloyd - hart , e. k. hege , and j. georges , `` sensing wave - front amplitude and phase with phase diversity , '' appl .
opt . * 41 * , 20952102 ( 2002 ) .
p. almoro , g. pedrini , and w. osten , `` complete wavefront reconstruction using sequential intensity measurements of a volume speckle field , '' appl
. opt . * 45 * , 85968605 ( 2006 ) .
m. agour , p. almoro , and c. falldorf , `` investigation of smooth wave fronts using slm - based phase retrieval and a phase diffuser , '' journal of the european optical society - rapid publications * 7 * ( 2012 ) .
s. t. thurman and j. r. fienup , `` complex pupil retrieval with undersampled data , '' j. opt .
soc .
am .
a * 26 * , 26402647 ( 2009 ) .
r. j. noll , `` zernike polynomials and atmospheric turbulence , '' j. opt .
. am . *
66 * , 207211 ( 1976 ) .
j. idier , ed . , _
bayesian approach to inverse problems _ , digital signal and image processing series ( iste / john wiley , london , 2008 ) .
e. thibaut , `` optimization issues in blind deconvolution algorithms , '' in `` astronomical data analysis ii , '' , vol .
4847 , j .-
l .
starck and f. d. murtagh , eds .
( proc .
soc . photo - opt .
instrum .
eng . , 2002 ) , vol . 4847 , pp .
. k. b. petersen and m. s. pedersen , `` the matrix cookbook , '' ( 2008 ) .
version 20081110 .
k. kreutz - delgado , `` the complex gradient operator and the cr - calculus , '' arxiv e - prints ( 2009 ) .
b. paul , j .- f .
sauvage , and l. m. mugnier , `` coronagraphic phase diversity : performance study and laboratory demonstration , '' astron .
astrophys .
* 552 * ( 2013 ) .
a. blanc , l. m. mugnier , and j. idier , `` marginal estimation of aberrations and image restoration by use of phase diversity , '' j. opt .
soc .
am .
a * 20 * , 10351045 ( 2003 ) .
v. bagnoud and j. d. zuegel , `` independent phase and amplitude control of a laser beam by use of a single - phase - only spatial light modulator , '' opt . lett .
* 29 * , 295297 ( 2004 ) .
l. m. mugnier , t. fusco , and j .-
m .
conan , `` mistral : a myopic edge - preserving image restoration method , with application to astronomical adaptive - optics - corrected long - exposure images . ''
j. opt .
soc .
am .
a * 21 * , 18411854 ( 2004 ) .
j. j. dolne , p. menicucci , d. miccolis , k. widen , h. seiden , f. vachss , and h. schall , `` advanced image processing and wavefront sensing with real - time phase diversity , '' appl
. opt . * 48 * , a30a34 ( 2009 ) .
t. nishitsuji , t. shimobaba , t. sakurai , n. takada , n. masuda , and t. ito , `` fast calculation of fresnel diffraction calculation using amd gpu and opencl , '' in `` digital holography and three - dimensional imaging , '' ( optical society of america , 2011 ) , p. dwc20 . |
the electronic structure of atoms and molecules is usefully probed by external electric and magnetic fields .
the development of intense lasers has permitted such probing in regimes where simple perturbative treatments are not valid and one must use a nonperturbative semiclassical approximation or a numerical approach .
experiments with molecules display a much richer range of phenomena than with atoms , due to the additional molecular degrees of freedom @xcite .
these include above threshold ionization @xcite , multiple ionization @xcite , alignment effects @xcite , electron localization @xcite , non - sequential double ionization @xcite , direct excitation @xcite , stabilization @xcite , dissociative recombination @xcite , and separation effects @xcite .
however , molecules ( even the simplest diatomic molecules ) and their ionization are clearly more difficult to treat theoretically .
many approximation techniques have been developed and applied to atomic ionization processes @xcite , but much less is known for molecular systems .
realistic calculations are rather complicated and one loses some of the physical intuition that can often be gained from simple models . in this paper
we present the exact analytical solution for a simple molecular ionization model .
the molecule is taken to be one - dimensional .
this approximation is remarkably good in the strong - field regime where the ionization is predominantly along the field direction , so that the system is effectively one - dimensional @xcite .
the simplest such one - dimensional molecule consists of two atomic wells represented by attractive delta function wells of strength @xmath0 , separated by a distance @xmath1 .
this is a well - known soluble model @xcite .
this molecule always supports an even parity bound ground state and a continuum , and if @xmath2 it also supports an odd parity bound excited state .
there is a long tradition of using model potentials such as zero - range potentials in atomic and molecular physics @xcite .
the presence of an external electrostatic field , of field strength @xmath3 , dramatically changes the basic character of the spectrum , converting the bound states into quasi - bound states and modifying the resonance structure of the continuum .
these spectral changes are seen directly in the spectral function @xmath4 : the quasi - bound states are poles of @xmath4 at complex values of the energy @xmath5 , where the real part of the pole gives the energy location of the quasi - bound level and the imaginary part gives the width , and hence lifetime , of the level .
since this molecular ionization model is exactly soluble , we can easily investigate the dependence of these quasi - bound levels on the relevant physical parameters the field strength @xmath3 , the atomic well depth @xmath0 , and the atomic separation parameter @xmath6 .
the same applies for the `` continuum '' , where resonance structures appear due to the delicate interplay between tunneling , binding and scattering effects .
the solution presented here uses the weyl - titchmarsh - kodaira ( wtk ) spectral theorem @xcite .
this spectral theorem expresses the completeness of the wavefunctions of the schrdinger equation in a general way that applies not just to the familiar discrete spectrum models ( such as the infinite square well or the harmonic oscillator ) , but also to systems with discrete and continuum spectra ( such as the finite square well or the hydrogen atom ) , and even to systems with a purely continuous spectrum , such as for ionization problems where there are no true bound states .
the wtk approach is well suited for numerical implementation and has been applied long ago to the stark effect in atomic hydrogen @xcite .
an interesting soluble one - dimensional atomic model consisting of a single finite square well on the half - line is solved using the wtk method in @xcite .
more recently , the numerical wtk approach has been used for studying strong - field ionization effects in effectively one - dimensional diatomic molecules @xcite .
various numerical and approximate methods for computing resonance locations and widths are compared in @xcite .
this current paper is complementary to @xcite , but the choice to represent the atomic wells by delta function wells makes the entire molecular ionization problem analytically solvable , thereby bypassing the numerical part of the computation .
we also mention that this one - dimensional molecular ionization model has been studied in @xcite using the solution of the associated lippman - schwinger equation .
this paper is organized as follows .
section 2 contains a summary of the implementation of the weyl - titchmarsh - kodaira ( wtk ) method for computing the spectral function .
this is applied in section 3 to the one - dimensional molecular model without an applied electric field . in section 4 the electric field
is applied and the exact solution for the spectral function is derived .
the dependence of the spectral function on the various physical parameters is explored through plots and also analytically .
section 5 is devoted to the simpler case of atomic ionization , obtained from the molecular solution by taking the atomic separation parameter , @xmath6 , to zero . in this case
we also study the large orders of perturbation theory for the stark shift and show its relation to the nonperturbative level width .
the final section contains some concluding comments .
the weyl - titchmarsh - kodaira ( wtk ) spectral theorem @xcite for quantum mechanical hamiltonians is very general , covering not just simple hamiltonians like the harmonic oscillator which have only bound states , but also hamiltonians with both bound and continuum states .
it also extends to ionization problems where the spectrum is purely continuum .
indeed , in his classic book @xcite , titchmarsh solves the half - line `` atomic '' problem of a binding delta function potential well plus a constant electric field .
this approach can be applied directly to any one - dimensional or radial schrdinger problem .
the wtk method can be summarized as follows .
consider the schrdinger equation ( we work in units where @xmath7 ) : -^(x ) + v(x ) = e(x ) [ schrod ] on the real line @xmath8 .
pick some point , chosen without loss of generality to be @xmath9 , and normalize the two independent solutions of ( [ schrod ] ) , @xmath10 and @xmath11 , so that their wronskian is equal to 1 at that point by choosing : u(0,e)=1 & , & u^(0,e)=0 + v(0,e)=0 & , & v^(0,e)=-1 [ norm ] next , integrate ( numerically or analytically ) each of these solutions out towards @xmath12 and @xmath13 , producing four functions @xmath14 , and @xmath15 . the wtk method involves constructing particular linear combinations of these functions such that these combinations are normalizable on the intervals @xmath16 and @xmath17 , when the energy @xmath5 has a small positive imaginary part : @xmath18 . specifically , going towards the right , we construct _
+ ( x , e)=u_+(x , e)+m_+(e ) v_+(x , e ) [ mp ] such that it is normalizable on the interval @xmath16 when the energy @xmath5 has a small positive imaginary part .
this determines the coefficient function @xmath19 .
similarly , going to the left , we construct the linear combination _ - ( x , e)=u_-(x , e)+m_-(e ) v_-(x , e ) [ mm ] such that it is normalizable on the interval @xmath17 when the energy @xmath5 has a small positive imaginary part .
this determines the coefficient function @xmath20 .
the spectral function , and the completeness of the wavefunctions , can be expressed in terms of these coefficient functions @xmath21 @xcite .
for example , the spectral function is ( e)=_0 ( ) [ spectral ] this wtk method is well suited to numerical implementation for arbitrary potential wells @xcite . in the models studied in this paper
the situation is even simpler , since the independent solutions @xmath10 and @xmath11 are known in analytic form for all @xmath22 ; the `` integration '' process simply involves applying the correct continuity and discontinuity boundary conditions at the locations of the two delta function potentials .
this will be shown in detail in section [ molelectric ] .
we first review the model without the electric field . we choose the following simple double - delta - function potential to represent the diatomic molecule : v(x)= - g[potential ] where @xmath23 .
this potential has two binding delta - function wells located at @xmath24 , and chosen for simplicity to have equal strength @xmath25 .
this problem , without an external electric field , is a standard problem in quantum mechanics courses @xcite .
the potential ( [ potential ] ) always supports a bound ground state which has even parity , and if @xmath26 it also supports a bound excited state which has odd parity . in this section
we present the wtk solution as an introductory illustration of the method . in the vicinity of @xmath9 the potential ( [ potential ] ) vanishes .
thus , the two independent solutions , @xmath10 and @xmath11 , of the schrdinger equation ( [ schrod ] ) , which satisfy the normalization conditions ( [ norm ] ) are u(x , e)= ( x ) , v(x , e)= - .
[ freeuv ] integrating to the right , these solutions remain valid until we reach the right - hand delta - function at @xmath27 , at which point we impose the standard boundary conditions @xcite : ( a-)=(a+),|_a-^a+=-g(a ) [ bcs ] these conditions determine the solutions for @xmath28 to be : u_+(x , e)=a ( x ) + b + v_+(x , e)=c ( x ) + d [ jump ] where the coefficients are a=1 + , b =- g ^2(a ) + c=-^2(a ) , d=-1 + there is only one bound state , but for @xmath29 there are two bound states . ] when the energy @xmath5 has a small positive imaginary part , the linear combination @xmath30 in ( [ mp ] ) will be normalizable on @xmath31 if the @xmath32 part is eliminated .
this determines the function @xmath19 in ( [ mp ] ) to be m_+(e)&= & - + & = & since the potential ( [ potential ] ) is symmetric , it follows that @xmath33 .
thus , the spectral function ( [ spectral ] ) is determined to be : ( e)=_0 ( -m_+(e+i)+ ) [ molspectral ] bound states appear on the negative real energy axis as poles of the spectral function ( [ molspectral ] ) .
the pole of @xmath34 , satisfying the transcendental equation 1+e^-2a&= & , [ moleven ] corresponds to the even parity bound state . from ( [ molspectral ] ) , the zero of @xmath34 , satisfying the transcendental equation 1-e^-2a&= & , [ molodd ] also gives a pole of the spectral function , and corresponds to the odd parity bound state ( if it exists ) .
there is always an even bound state , given by the pole of @xmath19 satisfying ( [ moleven ] ) . if @xmath26 there is also an odd bound state , given by the zero of @xmath19 satisfying ( [ molodd ] ) . , while the right - hand plot has @xmath35 . ]
some plots of the spectral function ( [ molspectral ] ) are shown in figs .
[ freebound ] and [ freecontinuum ] .
note the appearance of the single bound state when @xmath36 , but of two bound states when @xmath37 . in fig .
[ freebound ] the width of the bound state peaks in these plots is artificial , as we have kept the small imaginary part @xmath38 of the energy nonzero in order to show the peaks . in the true @xmath39 limit
these bound state peaks have zero width , and so would not show up on the plot .
[ freecontinuum ] shows the continuum part of the spectrum
notice the periodic behavior of the spectral function , due to resonant backscattering between the two delta - functions .
this periodicity is determined by the separation @xmath1 between the wells , and the well strength @xmath0 .
in this section we solve the problem of the one - dimensional molecular potential ( [ potential ] ) studied in the previous section , with an additional external static electric field , of magnitude @xmath3 .
thus , the potential is v(x)=-g - fx [ starkpotential ] where @xmath23 , and @xmath40 . the potential ( [ starkpotential ] )
has no bound states when @xmath41 , but it does have quasi - bound states .
the corresponding schrdinger equation ( [ schrod ] ) is analytically soluble since the linearly independent solutions are airy functions .
in fact , the wtk solution proceeds exactly as in the case without the electric field , except that the basic trigonometric solutions in ( [ freeuv ] ) are replaced by airy functions . in the vicinity of @xmath9 ,
the independent solutions satisfying the wronskian normalization condition ( [ norm ] ) are u(x , e)&=&a^(u ) ai(-)+b^(u ) bi(- ) + v(x , e)&=&a^(v ) ai(-)+b^(v ) bi(- ) [ airyuv ] where @xmath42 and @xmath43 are airy functions @xcite , and the coefficients needed to satisfy the normalization conditions ( [ norm ] ) are a^(u)(e)=&,&b^(u)(e)=- + a^(v)(e)=-f^-1/3 & , & b^(v)(e)=f^-1/3 here we have made use of the fundamental airy function wronskian identity @xcite : ai(x)bi^(x)-ai^(x)bi(x)= , x as before , integrating to the right , the solutions in ( [ airyuv ] ) remain valid until we reach the right - hand delta - function well at @xmath27 ,
at which point we apply the delta - function boundary conditions ( [ bcs ] ) .
this determines the solutions in the region @xmath28 to be : u_+(x , e)&=&a_+^(u ) ai(-)+b_+^(u ) bi(- ) + v_+(x , e)&=&a_+^(v ) ai(-)+b_+^(v ) bi(- ) where the coefficients are a_+^(u)(e)&= & -g^2 f^-1/3 + & & + g^2 f^-1/3 ^2
+ + b_+^(u)(e)&= & -g^2 f^-1/3 + & & + g^2 f^-1/3 ^2 - + a_+^(v)(e)&= & g^2 f^-2/3 + & & -g^2 f^-2/3 ^2 -f^-1/3 + b_+^(v)(e)&= & g^2 f^-2/3 + & & -g^2 f^-2/3
^2 + f^-1/3 [ pluscoeff ] similarly , integrating to the left , the solutions ( [ airyuv ] ) remain valid until we reach the left - hand delta - function
well at @xmath44 , at which point we apply the boundary conditions ( [ bcs ] ) .
this determines the solutions in the region @xmath45 to be : u_-(x ,
e)&=&a_-^(u ) ai(-)+b_-^(u ) bi(- ) + v_-(x , e)&=&a_-^(v ) ai(-)+b_-^(v ) bi(- ) where the coefficients are a_-^(u)(e)&=&g ^2 f^-1/3 + & & -g^2 f^-1/3 ^2 + + b_-^(u)(e)&= & g ^2 f^-1/3 + & & -g^2 f^-1/3 ^2 - + a_-^(v)(e)&=&-g ^2 f^-2/3 + & & + g^2 f^-2/3 ^2 -f^-1/3 + b_-^(v)(e)&=&-g ^2 f^-2/3 + & & + g^2 f^-2/3 ^2 + f^-1/3 [ minuscoeff ] thus , the independent solutions to the schrdinger equation for the potential ( [ starkpotential ] ) are known analytically for all @xmath22 .
as described in section [ wtk ] , the wtk method involves finding the functions @xmath21 such that the linear combinations @xmath46 are normalizable as @xmath47 , when the energy has a small positive imaginary part . in the region @xmath28 , the normalizable solution has the form _
+ ( x , e)ai(- ) -i bi(-)[rightdecay ] this determines @xmath19 to be m_+(e)=- [ mplus ] in the region @xmath45 , the normalizable solution has the form _
-(x , e)ai(- ) [ leftdecay ] this determines @xmath20 to be m_-(e)=- [ mminus ] given these expressions for @xmath21 , the spectral function is given by ( [ spectral ] ) , which we can write in terms of the coefficients ( [ pluscoeff ] ) and ( [ minuscoeff ] ) as ( e)=_0 i m ( ) [ starkspectral ] this is an analytic expression for the exact spectral function for the potential ( [ starkpotential ] ) . in the remainder of this section
we discuss the physical properties of this spectral function , using plots and analytical methods . before discussing the analytic properties of the spectral function ( [ starkspectral ] ) , we present some plots which illustrate how the spectral function depends on the physical parameters , in order to develop some intuition for the physical processes involved . in addition ,
some animations showing how the spectral function changes , both in the quasi - bound and in the quasi - continuum region , as we vary the field strength @xmath3 , the atomic separation parameter @xmath6 , or the atomic well depth parameter @xmath0 can be found at @xcite . illustrating how the quasi - bound states change as the electric field strength @xmath3 varies .
the dashed line shows the corresponding @xmath48 molecular spectrum ( . in these plots ,
the atomic well strength is @xmath49 , and the separation parameter is @xmath50 , so the free molecule has two bound states .
notice that as @xmath3 increases these two states are stark shifted and broadened . ] illustrating how the continuum part of the spectrum changes as the electric field strength @xmath3 varies .
the dashed line shows the corresponding @xmath48 molecular spectrum ( ) . in these plots , the atomic well strength is @xmath49 , and the separation parameter is @xmath50 .
notice the two oscillation scales the longer one is set by the free case [ dashed line ] , while the rapid oscillation is set by the field strength @xmath3 , with the free spectral function providing an average .
as @xmath3 decreases in magnitude , the oscillations become more rapid , eventually averaging out to the free case . ] in the absence of the electric field , the molecule has one or two bound states depending on the well separation parameter , @xmath6 , and well strength , @xmath0 .
and the continuum exhibits structure due to ramsauer - townsend resonances in the scattering between the two delta wells .
these two features are illustrated clearly in figures 1 and 2 .
when the field strength @xmath3 is nonzero , the bound states become quasi - bound states with a nonzero width , and their central values are stark shifted .
these effects are illustrated in figure 3 , which shows a molecule with two bound states subjected to an external electric field of strengths @xmath51 .
note that the quasi - bound states broaden as the field increases , with the higher state being broader since it is less deeply bound and so can tunnel more easily .
the quasi - bound states are stark shifted in opposite directions : the lower state is stark shifted down in energy , while the higher state is stark shifted up in energy . illustrating how the continuum part of the spectrum correlates with the zeros of the airy function @xmath52 [ dashed line ] , which are the energies of the wedge potential @xmath53 with an infinite wall at some point .
this is an illustration of the ramsauer effect .
these plots are for @xmath54 and @xmath49 , and various values of the electric field strength @xmath3 , as shown . ] the effect of the external electrostatic field on the continuum states is shown in figure 4 . in these plots
the dashed line shows the continuum spectral function when @xmath48 .
the solid lines show the spectral function for various values of @xmath3 : @xmath55 .
note that the free spectral function provides an average for the @xmath40 spectral function .
the scale of the oscillations of this average function is clearly independent of @xmath3 , being determined by the combination @xmath56 , as illustrated in figure 2 .
for @xmath40 there is an additional scale in the `` continuum '' spectrum , and as @xmath3 decreases to zero the rapid oscillations become more and more rapid , and eventually average out to the free spectral function .
as @xmath3 increases the period of this oscillation increases .
these oscillations can be correlated approximately with the zeros of the airy function @xmath52 , since these zeros give the energies of the half - wedge potential well which has @xmath53 for @xmath57 , with an infinite barrier at @xmath9 .
this is illustrated in figures 5 and 6 , for @xmath54 and @xmath50 , respectively .
note that the peaks of the spectral function ( the solid line ) coincide roughly with the zeros of @xmath52 , the dashed line .
the agreement is quite good , even for the molecular model having @xmath50 .
this is an example of ramsauer - townsend resonance , with the electron backscattering off the two delta wells providing the large period oscillations , and the electron scattering off the delta wells and the linear electrostatic potential providing the shorter period oscillations .
illustrating how the continuum part of the spectrum correlates with the zeros of the airy function @xmath52 [ dashed line ] , which are the energies of the wedge potential @xmath53 with an infinite wall at some point .
this is an illustration of the ramsauer effect .
these plots are for @xmath50 and @xmath49 , and various values of the electric field strength @xmath3 , as shown . ] illustrating how the quasi - bound part of the spectrum changes as the well separation parameter @xmath6 varies .
the dashed line shows the corresponding @xmath48 molecular spectrum ( ) . for the solid lines , the field strength is @xmath58 , the well strength is @xmath49 , and @xmath6 ranges through @xmath59 , @xmath60 , @xmath61 and @xmath62 .
notice that as @xmath6 increases the quasi - bound states become narrower in width and they move apart .
in contrast , the free bound states move together , eventually becoming degenerate as @xmath63 . ] illustrating how the continuum part of the spectrum changes as the well separation parameter @xmath6 varies .
the dashed line shows the corresponding @xmath48 molecular spectrum ( ) . in these plots , the field strength is @xmath64 , and the well strength is @xmath65 . notice that as @xmath6 increases a second quasi - bound state peels off the positive energy continuum and forms a stark - shifted pair around the two @xmath48 bound states which exist for @xmath66 .
also note that in the @xmath48 case , as @xmath6 increases the splitting between the two bound states becomes vanishingly small , as the tunneling between the two wells is suppressed . in the continuum ,
the average function varies with @xmath6 due to resonances in the backscattering between the two wells .
even with nonzero @xmath3 the spectral function follows this average closely as @xmath6 varies . ]
the dependence of the spectral function ( [ starkspectral ] ) on the atomic well separation parameter , @xmath6 , is illustrated in figures 7 and 8 , for the `` bound '' and `` continuum '' parts of the spectrum , respectively . in the zero field case the spectral function has two bound states if @xmath67 .
these are shown as the dashed curves in figure 7 .
as the separation increases these two bound states approach the same energy , becoming degenerate in the limit @xmath68 .
this is because in the large separation limit the tunneling which mixes the two levels becomes suppressed and the two atoms become essentially independent of one another .
thus the bound state spectrum approaches that of a single atomic well . on the other hand ,
if the field strength is nonzero the quasi - bound states do not become degenerate in the limit of large separation .
this is shown by the solid curves in figure 7 . instead ,
the two states move away from each other ; the even state is stark shifted further down in energy and the odd state is stark shifted further up in energy .
this is because for large @xmath6 the tunneling is essentially from each well independently , with one stark shifted up and the other down , depending on the parity of the original state .
these quasi - bound levels also become narrower in width as there is a larger barrier through which the electron must tunnel as @xmath6 increases . the effect on the `` continuum '' part of the spectrum
is shown in figure 8 . here
@xmath64 and the spectrum exhibits resonances which oscillate about the @xmath48 case , which is shown by the dashed lines . as the well separation changes , the average function changes , due to the ramsauer - townsend resonance between the two wells .
note that for nonzero field strength the spectral function oscillates rapidly with energy , but still follows the free field average .
thus , the @xmath40 spectral function represents a competition between the resonance between the two wells , which depends strongly on @xmath6 , and the resonance between a given well and the linear potential , which is not sensitive to @xmath6 . illustrating how the quasi - bound part of the spectrum changes as the well depth parameter @xmath0 varies .
the dashed line shows the corresponding @xmath48 molecular spectrum ( ) . in these plots , the field strength is @xmath58 , and the well separation parameter is @xmath69 .
] the dependence of the spectral function ( [ starkspectral ] ) on the atomic well depth , @xmath0 , is illustrated in figures 9 and 10 , for the `` bound '' and `` continuum '' parts of the spectrum , respectively . in the @xmath48 case , the difference in energy between the even and odd states gets smaller as the well strength increases , which is similar to what happens when the well separation increases .
this is because the tunneling mixing is more highly suppressed as the states become more deeply bound .
the difference between the limit of large well separation and the limit of large well strength is that as the well separation increases the two bounded states approach the same energy , whereas when the well strength is increased the states become degenerate but their energies tend to @xmath70 as @xmath71 .
however , if the electric field is applied , the two states do not become degenerate in the large well strength limit but instead keep a nonzero relative distance between each other .
the distance between the two states approaches @xmath72 as well strength increases , which is easily explained by the following argument .
if the well strength is very strong we can think of a particle being localized at a single well .
if a uniform electric field of strength @xmath3 pointing in the positive x - direction is applied , a particle localized at the left well will increase in energy by @xmath73 while the energy of a particle localized at the right well will decrease by @xmath73 . since the unperturbed states have the same energy the energy difference is just @xmath72 .
this can be seen in the last panels of figure 9 , for which @xmath58 and @xmath69 , so @xmath74 , which is roughly the separation between the two quasi - bound levels .
also note that the quasi - bound state peaks become narrower as @xmath0 increases , as the levels are more deeply bound .
backscattering resonances in the continuum become more prominent as the well strength increases , because the resonances are sharper , since the scattering potentials are deeper .
if the field is weak , the wkb approximation for a infinite well potential in a electric field can be used to find the energies of the backscattering states in the strong well strength limit . but
these peaks also follow the free field average ( the dashed lines in figure 10 ) , which is due to the resonance between the wells without the electric field .
thus , the @xmath40 spectral function represents a competition between the resonance between the two wells , which depends strongly on @xmath0 , and the resonance between a given well and the linear potential , which is also sensitive to the well strength @xmath0 . illustrating how the continuum part of the spectrum changes as the well depth parameter @xmath0 varies .
the dashed line shows the corresponding @xmath48 molecular spectrum ( ) . in these plots , the field strength is @xmath75 , and the well separation parameter is @xmath50 .
notice that as @xmath0 increases a second quasi - bound state peels off the positive energy continuum and forms a stark - shifted pair around the two @xmath48 bound states which exist for @xmath66 . ] with the electric field present there are no true bound states , but there are quasi - bound states .
the location of these quasi - bound states are given by the real parts of the poles of the spectral function ( [ starkspectral ] ) , and their widths are given by the imaginary part of these poles .
these poles are given by the zeros of : & & f^-1/3+i g^2 ^ 2f^-1 + & & -g f^-2/3[poles ] in the weak field limit we can use the following asymptotic expansions @xcite of the airy functions to find perturbative solutions for the zeros of ( [ poles ] ) : ai(z)&~ & _ k=0^ [ airyaasymptotic ] + bi(z)&~ & _ k=0^ [ airybasymptotic ] where @xmath76 , and the expansion coefficients @xmath77 are c_k= [ cs ] clearly , only even powers of @xmath3 will appear in the perturbative expansion for the real parts of the quasi - energies .
so , we define the expansion e =- g^2_n=0^a_n ( ) ^2n [ ptenergy ] to find the real parts of the quasi - energies , we can ignore the imaginary parts of ( [ poles ] ) , which are anyway exponentially suppressed in the weak field limit .
it is a straightforward exercise to expand ( [ poles ] ) in powers of the field strength @xmath3 , using for example _ mathematica _ @xcite .
the leading order @xmath78 term produces the equation 4a_0 - 4+(1-e^-4 a g ) = 0 [ lead ] whose solutions are just the solutions of the transcendental equations ( [ moleven ] ) and ( [ molodd ] ) derived in the previous section for the bound states in the free field case .
a solution to ( [ lead ] ) satisfying ( [ moleven ] ) is an even bound state of the @xmath48 potential ( [ potential ] ) , while a solution to ( [ lead ] ) satisfying ( [ molodd ] ) is an odd bound state of ( [ potential ] ) .
the first correction to these bound states comes from the @xmath79 term in the expansion of ( [ poles ] ) , which leads to the following expression for @xmath80 in terms of @xmath81 : a_1= [ next ] thus , to first nontrivial order , the stark shifted energy is e =- a_0 g^2 -a_1 + to find the shift for the even bound state , we find the solution @xmath81 of ( [ leading ] ) which also satisfies ( [ moleven ] ) , and then insert this value of @xmath81 into ( [ next ] ) to find the corresponding @xmath80 .
it is straightforward to continue this to higher orders .
( we study details of the higher orders in the section [ atomelectric ] for the special `` atomic '' case where the atomic separation parameter @xmath54 . ) for the odd bound state , we must first find if there is such a solution for @xmath81 to ( [ leading ] ) satisfying ( [ molodd ] ) .
this odd solution will exist if @xmath37 .
if it exists , then the corresponding stark shift is obtained by inserting this value of @xmath81 into the expression ( [ next ] ) for @xmath80 .
the widths of the quasi - bound states can be derived from the imaginary part of the poles of the spectral function . because of the complicated dependence of the quasi - bound state energies on the system parameters @xmath3 , @xmath0 and @xmath6 , it is difficult to derive simple analytical expressions for the line widths . however , in the large atomic separation limit , as @xmath63 , we can use wkb to approximate the tunneling rate , and hence the line width , as @xmath82 \label{width}\ ] ] where @xmath83 , and @xmath84 is the full stark - shifted energy of the lower quasi - bound state . in the limit of infinitely large well strength , @xmath85 , the @xmath86 term in ( [ poles ] ) dominates and
the zeros of ( [ poles ] ) lie on the positive real axis .
if the field strength @xmath3 is small , the first order asymptotic expansion of the airy functions @xcite : @xmath87 for @xmath88 can be used to estimate the location of the zeros of ( [ poles ] ) .
these approximate zeros are determined by the expressions : @xmath89 note that expression ( [ fwrexpression ] ) approximates the zeros of @xmath90 , while ( [ wwrexpression ] ) approximates the zeros of @xmath91 .
solutions for expression ( [ fwrexpression ] ) are the energies obtained using the wkb approximation for a half wedge potential with an infinitely high wall at @xmath44 . therefore , we interpret the energies satisfying ( [ fwrexpression ] ) as those of the backscattering states between the left well and the electric field , which is the ramsauer effect illustrated in figures 5 and 6 . similarly , expression ( [ wwrexpression ] ) yields the same set of energies obtained from the wkb approximation for a potential well in an electric field with infinitely high walls at @xmath44 and @xmath92 , and therefore ( [ wwrexpression ] ) determines the energies of backscattering states between the two wells , also in the presence of the electric field .
in general , the exact resonances reflect a competition of these scatterings amongst the linear potential and the atomic wells .
a nice feature of our molecular analysis is that the `` atomic '' analogue of the molecular model studied in section [ molelectric ] can be obtained simply by setting the separation parameter , @xmath6 , of the two wells to zero .
all the expressions carry over smoothly in this @xmath93 limit .
the corresponding potential is ( note that the delta function strength becomes @xmath94 in this limit ) : v(x)=-2 g ( x)-fx [ atompotential ] this atomic problem has also been discussed in terms of the corresponding green s functions in @xcite . the spectral function is given , as before in ( [ starkspectral ] ) , by ( e)=_0 i m ( ) [ atomspectral ] where the coefficient functions in ( [ pluscoeff ] ) and ( [ minuscoeff ] ) now simplify to : a_+^(u)(e)&= & ( bi^(- ) - g f^-1/3 bi(- ) ) + b_+^(u)(e)&=&- ( ai^(- ) -g f^-1/3 ai(- ) ) + a_+^(v)(e)&=&-f^-1/3 bi(- ) + b_+^(v)(e)&=&f^-1/3 ai(- ) + a_-^(u)(e)&= & ( bi^(- ) + g f^-1/3 bi(- ) ) + b_-^(u)(e)&=&- ( ai^(- ) + g f^-1/3 ai(- ) ) + a_-^(v)(e)&=&-f^-1/3 bi(- ) + b_-^(v)(e)&=&f^-1/3 ai(- ) [ atomcoeff ] taylor expanding in @xmath38 leads to a more explicit expression for the spectral function : ( e)&= & + & & -2 cm ( - ) it is well known that perturbation theory for an unstable quasi - bound state produces a divergent non - alternating series , with which an imaginary part can be associated using borel techniques .
this was first explicitly investigated for the unstable @xmath95 anharmonic potential @xcite , then in extensive detail for the @xmath96 potential @xcite , and is a very general property of perturbation theory @xcite .
the same ideas apply to the stark effect problem , as has been investigated exhaustively for atomic systems @xcite .
the quasi - bound states correspond to poles of the spectral function , namely solutions to ^2+^2=0 [ atompoles ] the real part of the quasi - energy can be found by making a perturbative expansion for the real part , @xmath97 , of the energy e_real
=- g^2_n=0^a_n ( ) ^2n [ atompt ] as in ( [ ptenergy ] ) . when the field strength @xmath3 vanishes , there is a single ( even ) bound state at @xmath98 .
thus , @xmath99 , as is consistent with ( [ lead ] ) when @xmath93 .
the width of the quasi - bound state can be estimated by writing e = e_real + i e_imag where we expect @xmath100 to be exponentially small .
indeed , expanding the imaginary part of ( [ atompoles ] ) immediately leads to the leading behavior e_imag~- g^2 [ atomimag ] in agreement with the @xmath93 limit of the molecular case ( [ width ] ) .
we now show how this is consistent with an analysis of the divergence of the perturbative expansion for the real part of the quasi - bound energy level .
first , note that the @xmath101 term in ( [ atompoles ] ) is exponentially small in the small @xmath3 limit , and so can be neglected .
thus the real part is determined by ai(_n=0^a_n ( ) ^2n ) bi(_n=0^a_n ( ) ^2n ) = [ atomcondition ] in the small @xmath3 limit we can use the asymptotic expansions ( [ airyaasymptotic],[airybasymptotic ] ) of the airy functions to make an expansion of ( [ atomcondition ] ) in powers of @xmath102 , thereby successively determining the coefficients @xmath103 .
it is trivial to program this expansion in _ mathematica _ , and the results for the first 21 expansion coefficients are shown in table 1 @xcite .
.this table lists the coefficients @xmath103 appearing in the perturbative expansion ( ) for the quasi - energy level fo the atomic system in an electric field .
note that these coefficients are non - alternating in sign , and that their magnitude , @xmath104 $ ] , grows very fast with the perturbative order @xmath105 , as is shown in the third column .
the fourth column shows the ratio of the @xmath103 to the leading factorial growth rate in ( ) , and the fifth column gives the fourth - order richardson extrapolation @xcite of this ratio , showing its rapid approach to unity .
[ cols="^,^,^,^,^",options="header " , ] notice that all the @xmath103 have the same sign , and their magnitude grows very rapidly .
in fact , the leading growth rate at large orders @xmath105 of the perturbation series is factorially fast : a_n^(lead)=()^2n(2n ) [ leading ] the last two columns of table 1 show how quickly this leading growth rate is achieved .
the final column uses fourth order richardson extrapolation @xcite for the ratio of the exact coefficients to the leading behavior in ( [ leading ] ) , and the rapid onset to the leading growth shows that after 10 terms one is already well and truly into the asymptotic regime described by ( [ leading ] ) .
since the perturbative series ( [ atompt ] ) for the energy is divergent and non - alternating , it is not borel summable @xcite .
however , one can associate an imaginary part with such a divergent series using standard borel dispersion relations @xcite .
this technique provides a bridge between perturbative and nonperturbative physics , and is well studied in a wide variety of physics contexts @xcite .
the argument can be motivated and illustrated roughly as follows .
consider the following _ alternating _ divergent series obtained by an asymptotic expansion of the integral _
0^ ~_n=0^ ( -1)^n c^2n(2n ) [ borel ] where @xmath106 .
the integral on the left - hand - side of ( [ borel ] ) is the borel sum of the alternating divergent series of the right - hand - side .
this argument does not work for the corresponding nonalternating series , @xmath107 , as one would encounter a pole on the @xmath108 integration axis .
the behavior at such a pole must be defined by some sort of prescription .
the standard principal parts prescription associates with this non - alternating divergent series a non - perturbative imaginary part @xcite ( _ n=0^c^2n(2n ) ) ~ [ pp ] thus , using the leading growth rate ( [ leading ] ) for the perturbative coefficients @xmath103 we see that this borel approach associates an imaginary part for the quasi - bound energy level equal to ( e)~-g^2 this agrees precisely with the imaginary part ( [ atomimag ] ) found directly from the properties of the airy functions , as well as with a simple wkb estimate .
in this paper we have presented the exact analytic solution for the spectral function for the simple one - dimensional molecular ionization model of a diatomic molecule represented by two attractive delta function wells in an external static electric field .
the weyl - titchmarsh - kodaira spectral theorem provides a simple construction for the spectral function in terms of suitably normalized solutions to the schrdinger equation . in this case
these solutions are airy functions , and the spectral function can be expressed in closed form in terms of airy functions .
thus , the spectral function can easily be plotted using a program such as _ mathematica _ @xcite .
the dependence of the spectral function on the relevant physical parameters , the field strength @xmath3 , the well strength @xmath0 , and the well separation parameter @xmath6 , is illustrated in section 4 by a collection of plots .
this helps develop a body of intuition for the behavior of the quasi - bound states as they are stark shifted and broadened , and also for the resonance structures in the `` continuum '' , which reflect a competition between ramsauer - townsend resonant scattering between the two atomic wells and between one or both atomic well(s ) and the linear field potential .
the most important extension of this model would be to consider the effect of time - dependence in the background electric field , which introduces yet another physical scale into the problem @xcite .
* acknowledgement : * we thank george gibson , misha ivanov and roman jackiw for helpful comments and suggestions .
we also thank stephen fulling and larry glasser for bringing references @xcite and @xcite to our attention .
k. codling , l. j. frashinski , and p. a. hatherly , `` on the field ionisation of diatomic molecules by intense laser fields '' , j. phys .
b , * 22 * , l321 ( 1989 ) ; j. h. posthumus , k. codling , l. j. frashinski and m. r. thompson , `` the field ionization , coulomb explosion of diatomic molecules in intense laser fields '' , laser phys . * 7 * , 813 ( 1997 ) .
p. b. corkum , n. h. burnett , and f. brunel , `` above - threshold ionization in the long - wavelength limit '' , phys .
lett . * 62 * , 1259 ( 1989 ) ; p. b. corkum , `` plasma perspective on strong field multiphoton ionization '' , phys .
lett . * 71 * , 1994 ( 1993 ) .
p. dietrich , d. t. strickland , m. laberge , and p. b. corkum , `` molecular reorientation during dissociative multiphoton ionization '' , phys .
a * 47 * , 2305 ( 1993 ) ; i. v. litvinyuk , k. f. lee , p. w. dooley , d. m. rayner , d. m. villeneuve , and p. b. corkum , `` alignment - dependent strong field ionization of molecules '' , phys . rev
* 90 * , 233003 ( 2003 ) .
t. seideman , m. yu .
ivanov , and p. b. corkum , `` role of electron localization in intense - field molecular ionization '' , phys .
* 75 * , 2819 ( 1995 ) ; m. ivanov , t. seideman , p. corkum , f. ilkov , and p. dietrich , `` explosive ionization of molecules in intense laser fields '' , phys .
rev . a * 54 * , 1541 ( 1996 )
. g. gibson , t. s. luk , a. mcpherson , k. boyer , and c. k. rhodes , `` observation of a new inner - orbital molecular transition at 55.8 nm in @xmath109 produced by multiphoton coupling '' , phys .
a * 40 * , 2378 ( 1989 ) .
w. p. reinhardt , `` method of complex coordinates : application to the stark effect in hydrogen '' , int .
chem . , symposium ( 1976 ) , 10 ( proc .
int . symp . at .
solid - state theory quantum stat . , 1976 ) , 359 . t. yamabe , a. tachibana and h. j. silverstone , `` theory of the ionization of the hydrogen atom by an external electrostatic field '' , phys . rev . a * 16 * ( 1977 ) 877 ; `` perturbation theory of resonant states induced by an electrostatic field : one - dimensional model '' , j. phys .
b * 10 * ( 1977 ) 2083 .
e. c. titchmarsh , _ eigenfunction expansions associated with second - order differential equations , vols .
i and ii _ , ( oxford university press , oxford , 1946 ) , `` some theorems on perturbation theory iii '' , proc roy . soc .
a * 207 * ( 1951 ) 321 .
g. n. gibson , g. v. dunne and k. j. bergquist , `` tunneling ionization rates from arbitrary potential wells , '' phys .
lett . * 81 * , 2663 ( 1998 ) [ arxiv : quant - ph/9810062 ] .
e. y. sidky and i. ben - itzhak , `` phase - amplitude method for calculating resonance energies and widths for one - dimensional potentials '' , phys . rev .
a * 60 * , 3586 ( 1999 ) .
http://www.physics.uconn.edu/~dunne/starkanimations/specmovies.html[animations ] of the spectral function dependence on the physical parameters @xmath3 , @xmath6 and @xmath0 can be found at http://www.phys.uconn.edu/@xmath112dunne/starkanimations/specmovies.html .
vainshtein , `` decaying systems and divergence of perturbation theory '' , novosibirsk report , december 1964 , reprinted in russian , with an english translation by m. shifman , in _ arkadyfest 2002 , continuous advances in qcd _ , k.a .
olive et al ( eds . ) , ( world scientific , singapore , 2002 ) . c. m. bender and t. t. wu , `` anharmonic oscillator , '' phys .
rev . * 184 * , 1231 ( 1969 ) ; `` large order behavior of perturbation theory , '' phys .
* 27 * , 461 ( 1971 ) ; `` anharmonic oscillator .
2 : a study of perturbation theory in large order , '' phys . rev .
d * 7 * , 1620 ( 1973 )
. g. v. dunne , `` perturbative nonperturbative connection in quantum mechanics and field theory , '' in proceedings of _ arkadyfest 2002 , continuous advances in qcd _ ,
olive et al ( eds . ) , ( world scientific , singapore , 2002 ) , pp 478 - 505 ; [ arxiv : hep - th/0207046 ] .
j. silverstone , `` perturbation theory of the stark effect in hydrogen to arbitrarily high order '' , phys .
a * 18 * ( 1978 ) 1853 .
v. franceschini , v. grecchi and h. j. silverstone , `` complex energies from real perturbation series for the losurdo - stark effect in hydrogen by borel - pad approximants '' , phys .
a * 32 * , 1338 ( 1985 ) . | we present a simple exact analytical solution , using the weyl - titchmarsh - kodaira spectral theorem , for the spectral function of the one - dimensional diatomic molecule model consisting of two attractive delta function wells in the presence of a static external electric field . for sufficiently deep and far apart wells ,
this molecule supports both an even and an odd state , and the introduction of a static electric field turns these bound states into quasi - bound states which are stark shifted and broadened .
the continuum spectrum also inherits an intricate pattern of resonances which reflect the competition between resonant scattering between the two atomic wells and between the linear potential and one or both atomic well(s ) .
all results are analytic and can be easily plotted .
the relation between the large orders of the divergent perturbative stark shift series and the non - perturbative widths of quasi - bound levels is studied .
[ theorem]corollary |
confinement and chiral symmetry breaking are supposed to follow from the dynamics of yang - mills fields .
these phenomena are highly non - perturbative and still have not been derived from first principles . in this paper
we will follow the strategy put forward by t hooft @xcite who considered yang - mills theories on a euclidean space - time torus @xmath2 .
the torus provides a gauge invariant infrared cut - off .
its non - trivial topology gives rise to a non - trivial structure in the space of yang - mills fields which yields additional information on the possible phases of yang - mills theories .
compared to other riemannian 4-dimensional compact manifolds the torus has many advantages ( besides being the ` space - time ' used in lattice simulations ) : * one can use a flat metric in which case curvature effect do not mix with finite size effects , * the circumference @xmath3 in the temporal direction can be identified with the inverse temperature @xmath4 @xcite , * gauge invariant periodic fields on @xmath5 can be viewed as fields on @xmath2 , * one may calculate non - perturbative quantities from finite size effects @xcite ; the string constant is directly related to the energy of a string winding around the torus @xcite , * one keeps the relevant part of the supersymmetry in susy - ym theories .
even the less ambitious goal to demonstrate confinement of static quarks without reliance on numerical simulations has not been achieved yet . without dynamical fermions
the relevant observables are products of wilson - loops @xcite . at finite temperature @xmath6
the gauge fields in the functional integrals are periodic in euclidean time i.e. @xmath7 and one may use polyakov loops @xcite as order parameters for confinement . here
@xmath8 is the representation of the gauge group which acts on the matter fields .
we shall assume that the _ gauge group @xmath0 is simply connected _ , e.g. @xmath9 rather than @xmath10 .
but since we allow for arbitrary representations @xmath8 of @xmath0 our results apply to general gauge groups @xmath11 , for example to @xmath12 .
the polyakov loop @xmath13 is invariant under gauge transformations which are periodic in time .
since it is a functional of @xmath1 only , one is motivated to seek a gauge fixing where @xmath1 is as simple as possible .
note that the weyl gauge , @xmath14 , is not compatible with time - periodicity . in a previous paper @xcite we discussed an extended abelian projection for @xmath15 gauge theories on the four torus in which @xmath1 is time independent and in the cartan subalgebra .
the gauge fixing procedure hinges on the diagonalization of the path ordered exponential , @xmath16 , whose trace is the polyakov loop .
in contrast to the two dimensional case investigated in @xcite the diagonalization procedure has unavoidable singularities @xcite . the singularities can be interpreted as dirac strings @xcite joining magnetically charged ` defects ' . here
we understand defects as points , loops ( not to be confused with the dirac strings ! ) , sheets and lumps where @xmath16 has degenerate eigenvalues . for the gauge group @xmath15
, the eigenvalues of @xmath16 are degenerate when @xmath17 .
thus there are two types of defect according to whether @xmath16 is plus or minus the identity . associated with the gauge fixing procedure one can define an abelian magnetic potential @xmath18 on @xmath19 @xcite . in @xcite we showed that the total magnetic charge of @xmath20 defects is equal to the instanton number @xmath21 .
moreover , the total magnetic charge of all defects is zero , i.e. the total magnetic charge of @xmath22 defects is minus that of the @xmath20 defects .
the relationship between magnetic charges and the instanton number was considered earlier by christ and jackiw @xcite , gross et.al .
@xcite and reinhardt @xcite who worked on @xmath23 or @xmath5 .
though here one requires ` charges at infinity ' to have overall magnetic charge neutrality . for an explicit discussion of the singularities emerging in the gauge fixing procedure at point like
monopoles see the recent paper by jahn and lenz @xcite . in this paper
we extend the defect analysis to gauge theories on @xmath2 with arbitrary gauge groups @xmath0 of rank @xmath24 .
we also consider arbitrary twists @xcite , which allows us to treat matter transforming according to any representation of the gauge group .
one has @xmath25 types of _ basic defects _ associated with the @xmath25 faces constituting the boundary of a ` fundamental domain ' ( these are essentially compactified weyl chambers ) in the root space . since the magnetic potential lies in the cartan subalgebra @xmath26 we now have a matrix @xmath27 of magnetic charges .
the possible magnetic charges are quantized and are in one to one correspondence with the points of the integral co - root lattice . for a basic defect
, @xmath28 is an integer multiple of a fixed matrix .
much as in the @xmath15 analysis there is a simple linear relation between the total magnetic charge of a given type of defect and the instanton number @xmath21 .
we have overall charge neutrality on @xmath19 unless there are non - orthogonal magnetic and electric twists .
the paper is organized as follows . in the remainder of this section we recall some basic facts concerning gauge fields on @xmath2 .
next we present a set of transition functions ( i.e. boundary conditions for the gauge fields ) where the instanton number is equal to the winding number of the mapping @xmath29 .
these transition functions serve as the starting point for our gauge fixing . in section three
we construct ` fundamental domains ' for all gauge groups .
our lie algebra conventions are stated here .
then we explain precisely what we mean by ` defects ' . in the next section
we define the magnetic charge of the defects . our key result is given in section six . here
we obtain the relationship between the magnetic charges and the instanton number .
next we rewrite @xmath30 in terms of ` higgs fields ' .
this enables us to tie up a loose end from section six , and also allows us to interpret the magnetic charges as higgs winding numbers . in section eight
we show how the ideas apply to @xmath31 and give our conclusions in section nine .
technicalities regarding our transition functions ( including a construction of magnetic twist eaters for all gauge groups ) can be found in appendix a. finally , an identity quoted in section six is derived in appendix b. we view the four torus as @xmath5 modulo the lattice generated by four orthogonal vectors @xmath32 , for a recent review see @xcite .
the euclidean lengths of the @xmath33 are denoted by @xmath34 ( we may identify @xmath3 with the inverse temperature @xmath4 ) .
local gauge invariants such as @xmath35 are periodic with respect to a shift by an arbitrary lattice vector .
however , the gauge fields have to be periodic only up to gauge transformations . in order to specify boundary conditions for gauge potentials @xmath36 on the torus one
requires a set of _ group valued transition functions _
@xmath37 , which are defined on the whole of @xmath5 .
the periodicity properties of @xmath36 are as follows @xmath38 where the summation convention is _ not _ applied .
it follows at once , that the path ordered exponential @xmath39 in ( [ defpol ] ) has the following periodicity properties the transition functions @xmath37 satisfy the cocycle conditions @xcite where the _ twists @xmath40 lie in the center @xmath41 of the group . from now on _ we assume that the transition functions belong to the universal covering group . in general , our matter fields will not transform according to the covering group . however , a matter field in some representation is equivalent to matter transforming according to the covering group _ provided we place suitable restrictions on the twists_. more precisely , consider a matter field which transforms under some representation @xmath11 of the gauge group
. a center element @xmath42 is an allowed twist if @xmath43 .
for example if we have matter fields in the defining representation of @xmath31 all the twists must be the identity , since the other two center elements are faithfully represented .
by contrast , if the matter fields are in the adjoint representation of any group then there is no restriction on the twists . _ _ under a gauge transformation , @xmath44 , the pair @xmath45 is mapped to the twists , @xmath40 , are gauge invariant .
we define the topological charge or instanton number as follows where the trace corresponds to the canonically normalized scalar product in the lie algebra .
note that @xmath21 is fully determined by the transition functions @xcite .
in particular , if we take all the transition functions to be the identity ( i.e. we assume the gauge fields are periodic in all directions ) then the instanton number is zero .
accordingly , if we are to describe the non - perturbative sectors , one must consider non - trivial transition functions .
for a given @xmath21 and set of twists , @xmath40 , we only require _ one _ set of transition functions . if we have two sets of transition functions with the same instanton number and twists then they are gauge equivalent @xcite .
first we construct a convenient set of transition function such that the instanton number is equal to the winding number of the map @xmath46 .
then we find the ( in general singular ) gauge transformation which transforms @xmath1 into a time - independent field in the cartan subalgebra.in the untwisted case , @xmath47 , we may assume that the transition functions have the following properties in @xcite it was shown by explicit construction that there exist untwisted ( i.e. @xmath48 ) transition functions satisfying ( [ condition ] ) in all instanton sectors .
the condition that @xmath49 is simply the statement that our gauge fields are periodic in time . since the transition functions are trivial on the time slice @xmath50 , and hence with ( [ condition ] ) also on the time slice @xmath51 , the path ordered exponential @xmath52 is periodic in the three spatial directions ( see ( [ polloopperiod ] ) ) . in the presence of magnetic twists ( i.e. at least one of the @xmath53 )
it is no longer possible to attain ( [ condition ] ) .
however , one can still arrange for the transition functions to be independent of @xmath54 on the time slice @xmath50 . in appendix
a we prove that there exist transition functions with the following properties where the @xmath55 are independent of @xmath54 and satisfy the ` twist eating ' conditions which follow from the cocycle conditions for the @xmath56 at time @xmath50 .
for example , consider @xmath15 gauge theory with the following magnetic twists @xmath57 . then a possible choice of @xmath55 s is @xmath58 , @xmath59 , where the @xmath60 are the pauli matrices .
twist eaters satisfying ( [ twisteaters ] ) are known to exist for arbitrary twists in @xmath61 gauge theories @xcite .
twist eaters for the other simple lie groups are constructed in appendix a. now we use the properties of the transition functions to obtain a relation for the instanton number in terms of the polyakov loop . consider the following gauge transformation @xmath62 where @xmath63 is the path ordered exponential in ( [ defpol ] ) which in general is non - periodic in time .
for brevity we use the notation using ( [ polloopperiod],[newu],[condition ] ) , the gauge transformed transition functions are @xmath64 the new @xmath65 is simply the path ordered exponential @xmath66 , while the transformed spatial transition functions are constant matrices .
applying the well know formula for the instanton number in terms of the transition functions @xcite yields where @xmath67 , and @xmath68 .
we emphasize that ( [ polloopindex ] ) is only valid when the ( original ) transition function satisfy ( [ twistcondition ] )
. another useful consequence of ( [ twistcondition ] ) is that @xmath69 has very simple periodicity properties ( + b_i)=z_0i_i^-1()_i , i=1,2,3.in particular , @xmath66 is completely periodic in the absence of twists .
now we follow @xcite and seek a ( time - periodic ) gauge transformation , @xmath44 , for which the gauge transformed @xmath1 is independent of time and in the cartan subalgebra . consider the time - periodic gauge transformation where @xmath70 is the path ordered exponential ( [ defpol ] ) , and @xmath71 diagonalizes @xmath66 , i.e. with @xmath72 in the cartan subalgebra @xmath73 . the fractional power of @xmath74 is defined via the diagonalization of @xmath74 .
it follows at once that the gauge transformed @xmath1 reads a_0^v = h(),which is indeed independent of time and in the cartan subalgebra . whereas @xmath66 is smooth the factors @xmath71 and @xmath75 in the decomposition ( [ diagonalization ] ) are in general not .
the classification and implications of these singularities are investigated in sections 4 - 7 .
[ sec - lie - conv ] the mapping @xmath76 in ( [ diagonalization ] ) from the cartan subalgebra to the toroidal ( cartan ) subgroup is not one to one . in this section
we shall find domains @xmath77 in the cartan subalgebra such that this mapping becomes bijective .
we shall choose domains which are left invariant under the action of the weyl group @xmath78 . if @xmath79 is a weyl reflection , then @xmath80 diagonalizes @xmath74 in ( [ diagonalization ] ) if @xmath81 does .
we shall fix this residual gauge freedom , under which @xmath82 , by restricting @xmath83 to one weyl chamber .
the intersection of a weyl chamber with the ` weyl invariant ' domain @xmath77 defines our fundamental domain @xmath84 .
@xmath84 is in one to one correspondence with the toroidal subgroup modulo weyl transformations or equivalently with the conjugacy classes of @xmath0 .
the main result of this section is that @xmath84 is the simplicial box with the extremal points ( [ wc4 ] ) .
our lie algebra conventions are as follows : let @xmath85 be an orthogonal basis of the cartan subalgebra @xmath26 , @xmath86 which are diagonal in a given representation for @xmath87 in any representation . ] , @xmath88=\al_k e_\al.\end{aligned}\ ] ] we normalize the roots such that the long roots have length @xmath89 , i.e. @xmath90 , and the @xmath87 become orthonormal .
_ throughout this paper we identify @xmath91 with @xmath92 .
_ let be the simple roots , fundamental weights , co - roots and co - weights , respectively : we used that the simple roots and fundamental weights are related by the cartan matrix , @xmath93 the fundamental weight - states ( which are the highest weight states of the @xmath24 fundamental representations ) and states in the adjoint representation obey the most negative root @xmath94 and its co - root @xmath95 define the integral _ coxeter numbers _ @xmath96 and _ dual coxeter numbers _
@xmath97 : @xmath98 where we have defined @xmath99 .
the ( dual ) coxeter numbers are listed in appendix a. for later convenience we assign to @xmath94 the co - weight @xmath100 . the fundamental domains we seek are intimately related to the center elements of the group .
thus it is useful to find conditions on @xmath101 such that @xmath102 is in the center @xmath103 .
center elements are the identity in the adjoint representation . because of the second set of equations in ( [ reps ] ) they must be powers of @xmath104 in an irreducible representation a center element acts the same way on all states .
hence , a necessary and sufficient condition for @xmath105 is that @xmath106 for at least one fundamental weight @xmath107 .
here we have used that the inner products of the weights with the co - weights yield the inverse cartan matrix , see ( [ gf1 ] ) .
the order of the center group is just @xmath108 .
the centers and their generators are listed in appendix a. let us now find a suitable domain in the cartan subalgebra which is mapped bijectively into the toroidal subgroup .
the elements @xmath109 in the toroidal subgroup are the identity if @xmath110 is in the integral co - root lattice , i.e. the lattice spanned by the simple co - roots @xmath111 ( see ( [ gf1 ] ) ) . thus , the convex region @xmath77 defined by the intersecting half - spaces @xmath112 , where @xmath113 is an arbitrary root , is in one to one , i.e. we have to remove half of the boundary to get a one to one correspondence . ]
correspondence with the toroidal subgroup of the gauge group is orthogonal to @xmath114 and goes through @xmath115 . ] .
this set is invariant under the action of the weyl group @xmath78 and is given by now we may fix the residual weyl reflections by further assuming that @xmath116 is in the weyl chamber defined by the inner product of a vector @xmath110 in this weyl chamber with the highest root @xmath117 is always greater or equal to the inner product with any other root .
it follows that the conditions ( [ wc1],[wc2 ] ) , which define the _ fundamental domain _ @xmath84 , simplify to @xmath84 is a simplex bounded by @xmath25 hyperplanes orthogonal to the roots @xmath118 . in what follows we call the plane orthogonal to @xmath119 the @xmath120-plane , @xmath121 .
the @xmath122-planes all meet at the origin .
since @xmath94 is a long root the last condition in ( [ weylch ] ) means that the @xmath123-plane orthogonal to @xmath94 goes through @xmath124 .
the roots @xmath119 point inside the box .
an equivalent definition of @xmath84 is that @xmath84 is the convex set with extremal points this can be seen by expanding @xmath110 in terms of the co - weights where @xmath125 being the @xmath24-vector formed from the coxeter labels .
for example , the fundamental domains @xmath84 for the @xmath126 and @xmath127 groups are the simplicial boxes with extremal points @xmath128 ( recall , that we have chosen @xmath129 ) . also , if @xmath130 and @xmath131 are the long and short roots at the endpoints of the @xmath132-dynkin - diagram , the fundamental domain for @xmath132 is the convex set with extremal points @xmath133 the fundamental domains @xmath84 and the center elements for the gauge groups of rank @xmath134 are depicted in fig.[su3 ] .
= 13 cm the fundamental domain of @xmath135 is an equilateral triangle , that of @xmath136 half a square , that of @xmath137 half of an equilateral triangle and that of @xmath138 is a square .
the reflections on the @xmath24 walls of @xmath84 through @xmath123 generate the weyl group @xmath78 of @xmath0 and give rise to @xmath77 . since @xmath139 ,
the highest root @xmath117 is always inside the weyl chamber ( [ weylch ] ) or on its boundary .
indeed , for all groups with the exception of @xmath140 @xmath141 lies on the boundary of @xmath84 . from the _
_ extended dynkin diagram _ _ to the system of simple roots @xmath142 and uses the well - known rules to draw the dynkin diagram of this extended system of roots.](see fig.[xdynkin ] ) = 3.8 cm one reads off that for all but the @xmath126 algebras the highest root is orthogonal to @xmath143 simple roots .
hence it must be proportional to the weight @xmath144 corresponding to the simple root @xmath142 with @xmath145 . although our strategy is to work in the covering group with suitably restricted twists rather than directly dealing with arbitrary representations , we could in principle do without twists if we used transition functions and fundamental domains @xmath146 appropriate to the representation @xmath8 .
actually it is quite straightforward to construct domains @xmath146 for any representation .
the volume of such domains is always less than or equal to that of @xmath84 ; more precisely @xmath147 where @xmath148 is the subgroup of the center @xmath149 which is mapped to the identity by going from the covering group to the representation @xmath150 and @xmath151 is its order . for a given group ,
the domain with the smallest volume is that for the adjoint representation since the center is trivial in this case .
the fundamental domains for the adjoint representation for the rank two groups are shown in figure [ su3 ] .
although the polyakov loop operator itself is smooth for smooth gauge potentials the factors @xmath71 and @xmath75 in the decomposition ( [ diagonalization ] ) are in general not .
in this section we shall see that singularities ( so called defects ) occur at points @xmath152 at which @xmath153 is on the boundary of the fundamental domain @xmath84 . at such defects
the residual gauge freedom is enlarged .
we shall explicitly determine the residual gauge groups at the various defects .
from now on we shall assume that @xmath153 is in the fundamental domain @xmath84 .
then ( [ diagonalization ] ) assigns a unique @xmath75 ( and thus a unique @xmath154 ) to each polyakov loop operator since we have fixed the weyl reflections .
however , the diagonalizing matrix @xmath71 in ( [ diagonalization ] ) is determined only up to right - multiplication with an arbitrary matrix commuting with @xmath75 at each point the residual gauge transformations @xmath155 form a subgroup of @xmath0 , the centralizer of @xmath75 in @xmath0 , denoted by @xmath156 .
the centralizer contains the toroidal subgroup of @xmath0 . at points where the centralizer is just the toroidal subgroup we can smoothly diagonalize the polyakov loop operator.however , at points where the centralizer is
non - abelian @xmath66 has degenerate eigenvalues and there are obstructions to diagonalizing @xmath66 smoothly @xcite . for what follows it
is useful to define the _ defect manifold _
on which the centralizer is non - abelian . in the special case @xmath9 the defect manifold is @xmath157 .
a _ defect _
@xmath158 is understood to be a connected subset of @xmath159 . in the neighborhood of a defect
the diagonalization is in general not smoothly possible and the gauge fixing will be singular .
note that @xmath159 is invariant under time - periodic gauge transformations so that the positions of the defects are gauge invariant.now we are going to classify the various defects which arise in our gauge fixing . to do that we expand @xmath153 in ( [ residual ] ) into a basis of the lie algebra as @xmath160 so that @xmath161 we see that @xmath75 commutes with the subgroup @xmath15 corresponding to @xmath113 if and only if @xmath162 is integer - valued . for @xmath163 in ( [ weylch ] )
this can only happen if @xmath110 lies on the boundary of the fundamental domain .
we parametrize @xmath164 as in ( [ wc3 ] ) so that @xmath165 therefore @xmath166 commutes with the @xmath15-subgroup corresponding to the simple roots @xmath142 if and only if @xmath167 and it commutes with the @xmath15-subgroup corresponding to @xmath94 if and only if @xmath168 .
in other words , the centralizer contains the @xmath15 corresponding to @xmath119 if the defect is on the @xmath120-plane , i.e. the plane perpendicular to @xmath119.the centralizer of @xmath75 generated by these @xmath15 subgroups can be read off from the extended dynkin diagram ( see fig.[xdynkin ] ) as follows : keep the vertex @xmath120 assigned to the root @xmath169 in the extended dynkin diagram if and only the defect lies on the @xmath120-plane . remove the other vertices and bonds attached to them .
the remaining diagram is then just the dynkin diagram belonging to the semisimple factor of the centralizer . to obtain the complete centralizer group we must multiply with as many @xmath170-factors as are needed to get a group of rank @xmath24.let us
illustrate how this works for the simply laced groups @xmath171 for which the fundamental domains @xmath84 can be parametrized as @xmath172 the extremal points of the fundamental domain are @xmath173 and they correspond to the @xmath174 center elements of @xmath126 .
if just one @xmath175 vanishes then @xmath110 lies inside the @xmath176-dimensional @xmath120-plane . and we must keep the vertex @xmath120 in the extended dynkin of @xmath126 , i.e. the leftmost diagram in fig.[xdynkin ] .
the corresponding centralizer is @xmath177 .
we call such defects with minimal non - abelian centralizers _ basic defects_. if @xmath175 and @xmath178 vanish in which case the defect lies both on the @xmath120- and @xmath179-plane , then we must keep the two vertices @xmath120 and @xmath179 in the extended dynkin diagram .
if they are neighbors in figure [ xdynkin ] , then the centralizer is @xmath180 , otherwise it is @xmath181 . in the extreme case where just one @xmath175 does not vanish
( then @xmath110 is one of the extremal points of @xmath84 ) we must retain all vertices with the exception of the vertex @xmath120 .
we get the dynkin diagram of @xmath126 and the centralizer is the whole gauge group . by scanning the whole boundary of @xmath84 comprising of @xmath182-dimensional , @xmath183-dimensional,
,@xmath184-dimensional simplices and the extremal points we obtain all stabilizer subgroups of @xmath0 .
in this section we define the abelian magnetic potential @xmath185 associated with the partial gauge fixing and show that the magnetic charge of any defect is quantized . away from the defects
the centralizer of @xmath75 is @xmath186 and @xmath71 in ( [ diagonalization ] ) is unique up to a residual abelian gauge transformation ( [ residual ] ) : if we append to each point in @xmath187 the set of all diagonalizing matrices @xmath71 we obtain a @xmath186 principal bundle over @xmath187 .
if we can find a smooth global section in this bundle then the diagonalization is smoothly possible outside of the defects , see also @xcite . to investigate the structure of the bundle we employ the abelian @xmath186 gauge potential , @xmath185 , obtained by projecting the pure gauge @xmath188 onto the cartan subalgebra , i.e. @xmath189 where the subscript @xmath190 denotes projection onto the cartan subalgebra of @xmath0
this potential is singular at the defects and on dirac strings joining the defects . under a residual gauge transformation ( [ diagonalize ] )
the gauge potentials transform as @xmath191 since @xmath192 is pure gauge the corresponding field strength is given by [ mfield ] f_mag = a_mag = i(aa)_c , and it is invariant under residual @xmath186-gauge transformations .
next we will show that a defect may carry @xmath24 quantized magnetic charges @xcite .
for each defect these charges form a matrix @xmath28 in the cartan subalgebra @xmath26 , here @xmath193 is a surface surrounding the defect @xmath158 .
= 5 cm excluding walls extending over the whole @xmath194-torus this surface is either a @xmath134-sphere or a @xmath134-torus ( see fig.[spheretorus ] ) . for each @xmath170
the magnetic charge is just the instanton number of an abelian gauge model on @xmath195 or @xmath196 @xcite and hence is quantized .
more explicitly , the magnetic charges are the winding numbers of the map @xmath197 , @xmath198 where @xmath199 is in the overlap of the two patches @xmath56 one needs to cover @xmath195 or @xmath196 .
since the gauge transformation @xmath200 is single valued on the overlap , @xmath27 must satisfy for simply connected @xmath0 this equality must hold on all states @xmath201 and we find thus we obtain the same magnetic charge quantization as uncovered by goddard , nuyts and olive @xcite in their pioneering work on electric - magnetic duality in yang - mills - higgs theories .
in this section we work with the simply connected covering group and exclude twists . ] . depending on the residual gauge symmetry in the defects we get different types of magnetic monopoles .
there are @xmath25 kinds of basic monopoles with minimal non - abelian centralizer @xmath202 , corresponding to the @xmath25 hyperplanes which make up the boundary of the fundamental domain
. we will show that a basic defect on the @xmath120-plane has magnetic charge with integer @xmath203 . if we have a defect which is on two or more of the hyperplanes ( which means that the polyakov loop has more than two degenerate eigenvalues ) then the magnetic charge of this defect is an integer combination of the co - roots perpendicular to these hyperplanes .
below we argue that in general the total magnetic charge of the defects on a given face gives the instanton number .
for example , the magnetic charge of a defect on the @xmath123-plane is @xmath204 , where @xmath205 is in the co - root lattice .
this decomposition of the magnetic charge is unique , see below . now
the instanton number is simply this is our main result .
some illustrative examples of the use of this formula are given in section 8 .
0.2truecmto derive the results ( [ magneticcharge],[instantonno ] ) we assume that : * there are no wall defects , i.e. @xmath206 there are no walls if @xmath207 is connected . ] * inside a defect the centralizer @xmath208 is uniform .
the first assumption is a reflection of the fact that one can not surround a wall defect with a closed surface and so it is not obvious how to define the magnetic charge of such a defect .
the second assumption is made to avoid the complication of ` defects within defects ' .
it may be possible to drop this requirement .
our arguments are based on the observation that and furthermore where the @xmath134-forms are here @xmath209 is the least common multiple of the coxeter labels @xmath96 and as before @xmath210 and @xmath211 .
we prove this crucial identity in appendix b. these 2-forms are well defined outside the defects , because they are invariant under the residual abelian gauge transformations ( [ diagonalize ] ) .
both terms in ( [ a ] ) may be singular at defects .
however , in the following section we will show that @xmath212 can be singular only at defects on the @xmath120-plane or equivalently at defects whose centralizers have @xmath119 as root , actually , in ( [ a ] ) we could have subtracted an arbitrary constant lie algebra element from @xmath153 and ( [ aaa ] ) would still hold true .
but the smoothness conditions ( [ regular ] ) only hold if this constant element is an extremal point of the fundamental domain and if @xmath213 is a center element .
thus we take for @xmath209 in ( [ aaa ] ) the least common multiple of the coxeter labels @xmath96 . for example @xmath214 for the @xmath126 series and @xmath215 for the other classical groups .
now we make use of ( [ aaa ] ) to relate the magnetic charges of the defects on the @xmath123-plane to the instanton number .
away from defects on the @xmath123-plane @xmath216 is regular .
now we surround each defect @xmath166 on the @xmath123-plane with a closed surface @xmath193 and pick a two form @xmath217 which is smooth inside @xmath193 , see fig.[instmon ] .
since a defect can lie on at most @xmath24 of the @xmath25 faces constituting the boundary of @xmath84 there is always at least one such regular two form .
= 5 cm with ( [ topological],[aaa ] ) the instanton number reads where , since @xmath216 is periodic on @xmath19 , we get no contributions from the ` boundary of the torus ' . using ( [ a ] )
we obtain @xmath218 since the magnetic field @xmath219 is the projection to the cartan of @xmath220 we find and end up with where we used ( [ magncharge ] ) .
the sum extends over defects on the inhomogeneous @xmath123-plane .
let us have a closer look at the contribution of a given defect on the @xmath123-plane .
consider first a _ basic defect _ with minimal non - abelian centralizer .
then all two forms @xmath221 are regular and must lead to the same contribution ( [ onedefect ] ) .
we see at once that the magnetic charge must be proportional to @xmath95 , @xmath222 and it contributes @xmath203 to the instanton number .
a non - basic defect on the inhomogeneous face must also lie on at least one of the homogeneous faces , say the @xmath122-plane .
for such a defect we must not take the corresponding singular @xmath217 in ( [ instantonnu ] ) or @xmath223 in ( [ onedefect ] ) .
we see that @xmath28 may be an integer linear combination of @xmath95 and @xmath224 .
more generally , if the defect lies on the @xmath123-plane and several homogeneous planes , then since a defect on the @xmath123-plane can at most sit on @xmath143 of the @xmath24 homogeneous planes , the representation ( [ chargedec ] ) for the magnetic charge is unique .
outside of the defects we could have taken any @xmath212 instead of @xmath216 .
then only defects on the @xmath120-plane would contribute to the instanton number and we would find @xmath225 again the contribution of a given defect must not depend on @xmath110 if the corresponding two form @xmath226 is regular on the defect . as above
we conclude that the magnetic charge of a defect is in the co - root lattice of the defect centralizer , and that the instanton number is @xmath227
in this section we consider a parametrization of @xmath228 in terms of static ` higgs ' fields .
this may seem to be a backward step since we are encoding a smooth group - valued object , @xmath228 , in terms of @xmath25 , in general singular , lie algebra - valued fields .
however the higgs fields facilitate a very direct proof that the @xmath229 @xmath134-forms introduced in the previous section have the stated smoothness properties .
moreover , we shall see that the magnetic charges of the defects can be related to higgs winding numbers around the defects .
one can define a ` basic ' higgs field , @xmath230 , as follows ( ) = ^(0 ) ( ) = w()h()w^-1().now , @xmath231 , is smooth everywhere except for the inhomogeneous @xmath123-plane .
this follows because the centralizer of @xmath75 commutes with @xmath153 unless @xmath232 .
it is possible to define ` alternative ' higgs fields which are smooth on the @xmath123-plane , but singular on one of the homogeneous @xmath122-planes , i.e. consider ^(i)=w()(h()- _ ( i)^h)w^-1(),i=1,2, ... ,r.@xmath96 being the @xmath122th coxeter label . the field @xmath233 is smooth everywhere except points on the @xmath122-plane .
the relation between the polyakov loop and the alternative higgs fields is as follows @xmath234^{n_i}z_i=\exp\left[2\pi i n_i \phi^{(i)}(\vx)\right],\end{aligned}\ ] ] where @xmath235 is the center element @xmath236 $ ] .
the @xmath25 higgs fields @xmath237 ` cover ' the group in the sense that it is possible to partition @xmath19 into patches , so that in each patch at least one of the higgs fields is smooth . in the previous section we wrote @xmath238 as the derivative of two forms @xmath212 .
we claimed that @xmath212 is only singular on the @xmath120-plane .
in other words , wherever @xmath239 is smooth @xmath212 is smooth .
this is obvious in the light of the following identity [ mainformula ] ^()= 12 ^ 2 ^2 ^1_0 ds ( s-1),where as before @xmath240 is the least common multiple of the coxeter labels into the integral and comparing with equation ( [ a ] ) .
alternatively , one can get it from the identity @xmath241 $ ] . ] .
we now show that the magnetic field , @xmath219 can be written in terms of the higgs fields . using the fields
@xmath233 one can construct normalized higgs fields @xmath242 as follows @xmath243 in terms of the normalized higgs fields , the magnetic fields are @xmath244{}\end{aligned}\ ] ] if the coxeter label @xmath96 is unity , the integral reduces to let @xmath193 be a closed surface surrounding a defect .
since the centralizer of @xmath245 in @xmath0 is @xmath246 , where @xmath247 is semi - simple , the normalized higgs field @xmath242 induces a map from @xmath193 into a coset space @xmath248 with @xmath249 .
that is to each normalized higgs field @xmath250 there is _ one _ associated winding number which can be identified with @xmath251 .
for @xmath61 all the coxeter labels are unity , and so @xmath252 for the groups @xmath132 , @xmath127 , @xmath253 , @xmath254 and @xmath255 it seems that the magnetic field can not be written trilinearly in normalised fields since ( [ final ] ) only applies if the relevant coxeter label is one .
for example the gauge group @xmath255 has only one unit coxeter label , @xmath256 .
however , the weyl orbit of @xmath257 contains a linearly independent basis of the root space . to make this more concrete , consider the field @xmath258 a simple calculation shows that @xmath259 if and only if clearly , @xmath260 is a solution of ( [ cubic ] ) if and only if @xmath261 .
but there are other solutions of ( [ cubic ] ) apart from the co - weights with unit coxeter ; these correspond to weyl reflections of the co - weights .
in fact for @xmath132 , @xmath127 , @xmath253 , @xmath254 and @xmath255 one can always find @xmath24 linearly independent solutions of ( [ cubic ] ) which we denote by @xmath262 .
thus we have @xmath263 where now @xmath264 , and the @xmath265 are dual to the @xmath266 in the sense that @xmath267 ( the @xmath265 are roots or weyl reflections thereof ) .
to each normalized higgs field @xmath250 there is _ one _ associated winding number which can be identified with @xmath268 . for the groups @xmath269 , @xmath270 and @xmath137 no solutions of ( [ cubic ] ) exist .
[ examples ] in this section we illustrate the ideas of the previous sections by considering the relevant gauge group @xmath31 . in the instanton number calculation of chapter 6 we assumed that our matter transformed according to the covering group . here
we will also consider the case of matter in the adjoint representation by allowing for twists .
first we consider @xmath31 with untwisted gauge fields , i.e. the polyakov loop operator in the defining representation .
the fundamental domain @xmath84 has been depicted in figs.([su3]a,[bild5 ] ) .
the magnetic charges of the three types of defects corresponding to the three edges of @xmath84 are integer multiples of @xmath271 because of overall charge neutrality the magnetic charges of all defects must add up to zero , @xmath272 = 4.5 cm any cluster of magnetic monopoles connected by a dirac string has vanishing magnetic charge .
for example , if a monopole pair is uncharged no dirac string , besides the one connecting the two monopoles , is needed . since defects on the @xmath123-plane for which @xmath273 ( ignoring ` higher defects ' ) contribute to the instanton number
as @xmath274 the monopole pair connected by a string in fig.[bild5 ] does not contribute to the instanton number .
the three monopoles connected by a dirac string contribute @xmath275 to the instanton number . what about defects with larger centralizers ? if @xmath276 , in which case @xmath153 lies at an extremal point of @xmath84 in fig.[bild5 ] , then @xmath74 has maximal degeneracy and the centralizer is @xmath135 .
such a defect has magnetic charge @xmath277 and contributes with @xmath278 to the instanton number .
finally , let us switch to the adjoint representation . in principle
we could do this by restricting @xmath153 to the fundamental domain for the adjoint representation , see the shaded regions in fig.[su3 ] . however
this would lead to walls on which @xmath279 is not smooth .
a much easier approach is to work in the covering group _ but now with arbitrary twists . _ in general this leads to a fractional instanton number .
such fractional instanton numbers are related to a loss of charge neutrality and nonperiodicity of @xmath280 engendered by the twists . for example
consider the following set of twists @xmath281 \id = \exp[2 \pi i \mu_{(1)}^\vee \cdot h ] , z_{23}=\exp[2\pi i/3 ] \id = \exp[2 \pi i \mu_{(2)}^\vee \cdot h ] $ ] , and all other twists the identity .
this is an example of non - orthogonal twists , and leads to an instanton number of the form @xmath282 where @xmath283 . from the periodicity properties of @xmath284 @xmath285 we obtain periodicity properties of @xmath286 ,
@xmath287 $ ] and @xmath288 . in our example
we get @xmath289 where @xmath290 and @xmath291 corresponds to an element of the weyl group , here a rotation of @xmath292 .
the equation can be understood as follows . by multiplying @xmath293 with @xmath294 we shift @xmath295 by the vector @xmath296 .
then we have to weyl reflect this shifted vector back into the fundamental domain @xmath297 with @xmath291 . in @xmath297 itself
this corresponds to a rotation with angle @xmath298 around the center of the equilateral triangle @xmath297 .
it follows that we get charge neutrality in the ` tripled ' torus obtained by taking three adjoining tori in the @xmath299-direction .
if we have in the first torus a defect of one type then in the adjoining torus in the @xmath300 direction we have a defect with the next type of charge and so on , see fig.[b2 ] .
= 3.5 cm in the @xmath301- and @xmath302-directions @xmath288 is periodic ( @xmath303 ) .
the periodicity properties of @xmath304 are given by @xmath305 where @xmath306 and @xmath307 are twist eaters such that @xmath308 and @xmath309 are functions with values in the cartan subgroup can not be chosen smooth on the whole torus . ] . from these conditions
we obtain periodicity of the magnetic field strength @xmath310 in the @xmath311- and @xmath312-directions and @xmath313 . to calculate the topological index @xmath314 we may use the 2-forms @xmath315 , but now we will get contributions from the ` boundary ' of the torus .
this is in contrast to the non twisted case where we have had no contributions from the boundary because of the periodicity of @xmath316 .
we assume that there are no defects on the boundary
. then we can integrate @xmath317 over the boundary .
one easily checks that @xmath317 is periodic in the @xmath311- and @xmath312-directions .
therefore we end up with @xmath318 this shows the relation between the noninteger boundary contribution and using the cocycle condition for @xmath319 and @xmath320 one easily sees that @xmath321 is indeed noninteger .
] to the instanton number and the total magnetic flux through the torus which results from the loss of charge neutrality on @xmath19 . in our example
the element @xmath291 of the weyl group is a rotation of @xmath322 in the cartan subalgebra .
therefore @xmath323 which shows together with the periodicity properties of @xmath324 that in the tripled torus we have no boundary contributions to the topological index .
in this paper we have considered gauge - fixing of yang - mills theory on the four torus for arbitrary gauge groups , instanton sectors and twists .
we have generalized our earlier results @xcite on the extended abelian projection with respect to the polyakov loop operator on the four torus .
we have constructed a complete set of non - abelian transition functions , which encode the ` boundary conditions ' for the gauge potentials , for all instanton sectors and arbitrary twists . with these transition functions
the path ordered exponential , @xmath325 , which is central to the gauge fixing , is periodic up to multiplication by constant matrices , even though of course the gauge field itself is non - periodic .
then we found an explicit gauge transformation which transforms @xmath1 into the cartan subalgebra and hence the polyakov loop operator into the toroidal subgroup of @xmath0 . the resulting gauge fixed @xmath1 is time independent .
we have fixed the freedom in choosing the gauge transformation by restricting @xmath1 to a fundamental domain in the cartan subalgebra . in the sectors with non - vanishing instanton number the final gauge fixed potential must have singularities @xcite .
these singularities are due to ambiguities in the diagonalization of @xmath325 at points where the centralizer of @xmath325 is non - abelian .
there is a close analogy between these defects and magnetic charges in yang - mills - higgs theories .
the defects are classified according to the non - abelian centralizer subgroups of @xmath66 .
a point @xmath152 belongs to a defect if the gauge fixed @xmath326 lies on the boundary of the fundamental domain . here
the results for @xmath15 may be misleading ; at the defects the polyakov loop operator need not be in the center of the gauge group as it must for @xmath15 . for example , for @xmath327 the center is trivial but there are many different types of defects corresponding to the different faces of the fundamental domain .
the magnetic charges of the defects are quantized and linearly related to the points of the integral co - root lattice .
for all groups with nontrivial centers we have constructed @xmath24 normalized higgs fields which wind around the magnetized defects .
finally we generalized earlier results in @xcite and related the magnetic charges of a given type of defect to the instanton number @xmath21 .
in particular , if @xmath328 then all possible magnetic defects must appear .
one may view our gauge fixing as the ` nearest ' fixing to the weyl gauge compatible with time periodicity .
yet unlike the weyl gauge we find monopole like singularities .
this is gratifying , since in those theories where we analytically understand confinement , the latter is due to the condensation of monopoles ; these examples are compact @xmath329 @xcite and supersymmetric yang - mills theories @xcite .
of course , there is a long way from the picture of condensed magnetic monopoles to real @xmath330 .
the treatment given here has been purely classical
. the next step would be to study the path integral within this gauge fixing . at this point
one would need a sensible approximation @xcite .
the balancing of the energy and the entropy of monopoles ( and/or loops ) may explain the occurrence of the deconfinement transition in @xmath330
. it would be interesting to clarify the role of the center of the gauge groups .
there are gauge groups with trivial centers but many different types of monopoles and other magnetic defects .
we are grateful to falk bruckmann , jan pawlowski and hugo reinhardt for helpful discussions .
we thank r. jackiw for bringing @xcite , in which the relation between magnetic charges and the instanton number has been discovered , to our attention .
we prove that for arbitrary twists and instanton number there exist transition functions with the following property u_0=,u_i(x^0=0)=_i , where the @xmath331 are twist eaters satisfying [ hcocycle ] _ i _ j = z_ij_j _ i , z_ij .we now start off with _ abelian transition functions _ where @xmath332 is a cartan sub - algebra valued lower triangular matrix
_ = 0&0&0&0 + * n*^1&0&0&0 + * n*^2&*m*^3&0&0 + * n*^3 & -*m*^2 & * m*^1&0.with this choice of @xmath332 we have @xmath49 .
the cocycle condition ensures that the @xmath333 and @xmath334 satisfy the constraints e^2i * n*^i = z_0i , e^2i * m*^1=z_23.the instanton number is simply @xmath335 now we claim that there exists a time - independent gauge transformation @xmath336 with the following properties to prove this consider the following two sets of transition functions .
firstly take the abelian transition functions ( [ abelian ] ) but with the @xmath333 all set to zero .
secondly take the set of transition functions @xmath337 , where the @xmath331 are defined as in ( [ hcocycle ] ) .
now both sets of transition functions have instanton number zero and an identical set of ( magnetic ) twists .
hence they must be gauge equivalent @xcite .
this establishes the existence of a smooth @xmath336 satisfying ( [ vprop ] ) .
now we perform this gauge transformation on the original abelian transition functions ( i.e. with the @xmath333 not necessarily zero ) u_0^v=,u_i^v = v^-1()v(+b_i).these transition functions have the stated properties .
this proof hinges on two assumptions : * the existence of abelian transition functions for arbitrary twists and instanton number . *
the existence of twist eaters for all possible magnetic twists @xmath338 .
it is well known that the first assumption breaks down in the odd instanton sectors of untwisted @xmath15 gauge theory .
this special case has been addressed in ref .
@xcite . in @xcite
it was shown that the second assumption is valid for @xmath61 .
we will show the existence of magnetic twist eaters also for the other simple lie groups .
for every group ( with the exception of the @xmath339-series , which will be considered separately ) the cyclic center is generated by @xmath340 in the table below we list the co - weights @xmath341 generating the centers .
we now argue that magnetic twist eaters can be constructed from an abelian element @xmath342 and an element @xmath291 in the weyl group .
the abelian element @xmath342 is given by where @xmath343 is the coxeter number ( see the table below ) and @xmath344 is the weyl vector latexmath:[\ ] ] as required .
now we prove that for given magnetic twists @xmath358 we can find twist eaters @xmath359 satisfying equation ( [ hcocycle ] ) .
we make the ansatz @xmath360 it follows that equation ( [ twisteaters ] ) is equivalent to where @xmath361 is the order of the center group . if all twists are the identity ( all @xmath96 are zero ) the solution is trivial .
so let us assume that at least one @xmath96 , say @xmath362 is not zero .
then we choose @xmath363 it remains to be shown that for a given @xmath364 and @xmath365 we can solve if the order of the center is a prime number , as it is for all but the @xmath192 and @xmath166 groups , then we can always find a @xmath366 solving this equation . for the @xmath253 groups with odd
@xmath24 the order of the center is not prime but @xmath367 .
if only one @xmath96 , say again @xmath362 is odd then we can again solve ( [ eee ] ) . in the other case
all @xmath96 must be even and ( [ eee ] ) can again be solved .
this proves the existence of twist eaters for all but the @xmath368-groups with even rank . for the @xmath368-groups with even rank the center comprises of @xmath369 where @xmath370 . as before one can find commuting weyl words @xmath371 such that for each center element for example
, @xmath372 now we make a case by case analysis to show the existence of twist eaters for arbitrary given twists . using ( [ twistss ] )
one finds the following solution for the possible choices for @xmath338 in ( [ hcocycle ] ) : [ cols="<,^,^,^,^,^,>",options="header " , ] table 1b : _ centers @xmath373 , generators @xmath341 of the centers : @xmath374 , coxeter labels @xmath96 , dual coxeter labels @xmath97 and coxeter number @xmath375 of the exceptional groups _ 0.5truecm
for @xmath376 this formula is easily checked if one uses @xmath377 and the definitions @xmath378 and @xmath379 . to prove the formula for @xmath380 is less trivial . as a first step consider two group valued fields @xmath381 .
then @xmath382 if the @xmath383 are smooth and periodic then @xmath384 with our choice for the transition functions the polyakov loop operator is indeed periodic and we conclude that now we can relate the instanton number in ( [ topological ] ) to the winding of @xmath385 as follows @xmath386 since @xmath387 we can now apply formula ( [ aaa ] ) with @xmath166 replaced by @xmath388 .
this then leads to @xmath389 where @xmath212 is smooth in @xmath390 and has been defined in ( [ a ] ) .
this proves ( [ aaa ] ) for @xmath380 as required . | we consider yang - mills theories with general gauge groups @xmath0 and twists on the four torus .
we find consistent boundary conditions for gauge fields in all instanton sectors .
an extended abelian projection with respect to the polyakov loop operator is presented , where @xmath1 is independent of time and in the cartan subalgebra .
fundamental domains for the gauge fixed @xmath1 are constructed for arbitrary gauge groups . in the sectors with non - vanishing instanton number
such gauge fixings are necessarily singular .
the singularities can be restricted to dirac strings joining magnetically charged defects .
the magnetic charges of these monopoles take their values in the co - root lattice of the gauge group .
we relate the magnetic charges of the defects and the windings of suitable higgs fields about these defects to the instanton number .
abelian projection on the torus for general gauge groups 1truecm * * c .
ford , t. tok and a. wipf * * + _ theor.phys .
institut , universitt jena + frbelstieg 1 , d07743 jena , germany _ 1 truecm pacs numbers : 11.10wx , 11.15tk , 11.15kc , 12.38aw .5 truecm * keywords : * gauge field theory at finite temperature , gauge fixing , abelian projection , magnetic monopoles , instanton number |
the observation of the cosmic microwave background radiation by the planck satellite is consistent with the predictions of simple , small - field slow - roll models : perturbations are adiabatic , gaussian , with a red spectral index and gravitational waves , which would rule out small field models if confirmed independently . ]
are not observed @xcite .
on the other hand , inflationary models in string theory are commonly of the multi - field type , with models on landscapes ( higher dimensional moduli spaces @xcite ) appearing to be common . extracting generic predictions of inflation on such landscapes
has received increased attention , as evident in the ongoing investigation of random landscapes @xcite the effect of decaying fields during inflation ( staggered inflation ) @xcite , cascade inflation @xcite or multi - field open inflation @xcite among others multi - field models @xcite .
it should be noted that the presence of eternal inflation , requiring the choice of a measure as well as anthropic reasoning ( see e.g. @xcite for a recent analysis of the measure problem and @xcite for a review of proposed measures ) , can hamper solid predictions . while planck is consistent with single field models , it does not preclude , but merely constrains most multi - field effects , see e.g. @xcite for a summary .
for instance , the absence of primordial non - gaussianities @xcite curtails the curvature of the end - of - inflation hypersurface @xcite , modulated reheating @xcite , curvatons @xcite , modulated trapping @xcite and particle production during inflation @xcite , among other effects , see @xcite for reviews .
furthermore , the absence of gravitational waves @xcite puts pressure on large - field models ( @xmath0 due to the lyth bound @xcite ) , whether they are of the single- or multi - field type . in this article
we consider trapped inflation in higher dimensional field spaces @xcite , which is a class of multi - field models operating on moduli spaces containing a dense distribution of extra - species points ( esps ) .
esps are locations at which additional degrees of freedom become light and can be produced kinematically ; once produced , they need to be incorporated into the low energy effective field theory .
esps are generic in string theory and often associated with additional symmetries , see @xcite for some examples ( see also @xcite for a discussion of the string higgs effect ) .
applications include moduli trapping @xcite , effects on inflation @xcite , trapped inflation @xcite ( see also @xcite ) , modulated trapping @xcite , and preheating @xcite ( see @xcite for a review and extensive references on preheating ) .
esp - distributions can be dense , as in trapped inflation @xcite , or sparse , as in @xcite , depending on the moduli space under consideration . in this paper we model the extra species particles phenomenologically by including additional scalar fields that are coupled quadratically to the inflatons , as in @xcite ( see @xcite for the incorporation of gauge fields ) .
trapped inflation is based on the inclusion of backreaction of the extra species particles onto inflatons @xcite : once particles are produced , an attractive force towards the esp results , affecting inflationary dynamics @xcite .
if the dimensionality of field space is large , @xmath1 , and esps are dense , a terminal velocity @xmath2 results @xcite .
@xmath3 becomes independent of the potential in the large @xmath4 limit .
as long as the potential is steep enough , the trajectory is traversed at this constant speed ; thus , functional fine tuning , i.e. , the @xmath5 problem , is relaxed .
preheating was discussed in @xcite , entailing qualitatively new resonance effects : efficient preheating is likely for dense distributions , @xmath6 , in contrast to preheating with a single esp at the vev of the inflatons @xcite .
intrinsic perturbations ( i.e. , solutions to the homogeneous equation for perturbations ) are expected to have a red spectral index and a suppressed amplitude similar to perturbation in single field trapped inflation @xcite .
we leave the computation of this spectrum to a future publication in lieu of focussing on additional curvature fluctuations sourced by backscattering of the produced particles of the inflaton condensate .
this effect was computed for distinguishable esp encounters in a series of papers by barnaby et al .
@xcite : in the power - spectrum , a bump ( with a suppressed ringing pattern ) results at a wave - number corresponding to the hubble radius at the time of the encounter . in our setup , several esps are encountered at any given time , leading to a superposition of bumps . in sec .
[ sec : powerback ] , we compute analytically a nearly scale - invariant contribution to the power - spectrum , @xmath7 , resulting from this superposition .
we find a small blue tilt that is ruled out by planck ( a red index was found at the @xmath8 level ) .
consequently , the amplitude of @xmath7 needs to be sub - dominant , leading to an upper bound on the coupling constant , @xmath9 .
we discuss the implication for trapped inflation in sec .
[ sec : discussion ] , concluding that only a narrow region of parameter space remains viable .
we comment on non - gaussianities in sec .
[ sec : nong ] , which receive an additional contribution from the same effect .
we expect a similar constraint on @xmath10 from bounds on the bi - spectrum , but leave a computation to future work .
the detailed outline of this paper is as follows : we start with a brief recap of trapped inflation in sec .
[ sec : recap ] , providing a heuristic argument motivating the presence of a terminal velocity .
the conditions to drive trapped inflation at @xmath3 are summarized in sec .
[ sec : conditions ] .
readers familiar with @xcite may skip these sections . after commenting briefly on the intrinsic power - spectrum in sec .
[ sec : intrinsic ] , we derive the additional contribution from backscattering in sec . [
sec : powerback ] , followed by a discussion of the allowed values of @xmath10 and @xmath11 in sec .
[ sec : discussion ] .
we comment on the effectiveness of preheating and non - gaussianities thereafter , before concluding in sec .
[ sec : conslusion ] . throughout this article
we set the reduced planck mass to unity , @xmath12
in @xcite the mechanism of trapped inflation @xcite has been generalized to higher dimensional field spaces and preheating has been discussed in @xcite .
we refer the interested reader to @xcite for a thorough derivation of the results stated in this section , but would like to provide a heuristic argument to aid the reader s intuition .
extra species points ( esps ) are locations in field space at which additional degrees of freedom become light , so that they can be produced kinematically . here
, we wish to consider a higher dimensional field space with potential @xmath13 , @xmath14 , @xmath1 and a dense , homogeneous and isotropic esp distribution , characterised by an average inter - esp distance @xmath11 .
we model the extra degree of freedom at a single esp at @xmath15 by an additional light scalar field coupled to the inflaton via the interaction lagrangian @xmath16 for simplicity , we assume canonical kinetic terms and an identical coupling @xmath10 between the inflatons and the extra fields . if the bare mass @xmath17 is small and @xcite @xmath18 with @xmath19 and @xmath20 the hubble parameter , particle production during an esp encounter can be computed analytically , identical to particle production during preheating @xcite .
the produced particle density depends sensitively on the distance of closest approach , i.e. the impact parameter @xmath21 so that @xmath22 evidently , if @xmath23 , particle production is suppressed and this particular esp can be neglected .
however , if the trajectory comes closer , a fraction of the inflatons kinetic energy is transferred to @xmath24-particles . as the trajectory moves away from the esp ,
these particles become heavy and , via the coupling in ( [ lint ] ) , lead to an attractive force towards the esp that diminishes over time in an expanding universe , @xmath25 .
such a single encounter can slow down , bend and , for strong coupling , temporarily trap the trajectory @xcite . while interesting effects similar to the curvaton scenario can result for such a single encounter @xcite
, we wish to consider the other extreme of a dense esp distribution , so that individual encounters become indistinguishable from each other . due to the exponential suppression in ( [ suppression ] ) , only esps in a cylinder around the trajectory with radius of order @xmath26 need to be considered . if @xmath27 , esps are encountered individually as in @xcite , if at all . on the other hand , if @xmath28 , many esps are close enough to the trajectory at any given time for particle production to take place .
since we distribute esps homogeneously , no particular direction is singled out .
hence , if the trajectory in field space is sufficiently straight in the absence of esps , the resulting backreaction of the produced @xmath24 particles onto the inflatons is opposite to the velocity @xmath29
. let s consider inflatons placed somewhere high up in a steep potential @xmath30 : initially at rest , the fields accelerate due to the gradient . as long as @xmath31 , hardly any esps are in reach and the inflatons keep accelerating .
once @xmath32 is reached , particles at many esps are produced , resulting in an opposing force to the speed , and thus the gradient of the potential if the trajectory is sufficiently straight .
this force becomes exceedingly strong as @xmath33 increases , subsequently reducing @xmath33 . on the other hand ,
if @xmath33 dips below @xmath34 , the speed picks up again due to the gradient of the potential .
thus , after initial oscillations , balancing these two forces leads to a movement at a terminal velocity of order @xmath34 , see fig .
[ pic:1 ] for an illustration . distributed on a @xmath35 dimensional field space
; the arrow denotes an approximately straight trajectory given by @xmath36 .
the blue circles have radius @xmath26 ; only those esp that have an imact parameter of @xmath37 ( dark blue ) can affect inflationary dynamics , due to the exponential suppression in ( [ rhochi ] ) .
top : the speed in field space @xmath33 is small , so that @xmath38 for almost all esps . except for rare encounters with single esps , no particle production takes place .
bottom : as @xmath33 increases , @xmath39 , and many esps become within reach of the trajectory .
back - reaction provides a friction force , leading to a movement at the terminal velocity @xmath3 in ( [ approximatevt ] ) .
[ pic:1 ] ] this heuristic argument can be made precise , see @xcite , leading to the terminal velocity @xmath40 in the large @xmath4 limit ( this result ignores factors of order @xmath41 ) , where we defined @xmath42 as the field along the trajectory . since @xmath43 for large @xmath4 , we can approximate @xmath44 in the large @xmath4 limit , which is independent of the potential ( assuming that the slope of the potential is steep enough to drive @xmath33 towards @xmath3 ) . in order for ( [ approximatevt ] ) to be self - consistent and inflation to be driven at the terminal velocity ,
several conditions need to be satisfied ( see @xcite for more details ) : 1 . the potential needs to be steep enough to ignore hubble friction , yielding the condition @xmath45 .
if @xmath46 , the slow roll regime is entered before the terminal velocity is reached .
2 . to avoid prolonged oscillations around @xmath3 , the time for backreaction to act needs to be smaller than the characteristic time needed to change the speed due to the potential .
this entails the lower bound @xmath47 .
3 . the trajectory should not be strongly curved over a few hubble times , so that backreaction is approximately anti - parallel to the velocity and the gradient , yielding the upper bound @xmath48 .
4 . the potential energy needs to dominate over the kinetic energy which in turn needs to be bigger than the energy in produced particles , to guarantee that inflation takes place .
at least @xmath49 e - folds of inflation are needed to solve the standard problems of the big bang .
the study of perturbations in trapped inflation is subtle , since backreaction of @xmath24 particles onto the background is included at the level of the equations of motion , not the lagrangian . in the one dimensional case ,
an analysis of perturbations in @xmath42 in the uniform curvature gauge for a desitter background is given in @xcite , which can be generalized to our case .
this computation takes place in two steps : first , the solution of the homogeneous equation leads to the intrinsic power - spectrum . in v1 of the this article on the arxiv ,
a discussion of the intrinsic power - spectrum was included , following the standard approach to cosmological perturbation theory as used in @xcite .
this computation , as well as the section pertaining to perturbations in @xcite , is not applicable , since it is based on the standard lagrangian for cosmological perturbations ; instead , the methods employed in @xcite should be used .
we plan to correct this computation in a forthcoming separate publication and focus on a second contribution to the power - spectrum induced by the peculiar solution to the inhomogeneous equation , i.e. by the inclusion of a source term due to backscattering : perturbations in the produced particle density ( responsible for the speed limit and thus crucial for the background solution ) back - scatter off the inflaton condensate . for a single esp encounter
, this back - scattering effect leads to a bump in the power - spectrum , accompanied by a ( suppressed ) ringing pattern @xcite .
a dense superposition of these bumps should lead to a nearly scale invariant contribution to the power - spectrum , that can reach the cobe limit if @xmath10 is sufficiently large , as mentioned in @xcite .
we expect the this second contribution to dominate over the intrinsic power - spectrum without tuning , in line with the results of @xcite .
the computation of this contribution and its scalar spectral index is the aim of this paper .
we will show that this contribution @xmath7 carries an observationally ruled out blue spectral index so that it must be sub - dominant , leading to an upper bound on @xmath10 .
thus , the intrinsic power - spectrum has to saturate the cobe bound while satisfying observational constraints on the scalar spectral index and the tensor to scalar ratio or an additional mechanism has to be invoked to provide the power - spectrum .
in the former case , fields move at the terminal velocity while scales in the observational window leave the horizon ; a regular slow roll regime is entered thereafter .
this setup can be valid for complicated potentials as long as @xmath30 is steep enough and the conditions in sec .
[ sec : conditions ] are satisfied . if the spectral index or amplitude fail to meet observational requirements one may use a curvaton @xcite or modulated reheating @xcite , to provide the power - spectrum .
such an addition would render the model rather complicated and one may argue to discard it based on ockhams razor
. if trapped inflation is not operational during the last sixty e - folds of inflation , it may be present and useful at earlier times : if fields start out high up in a steep potential on some complicated ( random ) landscape in string theory inflation is rare in the absence of a speed limit ; if it occurs , it usually takes place near a saddle point @xcite .
if a speed limit is present , inflation is already taking place when the saddle is encountered , avoiding the over - shoot problem and addressing the initial value problem of inflationary cosmology ( why is @xmath50 ? ) . in this setup ,
the initial inflationary phase at the terminal velocity takes place before scales relevant for observations left the horizon and has therefore no observable impact .
) , denoting the region of slow roll dynamics for @xmath51 and trapped inflation at the terminal velocity @xmath3 in ( [ approximatevt ] ) for @xmath52 , with the transitional field value @xmath53 in ( [ defphit ] ) .
[ pic:2 ] ] in @xcite the effect of back - scattering of a single esp onto fluctuations of a single inflaton field were computed , yielding a bump - like contribution to the power - spectrum that can be approximated by @xmath54 which ignores a ringing pattern in the large @xmath55 tail ( we are not interested in these oscillations , since they will be averaged out via the superposition of bumps ) .
the amplitude was computed in @xcite by comparison to lattice field simulations to @xmath56 which was tested for a quadratic potential and @xmath57 ; it was found to be a good approximation up to factors of order unity @xcite .
we shall use this estimate to put constraints on @xmath10 subsequently , keeping this theoretical uncertainty in mind .
the location of the peak is set by the hubble scale at the esp - encounter , @xmath58 where @xmath59 is the number of e - folds between @xmath60 at @xmath49 e - folds before the end of inflation and the esp encounter at @xmath61 , and the proportionality constant is given by @xmath62 @xcite ( note that our notation differs from the one in @xcite ) . for us
, there are two important differences compared to @xcite : firstly , while @xmath33 is given by the slow roll speed in @xcite , it is the terminal velocity @xmath63 for us . as a consequence , we get @xmath64 so that @xmath65 with @xmath66 where we defined @xmath67 and used the monomial potential @xmath68 with @xmath69 in the last step as an explicit example .
we only consider esp encounters in the trapped inflation regime , i.e. @xmath70 ; the latter results by equating the slow roll speed @xmath71 with the terminal velocity @xmath3 , so that the transitional field value becomes @xmath72 for @xmath73 ( for @xmath74 the slow roll speed is constant , so that either slow roll or movement at at the terminal velocity is operational during the inflationary regime ) .
thus , the slow roll regime is operational for @xmath75 with @xmath76 at which the potential slow roll parameter equals one , while inflation is driven at the terminal velocity for @xmath77 , see fig .
[ pic:2 ] for an illustration .
the transition between the two regimes is smooth dips below @xmath78 , particle productions ceases , and the effect of the already produced particles red - shifts away since @xmath25 .
hence , the velocity slowly decreases from @xmath3 tracking @xmath79 . ] so that no sharp features in observables are expected .
the total number of e - folds @xmath80 can be computed to @xmath81 for @xmath73 , where we used the slow roll approximation for @xmath82 and @xmath83 as well as @xmath84 for @xmath85 . since @xmath86 and @xmath87 via ( [ defphit ] ) , it is usually not possible to solve @xmath88 for @xmath60 analytically , but once the inflationary energy scale @xmath89 is specified , it may always be done numerically .
a second difference compared to @xcite is that esps are not encountered head on , but distributed around the trajectory and passed in rapid succession . as a result ,
particle production at a single esp is suppressed by an impact parameter @xmath90 , i.e. , the distance of closest approach to the esp , leading to @xmath91 , which carries over to the energy density in ( [ rhochi ] ) and all associated effects .
these consequences were computed in @xcite , where it was shown that particle production , and thus backreaction , is dominated by esps with impact parameters close to @xmath92 particle production of esps further away is exponentially suppressed , while esps closer to the trajectory are too few to affect the dynamics strongly if @xmath4 is large ; the latter is a direct consequence of the fact that most of the volume , and thus esps , in a hypersphere is close to its surface in the large d limit .
thus , the corresponding particle density @xmath93 carries a suppression factor of @xmath94 , which in turn carries directly over to @xmath95 ; we may thus write the cumulative contribution to the power - spectrum from back - scattering as @xmath96 with @xmath97 and we defined the effective esp line - density @xmath98 where @xmath99 is the volume of a @xmath100 dimensional unit - sphere .
see fig .
[ pic:3 ] for a schematic of the infinitesimal volume element to be integrated along the trajectory
. in ( [ defmu0 ] ) affect inflationary dynamics via backreaction and potentially curvature perturbations via back - scattering ( see fig .
[ pic:1 ] ) .
[ pic:3 ] ] replacing @xmath42 by @xmath101 as the integration variable and defining the dimensionless function @xmath102 we get @xmath103 were we used ( [ defmu0 ] ) so that @xmath104 and extracted the leading order result in a @xmath41 expansion . to be concrete
, we used @xmath105 and the large argument limit of the @xmath106-function .
note that the suppression factor of @xmath107 of a single esp encounter with impact parameter @xmath108 is compensated by the large number of traversed esps .
replacing @xmath101 by @xmath109 via ( [ kesp ] ) and ( [ nespofy ] ) , and defining @xmath110 we can write the integral explicitly as @xmath111 to compute the scalar spectral index of the additional contribution to the power - spectrum , @xmath112 we need @xmath113 and its derivative at @xmath114 . let us focus on @xmath115 first : the integrand peaks at @xmath116 so that we may approximate the logarithm in @xmath117 by its value at @xmath118 .
the remaining integral can be approximated by @xmath119 similarly , we can compute @xmath120 by first approximating the logarithm in @xmath117 by its value at @xmath121 , computing the partial derivative of the entire integrand with respect to @xmath122 , approximating the remaining elementary integrals and finally setting @xmath114 , yielding after some algebra @xmath123 we checked the analytic approximations numerically , and we found them to be accurate at the percent level . with these approximations
, the scalar spectral index simplifies to @xmath124 evidently , this additional contribution to the power - spectrum has a blue spectral index @xmath125 for all @xmath126 ( the logarithm gives a factor of order one for the allowed values of @xmath10 in ( [ allowedintervallg ] ) ) , which is ruled out by the planck satellite at the @xmath8-level .
hence @xmath127 where we used @xmath128 , ca nt provide the dominant contribution to the power - spectrum , but has to satisfy @xmath129 demanding @xmath130 , yields the upper bound @xmath131 by using @xmath7 from ( [ pbsfinal ] ) , the approximate amplitude @xmath132 from ( [ aesp ] ) and @xmath60 from solving @xmath88 .
this bound depends on the inflationary energy scale @xmath89 and the terminal velocity @xmath3 .
if the intrinsic power - spectrum is computed , these parameters can be expressed in terms of observables such as the scalar spectral index and the tensor to scalar ration .
fortunately , for reasonable assumes an inflationary energy scale in line with slow roll models where @xmath133 is set by the cobe normalization ( using @xmath134 as in the estimate that follow ) .
this slow roll result _ does not _
apply for trapped inflation at the terminal velocity , which requires the computation of the intrinsic power - spectrum as in @xcite ( not provided in this paper ) .
if @xmath89 were considerably below @xmath135 , the lower bound on @xmath10 in ( [ allowedintervallg ] ) would be relaxed , but the upper bound would hardly be affected due to the logarithmic dependence . ] values of @xmath89 the logarithm in ( [ pbsfinal ] ) can be ignored , so that @xmath136 as @xmath3 has to be below the slow roll speed to be relevant , @xmath60 needs to be below the value it would have if esps were absent .
thus , taking @xmath137 and @xmath138 as a concrete example , we get @xmath139 taking @xmath134 or smaller so that @xmath3 is below @xmath79 yields @xmath140 as an upper bound . along with the lower bound on @xmath10 in ( [ lowerboundg ] ) , @xmath141 if @xmath142 and @xmath143 during inflation at the terminal velocity , we arrive at the allowed interval @xmath144 the upper bound on @xmath10 is comparable to the one originating from a single bump in the power - spectrum in @xcite , and somewhat lower than the estimate in @xcite . whereas individual bumps may be used to match outliers / oscillating features in the power - spectrum , and thus improve a fit , a seizable contribution of @xmath7 is not desirable due to its blue tilt .
the interval for @xmath10 is not particularly wide , but allowed values do not appear to be overly fine tuned either and cover the same range commonly considered for studies of preheating after inflation @xcite .
in fact , the fields at esps in the vicinity of the final resting place can act as preheat matter fields , as investigated in @xcite : there , it was found that preheating is qualitatively different than in models that have an esp at the vev of the inflatons ; while de - phasing of inflatons tends to suppress parametric resonance @xcite ( see @xcite for the case of two inflations , that still permits resonances ) , two new effects leading to efficient preheating were found : particle production during the first in - fall can already comprise a seizable energy transfer if the trapped inflation regime lasts until preheating commences , but it is never complete .
subsequent broad resonance is generically suppressed due to de - phasing of the fields , but if an esp happens to lie at a well defined distance from the vev , a prolonged narrow resonance regime can complete preheating . in @xcite
it was concluded that both effects are important / likely if the average inter esp distance is of order @xmath145 or smaller ( @xmath146 was used ) .
since @xmath128 the allowed interval of @xmath10 corresponds to @xmath147 such values of the average inter esp separation appear reasonable in light of the known examples of moduli - spaces in string theory . since @xmath145 lies outside of this interval
, it is unlikely for an esp to be at the right position for a prolonged narrow resonance regime .
furthermore , particle production during the first in - fall does not take place if the slow roll regime is operational towards the end of inflation .
the latter is the case for monomial potentials if @xmath148 in ( [ defphit ] ) is bigger than @xmath149 .
however , if the inflationary model is a large - field one , one may / should expect @xmath11 to differ by a factor of order one over the course of inflation , which might boost preheating ; if preheating is absent , fields either decay via tachyonic instabilities @xcite or the standard theory of reheating . to summarize , while a small value of @xmath150 still requires fine tuning of the inflationary energy scale can be large enough to saturate the cobe bound ; if this is the case , one can relate @xmath150 to @xmath89 and tune @xmath89 accordingly .
] , the functional fine tuning of the potential needed for large - field models in the absence of additional symmetries is absent for inflation at the terminal velocity , since inflation is due to a small @xmath151 , which in turn can be achieved by reasonable values of the coupling constant @xmath10 and inter esp - distance @xmath11 ; the main requirement is a steep enough potential to drive @xmath152 towards the terminal velocity , i.e. , @xmath153 ( see sec . [
sec : conditions ] for additional , mild conditions ) . in this regime
, we showed that the additional contribution to the power - spectrum from backscattering is observationally ruled out by planck , since it caries a blue spectral index .
as a consequence , only the narrow interval in ( [ allowedintervallg ] ) remains viable for trapped inflation at the terminal velocity , severely reducing the motivation to consider trapped inflation in the first place .
let us comment briefly on non - gaussianities : conclusions of @xcite , we do not expect large intrinsic non - gaussianities to be generated during inflation even if the potential is more structured than in the simple monomial cases considered here , since the trajectory has to be reasonable straight ( see point 3 in sec .
[ sec : conditions ] ) .
however , another source of non - gaussianities is present : as back - scattering leads to an additional contribution to the power - spectrum , it also acts as a source for higher order correlation functions , such as the bi - spectrum . in @xcite , the shape - function due to a distinguishable esp encounter
was computed ( see also @xcite ) , entailing a localised feature as well as oscillatory components .
such a highly structured bi - spectrum ( two integrals ca nt be expressed in closed form ) is difficult to compare with observations : non - linearity parameters @xmath154 for smooth shapes ( local , equilateral , orthogonal , etc . )
are essentially blind to localized and/or oscillatory features , and even modal expansions @xcite reach their limit of applicability fast for such oscillating signals , as discussed in @xcite . however , in our case many copies of the shape function in @xcite are superimposed , similar to the individual bumps in the power - spectrum .
hence , a smooth , nearly scale invariant bi - spectrum results , which should be much simpler to constrain .
it would be interesting to perform this computation and compare the resulting upper bound on @xmath10 with the one stemming from the power - spectrum ( we expect them to be comparable ) .
we leave this interesting project to a future study .
the presence of extra species points ( esps ) on moduli spaces in string theory is a generic phenomenon .
if fields evolve on such a higher dimensional landscape ( @xmath1 ) with a dense distribution of esps , a speed limit at the terminal velocity @xmath155 is present ( @xmath10 is the coupling constant to the extra species particles and @xmath11 the average inter - esp distance ) , potentially leading to trapped inflation .
we investigated the feasibility of this inflationary scenario in light of planck s observation of a red - spectral index , @xmath156 .
in addition to the intrinsic curvature perturbation which has a suppressed amplitude and is expected to carry a red spectral index ( not computed here ) , another contribution to the power - spectrum is present in trapped inflation : the produced extra species particles backscatter off the inflaton condensate , leading to additional curvature fluctuations .
a single esp encounter leads to a bump in the power - spectrum at a wave - number set by the hubble radius at the time of the event , as computed by barnaby et.al .. since the terminal velocity is due to the superposition of many esp encounters at any given time , the resulting superposition of bumps leads to an additional , nearly scale invariant contribution to the power - spectrum , @xmath7 .
we computed analytically the amplitude ( [ pbsfinal ] ) and the scalar spectral index ( [ nbsfinal ] ) of @xmath7 .
the index is always blue and therefore observationally ruled out by planck . as a consequence , @xmath7 needs to be sub - dominant , which leads to an upper bound on the coupling constant , @xmath157 for @xmath158 ( the lower bound is required by the employed method to compute particle production at esps ; it may be relaxed if the inflationary energy scale turns out to be considerably lower in trapped inflation compared to slow roll models , while still consistent with the cobe normalization . ) ; correspondingly , the average inter - esp distance needs to satisfy @xmath159 .
values for @xmath10 and @xmath11 in these narrow intervals provide the remaining parameter space for trapped inflation at a terminal velocity . given these narrow ranges , the appeal of trapped inflation is reduced as fine - tuning re - emerges to suppress the contribution to the power - spectrum from backscattering .
further , it is not clear if observational constraints can be met in this narrow range : the intrinsic power - spectrum needs to be computed to check whether or not the amplitude and spectral index of the power - spectrum as well as the tensor to scalar ratio are consistent with observations .
further , the lower bound on @xmath11 indicates that preheating at esps after inflation is most likely ineffective ; hence , inflatons decay either via tachyonic instabilities or the standard theory of reheating . an additional contribution to non - gaussianities from backscattering
is present , but given that @xmath7 needs to be sub - dominant , we expect these non - gaussianities to be below observational bounds . j. c. bueno sanchez and k. dimopoulos , phys .
b * 642 * , 294 ( 2006 ) [ erratum - ibid .
b * 647 * , 526 ( 2007 ) ] [ arxiv : hep - th/0605258 ] .
j. c. bueno sanchez and k. dimopoulos , jcap * 0710 * , 002 ( 2007 ) [ arxiv : hep - th/0606223 ] .
p. brax and e. cluzel , jcap * 1003 * , 016 ( 2010 ) [ arxiv:0912.0806 [ hep - th ] ] . | we consider trapped inflation in a higher dimensional field space : particle production at a dense distribution of extra species points leads to a terminal velocity at which inflation can be driven in steep potentials .
we compute an additional , nearly scale invariant contribution to the power - spectrum , caused by back - scattering of the continuously produced particles .
since this contribution has a blue tilt , it has to be sub - dominant , leading to an upper bound on the coupling constant between the inflatons and the extra species particles .
the remaining allowed parameter space is narrow . |
the abundance anomalies of the am stars ( and , as a rule , of most of the chemically peculiar stars ) are nowadays broadly considered to be produced by microscopic diffusion .
indeed , the observed anomalies are consistent with the computed diffusion velocities for many elements .
nevertheless , the detailed understanding of the am phenomenon requires more thorough studies since the stratification process is time - dependent and affected by the other physical processes at play in the stellar medium .
calcium and scandium are elements of special interest since their deficiencies are usually used to detect am stars ( conti 1970 ; preston 1974 ) .
the evolution of the abundances of ca and sc in the superficial layers of an a star has been computed by alecian ( 1996 ) assuming no helium convection zone , as suggested by the diffusion model for am stars .
his results show that the behaviour of both elements are strongly dependent on the strength of the large scale motions introduced in the computations , namely , a mass - loss and the extension of the superficial mixing zone beneath the convective zone . in some cases , phases of overabundance occur when the star is around 10 million years old .
so , the classification criterion of am stars based on the calcium or scandium deficiency is questionable for young a stars . abundance determinations in young main - sequence a stars can constrain such computations .
open cluster stars are the best candidates since their age is known with much greater accuracy than for field stars .
few abundance studies in open clusters have been done up to now ( see burkhart & coupry 1997 and references therein ) .
the use of electronic detectors ( reticon and ccds ) and the subsequent improvement in the quality of the spectra made possible the study of the lithium abundance and , by the way , renewed the interest for open clusters : observations of a - type stars were carried out by boesgaard ( 1987 ) in coma , and by burkhart & coupry ( 1989 ; 1997 ) in the hyades and pleiades .
this paper summarizes the work undertaken in collaboration with g. alecian and c. burkhart concerning the abundance of calcium and scandium in open cluster a stars .
more details are provided in hui - bon - hoa et al .
( 1997 ) and hui - bon - hoa & alecian ( 1998 ) .
we chose 34 a - type stars belonging to the northern open clusters : @xmath0 per , pleiades , coma , hyades , and praesepe ( see table 1 ) .
the ages of these clusters are well spread over the time interval covered by the simulations of alecian ( 1996 ) .
the selection criteria for each star were : - rotational velocity compatible with abundance determinations using the lines of our wavelength domain ; - availability of _ uvby @xmath1 _
photometry for the estimation of effective temperature and surface gravity .
we have included all a stars that match these criteria in our sample since alecian ( 1996 ) suggests that a phase of overabundance of ca and sc can occur for young am stars .
three field stars have also been observed : hr 178 ( hd 3883 ) , sirius ( @xmath0 cma , hd 48915 ) , and procyon ( @xmath0 cmi , hd 61421 ) .
.basic data for the clusters . [ cols="^,^,^",options="header " , ] references : ( 1 ) meynet et al .
( 1993 ) ; ( 2 ) boesgaard ( 1989 ) the spectra were obtained during two runs in december 1995 and december 1996 at the observatoire de haute - provence using the aurelie spectrograph at the coud focus of the 152 cm telescope .
this instrument was set to yield a spectral resolution of about 34000 ( linear dispersion of 5 @xmath2 ) .
the limiting _ v _ magnitude was around 9 for an integration time of 3 hours and a typical signal - to - noise ratio of 300 .
we chose the spectral interval 54955620 , which allows abundance determinations of mg , ca , sc , cr , fe , and ni .
we used model atmospheres computed with the atlas9 code ( kurucz 1992a , b ) .
the atmospheric parameters are derived from _
uvby @xmath1 _ photometry through the grids of moon & dworetsky ( 1985 ) which are still quite reliable for single stars as shown by napiwotski et al . (
1993 ) .
the abundance analysis is carried out assuming lte with a series of codes written by m. spite ( 1967 , 1996 private communication ) .
we used solar oscillator strengths deduced from a solar spectrum obtained with the same instrumentation as for the programme stars .
we mostly used a curve of growth method and the equivalent widths are measured with a procedure developed by cayrel et al .
( 1985 ) , which fits gaussian profiles to the observed spectrum . when the rotational velocity is above 30 km / s , the gaussian profile is not suitable anymore and the lines are fitted with a rotational profile .
as usual , the script [ x ] for any quantity x means @xmath3 . in fig . 1 , we can see a loose correlation between [ ca / h ] and [ sc / h ] for our cluster stars .
this suggests that the anomalies of these two elements come from the same physical process . besides
, the more pronounced deficiencies ( left part of the graph ) are found in the oldest stars ( members of the hyades and praesepe ) .
younger targets ( in @xmath0 per , pleiades ) show weak underabundances or marginal overabundances ( right part of the graph ) .
the abundance patterns for our sample stars are shown in fig . 2 . in ordinates
are the logarithmic differences between the abundance value in the star of concern and the solar one for mg , ca , sc , cr , fe , and ni .
the corresponding points are linked for each star .
an arrow means upper limit and error bars ended by arrows denote very uncertain values .
we can see that the am pattern is well - marked in the oldest cluster stars ( hyades and praesepe ) of our sample as well as for the field am stars . in the youngest clusters ( @xmath0 per , pleiades ) , the am pattern is almost absent and several objects show atypical patterns with marginal overabundances of ca and/or sc .
this would suggest that stars of these clusters are in transient phases of the stratification process .
the marginal overabundances of ca and sc in @xmath0 per and pleiades are not strong enough to confirm the phase of overabundance predicted by alecian ( 1996 ) .
either the youngest clusters of our sample are already too old and their stars have passed the phase of overabundance or this phase does not exist . in this last case ,
the extension of the mixing zone should be less than one pressure scale height and the mass - loss rate about @xmath4 .
these conclusions need to be confirmed by observations of more clusters and younger targets .
also , elements that could help a better understanding of the am phenomenon deserve to be studied ( rare earths for instance ) .
many thanks to m. spite who has kindly provided us with her codes and to c. vant veer for having introduced us to the atlas9 code .
thanks also to g. michaud and s. j. adelman for useful suggestions and to the ohp staff for their excellent support .
alecian g. : 1996 , _
a&a _ , * 310 * , 872 boesgaard a. m. : 1987 , _ apj _ , * 321 * , 967 boesgaard a. m. : 1989 , _ apj _ , * 336 * , 798 burkhart c. , coupry m .- f .
: 1989 , _ a&a _ , * 220 * , 197 burkhart c. , coupry m .- f . : 1997 , _ a&a _ , * 318 * , 870 cayrel r. , cayrel de strobel g. , campbell b. : 1985 , _ a&a _ , * 146 * , 249 conti p. s. : 1970 , _ pasp _ , * 82 * , 781 hui - bon - hoa a. , burkhart c. , alecian g. : 1997 , _ a&a _ , * 323 * , 901 hui - bon - hoa a. , alecian g. : 1998 , _ a&a _ , in press kurucz r. l. : 1992a , _ rev .
_ , * 23 * , 45 kurucz r. l. : 1992b , _ rev
_ , * 23 * , 181 meynet g. , mermilliod j .- c . ,
maeder a. : 1993 , _
a&as _ , * 98 * , 477 moon t. t. , dworetsky m. m. : 1985 , _ mnras _ , * 217 * , 305 napiwotski r. , schnberner d. , wenske v. : 1993 , _ a&a _ , * 268 * , 653 preston g. w. : 1974 , _ ara&a _ , * 12 * , 257 spite m. : 1967 , _ ann . dastrophys .
_ , * 30 * , 211 | abundances of mg , ca , sc , cr , fe , and ni are derived for a stars of five nearby open clusters of various ages using high resolution spectroscopy .
we point out a correlation between the abundance of ca and that of sc , suggesting that the abundance anomalies of these elements arise from the same physical process .
pronounced am patterns are rather found in the oldest cluster stars whereas younger targets show weaker am anomalies and atypical patterns for some of them . |
i am grateful to the organizing committee for the opportunity to speak on searches for high - redshift quasars at this meeting .
although i have not worked directly with hy , we have known each other for many years , and i have admired his research .
i think three qualities stand out : 1 ) his determination and persistence at pushing telescopes and instruments to their limits , 2 ) his evident success at finding and studying successively more distant galaxies , and 3 ) his cheerfulness and optimism , which i think inspired his students and collaborators to be so successful over the many years of his career .
i have also heard that hy never liked quasars very much , so i feel honored that the topic made it on the program .
maybe the organizing committee decided it was safe to do so , now that galaxies have overtaken quasars as the objects of highest known redshift since 1997 , when they displaced quasars for the first time since 1965 . turning to the subject of this paper ,
the goals of surveys for high - redshift quasars include the observational determination of : * the epoch of quasar formation * the role of dust obscuration * the relation of quasars to galaxy formation and evolution * the contribution of quasars to the ionization of the intergalactic medium at high redshifts according to the picture that quasars are powered by the accretion of matter onto massive black holes in the nuclei of galaxies , the formation and evolution of black holes , quasars , and galaxies are closely related .
the evolution of the space density of luminous quasars shows a strong peak at lookback times of 0.8 to 0.85 ( fig .
the evolution of galaxies is very much under debate , as we are hearing at this meeting , but it appears to extend over a broader range of redshift . among the questions we would like to address are : which formed first , galaxies ( or at least parts of galaxies ) , or quasars ?
what roles do interactions and mergers play in the fueling of quasars ? and how do we account for the chemical abundances in high - redshift quasars , whose emission - line spectra look very similar to those at lower redshift ?
we now have the observational capabilities to map the evolution of both quasars and galaxies to redshifts beyond 6 , when the universe was 5% or less of its present age .
hubble space telescope and 8 - 10-m ground - based telescopes have already enabled us to discover more objects at @xmath2 than seemed possible just a few years ago .
now the challenge is to carry out sufficiently large , systematic , and quantifiable surveys to answer the questions posed above . in this paper
i will discuss how to search for quasars at @xmath2 , concentrating on optical methods .
next i will review the status of current surveys and then compare the spectra of the quasars and galaxies with @xmath3 that have been found to date .
i will close by making some speculations based on the current results and considering what we may expect in the future .
since the discovery of the first quasar with @xmath5 in 1987 ( warren et al .
1987 ) , more than 150 such objects have been found ( g. djorgovski and r. mcmahon , private communications ) , and several search techniques are by now well developed .
they include : * multi - color imaging * slitless spectroscopy * deep radio surveys * deep x - ray surveys the multicolor technique pioneered by warren et al .
( 1987 ) and now applied to several surveys ( see , e.g. , warren et al . 1991 , irwin et al .
1991 , kennefick et al .
1995 , djorgovski 1999 ) has contributed the majority of the known @xmath5 quasars .
it uses the large @xmath6 color of quasars as a discriminant ; the presence of ly@xmath7 emission in the @xmath8 filter band combined with the continuum depression caused by intervening ly@xmath7 absorption in the @xmath9 filter band makes quasars stand out from cool stars and yields an effective search criterion .
the addition of @xmath10 band photometry , which is an indicator of the continuum level longward of ly@xmath7 , aids in distinguishing high - redshift quasars from late - type stars .
the multicolor technique was first used with photographic plates and schmidt telescopes .
the subsequent development of large format ccd cameras is now enabling wide - angle surveys to significantly fainter magnitudes , as discussed in more detail below . at higher redshifts , e.g. @xmath2 ,
the same principle applies , with ly@xmath7 shifting into the @xmath10 band .
however , as we shall see below , it is critical to have observations in the @xmath11 band , which is in the continuum longward of ly@xmath7 , to separate quasars from late - type stars .
the slitless spectroscopy technique is an effective way to discover high - redshift quasars through the direct detection of ly@xmath7 emission in low - resolution slitless spectra ( smith 1975 , osmer 1982 ) and has been used by schmidt , schneider , and gunn ( ssg , 1995 and references therein ) in their palomar grism surveys .
the selection effects and survey efficiencies are more straightforward to model for slitless spectroscopy in many cases than they are for the multicolor technique .
the ssg survey is one of the cornerstones of our knowledge of the evolution of optically selected quasars at @xmath12 .
deep radio surveys offer a different way to discover high - redshift radio galaxies and quasars .
a key property of radio - selected objects is that they are significantly less affected by surrounding or intervening dust that blocks optical radiation .
although radio - loud quasars in general constitute about 10% of the total quasar population , the availability of deep radio surveys that cover a significant fraction of the sky offers the opportunity to find significant numbers of radio - loud quasars and radio galaxies at high redshift .
current radio searches for high - redshift quasars make use of deep surveys for flat - spectrum sources ( e.g. hook and mcmahon 1998 ) while searches for high - redshift radio galaxies focus on steep - spectrum sources ( e.g. van breugel et al .
1999 , also this volume ) . in both cases , subsequent selection concentrates on optically faint objects to help weed out objects of lower redshift .
hook and mcmahon ( 1998 ) select objects that are red in @xmath6 to enhance the selection of high - redshift objects .
van breugel et al . ( 1999 ) make use of the well - defined @xmath13 diagram for radio galaxies and concentrate on objects with faint @xmath14 magnitudes
. the discoveries of the radio - selected quasar of highest known redshift , @xmath15 by hook and mcmahon ( 1998 ) and the radio galaxy of highest known redshift , @xmath16 by van breugel et al .
( 1999 ) demonstrate the power of their approaches .
it may be argued that x - ray emission is the key defining property of the global quasar / agn population and should be the primary search criterion for discovering them . however , until recently , x - ray observatories have not had sufficient sensitivity to reach the objects of highest redshift . the rosat deep survey ( hasinger et al .
1998 ) is an indicator of the advances we may expect from chandra ( e.g. , mathur , this volume ) and xmm as they come into operation ; it has yielded the x - ray selected quasar of highest known redshift , @xmath17 ( schneider et al . 1998 ) .
the requirements are a limiting x - ray sensitivity of @xmath18 erg s@xmath19 @xmath20 ( 0.5 2kev ) and high accuracy , @xmath21 arcsec positions
( to avoid confusion with foreground objects ) .
then follow - up optical imaging and spectroscopy , presumably concentrating on faint , red objects , should be an effective way of isolating high - redshift objects . to summarize , all approaches to find @xmath2 quasars need to cover a wide area on the sky to faint limiting sensitivities because the objects are so rare .
optical surveys need to have wavelength coverage extending to 0.9@xmath22 ; indeed all approaches benefit from such coverage .
searches at even higher redshifts will push the requirements into the infrared ( @xmath23 ) .
i would like to give a status report on two multicolor surveys my collaborators and i are carrying out : the btc40 and the bfqs .
i will also describe two other major optical surveys , cadis and sdss , and note the important contributions they are making to the search for quasars at @xmath2 .
the btc40 is a large collaborative effort led by e. falco and is aimed at finding gravitational lenses and distant galaxy clusters as well as high - redshift quasars .
it is an optical , multicolor survey that makes use of the btc camera and the ctio 4-m telescope .
julia kennefick is leading the quasar survey team , whose other members include alberto conti , richard green , pat hall , eric monier , malcolm smith , and myself .
btc40 has imaging data at high galactic latitude for 40 deg@xmath24 in the @xmath25 bands and reaches to @xmath26 25th magnitude at its deepest limit .
all the imaging data are in hand and are being analyzed .
the goal is to select @xmath2 quasar candidates to @xmath27 .
the bfqs survey is being led by pat hall in collaboration with the other team members mentioned above .
it also uses the btc camera and reaches to @xmath28 over 7 deg@xmath24 .
it uses the @xmath29 bands and is aimed at quasars with @xmath30 down to l * luminosities .
the bfqs imaging data are also in hand and are being analyzed .
together , the btc40 and bfqs surveys extend the multicolor technique to fainter magnitudes and wider areas than have been covered before . the cadis ( calar alto deep imaging survey , meisenheimer et al .
1998 ) survey extends the multicolor concept to higher spectral resolution and broader wavelength coverage by using 13 medium - band filters over the 0.39 to 0.93@xmath22 range and 3 broad - band filters , including @xmath31 .
the filters are designed to give both improved sensitivity to the quasars being sought and improved rejection of non - quasars , such as emission - line galaxies ( the veto filter concept ) .
cadis plans to cover 9 fields , each @xmath32 in area . to date the cadis team ( wolf et al .
1999 ) have shown the effectiveness of their approach by finding 6 quasars with @xmath33 to @xmath34 mag in one field , about 6 times the expected number .
the cadis technique is more powerful than the original multicolor approaches because of its increased spectral resolution and greater number of filters . on the other hand ,
it requires more observing time per unit area on the sky to reach a given limit in sensitivity .
thus , it is a complementary and valuable addition to techniques for finding high - redshift quasars
. it will be very important to see if the initial cadis results on the surface density apply to the additional fields in their survey .
previous surveys have been subject to considerable field - to - field fluctuations , and it will be very interesting to see how the cadis results come out .
our knowledge of the luminosity function of quasars to 22nd magnitude and fainter is still rudimentary .
the sdss ( sloan digital sky survey , gunn & weinberg 1995 ) will be the definitive survey for high - redshift quasars ( and also for quasar clustering ) down to 20th magnitude because of its large area coverage , 10,000 deg@xmath24 .
sdss is expected to yield 10@xmath35 quasars when completed .
already the initial results are very exciting .
fan et al .
( 1999 ) report the discovery of 15 quasars with @xmath36 in 140 deg@xmath24 , including quasars with @xmath37 and @xmath38 .
they extend the high - redshift record for quasars for the first time since 1991 . ] .
initial spectroscopic results from the btc40 survey confirm the importance of using the @xmath11 band for isolating @xmath2 quasar candidates .
unlike the situation at @xmath39 , where quasars can be selected with reasonable efficiency solely on their large values of @xmath40 , the analogous approach using @xmath41 does not work for @xmath0 quasars .
julia kennefick found that samples selected on the basis of large @xmath41 are overwhelmed by late - type stars .
the @xmath11 band is necessary because it is a measure of the continuum level at wavelengths longward of the ly@xmath7 emission , and the behavior of late - type stars and @xmath0 quasars is significantly different in , for example , the @xmath42 two - color diagram , as kennefick s simulations show ( fig .
2 , left ) .
we are now in the process of selecting @xmath0 candidates for follow up with spectroscopic observations .
the importance of the @xmath11 band is amply demonstrated in the sdss results ( fig .
2 , right ) , where fan et al . ( 1999 ) illustrate the stellar locus in the @xmath43 plane and the location of their @xmath44 quasars , which are well separated from the stars .
their results confirm that we now have well established techniques for finding the most distant quasars .
it is of interest to note that the spectra of the @xmath1 quasars discovered to date ( fig .
3 ) show the characteristically strong emission features of ly@xmath7 , c iv , nv and si iv + o iv ] seen in quasars at lower redshift and look remarkably similar to them .
furthermore , the spectra provide information on the chemical abundances in the broad - line region that have important implications for the evolutionary history of the host galaxies of quasars ( see review by hamann & ferland 1999 ) .
on the other hand , the spectra of the galaxies found so far at @xmath2 are very different from those of quasars ( figs . 4 and 5 ) .
for example , the dey et al .
( 1998 ) galaxy at @xmath45 has a very strong but narrow ly@xmath7 emission line on a weak continuum .
the weymann et al .
( 1998 ) galaxy at @xmath46 and the van breugel et al .
( 1999 ) radio galaxy show ly@xmath7 emission that is considerably weaker , while the spinrad et al .
( 1998 ) galaxy pair at @xmath45 has very weak , if any , emission and is distinguished primarily by a break in the continuum .
let us consider for a moment the faintness of these galaxies and the large amount of spectroscopic observing time that has been dedicated by several groups of observers with the keck telescopes to searching for faint galaxies at high redshift .
it is surely impressive that galaxies with @xmath0 are being found .
the numerous discoveries exceed what i thought only a few years ago would occur .
but the weakness of the spectral features of the galaxies shown in figs . 4 and 5 compared to the strength of quasar emission lines in fig .
3 do make one wonder if quasars / agns with broad - line spectra are truly very scarce at @xmath0 .
their spectra are much easier to identify than those of the galaxies seen to date .
what is their absence telling us ?
of course , it may just be too early to tell , and we must pursue the spectroscopic follow up of the @xmath47 surveys described above . in any case , the observational opportunities that we now have to study the universe at @xmath47 are very exciting , and we may expect continued important discoveries in the near future . let s meet again on the occasion of hy s 70th birthday to see what we have learned .
between june , 1999 , when the meeting was held , and may , 2000 , more quasars with @xmath2 have been discovered , including ones at @xmath48 ( zheng et al .
2000 ) , at @xmath49 ( stern et al .
2000 ) and at @xmath50 ( fan et al .
the @xmath51 object , which has @xmath52 , is the faintest known quasar at high redshift .
the one at @xmath53 , with @xmath54 on the ab system , is very luminous .
all have the strong emission lines characteristic of quasars at lower redshifts , and all have strong ly@xmath7 forest absorption at wavelengths shortward of their ly@xmath7 emission lines .
however , the universe is still highly ionized at @xmath53 .
fan et al . ( 2000b ) report that they have covered 600 deg@xmath24 so far in their work on the sloan digital sky survey and that the discovery of the @xmath50 quasar is consistent with the expectations of the schmidt et al .
( 1995 ) parameterization of the decline of the quasar luminosity function with increasing redshift .
they also note that for @xmath55 it is difficult to distinguish quasars from very cool stars and brown dwarfs on the basis of the sdss filters alone .
observations in the near infrared , where the cool objects have redder energy distributions than the quasars , aid in the selection of the quasars .
i am grateful to the organizing committee for the invitation to speak and for their financial support .
this work has also been supported by the national science foundation under grant ast-9802658 .
figures 2b , 3 & 5b are reproduced from the astronomical journal , and figures 4 & 5a are from the astrophysical journal ( letters ) , courtesy of the american astronomical society and the authors .
arjun dey , xiaohui fan , donald schneider , hyron spinrad and wil van breugel are thanked for giving permission to include their figures in this review article .
meisenheimer , k. , beckwith , s. , fockenbrock , h. , huang , j. , leinert , ch . , phleps , s. , roser , h .- j .
, thommes , e. , thompson , d. , wolf , c. , & chaffee , f. 1998 , in the young universe , eds . s. dodorico , a. fontana , e. giallongo , asp conf .
series , 146 , 134 | quasars continue to be the most luminous objects known in the universe but no longer have the largest redshifts .
i review current techniques for finding quasars at @xmath0 and the status of current optical surveys .
i compare the spectra of known quasars with @xmath1 with the spectra of some recently discovered galaxies with @xmath0 to see what we may expect in the future from surveys for high redshift quasars and galaxies .
the prominent emission lines of quasars should make them easier to detect and confirm spectroscopically than the @xmath0 galaxies discovered so far . |
over the past fifteen years , the problem of pattern formation and dynamics in non - equilibrium , spatially extended systems has attracted much attention .
although the most thorough and precise work has been on rayleigh - bnard convection ( rbc ) and taylor - vortex flow , another hydrodynamic system , the faraday instability @xcite , shows great promise . here
, surface waves are excited on a fluid by vertically vibrating its container .
the instability is due to parametric resonance : in the reference frame moving with the container , gravity is modulated periodically at a frequency @xmath2 , leading to surface waves vibrating at @xmath3 . among the allures of the faraday instability
are its short time scales , large aspect ratios , and small number of material parameters .
large aspect ratios are important , for distant boundaries can still strongly influence pattern formation @xcite .
a figure of merit is the aspect ratio @xmath4 , where @xmath5 is the system size and @xmath6 the spatial correlation length . in many experiments , @xmath7 , the pattern wavelength .
aspect ratios of order 100 have been achieved recently in rbc @xcite and in chemical reactions @xcite .
a number of experiments on the faraday instability using low - viscosity fluids have been conducted in containers with geometrical aspect ratios @xmath8 @xcite . however , as pointed out by edwards and fauve @xcite , the damping length ( @xmath9 ) of waves in low - viscosity fluids such as water and butanol can be much larger than @xmath10 ( one can have @xmath11 while @xmath12 ) , implying that patterns in small-@xmath13 experiments should be sensitive to details of the boundary conditions well above threshold , even when @xmath14 is large . indeed , they are @xcite . in this letter , we report observations of secondary instabilities for varying viscosities .
we focus on high - viscosity fluids where @xmath15 , in a box of moderate aspect ratio ( @xmath13 ranges up to 12 , depending on @xmath16 ) . at high viscosities ,
the primary instability gives a stripe pattern , in contrast with the squares observed for low viscosities @xcite .
the sequence of secondary instabilities of stripe patterns is quite different from that observed for squares . at moderate viscosities , the competition between square and stripe states makes the situation more complex still
. in low - viscosity fluids , squares destabilize via low - frequency modulations that ezersky _ et al . _ interpreted as transverse amplitude modulations ( tam ) of one component of the two sets of standing waves making up a square pattern @xcite .
amplitude equations for standing waves do indeed predict a secondary tam instability @xcite .
a similar analysis was assumed to hold for square patterns , but douady @xcite showed by stroboscopic methods that the supposed tam instability was actually a set of kink defects separating two sets of waves 180@xmath17 out of phase , an observation that we have confirmed . here , we show that when the primary instability is a stripe state , the secondary instability is an oscillatory tam instability .
a stationary tam instability of very small amplitude is , however , seen in moderate - viscosity fluids .
the above concerns control - parameter ramps , where the acceleration of the fluid container is slowly increased from zero .
quite different phenomena , such as converging target patterns and spirals , are observed when the control parameter suddenly jumps from zero to a large value @xcite .
our apparatus has been described at length in a previous study of acceleration thresholds and onset wavelengths @xcite .
that study showed that under the proper conditions , we could predict quantitatively thresholds and onset wavelengths using a theory that assumed infinite lateral extent , infinite depth , but finite viscosity @xcite .
we used a square sample container of side @xmath18 cm and depth 1.0 cm .
the fluid temperature could be controlled between 0 and 60 @xmath17c @xcite , with a stability of @xmath19c .
temperature variation due to vertical gradients was 0.5 @xmath17c in the stationary fluid but was less than 0.05 @xmath17c once surface waves appeared .
our control parameters were the driving frequency @xmath16 and the acceleration @xmath1 , which was measured directly by an attached accelerometer .
the acceleration could be varied from 0 to 17 g s and was regulated to 1% . a spatial homogeneity of @xmath20 1% variation was achieved by careful leveling and dynamic balancing .
soft " boundary conditions were created by using sloping sidewalls , which greatly reduced wave reflections from the boundaries .
we used three different working fluids : a vacuum - pump oil made of paraffins ( fisher scientific , brand 19 ) , propylene glycol , and glycerine .
the properties of the pump oil were stable over days and allowed repeatable threshold measurements .
propylene glycol and glycerine could be doped with paint pigment , as described below , and allowed us to scan a wide range of viscosities ( 0.2 - 5 @xmath21/sec ) .
polar fluids , however , have much less stable material properties : their surface tension decreases upon contamination by small amounts of impurities , and their viscosity decreases if small amounts of water are absorbed .
the sensitivity of the surface tension to impurities implied that different runs for glycerine and propylene glycol samples gave values of the surface tension that differed by as much as 30% . in data quoted below , the material parameters were determined by fits to acceleration thresholds and onset wavelengths .
these values were stable to about 5% over the course of a day .
again , the oil samples did not display such variability .
all phenomena observed in the polar fluids were also present in the oil .
we used three methods for visualizing the fluid s surface . because the waves are very steep ( surface inclinations can exceed 90@xmath17 ; see fig . [
fig : morphology]f ) , methods based upon specular reflection , such as the shadowgraph technique , can not give much information about the detailed structure of the waves .
( 1 ) we doped the fluid with paint - pigment particles in order to turn the fluid surface into a diffuse light source @xcite .
we added 2% ( by weight ) polymer - stabilized titanium dioxide particles to the glycerine and propylene glycol ( the particles could not be suspended in the pump oil ) .
we then illuminated the fluid surface using a stroboscope whose light was directed from the side of the container at an angle that minimized specular reflection into the ccd camera .
( see fig .
[ fig : morphology ] . )
( 2 ) we projected a known shape ( straight edge ) onto the pigment - doped fluid surface . from its apparent form , we could infer the surface profile .
( 3 ) we suspended a ring of leds above the container @xcite .
the leds were strobed with millisecond pulses and gave the positions of wave peaks and troughs .
this method showed that all of the phenomena observed in glycerine and propylene glycol ( with or without the paint particles ) were also present in the oil .
figs . [ fig : morphology ] and [ fig : domains ] summarize the patterns seen in high - viscosity fluids when the acceleration is increased slowly from zero . below a critical acceleration @xmath22 , the surface remained flat . just above @xmath22 , for @xmath23 , we observed complicated , slowly shifting patterns ( fig .
[ fig : morphology]a ) . because the bifurcation away from a flat interface is ( nearly ) supercritical @xcite ,
the correlation length is large when @xmath24 .
( one expects @xmath25 , where @xmath26 is the correlation length far from the transition . ) in fig . [
fig : morphology]a , @xmath27 .
( @xmath28 . )
the effects of spatial gradients in @xmath29 are important in this region , further complicating the observed patterns . at higher @xmath1
, the surface waves spontaneously order into standing waves of stripes when @xmath30 .
( the @xmath13-value where stripes order depends on the boundary conditions . )
[ fig : morphology]b ) . in different runs ,
the stripes were parallel to one or the other of the sides of the square container . above a second acceleration threshold , @xmath31 ( the value of @xmath29 depends greatly on @xmath16 )
, we observed a supercritical bifurcation to an oscillatory tam instability of the stripe state .
[ fig : morphology]c ) . in one side of the cell ,
left - going traveling tam waves dominate ; in the other side , right - going tam waves dominate . in the center ,
the modulations formed a standing wave .
the frequency @xmath32 of the standing wave was incommensurate with the driving frequency @xmath33 .
( see fig .
[ fig : tam ] . )
the tam frequency depends approximately linearly on @xmath33 , but systematic deviations are clearly seen . the solid line in fig .
[ fig : tam ] shows a quadratic fit , which needs theoretical justification .
the wavelength of the tam was ( 3.30 @xmath34 ) @xmath10 over the range of frequencies studied ; the ratio @xmath35 varied from 5.15 to 5.39 ( _ i.e. , _
@xmath36 to @xmath37 ) . at a third acceleration threshold , @xmath38
, we observed the spontaneous formation of defects .
the line in fig .
[ fig : domains ] represents the acceleration where isolated single defects edge dislocations first appeared spontaneously .
[ fig : morphology]e . )
they moved to the cell edge by a combination of climb and glide . above @xmath38 ,
the number of defects increased rapidly , resulting in very complicated dynamics .
[ fig : morphology]f . )
we have not analyzed this regime , but it might be useful to connect it with studies of defect - mediated turbulence @xcite . in the defect regime ,
droplets of fluid were spontaneously ejected upwards .
droplet emission was correlated with pattern defects , and their frequency rapidly increased with @xmath29 . between @xmath31 and @xmath38 ,
there is clear evidence for another threshold where the tam modulations dephase ( fig .
[ fig : morphology]d . )
however , the dynamics of this regime are complicated , and our cell was too small to study this tertiary instability quantitatively .
we also investigated the effect of changing viscosity by changing the temperature of propylene glycol and glycerine samples . at 50 hz and @xmath39 @xmath40/sec@xmath41
, for example , we found that in propylene glycol , squares are stable for @xmath42 @xmath21/sec while stripes are stable for @xmath43 @xmath21/sec .
the crossover region between squares and stripes is complex and deserves careful study in its own right .
for example , in the same fluid sample , at @xmath0 = 0.63 @xmath21/sec , we destabilize first to a square state and then , at @xmath44 , to a time - dependent , probably chaotic state . at @xmath45 ,
this state gives way to a stripe state with stationary tam modulations . at @xmath46 , the oscillatory tam sets in .
the wavelength of the stationary tam is 1.4 times the stripe wavelength @xmath10 , while that of the oscillatory tam is , as mentioned above , 3.3@xmath10 . also , by creating a stripe state at high viscosity and then slowly decreasing the temperature of the fluid , it is possible to preserve the state even at temperatures where ramping from below threshold leads unambiguously to squares .
( the reverse is true , as well . )
the crossover region is also distinguished by low - frequency dynamics for @xmath47 , driven presumably by competition between the two primary states . in this regime , we confirmed that stripe states are born with a stationary tam and then further destabilize via oscillatory tams at higher @xmath29 , while square states show 180@xmath17 phase defects .
a promising feature of the intermediate - viscosity regime is the possibility , in a larger - aspect cell , of finding a spatiotemporally chaotic state at the onset of the primary instability @xcite .
secondary instabilities in viscous fluids that have stripe patterns differ markedly from those observed in the low - viscosity , square - pattern case . the tam instability predicted for the latter is in fact only observed for the former . while theory predicts a stationary tam , the instability is oscillatory with a frequency that is generically incommensurate with the driving frequency . in the crossover regime , both stationary and oscillatory tams are seen .
it would be interesting to explore the case where @xmath32 is commensurate with @xmath48 .
the closest integer ratio , @xmath49 , requires higher combinations of @xmath16 and @xmath1 than were possible with our shaker table .
an obvious theoretical challenge is to see whether the four coupled amplitude equations describing standing waves can account for the oscillatory tam instability .
the coefficients in the amplitude equations were derived assuming small damping and small surface slopes @xcite .
neither of these conditions is met for high - viscosity fluids @xcite , although the approximations are more reasonable for fluids of moderate viscosity .
this suggests difficulties for perturbation expansions based on a flat interface and small damping , but numerical studies remain feasible , as do studies based on models analogous to the swift - hohenberg equation @xcite . in the intermediate - damping regime , a small number of amplitude equations may adequately describe what appear to be spatiotemporally chaotic states .
such a situation has recently been discovered in a rotationally anisotropic system @xcite , but no isotropic system has heretofore been found .
all of these phenomena tertiary phase instabilities , locking between primary and secondary instabilities , square - stripe competition , the crossover regime will be much easier to explore if one can increase the aspect ratio .
since much larger shaker tables ( albeit expensive ) are available commercially , there is no technical limitation to increasing the aspect ratio at least tenfold .
we thank a. rawicz for the loan of the vibration exciter and w. gonzlez - vias for help with many aspects of the experiment . this work was supported by nserc ( canada ) .
j. b. is an alfred p. sloan foundation fellow .
l. d. , v. e. and s. m. acknowledge support from the ens ( paris ) . a more familiar example of an instability with @xmath50 well above threshold is the laser : even though the cavity may be millions of wavelengths long , it determines the output modes . for low - viscosity surface waves , the sensitivity to boundary conditions for is clearly seen in time averages of images taken in nominally chaotic regimes .
kumar k. and tuckerman l. s. , _ j. fluid mech . _
* 279 * 49 ( 1994 )
; beyer j. and friedrich r. , _ phys . rev . e _ * 51 * , 1162 ( 1995 ) .
our calculations also took into account finite depth , but its effects were negligible . for @xmath53 @xmath21/sec , @xmath54 @xmath55 , @xmath56 hz
, we have @xmath57 .
perturbation expansions assume @xmath58 .
physically , @xmath59 is the ratio of the period of the waves to their decay time ; it is also the inverse of the waves quality factor . using our second visualization method
, we find that at @xmath31 , the stripe profile @xmath60 $ ] , implying a maximum surface slope of 80@xmath17 . | secondary instabilities of faraday waves show three regimes : ( 1 ) as seen previously , low - viscosity ( @xmath0 ) fluids destabilize first into squares . at higher driving accelerations @xmath1 ,
squares show low - frequency modulations corresponding to the motion of phase defects , while theory predicts a stationary transverse amplitude modulation ( tam ) .
( 2 ) high-@xmath0 fluids destabilize first to stripes .
stripes then show an _ oscillatory _ tam whose frequency is incommensurate with the driving frequency . at higher @xmath1 ,
the tam undergoes a phase instability . at still higher @xmath1 , edge dislocations form and fluid droplets
are ejected .
( 3 ) intermediate-@xmath0 fluids show a complex coexistence of squares and stripes , as well as stationary _ and _ oscillatory tam instabilities of the stripes . |
since @xcite suggested that a galaxy might lose its interstellar medium ( ism ) by interacting with the intracluster medium ( icm ) in the cluster environment , much evidence for ram pressure stripping has been found to date . in early days , @xcite have shown that virgo galaxies are generally more deficient in hi compared to field galaxies based on their single - dish observations .
in addition , @xcite have found that the hi content is well correlated with the location of galaxies in a sense that hi is more deficient at smaller distances from the cluster centre .
then the hi synthesis imaging studies such as @xcite and @xcite have shown that hi - deficient galaxies near the cluster centre have small hi - to - optical extents , truncated within the stellar disc in many cases . in a more recent high - resolution hi imaging study of @xmath150 selected virgo galaxies by @xcite , a number of galaxies have been found with various scales of extraplanar hi gas or long hi tails , indicating that ram pressure stripping is indeed acting in the cluster environment . as a galaxy loses its ism ,
the star formation rate is expected to be suppressed , which agrees well with the observations . as a good example , @xcite find that the massive star formation rate of virgo spirals is lower than their field counterpart by a factor of 2.5 on average .
@xcite also show that many virgo spirals have a smaller h@xmath0 extent compared to their stellar disc , reflecting that the process truncating hi discs may also be acting on the star - forming disc and h@xmath0 emission .
on the other hand , molecular gas is unlikely to be as easily stripped as atomic hydrogen , since it is more tightly bound to the galactic centre and the density is higher .
in fact , most previous studies find that the molecular gas mass of the cluster population is not significantly different from that of field galaxies @xcite .
in addition , more recent studies are finding clumpy dust features in the upstream side of hi gas stripped galaxies , which are likely to be surviving dense clouds that are unveiled after diffuse atomic gas is removed ( @xcite ; kenney , abramson & bravo - alfaro 2015 ) .
on the other hand , however , the opposite results also have been reported .
@xcite find that h@xmath2 mass normalized by the dynamical mass of galaxies tends to get larger with increasing clustercentric distance , which supports that molecular gas can be also deficient in the cluster environment .
more recently , @xcite show that hi - deficient galaxies in the cluster environment tend to be also modestly deficient in molecular gas
. however , the molecular gas fraction to the optical luminosity of spiral galaxies measured using 12co ( 1@xmath30 ) ranges quite widely for all environments @xcite .
this implies that the molecular gas fraction alone may not be a good tracer of molecular gas deficiency . due to the scatter in this relation and opposing observational results
, it is still arguable whether molecular gas can be affected by the icm pressure in a similar way to hi gas , and it is then still puzzling why star formation appears to be quenched in hi stripped galaxies if molecular gas , which is the more direct ingredient for star formation , is not deficient . of the sample in the _ b _ band .
ngc 4569 that we have not observed using the sma but is included in our discussion as a good representative of a galaxy at post - peak pressure , is shown in green .
the sample galaxies , located at 0.41 mpc from m87 , adopting a virgo distance of 16 mpc @xcite , make a nice sequence of ram pressure stripping from early stage , close to peak pressure and post - peak pressure in their orbit . ]
no clear evidence for molecular gas stripping yet low star formation activities in cluster galaxies may imply distinct molecular gas properties in high - density environments as supported by some previous observations .
for example , @xcite have found a highly asymmetric co morphology in an hi deficient virgo cluster galaxy .
@xcite also have found very peculiar co distributions in some of virgo spirals undergoing active hi stripping .
these studies indicate that molecular gas can be potentially disturbed by strong icm pressure whether it is stripped or not . particularly along the side experiencing icm pressure , it has been suggested that interstellar gas including the molecular phase can be pushed up against the centre of a galaxy .
this leads to ism compression and increasing h@xmath2 formation ; hence , it locally triggers intensive star formation @xcite .
indeed , ism compression with high molecular fraction in the upstream side is seen in a number of galaxies experiencing icm pressure ( @xcite ; nehlig , vollmer & braine 2016 ) , which are often accompanied by an enhancement of star formation .
molecular gas enhancement among the virgo cluster members has been also recently reported by @xcite .
these observations clearly show that molecular gas properties and hence star formation activity within a stellar disc can be affected by ram pressure . therefore , in order to get a deeper understanding on how galaxies become passive after hi stripping , not only the possibility of stripping but also the detailed properties of molecular gas inside a galaxy under strong icm pressure need to be probed .
hence , using the submillimeter array ( sma ) , we have taken high - resolution 12co ( 2@xmath31 ) imaging data of a subsample of virgo galaxies that have lost atomic gas significantly by icm pressure .
some preliminary results are published in @xcite , and in this work , we present more complete analysis on both morphology and kinematics of three virgo galaxies based on the sma data . this paper is organized as follows .
we introduce the sample in section [ sec2 ] .
details of observations and data reduction procedure are provided in section [ sec3 ] . in section [ sec4 ]
, we present the sma data , describing the co morphology and kinematics . in section [ sec5 ]
, we compare the co and other wavelength data to discuss how the molecular gas in these galaxies has been affected by the icm pressure . in section [ sec6 ] , we summarize the results and conclude . a distance of 16 mpc ( 1 arcsec @xmath178 pc ) to virgo cluster is adopted in this work ( yasuda , fukugita & okamura 1997 ) . [
cols="<,^,^,^,^",options="header " , ] [ tab3 ] _ note .
_ @xmath4the co - to - h@xmath5 conversion factor of 3.2 @xmath6 pc@xmath7 ( k km s@xmath8)@xmath8 is adopted from @xcite .
+ in this section , we present the results of our sma observations .
the 12co ( 2@xmath31 ) flux and its uncertainty are measured in jy km s@xmath8 as follows , @xmath9 where @xmath10 is the total flux of co emission in each channel , @xmath11 is the channel separation of the final cube ( 5 km s@xmath8 ) , and @xmath12 is the rms of each channel outside co emission .
continuous features above @xmath13 are considered as real signal in individual channels .
co linewidths are calculated using the velocities , where the flux density corresponds to 20 per cent and 50 per cent of the peaks in the receding and the approaching side , adopting the definition of @xcite : @xmath14 @xmath15 the co velocity is determined following @xcite : @xmath16 following the recipe from @xcite , 12co ( 2@xmath31 ) line luminosity is measured as follows : @xmath17 in @xmath18 , where @xmath19 is an integrated total 12co ( 2@xmath31 ) flux in jy km s@xmath8 , @xmath20 is the luminosity distance in mpc , @xmath21 is the observation frequency in ghz , and _ z _ is the redshift . adopting a typical 12co ( 2@xmath31)/(1@xmath30 ) ratio of nearby galaxies ( @[email protected] ; @xcite ) and using the conversion relation of normal spirals ( @xmath24 = 3.2 @xmath6 pc@xmath7 ( k km s@xmath8)@xmath8 ; @xcite )
, the molecular gas mass can be calculated by @xmath25 where @xmath26(2@xmath31 ) is the luminosity of 12co ( 2@xmath31 ) transition .
the global co properties of the sample are summarized in table [ tab3 ] .
on the left of figs [ n4330][n4522 ] , the co intensity map , velocity field and position - velocity diagram ( pvd ) with global and radial profile are presented in ( a ) , ( b ) and ( c ) for each galaxy . the overlays with other wavelength data shown on the right - hand side of figs [ n4330][n4522 ] are discussed in section [ sec5 ] . as seen in figs [ n4330](a)(c ) , the co morphology of ngc 4330 is highly asymmetric , with the south - west extent in the downstream side being about 75 per cent ( 40 arcsec versus 53 arcsec ) of the northeast but only 68 per cent when the outermost 4 arcsec of the south - west disc , which is bent down to the south ( indicated by arrow in fig . [
n4330]a ) is excluded ( 36 arcsec versus 53 arcsec ) .
the central part , within 30 arcsec in diameter ( @xmath12 kpc at the distance of virgo ) , is also asymmetric due to the strong local peak in the north - east from the stellar disc centre .
the distinct extents and surface densities are also clearly seen in radial profiles . in the end of the southwest disc of the downstream side , co is found to be slightly bent ( fig . [
n4330]a ) .
this bending is also found in many other wavelengths such as uv , h@xmath0 and hi , but all are different in scale and angle from one another as further discussed in section [ sec5 ] . in the case of co , this bending part is clumpy , almost identified as an independent blob or clump ( see the arrow in fig .
[ n4330]c ) .
this co clump corresponds to 2 per cent of the total in flux ( @xmath11.9@xmath2710@xmath28 @xmath6 , comparable to that of a large molecular cloud ) .
the centre of the clump is off from the mid - plane of the main disc toward the south by @xmath14 arcsec ( @xmath1312 pc ) .
the co kinematics also shows peculiar structures , especially along the south - west concentration and the end of the tail as shown in fig .
[ n4330](b ) .
the velocity gradient of the south - west is quite steep , while it is more slowly rising on the other side within the small radii from the centre , reaching 72 km s@xmath8 on the approaching side but only 43 km s@xmath8 on the receding at 15 arcsec radii . in the south - west clump , the velocity gradient is inverted .
the kinematical complexity is also clearly seen in the pvd ( fig .
[ n4330]c ) . a distinct component of a steep velocity gradient in the central @xmath113 arcsec ( @xmath11 kpc ) is quite noticeable .
this may indicate a molecular ring or bar .
in addition , the co peak is off from the stellar disc centre by @xmath1270 pc ( fig .
[ n4330]a ) , while the co kinematic centre is more or less consistent with the stellar disc centre ( fig .
[ n4330]c ) . as shown in fig .
[ n4402](a ) , the co disc of ngc 4402 is slightly more extended in the west ( 62 arcsec in the west versus 58 arcsec in the east ) , but the difference is subtle and not as significant as in hi .
what makes this case look highly asymmetric is the north - west quarter of the downstream side ( the west side of the disc ) . in this region ,
a modest co bump ( indicated by arrow in fig .
[ n4402]a ) is found as in hi ( fig .
[ n4402]d ) .
meanwhile , the southern part of the co disc looks quite compressed along the enhanced far - ultraviolet ( fuv ) emission ( fig . [ n4402]e ) , which will be discussed more in detail in section [ sec5 ] .
the northern part above the major axis is measured to be 91 per cent of that of the south in flux .
although the co peak agrees well with the stellar disc centre , the inner co disc also appears to be slightly more stretched toward the west , reflecting the outer co disc .
the co kinematics reveals the evidence for non - circular motions as shown in fig .
[ n4402](b ) . in the inner region ,
the iso - velocity curves are highly skewed and not perfectly parallel with the minor axis .
the skewness in velocity must be the result of two distinct disc components in the inner @xmath115 arcsec ( @xmath11 kpc ) being projected on the sky as seen in its pvd ( fig .
[ n4402]c ) .
the velocity gradient in the outer part is quite different in the two sides of the disc along the major axis .
in the north - west quarter ( in the downstream side ) , the kinematics of modest co bump is slightly deviated from the main disc .
it also has a distinct co component with steep velocity gradient in the inner @xmath115 arcsec .
this likely indicates a nuclear bar or ring , although we favor a nuclear bar as this would also explain the skewed isovelocity contours . in ngc 4402
, @xcite also find a nuclear molecular disc traced by 12co ( 1@xmath30 ) in the inner @xmath110 arcsec .
we do not find any direct relevance of these nuclear structures with ram pressure stripping . while the overall size of 12co ( 2@xmath31 ) disc in the sma image is consistent with the extent measured by a single dish for ngc 4330 @xcite and ngc 4402 ( lee et al . in preparation ) , our sma data of ngc 4522 reveal only the inner @xmath157 per cent of its single - dish map in size @xcite .
while @xcite have detected co along the extraplanar h@xmath0 and hi in both ends of the disc , we did not detect any such features in the south - west in our sma data due to the lack of sensitivity and/or potentially due to the diffuse nature of gas in the outer molecular disc .
the north - east end was not covered in the sma observations .
therefore , we limit our discussion to only the inner co disc for this particular case in this section and we cite the iram single - dish data when it is needed for comparisons with other wavelength data in section [ sec5 ] . as shown in fig . [ n4522 ] ,
the co extent is comparable in both sides ( 27.4 arcsec in the north - east versus 27.6 arcsec in the south - west ) .
the inner co disc of ngc 4522 is found to be curved in the opposite way to the outer co and hi disc , i.e. to the south - east as the inner h@xmath0 disc .
we find two features sticking out from the inner part , one from the end of north - east disc and another almost to the same direction but connected from the centre of the main disc , more to the south .
the north - east blob coincides with the morphology of the inner co disc in the iram image ( fig .
[ multi]c ; @xcite ) , which is smoothly connected to the disc farther extended in the north - east , coinciding with a dust loop in this region . while the downward bending coincides with the inner spiral structure including the dust feature ( see fig . [ hst ] in appendix a ) .
the co velocity structure within the main disc generally shows that of a regularly rotating disc as seen in figs [ n4522](b ) and ( c ) .
the velocity keeps rising up to @xmath110 arcsec on both sides , then almost flattens out , and toward the end of the disc , the velocity rises again .
the velocity gradient along the north - east branch is overall consistent with that of the main disc on the same side .
however , the gradient is smaller and the velocity rises more slowly compared to the other side .
in this section , we compare 12co ( 2@xmath31 ) properties with other wavelength data including hi @xcite , optical ( dss2 red ) , fuv @xcite , and h@xmath0 ( ngc 4330 and ngc 4569 from @xcite , ngc 4402 from @xcite , ngc 4522 from koopmann , kenney , & young 2001 ) .
co traces star - forming gas , and h@xmath0 and fuv emission are good indicators of star formation with time - scales of @xmath120 and 100 myr , respectively @xcite .
we also compare the radial gas surface density between atomic and molecular hydrogen .
the hi column density is calculated using , @xmath29 in cm@xmath7 , where @xmath31 is the integrated intensity ( brightness temperature ) in @xmath32 @xcite .
the h@xmath2 surface density is estimated by the following relation : @xmath33 in @xmath6 pc@xmath7 , where @xmath34 is 12co ( 2@xmath31 ) integrated intensity , and @xmath35 is the same as in equation ( 6 ) .
considering the high inclination of all three galaxies in our sample , we utilize the strip integral method instead of ellipse fitting to derive radial surface density because ellipse fitting is usually inappropriate for edge - on galaxies when the spatial resolution is limited @xcite .
this method calculates _ strip integrals _ by integrating hi column density perpendicularly to the major axis , using lucy s ( 1974 ) iterative deconvolution method .
then the strip integrals of approaching and receding sides are deprojected separately to infer the face - on hi surface density @xcite .
the surface density of both atomic and molecular hydrogen is measured in @xmath6 pc@xmath7 .
the comparison between atomic and molecular gas radial distributions is found in the bottom on the left - hand column of figs [ n4330][n4522 ] .
overlays between co and other multiwavelength data are presented on the right - hand column of figs [ n4330][n4522 ] and fig .
[ multi ] . for the reference , the iram 12co ( 2@xmath31 ) maps
are also overlaid in the composite map .
the sensitivity of the iram cubes is @xmath110 mk per 5 km s@xmath8 channel ( the same channel width as ours ) , which is comparable to the sma cubes or slightly better in the cases of ngc 4330 and ngc 4522 when the different spatial resolutions are taken into account ( @xmath111 arcsec versus @xmath15 arcsec for iram and sma , respectively ) . for ngc 4402 , the sensitivity of sma data is somewhat better than the iram data not only because of the sufficient integration time with the sma but also because of the poor weather condition when the iram data were taken . for ngc 4330 and ngc 4402 ,
the iram and sma maps are generally in good agreement in morphology , considering the beam sizes , while we are missing most of the extraplanar molecular of ngc 4522 in the sma data due to the lack in coverage , sensitivity , and/or potentially due to the diffuse nature of molecular gas in the extragalactic space as further discussed in section 5.1.3 .
asymmetry is ubiquitous in a range of wavelengths data except in the red optical image ( dss2 red ) . in fig .
[ multi](a ) , we see that the hi is truncated within the stellar disc in the north - east , where h@xmath0 and fuv emission reveals an upturn feature @xcite . a hint of upturn is also found in the north - east end of the upstream of radio continuum at 6 cm and 20 cm @xcite .
the single - dish 12co ( 2@xmath31 ) data @xcite appear to be similar to the radio morphology , also with some hint of upturn , although the extent of the molecular gas is much less than that of hi . in our sma data , we are missing some outermost features including the tip of the nose in the north - east . instead
, the sma data clearly reveal very detailed structures of the inner co disc such as the difference in scaleheight between the north and the south along the major axis ( up to @xmath10.36 kpc , fig .
[ n4330]a ) which is not as clear as this in the previous single - dish data . the south - west side , i.e. the downstream of the galaxy where an hi gas tail is present , is bent toward the south .
on this side , we also find fuv , h@xmath0 and radio continuum tail but the extents and bending angles are all different in various wavelengths ( fig [ multi]a ) .
the distribution becomes patchy in the outer co disc as seen in both iram single - dish data @xcite as well as our sma data .
intriguingly , the co blob toward the end is located in almost the same galactic radius as one of the distinct h@xmath0 blobs and off from the mid - plane in the same direction as the tails of the other wavelengths ( fig [ multi]a ) .
as clearly seen , the molecular disc , which extends to only half the stellar disc or less , shows many properties in common with the morphologies in the other wavelengths , reflecting the impact of the icm pressure .
although there are subtle differences , this suggests that the molecular gas in this galaxy has been also affected in similar ways by the same mechanism that is responsible for the peculiarities found in the other wavelengths .
this indicates that icm pressure can change molecular gas properties well inside the stellar disc .
the co emission shows a number of properties in common with the hi gas ( fig .
[ n4402]d ) .
it is somewhat extended in the west , i.e. the same side where the hi is more extended .
both phases show a bump in the west ( indicated by arrow in fig .
[ n4402]a ) , although the scale is distinct because the hi bump is visible in extraplanar gas .
hi is truncated within the mid - plane of the stellar disc on both sides , with a sharp cut - off in the east to south along the upstream . meanwhile ,
the northern side is more extended with a short tail pointing the north - west in the downstream side .
the extent of the molecular gas is not significantly different in the receding and the approaching side , yet its distribution is quite distinct
. these differences between the two sides of the co disc can be also clearly seen in the radial velocity distributions ( fig .
[ n4402]c ) .
note that the hi extent appears to be almost twice as long as the co in the west due to the tail ( figs [ n4402]c and d ) .
the molecular gas disc is thicker in the east ( especially third contour level from the outermost contour ) , reflecting the change in fuv disc thickness across the disc .
although the overall morphology of fuv is quite different from that of co , both the fuv enhancement and co compression are present in the east and south - east ( fig .
[ n4402]e ) .
h@xmath0 is tightly correlated with co in general , with a similar distribution and extent .
also , strong h@xmath0 knots are found near or within the local co peaks ( fig [ n4402]f ) . as with ngc 4330
, the co morphology is surprisingly similar to that of hi , strongly suggesting that the impact of the icm has reached the inner interstellar gas disc .
with given sensitivity limit , we are missing the outer part of the molecular gas disc in the sma data for this case .
as shown by @xcite , however , the overall morphology of co is quite similar to that of hi , revealing a significant fraction of molecular gas outside the stellar disc ( fig [ multi]c )
. this galaxy must be one of the best examples found to date where extraplanar molecular gas is present .
in addition to the sensitivity , the molecular gas could be more diffuse in nature compared to the gas within the disc , which could make the detection more difficult .
in fact , we recently have detected some extragalactic 13co ( 1@xmath30 ) in ngc 4522 using the atacama large millimeter array ( alma ) - cycle 3 .
the nature of this extraplanar co gas will be further analyzed using 12/13co ratio , and will be presented in a separate work .
the inner co disc is slightly bent down to the south as h@xmath0 emission .
the strongest h@xmath0 blobs near the galactic centre are located within the inner co disc ( fig [ n4522]f ) , but there is an offset between the inner fuv enhancement and the inner co disc ( fig [ n4522]e ) .
however , this offset in the inner region might be simply due to the dust extinction effect .
on the other hand , the offset between co ( the iram data)/h@xmath0 and fuv at large radii is real since there is almost no dust beyond the co extent .
the extent of fuv is larger than the co and h@xmath0 extents within the stellar disc by a factor of 1.5 ( fig .
[ multi]c ) . however , the fuv enhancement is roughly consistent with the co compression in the downstream side as in ngc 4402 .
this implies that there is significant change in the recent star formation activity and the distribution of molecular gas in a very short time as further discussed in section [ sec5].3 .
most previous studies using single - dish radio telescopes did not find a significant difference in co luminosities between field galaxies and cluster members @xcite .
however , some individual galaxies in the cluster environment have been reported to show very distinct co distributions from those with normal hi content in the field ( e.g. @xcite ) .
more recently , @xcite have shown that the molecular gas surface density of galaxies that are hi - deficient tends to be low .
these suggest that the molecular gas of cluster galaxies can be affected in some ways by ram pressure even if it is not stripped .
indeed , our high - resolution co data clearly show that molecular gas has been affected in a similar manner as diffuse gas by ram pressure , and the impact of icm winds reaches as deeply as a few kpc from the galactic centre , as we further elaborate below .
firstly , the co morphology in our sample is all found to be highly asymmetric and disturbed .
the inner molecular gas disc of many field galaxies does show some peculiarities to a certain degree , including asymmetry ( e.g. bima song survey ; @xcite ) .
however , the overall asymmetry found in the co morphology of all three galaxies is similar to that of the hi morphology . in figs [ multi](a)(c )
, the composite maps of cool ism contents ( co and hi ) , star formation indicators ( fuv and h@xmath0 ) and the old stellar disc ( dss2 red ) of our sma sample are presented . a rough direction of the icm wind based on the hi morphology for each case
is shown with thick arrows .
secondly , the detailed co structure of the sample shows good connections with the hi features which have been pushed and/or stripped by the icm pressure .
in ngc 4330 , the co clump in the south - west end seems to have been pushed down to the south from the mid - plane of the stellar disc , similarly to its hi tail ( fig .
[ n4330]d ) .
in addition , similar kinematical structures are found in this side , i.e. in the downstream in both hi and co ( fig . [ pvd ] ) .
for example , the co gas reveals an inverted velocity gradient in the end of south - west like hi which shows an anomalous velocity component in the same side @xcite .
this implies that dense molecular phase gas can be pushed by ram pressure in addition to the diffuse atomic phase gas .
although the morphology is suggestive of gas being pushed away , one problem with this possibility is that both the hi tail and the co clump in the downstream are found with higher velocities than the main disc , i.e. the opposite direction of the system velocity of cluster centre .
this is inconsistent with what is generally expected in the case of the gas pushed by the icm as found in other virgo spirals ( e.g. ngc 4522 ; kenney , van gorkom & vollmer 2004 ) .
however , this is still feasible if the gas has been losing angular momentum in extraplanar space as suggested by @xcite .
alternatively , the co clump might be an example of a dense cloud unveiled after the stripping of surrounding diffuse gas , which hence appears to stand alone , decoupled from the rest of the disc , similar to the ones seen as narrow dusty plumes found in virgo galaxies such as ngc 4402 and ngc 4522 @xcite or ngc 4921 , a coma spiral @xcite .
if this is an example of those decoupled clouds , it is the first time it has been seen clearly in co. however , one aspect that makes this case particularly interesting is its gas kinematics . rather than being smoothly connected from the inner disc ,
the velocity gradient at the location of the co clump turns back .
this velocity structure strongly suggests that this clump is the molecular gas that has been pushed off from the main disc . in ngc 4402 , both hi and co are compressed from the south - east to south along the upstream .
meanwhile in the north - west , a modest co bump is found to be pointing to the same direction as the hi bump ( fig .
[ n4402]d ) .
the compressed co in the upstream side reveals a steeper velocity gradient than the other side .
meanwhile , the co bump in the north - west downstream side ( fig .
[ n4402]b ) is found to have larger velocities than the main disc ( fig .
[ n4402]c ) , i.e. closer to the cluster centre .
this is consistent with our expectations if the gas is accelerated by ram pressure due to the icm . in ngc 4522 , the overall co morphology from the iram data
@xcite is very similar to the hi morphology , with the extraplanar gas in the downstream side and compression in the upstream side ( fig .
[ multi]c ) .
this strongly suggests that the molecular gas has been influenced by the same mechanism that affects the atomic gas , i.e. ram pressure due to the icm wind . in the upstream side , the interstellar gas can also get compressed by the icm pressure , which potentially results in the formation of molecules as inferred by the excess of co in this side ( see also the example of ngc 4654 ; @xcite ) . as shown in fig .
[ pvd ] , the co kinematics also suggests that the icm pressure can disturb the molecular gas , deep inside a galactic potential well . in ngc 4330 and ngc 4402
, the co velocity gradients are quite distinct in the receding and the approaching side , an asymmetry also observed in hi . in the outer part of ngc 4330 and ngc 4402 ( the co clump and the modest co bump , respectively ) where the molecular gas is off from the mid - plane , we also find the velocity gradient deviates from the overall gas flow of the gas disc .
the comparisons with the hi pvd is also intriguing ( the bottom row of fig . [ pvd ] ) . in the upstream of ngc 4330 and ngc 4402 ,
the velocity gradients of the molecular gas and atomic gas are similar to each other , but they deviate in the downstream . in ngc 4522 ,
the deviation starts at @xmath10.8 kpc from the centre on both sides .
this may indicate a delay in stripping of gas in distinct phases , i.e. more diffuse gas dragged prior to relatively dense gas .
our results clearly show that the molecular gas in galaxies experiencing active hi stripping can become highly asymmetric , as demonstrated by simulations ( e.g. @xcite ) .
in addition , our high - resolution data indicate that the influence of strong icm pressure reaches quite deep inside of the galactic disc .
although molecular gas may be pushed off from the stellar disc together with atomic gas as in ngc 4522 @xcite , active ram pressure does not seem to always cause molecular gas stripping ( ngc 4330 and ngc 4402 ) .
indeed , ngc 4569 , which is thought to have lost more than 90 per cent of its hi gas while crossing the cluster core a few hundred myr ago ( hi deficiency : 1.47 , a projected distance of 1.7 degree ; @xmath10.5 mpc to m87 , the bottom right of fig .
[ multi ] ) , has a comparable molecular gas fraction to optical luminosity of our sma sample ( ngc 4330 , ngc 4402 and ngc 4522 ) . adopting the h@xmath2 deficiency of @xcite , single - dish measured co fluxes and the mw conversion factor result in @xmath36 of 0.48 , 1.25 , 0.97 and 1.19 for ngc 4330 , ngc 4402 , ngc 4522 and ngc 4569 , respectively .
considering that the sample of @xcite does not include galaxies as faint as ngc 4330 or ngc 4522 , the expected h@xmath2 mass could have been overestimated for these two galaxies .
this supports that molecular gas mass does not necessarily change significantly during and after severe hi stripping , while its detailed molecular gas properties may change .
these changes to the co gas properties are consistent with those from @xcite , who found enhanced levels of warm molecular gas which resulted from additional ism - heating due to ram pressure driven shocks in ngc 4330 , ngc 4402 and ngc 4522 .
all three galaxies have lost hi gas significantly , containing only @xmath37 @xmath120 per cent of hi mass compared to field galaxies of similar size .
the star formation quenching time - scale , @xmath38 , measured based on the youngest stellar age at the location , where hi gas is truncated ( i.e. one way to estimate how long ago star formation was suppressed in a galaxy ) , ranges from a few tens myr to a few hundreds myr for our sample @xcite .
it is however comparable among the sample within the uncertainties , and they share several properties in common in star formation activity in spite of subtle differences due to the icm wind angle and the projection on the sky as shown in fig .
[ multi ] .
the spatial distributions of co and h@xmath0 ( a tracer of recent star formation with a timescale of @xmath3920 myr ; @xcite ) are remarkably well correlated in the inner few kpc radius ( fig .
[ multi ] ) .
the locations of local co peaks are generally coincident with the h@xmath0 knots as shown by high - resolution data .
the extents and overall morphology are also similar to each other as seen in the comparison with single - dish data , although the h@xmath0 blobs outside the old stellar disc do not always have a co counterpart ( e.g. in both ends of ngc 4330 s fuv disc ; fig .
[ multi]a ) . meanwhile , fuv ( a star formation tracer of a time - scale of @xmath1100 myr ; @xcite ) is somewhat distinct from co and h@xmath0 in extent and morphology , at least on one side , with fuv being more extended than co and h@xmath0 . although the fuv disc of many field galaxies are larger than their h@xmath0 disc ( e.g. @xcite ) , these three galaxies are distinct from the field population in a sense that both fuv and h@xmath0 are highly truncated within the stellar disc .
these observations indicate that star - forming discs have been shrunk considerably ( by a factor of up to 1.5 ) while undergoing strong ram pressure .
then the question is what causes the quenching of star formation , whether it is molecular gas stripping , the change of molecular gas properties , or both . as seen by the morphological / kinematical peculiarities of molecular gas in our sample ,
the icm pressure is strong enough to strip the hi gas in the outer disc or push the hi disc to one side , and seems to be able to also affect the molecular gas at a few kpc radii of the stellar disc .
this can cause a number of consequences .
first , in the outer disc where stars must have been formed mostly from low density interstellar gas , star formation activity will gradually decrease as the atomic gas is stripped .
even when quenching starts , some young stars from the last star formation event may be still observable , and indeed , the fuv emission outside co and h@xmath0 discs yet inside the old stellar disc could be good examples of this process .
our sma data suggest that these galaxies are close to or approaching peak pressure @xcite , which is also supported by the results of simulations in the cases of ngc 4330 and ngc 4522 @xcite .
therefore , they will experience strong(er ) ram pressure for a while , losing more hi . as long as hi
is supplied into the disc , the galaxy should be capable of forming molecular gas , keeping up with star formation .
hence , within one galaxy where star formation can be regulated , the fuv morphology should be more or less in good agreement with those of h@xmath0 , which traces more recent star formation on a time - scale of @xmath37 20 myr @xcite .
this is indeed true for isolated galaxies such as the sample of @xcite .
in addition , the fuv extent in isolated galaxies is usually smaller than the hi extent as shown in the comparison of radial profiles @xcite . in our galaxies
, however , the h@xmath0 looks much smaller than the fuv disc , and quite distinct in morphology .
this implies that star - forming disc in these galaxies has been shrunken at least in the last 100 myr . while the outer disc undergoes star formation truncation , the inner disc is also less likely to form stars in the same manner as field galaxies , despite enough molecular gas present . in the upstream side ,
molecular gas is found to be compressed as if it is pushed by the icm pressure .
this can be the result of direct molecular gas compression but also molecular gas excess due to formation from compressed diffuse gas @xcite .
consequently , the star formation rate is likely to be locally increased as @xcite suggested ( see also @xcite ) .
indeed , we find strong fuv enhancement on the same side where the outer co disc is found to be compressed . alternatively , this might be the result of the removal of dust due to the icm pressure and hence the upstream side being unveiled .
however , the fuv disc also shows morphological peculiarities and it is quite distinct from the old stellar disc ( e.g. a tail in ngc 4330 , different extents and scale heights of ngc 4402 , extraplanar fuv emission in ngc 4522 ) , which can not be explained by dust stripping alone .
in addition , more evidence for increased star formation due to gas compression by the icm pressure has been found in other galaxies ( e.g. @xcite ) , which also seems to be the case in our sample .
local fuv enhancements found near co compression within the old stellar disc ( such as the north - east of ngc 4330 , the east of ngc 4402 and the south - east of ngc 4522 ) can be good examples of induced star formation by ram pressure ( figs [ n4330]e[n4522]e and figs [ multi]a c )
. some interstellar gas removed from the main stellar disc yet lingering around in a galactic halo may be re - accreted onto the disc after core crossing .
ngc 4569 shown in fig .
[ multi](d ) , is known to be such case .
the single - arm structure in hi agrees well in morphology and kinematics with a model of stripped gas falling back to the galaxy , which assumes that the galaxy is currently moving away from the cluster centre after crossing the core @xmath1a few hundreds myr ago @xcite .
its star formation quenching time - scale estimated by @xcite is also somewhat longer than our sma sample by @xmath1100 myr or so , indicating that this galaxy has experienced strong icm pressure relatively longer ago than the other three galaxies .
what is intriguing in the case of ngc 4569 is that the extent of the fuv disc is comparable to that of the h@xmath0 and co ( fig . [
multi]d ) . in this case , the star formation is limited in the central region of the galaxy as most of the hi gas at larger galactic radii is stripped .
some gas can be re - obtained but not enough gas is left to replenish the entire disc .
therefore , the fuv extent can decrease to a similar size as the h@xmath0 extent because fuv emission may fade as time goes by , if molecular gas can not be replenished because the atomic gas has been stripped . at this stage
, it is similar to isolated galaxies in a sense that fuv is overall well correlated with h@xmath0/co , but now the star - forming disc is somewhat shrunken in comparison to what it used to be during active hi stripping or before .
the three galaxies that we observed using the sma are expected to go through a similar evolutionary sequence and will be found with a truncated fuv disc of a comparable size to the h@xmath0 and co disc within their old stellar disc .
judging from the star formation quenching timescales by @xcite and simulations by @xcite , the time it will take for our sma sample to become a galaxy like ngc 4569 is expected to be only a few hundreds myr .
lccc & ngc 4330 & ngc 4402 & ngc 4522 + & + @xmath40)@xmath4 & 2.1 & 1.4 & 3.3 + @xmath41@xmath42 & 8.2 & 63.8 & 20.4 + @xmath43@xmath44 & 12551775 & 13381892 & 11381609 + + & + @xmath45 ( kpc)@xmath46 & 0.24.2 & 0.24.9 & 0.22.1 + @xmath47@xmath48 & 1.4109 & 6.6568 & 1.5187 + @xmath49@xmath50 & 140 & 145 & 122 + [ tab4 ] @xmath51projected distance from m87 . + @xmath52icm density .
+ @xmath53range of possible orbital velocities .
+ @xmath54radius from the galactic centre ( minimum and maximum values from either side )
. + @xmath55molecular gas surface density ( minimum and maximum values from either side ) . + @xmath56rotational velocity , @xmath57 , @xmath58 : the inclination angle , @xmath59 : the hi linewidth @xcite .
+ while a galaxy is undergoing very strong ram pressure , some molecular gas might be pushed away from the stellar disc as clearly seen in ngc 4522 . in order to test whether dense molecular gas can get stripped from a galaxy or
not , we compare the ram pressure acting on our sample due to the icm wind with the restoring force ( per unit area ) by the galactic potential as a function of galactic radii ( i.e. @xmath60 versus @xmath61 ) in fig .
[ restore ] . the icm density and the orbital velocity of each galaxy
are adopted from the model of the virgo cluster @xcite .
the shaded region indicates the range of possible orbital velocities at a given projected clustercentric distance . to estimate the restoring force
at given galactic radii , we take the mean molecular gas and the hi gas surface densities along the major axis from the radial profiles ( i.e. figs [ n4330]c[n4522]c ) using our sma co data and the hi data ( viva ; @xcite ) . hence , in this calculation we do not include the iram data of ngc 4522 , and therefore the restoring force of this case in fig .
[ restore ] is only the lower limit of the lowest possible value .
the parameters used to estimate the restoring force and ram pressure are summarized in table [ tab4 ] .
our rough estimate suggests that it is not impossible to remove the molecular gas in the very outer region in some circumstances .
however , by adopting the rotational velocity of the co disc , which in all cases has a radius less than 5 kpc , we are focused on a region whose potential is dominated by the stellar component , not the dark matter halo . therefore , in practice , the actual stripping of molecular gas must be more difficult than our estimation . also , depending on the encounter angle between the icm wind and the galactic disc , the inner molecular gas can be better shielded , which will make the stripping even harder .
lastly , considering that the ism is one entity after all and the total gas surface density of the two phases is more appropriate for more realistic estimation , the chance for the molecular gas to get stripped can be even less .
in addition , even if molecular gas is pushed off from the galactic disc , whether the extraplanar molecular gas leaves the galactic halo for good or partially comes back is still debatable . considering that cluster galaxies are not significantly different from their field counterpart in co luminosity @xcite , the molecular gas completely stripped out of a galaxy is inferred to be a very small fraction .
our estimate is , however , only the prediction for how easy or hard it is to remove the molecular gas from a galaxy completely , and the molecular gas can be still disturbed without stripping , as our data clearly show .
in fact , the recent result by @xcite that severely hi - deficient galaxies tend to be found with lower molecular gas surface density may simply reflect the fact that the molecular gas of galaxies under strong icm pressure is perturbed , not necessarily implying molecular gas being deficient .
extraplanar molecular gas may also form new stars ( e.g. eso 137 - 001 with a prominent rps tail found in a massive cluster , abell 3627 ; @xcite ) .
the fuv emission with several h@xmath0 knots along the hi tail of ngc 4330 and the extraplanar fuv / h@xmath0 emission in ngc 4522 might be such a case .
outside the old stellar disc , stars may form in two ways , from stripped molecular gas , or alternatively from newly formed molecular gas . indeed , @xcite in their simulations show that stripped diffuse gas can form dense molecular gas by cooling process , which can possibly form new stars .
however , in extraplanar space , it is questionable how efficiently stripped atomic hydrogen can be pressurized , turning into molecular form @xcite .
therefore , star formation taking place outside the main disc is more likely to be utilizing the molecular gas that is pushed from the main disc but has not left the galactic halo . as in ngc 4522 , however , there are several extraplanar h@xmath0 regions with no detectable co. this suggests that the h@xmath2 gas depletion time - scale ( @xmath62 ) in extraplanar space can be much shorter than what it is on the disc ( of the order of a few gyr ; @xcite ) .
meanwhile , @xmath62 at the location of h@xmath0/co blob is measured to be comparable to what is normally found in the galactic disc ( see appendix b ) .
therefore , there might be some extraplanar co even in ngc 4330 and ngc 4402 , but with mass smaller than our detectability or that of the iram @xcite . together with this
, there is also an important difference between ngc 4330 and ngc 4522 , which is the presence of co in the extraplanar star formation region . in the case of ngc 4330 ,
more rapidly increasing ram pressure compared to ngc 4522 might be responsible for faster dissociation of molecular cloud , and hence absence of co along the fuv tail .
alternatively , some fraction of stripped molecular gas might be destroyed by high - energy photon emitted from massive young stars @xcite
. therefore extraplanar or stripped co may not be always observable even if it is pushed off from the stellar disc .
in this work , we have presented high - resolution 12co ( 2@xmath31 ) data obtained using the sma of three virgo spirals , ngc 4330 , ngc 4402 and ngc 4522 , which are undergoing active hi stripping due to ram pressure . by comparing with single - dish 12co ( 2@xmath31 ) ,
hi , fuv , h@xmath0 and dss2 red data , we have discussed how icm pressure changes molecular gas and star formation properties .
most important results from our study can be summarized as follows : 1 .
the overall co morphology and kinematics are quite asymmetric and disturbed .
morphological peculiarities found in co are closely related to hi morphology in these galaxies , indicating that molecular gas is also affected by the ram pressure .
we find the co clump in the south - west of ngc 4330 ( figs [ n4330]a c ) and the modest co bump in the north - west of ngc 4402 ( figs [ n4402]a c ) , with distinct velocity structure from the overall velocity gradient of the molecular disc .
this also supports the idea that molecular gas can be displaced by the external pressure inside a stellar disc .
3 . both fuv and h@xmath0 are enhanced where hi and co are compressed , supporting that ram pressure can also trigger star formation temporarily ( e.g. figs [ n4402]d f of ngc 4402 ) . in these galaxies
, however , fuv shows distinct morphology and extent from those of co , while h@xmath0 and co are overall in good agreement ( fig .
[ multi ] ) .
the large offset between fuv and h@xmath0 must indicate the star formation has been recently quenched , likely over the last 100 myr as the molecular gas properties have changed .
4 . to compare with ngc 4569 ( fig .
[ multi]d ) , which is thought to have already crossed the cluster core , the offset between fuv and h@xmath0/co is expected to smooth out over time ( @xmath1a few hundreds myr ) after the core crossing as the galaxy further loses hi , i.e. the capability of forming molecular gas , and star formation is completely shut down at large galactic radii . in summary ,
our results suggest that a strong icm pressure not only strips diffuse atomic gas but also changes the properties of dense molecular gas in the inner few kpc of a galaxy .
molecular gas can be pushed from the stellar disc in the outer part as reported in previous studies , yet we find that molecular gas is less likely to be completely stripped from a galaxy . on the side where the icm pressure acts , molecular gas can be enhanced by compression and molecular formation from atomic gas , locally increasing star formation .
as the gas is rapidly used up , the star - forming disc will decrease in size as no sufficient atomic gas is left to form more molecular gas . on the opposite side , molecular gas can be perturbed in a way that reflects the atomic gas , which may prevent a galaxy from forming stars as before , i.e. injecting turbulence into molecular gas is likely to cause star formation rates to be inefficient @xcite .
the combination of these two effects must be an important driver that makes galaxies evolve to become passive in the cluster environment . in order to investigate the molecular gas properties and its conditions for star formation under strong icm pressure more quantitatively
, we are currently probing the density and temperature of molecular gas using multitransitions and isotopes of co ( lee et al . in preparation ) .
our future work is expected to provide further insights on the evolution of molecular gas and thus star formation efficiency in galaxies undergoing ram pressure .
support for this work was provided by the national research foundation of korea to the center for galaxy evolution research ( no .
2010 - 0027910 ) and science fellowship of posco tj park foundation .
this work has been also supported by nrf grant no .
2015r1d1a1a01060516 . the submillimeter array is a joint project between the smithsonian astrophysical observatory and the academia sinica institute of astronomy and astrophysics and is funded by the smithsonian institution and the academia sinica .
this work made use of heracles , ` the hera co - line extragalactic survey ' @xcite .
we acknowledge the usage of the hyperleda database ( http://leda.univ-lyon1.fr ) . this research has made use of the goldmine database @xcite , and it made use of aplpy , an open - source plotting package for python hosted at http://aplpy.github.com .
st was supported by the alvin e. nashman fellowship in theoretical astrophysics .
bl is grateful to e .- j .
chung for the technical help with gipsy .
99 abramson a. , kenney j. d. p. , crowl h. h. , chung a. , van gorkom j. h. , vollmer b. , schiminovich d. , 2011 , aj , 141 , 164 abramson a. , kenney j. d. p. , 2014
, aj , 147 , 63 alatalo k. , et al . , 2015 , apj , 812 , 117 barmby p. , et al . , 2006 , apj , 650 , l45 bekki k. , 2014 , mnras , 438 , 444 bigiel f. , leroy a. , walter f. , brinks e. , de blok w. j. g. , madore b. , thornley m. d. , 2008 , aj , 136 , 2846 bigiel f. , et al . , 2011 , apj , 730 , l13 blitz l. , rosolowsky e. , 2006 , apj , 650 , 933 bhringer h. , briel u. g. , schwarz r. a. , voges w. , hartner g. , trmper j. , 1994 , nature , 368 , 828 boselli a. , cortese l. , boquien m. , boissier s. , catinella b. , gavazzi g. , lagos c. , saintonge a. , 2014 , a&a , 564 , a67 cayatte v. , van gorkom j. h. , balkowski c. , kotanyi c. , 1990 , aj , 100 , 604 chung a. , van gorkom j. h. , kenney j. d. p. , vollmer b. , 2007 , apj , 659 , l115 chung a. , van gorkom j. h. , kenney j. d. p. , crowl h. , vollmer b. , 2009 , aj , 138 , 1741 chung e. j. , 2012 , phd thesis , yonsei university chung e. j. , kim s. , 2014 , pasj , 66 , 11 crowl h. h. , kenney j. d. p. , van gorkom j. h. , vollmer b. , 2005 , aj , 130 , 65 crowl h. h. , kenney j. d. p. , 2008
, aj , 136 , 1623 davies r. d. , lewis b. m. , 1973 , mnras , 165 , 231 drimmel r. , spergel d. n. , 2001 , apj , 556 , 181 fujita y. , nagashima m. , 1999 , apj , 516 , 619 fumagalli m. , krumholz m. r. , prochaska j. x. , gavazzi g. , boselli a. , 2009 , apj , 697 , 1811 gavazzi g. , boselli a. , donati a. , franzetti p. , scodeggio m. , 2003 , a&a , 400 , 451 gil de paz a. , et al . , 2007 , apjs , 173 , 185 giovanelli r. , haynes m. p. , 1985
, apj , 292 , 404 gunn j. e. , gott j. r. , iii , 1972 , apj , 176 , 1 haynes m. p. , giovanelli r. , 1984 , aj , 89 , 758 helfer t. t. , thornley , m. d. , regan , m. w. , wong t. , sheth k. , vogel s. n. , blitz l. , bock d. c .- j . , 2003 , apjs , 145 , 259 henderson b. , bekki k. , 2016 , apj , 822 , l33 hidaka m. , sofue y. , 2002 , pasj , 54 , 33 jchym p. , combes f. , cortese l. , sun m. , kenney j. d. p. , 2014 ,
apj , 792 , 11 kenney j. d. p. , young j. s. , 1989 , apj , 344 , 171 kenney j. d. p. , young j. s. , hasegawa t. , nakai n. , 1990 , apj , 353 , 460 kenney j. d. p. , van gorkom j. h. , vollmer b. , 2004 , aj , 127 , 3361 kenney j. d. p. , abramson a. , bravo - alfaro h. , 2015 , aj , 150 , 59 kennicutt r. c. , jr .
, 1998 , ara&a , 36 , 189 koopmann r. a. , kenney j. d. p. , young j. , 2001 , apjs , 135 , 125 koopmann r. a. , kenney j. d. p. , 2004a
, apj , 613 , 851 koopmann r. a. , kenney j. d. p. , 2004b
, apj , 613 , 866 kregel m. , van der kruit p. c. , de blok w. j. g. , 2004 , mnras , 352 , 768 kronberger t. , kapferer w. , ferrari c. , unterguggenberger s. , schindler s. , 2008 , a&a , 481 , 337 lee b. , chung a. , 2015 , pkas , 30 , 491 leroy a. k. , walter f. , brinks e. , bigiel f. , de blok w. j. g. , madore b. , thornley m. d. , 2008 , aj , 136 , 2782 leroy a. k. , et al . , 2009 , aj , 137 , 4670 lucy l. b. , 1974 , aj , 79 , 745 mei s. , et al . , 2007 ,
apj , 655 , 144 merluzzi p. , et al . , 2013 , mnras , 429 , 1747 mok a. , et al . , 2016 , mnras , 456 , 4384 nehlig f. , vollmer b. , braine j. , 2016 , a&a , 587 , a108 paturel g. , petit c. , prugniel p. , theureau g. , rousseau j. , brouty m. , dubois p. , cambrsy l. , 2003 , a&a , 412 , 45 qi c. , 2012 , available at : http://www.cfa.harvard.edu/@xmath1cqi/mircook.html rengarajan t. n. , iyengar k. v. k. , 1992 , mnras , 259 , 559 rhee m .- h . , van albada t. s. , 1996 , a&as , 115 , 407 salo h. , et al . , 2015 , apjs , 219 , 4 skrutskie m. f. , et al . , 2006 , aj , 131 , 1163 sofue y. , koda j. , nakanishi h. , onodera s. , kohno k. , tomita a. , okumura s. k. , 2003a , pasj , 55 , 17 sofue y. , koda j. , nakanishi h. , onodera s. , 2003b , pasj , 55 , 59 sofue y. , koda j. , nakanishi h. , hidaka m. , 2003c , pasj , 55 , 75 solomon p. m. , downes d. , radford s. j. e. , barrett j. w. , 1997 , apj , 478 , 144 stark a. a. , knapp g. r. , bally j. , wilson r. w. , penzias a. a. , rowe h. e. , 1986 , apj , 310 , 660 strong a. w. , mattox j. r. , 1996 , a&a , 308 , l21 sun m. , donahue m. , voit g. m. , 2007 , apj , 671 , 190 sun m. , donahue m. , roediger e. , nulsen p. e. j. , voit g. m. , sarazin c. , forman w. , jones c. , 2010 , apj , 708 , 946 swaters r. a. , 1999 , phd thesis , university of groningen tonnesen s. , bryan g. l. , 2012 , mnras , 422 , 1609 verheijen m. a. w. , 1997 , phd thesis , university of groningen vollmer b. , cayatte v. , balkowski c. , duschl w. j. , 2001 , apj , 561 , 708 vollmer b. , balkowski c. , cayatte v. , van driel w. , huchtmeier w. , 2004 , a&a , 419 , 35 vollmer b. , soida m. , otmianowska - mazur k. , kenney j. d. p. , van gorkom j. h. , beck r. , 2006 , a&a , 453 , 883 vollmer b. , braine j. , pappalardo c. , hily - blant p. , 2008
, a&a , 491 , 455 vollmer b. , 2009 , a&a , 502 , 427 vollmer b. , et al .
, 2012a , a&a , 537 , a143 vollmer b. , wong o. i. , braine j. , chung a. , kenney j. d. p. , 2012b
, a&a , 543 , a33 walter f. , brinks e. , de blok w. j. g. , bigiel f. , kennicutt r. c. , jr .
, thornley m. d. , leroy a. , 2008 , aj , 136 , 2563 warmels r. h. , 1988a , a&as , 72 , 19 warmels r. h. , 1988b , a&as , 72 , 427 wong o. i. , kenney j. d. p. , murphy e. j. , helou g. , 2014 , apj , 783 , 109 yasuda n. , fukugita m. , okamura s. , 1997 , apjs , 108 , 417 yoon h. , 2013 , ms thesis , yonsei university
we estimate the molecular gas depletion time - scale ( the ratio of molecular gas mass and star formation rate ) in extraplanar region of ngc 4522 following the prescription below .
we first calculate molecular gas mass in two regions ( figure [ deple ] , red squares ) using iram 12co ( 2@xmath31 ) data of @xcite . using h@xmath0 image @xcite , the star formation rate in each region ( fig .
[ deple ] , red squares ) is estimated as follows : @xmath63 in @xmath6 yr@xmath8 where @xmath64 is the h@xmath0 luminosity in erg s@xmath8 @xcite .
measured depletion time - scales are 1.1 gyr ( region 1 ) and 0.8 gyr ( region 2 ) , respectively . | we present 12co ( 21 ) data of three virgo spirals ngc 4330 , ngc 4402 and ngc 4522 obtained using the submillimeter array .
these three galaxies show clear evidence of ram pressure stripping due to the cluster medium as found in previous hi imaging studies . using high - resolution co data ,
we investigate how the properties of the inner molecular gas disc change while a galaxy is undergoing hi stripping in the cluster . at given sensitivity limits
, we do not find any clear signs of molecular gas stripping .
however , both its morphology and kinematics appear to be quite disturbed as those of hi .
morphological peculiarities present in the molecular and atomic gas are closely related with each other , suggesting that molecular gas can be also affected by strong icm pressure even if it is not stripped .
co is found to be modestly enhanced along the upstream sides in these galaxies , which may change the local star formation activity in the disc .
indeed , the distribution of h@xmath0 emission , a tracer of recent star formation , well coincides with that of the molecular gas , revealing enhancements near the local co peak or along the co compression .
fuv and h@xmath0 share some properties in common , but fuv is always more extended than co / h@xmath0 in the three galaxies , implying that the star - forming disc is rapidly shrinking as the molecular gas properties have changed .
we discuss how icm pressure affects dense molecular gas and hence star formation properties while diffuse atomic gas is being removed from a galaxy .
[ firstpage ] ism : molecules galaxies : clusters : intracluster medium galaxies : evolution galaxies : ism galaxies : spiral . |
the effect of electron correlations on the properties of different materials is currently among the most burning problems of the condensed matter physics . as a rule ,
strong electron correlations are accompanied by the formation of nanoscale inhomogeneous states @xcite .
such inhomogeneities have been already studied for several decades . in particular ,
they were widely discussed for high-@xmath2 superconductors @xcite and heavy fermion materials @xcite . in the recent years
, they attract a special attention owing to the discovery of the colossal magnetoresistance effect in manganites ( the nature of which is believed to be closely related to inhomogeneous structures @xcite ) .
the inhomogeneities manifest themselves in other magnetic materials such as cobaltites @xcite , nickelates @xcite and also in low - dimensional magnets @xcite .
all these systems are characterized by a strong interplay of spin , charge , and orbital degrees of freedom leading to rather rich phase diagrams .
one of the first most spectacular examples of such kind of inhomogeneities is the formation of ferromagnetic ( fm ) droplets ( magnetic polarons or ferrons ) in antiferromagnetic ( af ) semiconductors at low doping levels as well as fm spin polarons in the paramagnetic state @xcite .
these examples correspond to the case of so called electron phase separation caused by self - trapping of charge carriers , which change their local environment .
in addition to such a small - scale phase separation , in manganites , as well as in other compounds exhibiting first order transitions ( e.g. , between fm and af phases ) , there also arises the phase separation of another type related to rather wide region where different phases coexist .
an example of such large - scale phase separation is the formation of rather large fm droplets with the size of the order of ( 100 - 1000 ) @xmath3 inside the af matrix @xcite . at higher doping levels close to the
half - filling , there appears one more threshold for the phase separation in the system corresponding again to the formation of ferromagnetic droplets , but now in a charge - ordered insulating matrix @xcite .
the interaction of spin , charge , and orbital degrees of freedom can also lead to the formation of stripe structures instead of droplets at high content of the alkaline - earth element @xcite . in manganites , owing to the strong electron - lattice interactions , such structures are related to the lattice distortions and can be observed by the electron diffraction and the low - angle neutron scattering @xcite .
both analytical and numerical studies in various models related to the strongly correlated electrons exhibit a pronounced tendency toward phase separation in a wide range of temperatures and electron or hole concentrations . among these , we can mention @xmath4 @xcite , @xmath5 @xcite , hubbard @xcite , and falicov - kimball @xcite models . the theoretical models usually imply that the number of charge carriers introduced by doping is equal to the number of itinerant electrons which take part in the formation of nanoscale inhomogeneities .
however , the comparison of experimental data with the theoretical results suggests that such an approach is insufficient @xcite and the number of self - trapped carriers can significantly differ from the doping level @xcite . here
, we analyze the model proposed in our paper ref . , which relates the doping level and the number of charge carriers .
we take into account the jahn - teller ( jt ) nature of magnetic ions , which could give rise to the localization of charge carriers at the lattice distortions .
we introduce this localization effect to the kondo - lattice model in the double exchange limit . beginning from seminal paper , ref .
, the role of jt distortions was widely discussed in literature @xcite . in particular ,
such distortions were taken into consideration in refs . in the analysis of the phase diagram of doped manganites . however , none of these papers dealt with the phase separation phenomena . in our paper ,
the main emphasis is made on the studying the phase diagram in the doping
temperature ( @xmath1 ) plane and its evolution under effect of the applied magnetic field .
first , let us note that the doped manganites , ( re@xmath6mn@xmath6)(a@xmath7mn@xmath8)o@xmath9 , are the compounds with mixed valence . here
re is a trivalent rare - earth element and a is a bivalent alkaline - earth element .
the ions mn@xmath10 and mn@xmath11 have @xmath12 and @xmath13 electron configurations , respectively . in such compounds ,
manganese ions are located in the centers of o@xmath14 octahedra . in a regular octahedron ,
a five - fold degenerate @xmath15 level is split into triple- and double - degenerate levels , @xmath16 ( @xmath17 , @xmath18 , @xmath19 ) and @xmath20 ( @xmath21 , @xmath22 ) , respectively .
the @xmath16 level lies lower than @xmath20 level .
manganese ions are characterized by the strong hund s rule coupling giving rise to the parallel alignment of intra - atomic electron spins .
so , the spins of @xmath16 electrons form local spin @xmath23 . in a regular mno@xmath14 octahedron ,
mn@xmath10 ion has one electron at double - degenerate @xmath20 level @xcite . according to the jahn - teller theorem , the latter configuration is unstable and the degeneracy is lifted by a distortion of the octahedron .
the mn@xmath11 is not a jahn - teller ion and mn@xmath11o@xmath14 octahedron remains undistorted .
the distortion of mn@xmath10o@xmath14 octahedron leads to the energy lowering by @xmath24 . in doped manganites ,
the @xmath20 electron can hop from mn@xmath10 to mn@xmath11 ion producing a gain in the kinetic energy due to electron delocalization .
therefore , an electron can lower its energy either due to the jahn - teller - induced localization at distorted octahedra or by the delocalization related to the inter - atomic hopping .
the strong hund s rule coupling favors the hopping of an electron when its spin is parallel to the spin of core ( @xmath16 ) electrons .
this is the origin of the well - known double - exchange mechanism of ferromagnetic interaction between the localized spins @xcite .
therefore , it is natural to assume that the @xmath25 electron can be either localized due to jahn - teller distortions with the energy gain @xmath24 ( @xmath26 electron ) or to decrease its energy due to band broadening ( @xmath27 electron ) @xcite .
so , there exists a competition between the localization and delocalization .
such a system with localized and band electrons can be analyzed using the hubbard hamiltonian taking into account the electron - lattice interaction , the hund s rule coupling , and the exchange interaction between core electrons @xcite @xmath28a_{{\bf n}b\sigma } \nonumber\\&&+\frac{k}{2}\sum_{{\bf n}}\left(q_{2{\bf n}}^{\,2}+q_{3{\bf n}}^{\,2}\right)\,,\nonumber\\ h_{\text{el - el}}&=&\frac{u_1}{2}\sum_{{\bf n}a\sigma}n_{{\bf n}a\sigma}n_{{\bf n}a\bar{\sigma}}+\frac{u_2}{2}\sum_{{\bf n}a\sigma\sigma'}n_{{\bf n}a\sigma}n_{{\bf n}\bar{a}\sigma'}\,.\nonumber\end{aligned}\ ] ] in this hamiltonian , @xmath29 , @xmath30 are creation and annihilation operators for @xmath20 electrons at site @xmath31 with orbital index @xmath32 ( @xmath33 or @xmath34 ) and spin projection @xmath35 , @xmath36 is a local spin of @xmath16 electrons . below we will consider @xmath36 as classical vectors .
@xmath37 are the pauli matrices and @xmath38 , @xmath39 are normal modes of vibration of mno@xmath14 octahedron .
the symbol @xmath40 denotes the summation over nearest sites .
the electron part @xmath41 of hamiltonian includes the kinetic energy of @xmath20 electrons and the hund s rule coupling between the spins of @xmath20 and @xmath16 electrons .
@xmath42 is the antiferromagnetic ( @xmath43 ) exchange interaction between local spins .
the @xmath44 term takes into account interactions between @xmath20 electrons and vibrational modes for the mno@xmath14 octahedra , here @xmath45 is the elastic energy and @xmath46 is the electron - lattice coupling constant .
the on - site coulomb repulsion @xmath47 includes the terms corresponding to @xmath20 electrons at the same and different orbitals , where the bar above @xmath32 or @xmath35 means _ not _ @xmath32 or _ not _
@xmath35 , respectively .
we consider the limit @xmath48 characteristic of manganites . in this case , the spin of @xmath25 electrons is parallel to @xmath49 and we can eliminate the spin indices by the transformation of @xmath50 to operators @xmath51 with spin projection @xmath52 onto the direction of @xmath36 accompanied by the transformation of hopping amplitudes @xcite : @xmath53 , where @xmath54 .
in addition , we assume that @xmath55 for @xmath26 electrons , which produce maximum splitting of @xmath20 level with the energy gain @xmath56 , whereas @xmath27 electrons with non - zero hopping integrals @xmath57 produce smaller distortions of mno@xmath14 octahedra .
preliminary calculations for the case @xmath58 demonstrated that the results are not significantly affected if the localized band is much narrower than itinerant one .
therefore , in this paper , we consider the limiting case of zero hopping integral for localized electrons .
hamiltonian then reads @xmath59 where @xmath60 and @xmath61 are the numbers of @xmath27 and @xmath26 electrons at site @xmath31 , @xmath62 , @xmath63 and @xmath64 , @xmath65 are the creation and annihilation operators for the @xmath27 and @xmath26 electrons , respectively , and @xmath66 is the chemical potential .
the first three terms in @xmath67 correspond , respectively , to the kinetic energy of @xmath27 electrons , jt energy of localized electrons , and on - site coulomb repulsion between @xmath27 and @xmath26 electrons .
the last term in @xmath67 is the heisenberg antiferromagnetic exchange between local spins .
the effective on - site coulomb repulsion @xmath68 in eq . can differ from @xmath69 , but has the same order of magnitude ( @xmath70 ) .
@xmath71 is the jt energy gain for @xmath26 electrons counting from the center of @xmath27 electron band .
the number of localized , @xmath72 , and band , @xmath73 , electrons per lattice site obeys an obvious relation @xmath74 , where @xmath0 is the doping level . in our paper
, we limit ourselves by the case of the large on - site coulomb repulsion @xmath68 , which strongly suppresses double occupancy of the site . moreover , at large @xmath68 , the characteristic time of existence for a double electron state is of the order of @xmath75 , where @xmath76 is the characteristic jt phonon frequency ( @xmath77 ) , which is of the order of debye frequency . therefore , the adiabatic approximation for the jahn - teller distortions is applicable . as a result , the jt term in eq
is determined only by localized electrons .
hamiltonian was analyzed at zero temperature in ref . .
the homogenous ferromagnetic and antiferromagnetic states as well as the phase - separated fm - af state were studied .
the effective parameters @xmath78 and @xmath79 were considered to be independent of the densities of band @xmath73 and localized @xmath72 electrons . in the present analysis , we also neglect the dependence of parameters @xmath79 and @xmath78 on @xmath73 and @xmath72 , but take into account the temperature dependence of hopping integral @xmath78 , which can be rather strong due to polaron band narrowing @xcite .
following refs .
we write the expression for the hopping integral in the form @xmath80 in this expression , @xmath81 is the dimensionless electron - phonon coupling constant . from eq .
, it is clear that the hopping integral @xmath82 decreases with temperature .
even if we take into account the finite bandwidth for localized electrons , it will also decrease with @xmath83 .
however , the parameters describing this behavior could be different for different bands .
we can not assert that at high temperatures the ratio of widths for the narrow and wide bands would be larger or smaller than at low temperatures .
in this paper , we consider the temperature range , which is relatively small as compared to characteristic value of @xmath76 .
thus , @xmath82 does not vary by orders of magnitude and the bandwidth ratio remains small .
to study the effects of temperature , we use the mean field ( mf ) approximation .
for this purpose , we make a decoupling procedure in the first term of @xmath67 , eq .
, in the following way .
the values of @xmath84 and @xmath85 can be written as @xmath86 where angular brackets mean thermal averaging , @xmath87 , and @xmath88 is the unit vector . omitting the products proportional to @xmath89 , we write the first term in @xmath67 as @xmath90 where @xmath91 .
note that in homogenous state , @xmath92 does not depend on indices @xmath31 and @xmath93 ( for sites @xmath94 nearest to the site @xmath95 ) .
now , the effective hamiltonian can be represented as a sum of electronic and magnetic parts @xmath96\label{hm}.\end{aligned}\ ] ]
in the ferromagnetic state , we have @xmath97 far below the curie temperature .
first , we consider the electronic sector of the problem .
the hamiltonian @xmath41 is similar to that considered in ref .
and the temperature @xmath83 enters only the effective hopping integral @xmath98 . to calculate the free energy of the electronic subsystem
, we use the hubbard i decoupling for the one-@xmath27-electron green function @xmath99 , as in ref . .
here @xmath100 is the time variable and @xmath101 is the time - ordering operator .
the equation of motion for @xmath102 can be written in the form : @xmath103 where summation is performed over sites nearest to site @xmath31 , and @xmath104 is the two - particle green function @xmath105 .
the equation of motion for @xmath104 is @xmath106 the decoupling in the first term in curly brackets gives @xmath107 . making similar decoupling in the next two terms
, we get @xmath108 in the homogeneous ferromagnetic state , the sum evidently vanishes . as a result , we obtain the closed system of equations for @xmath109 and @xmath104 . in the frequency - momentum representation ,
the solutions for @xmath110 and @xmath104 are as follows @xmath111 where @xmath112 in these expressions , @xmath113 and @xmath114 where @xmath115 is the number of nearest neighbors . in the case of the simple cubic lattice
, we have @xmath116 and in the tight - binding approximation @xmath117,\ ] ] where @xmath118 is the lattice constant and @xmath119 are the components of the wave vector . from eq .
, it follows that the energy spectrum of @xmath27 electrons includes two sub - bands given by eq .
and the number of states in each sub - band depends on @xmath72 . in the limit of large @xmath68 , which is relevant to magnetic oxides , eq .
can be written as @xmath120 it is clear from eq . that the width of the lower sub - band is @xmath121 while the width of the upper sub - band is @xmath122 .
the total number of states in two sub - bands per site is equal to one and the number of states in the lower and upper sub - bands is equal to @xmath123 and @xmath72 , respectively .
note that this result is valid for any value of @xmath68 as it can be demonstrated by integration of the corresponding terms in the green function eq . .
thus , at any doping level @xmath0 and temperature @xmath124 the upper sub - band is empty since @xmath125 . in this case
, it is reasonable to use the @xmath126 limit .
the green function @xmath110 then becomes @xmath127 at low temperatures , @xmath128 , it is reasonable to represenr the fermi - dirac distribution function for @xmath27 electrons @xmath129^{-1}$ ] by the step function @xmath130 .
it can be shown that this approximation works well at @xmath131 , and @xmath132 , where @xmath133 .
an appreciable discrepancy could arise only at doping levels @xmath134 and @xmath135 .
however , at these doping levels , the homogeneous ferromagnetic state is unfavorable ( see below ) .
therefore , the number of @xmath27 electrons can be expressed through the green function as @xmath136 this relationship defines @xmath73 as a function of chemical potential @xmath66 and @xmath72 .
the value of @xmath72 depends on relative positions of @xmath66 and @xmath24 .
if @xmath137 , then the jahn - teller - induced localization is unfavorable , @xmath138 , @xmath139 , and the chemical potential is found from eq . .
with the increase of @xmath73 , @xmath66 becomes equal to @xmath140 , the further growth in the number of itinerant charge carriers is ceased , @xmath72 becomes nonzero , and @xmath66 is pinned at the level @xmath140 . in the latter case , the number of localized electrons can be found from eq . at @xmath141 using the relation @xmath142 . as a result , we get @xmath143=x\,,\ ] ] where @xmath144 and @xmath145 is the density of states of free electrons . at zero doping , @xmath146 , the number of localized electrons @xmath147 , and the bandwidth @xmath148 .
at low doping , all electrons are localized , @xmath149 and @xmath150 , until the bottom of the band reaches the energy of @xmath26 electrons @xmath151 at some critical concentration @xmath152 . at @xmath153
the localized and band electrons coexist as long as @xmath0 is smaller than the second critical doping level @xmath154 at which the existence of localized electrons becomes unfavorable , that is , @xmath138 if @xmath155 . from eq .
we find @xmath156 .
naturally , such a picture can exist only for a certain relationship between the parameters of the model , in particular , if @xmath157 .
since @xmath158 decreases with the growth of temperature , both these critical concentrations , @xmath159 and @xmath160 , become larger when temperature increases .
note that the homogeneous ferromagnetic state can exist only at @xmath161 if @xmath162 .
the values of @xmath72 and @xmath73 as the functions of doping @xmath0 are shown in fig .
[ fignl ] at typical parameters of the model . these calculations and calculations below
are performed for a simple cubic lattice with @xmath163 given by eq . .
in this case , the expressions for @xmath164 and @xmath165 in eqs . and are found in appendix [ se ] , see eqs . and , respectively . at @xmath166 ,
the kinetic energy of @xmath27 electrons per site can be written as @xmath167 for further purposes we introduce the function @xmath168 where @xmath169
the kinetic energy of @xmath27 electrons then reads @xmath170 the energy of the on - site coulomb repulsion can be found using the two - particle green function @xmath104 @xmath171 since @xmath172 , the pole coming from the denominator of the function @xmath104 ( see second eq . )
does not contribute to the integral over @xmath173 , and in the limit @xmath126 , we get @xmath174 note that we replace the fermi - dirac distribution by the step function . therefore , we can omit the electron entropy term and write the free energy of electrons per site as a sum of eqs .
, , and the jt term @xmath175 for @xmath155 when @xmath138 , we should replace @xmath176 by @xmath177 in formulas , , and , where @xmath178 is found from the equation @xmath179 .
now , we consider the magnetic part of the hamiltonian , @xmath180 , eq . . following the conventional mean field approach for spin systems @xcite , we replace @xmath181 in eq . by its mean value @xmath182 ,
where @xmath183 . as a result ,
@xmath180 decouples into the sum of @xmath184 independent one - site hamiltonians , @xmath185 , where @xmath186 @xmath187 .
the value of @xmath188 is determined by the self - consistency condition @xmath189 where @xmath190 .
we should also take into account that @xmath158 is related to @xmath191 and @xmath188 by eq .
@xmath192 in order to find the curie temperature @xmath2 , we expand right hand side of eq .
in a power series in @xmath188 . using eqs .
and we find @xmath193 the curie temperature is found then from the condition @xmath194=t_c\,.\ ] ] if we neglect the effect of polaron band narrowing , @xmath195 , we find the explicit expression for the curie temperature @xmath196/3 $ ] .
it is clear that polaron band narrowing reduces @xmath197 .
let us now analyze the order of the phase transition at @xmath198 , which depends on the sign of @xmath199 , see ref . .
if @xmath200 , then @xmath188 tends to zero at @xmath201 as @xmath202 in this case , we have second order magnetic phase transition . in the opposite case @xmath203
, @xmath188 behaves approximately as ( @xmath204 ) @xmath205 and the transition to paramagnetic ( pm ) state is of the first order .
the analysis shows that @xmath204 for any values of parameters of the model , but the sign of @xmath199 can be both positive or negative depending on the doping level @xmath0 . at some @xmath206 ,
the coefficient @xmath199 changes its sign . at low doping ,
@xmath207 , we have @xmath203 and the transition from the fm to pm state is of the first order . at @xmath208
, we have the second order phase transition .
the free energy of the system per site is equal to @xmath209 at @xmath210 , @xmath211 , @xmath212 , @xmath213 , and the free energy @xmath214 in the pm state , we have @xmath215 , @xmath216 , @xmath217 , and @xmath218 the transition from fm to pm state does not mean that @xmath73 becomes zero . as it was mentioned above , @xmath162 at @xmath153 and the value of @xmath159 increases with the temperature . at a certain temperature @xmath219 , @xmath220 exceeds @xmath0 .
it is clear that @xmath221 since the fm state can exist only at @xmath162 .
therefore , in our model , in addition to magnetic phase transitions , there should exist an electronic transition to the state without itinerant electrons .
the temperature @xmath222 is determined by the evident condition @xmath223 . in our model
, there can exist other homogeneous states competing with fm and pm states .
it is natural to consider the two - sublattice antiferromagnetic and canted states . in the canted state ,
the angle @xmath224 ( canting angle ) between the local spins belonging to two sublattices varies from @xmath225 ( af state ) to @xmath226 ( fm state ) .
here we consider af and canting states of @xmath227 type , that is , in the cubic lattice each site of one sublattice is surrounded by sites of the second sublattice . at finite temperatures ,
the canting angle is defined as @xmath228 . in the mean field approximation
used above , we have @xmath229 .
thus , we can define variable @xmath188 similar to that introduced in the previous subsection , @xmath230 if @xmath231 or @xmath232 if @xmath233 .
in the first case , the value of @xmath234 and the free energy of the system is found in the same way as for fm state from eqs . .
the only difference is that @xmath235 at @xmath236 in fm state ( or @xmath237 ) while @xmath238 ( or @xmath239 ) in the canted state .
note that the temperature of the transition from canted to paramagnetic state , @xmath240 , is given by eq . , where @xmath197 should be replaced by @xmath240 .
in the case @xmath241 , instead of eq . and , we have @xmath242 @xmath243 where mean - field hamiltonian @xmath191 is given by eq . .
the temperature of the phase transition from canting to pm state is found now from the equation @xmath244=t_{cant}\,.\ ] ] the canted state can exist at @xmath162 since it arises in our model due to the motion of conduction electrons . at doping levels @xmath245 , we have @xmath246 and the af ordering is favorable . at rather low doping , @xmath247 , @xmath246 at any temperature . in this case
, the nel temperature is independent of @xmath0 and is determined by eq . as @xmath248 . at higher doping
, there can occur a transition from the af state to the pm state with @xmath162 , and the nel temperature , @xmath249 , is determined by the comparison of free energies of corresponding states . in this doping range
, the af state turns out to be more favorable than the canted state . with the further increase of @xmath0 ,
the phase with @xmath162 has the lower energy , but this phase is fm rather than canted . the temperature of the af
fm transition is found from the comparison of corresponding free energies .
this transition is of the first order .
note that the phase diagram exhibits af fm
pm triple point at @xmath250 ( @xmath251 ) .
the canted state can exist in the doping range @xmath135 , where the number of itinerant electrons is too small to stabilize the fm state .
the corresponding phase diagram in the @xmath1 plane is shown in fig .
[ fighomphdia ] .
we see that the competition between the itinerant and localized electrons gives rise to a rather complicated magnetic phase diagram of the system .
up to this point , we considered only homogeneous states , however , it is well known that different inhomogeneous states are possible in the systems with strongly correlated electrons .
so , we should compare the free energies of the states studied in the previous sections with those for inhomogeneous states . as a typical example of an inhomogeneous state ,
we analyze here the droplet model of electronic phase separation widely discussed in connection with manganites and other magnetic oxides . among the possible types of phase separation
, we treat below the coexistence of different phases : af fm , fm pm , and pm phases with different values of @xmath73 . for simplicity
, we do not include into consideration the phase separation involving the canted state since it exists only in the narrow doping range .
note that our model always leads to such kind of phase separation , where we have @xmath246 in one of the phases .
we consider a system separated into two phases with the volume concentrations @xmath252 and @xmath253 . in the homogeneous phases ,
the electron concentration per site coincides with the doping level @xmath0 . in the inhomogeneous states
, the electrons can be redistributed between the regions with different phases .
let @xmath162 in the first ( f ) phase and @xmath246 in the second ( a ) phase , the electron density per site in the first phase is @xmath254 and in the second phase is @xmath255 .
so , doping level @xmath0 lies between @xmath255 and @xmath254 .
the charge conservation requires @xmath256 . in fig .
[ figfenergy ] , we show the dependence of the free energy of the most favorable homogeneous state @xmath257 on the doping level at different temperatures .
the @xmath258 curves have two minima : one at @xmath146 and another near @xmath259
. then we could expect that @xmath255 should be around zero , while @xmath254 should be close to @xmath259 .
the phase separation corresponds to the non - uniform charge density and we should take into account the coulomb contribution to the total energy .
this contribution depends on the structure of inhomogeneous state . to evaluate the coulomb energy
, we assume the spherical geometry of the phase - separated state .
namely , at @xmath260 , the sample is modelled as an aggregate of spheres of f phase embedded into a matrix or that of a spheres in f phase for @xmath261 . for this geometry ,
it is reasonable to calculate the coulomb energy using the wigner - seitz approximation : each f or a sphere of radius @xmath262 is surrounded by a spherical cell of radius @xmath263 , such as the volume of the cell is @xmath264 , where @xmath265 is the volume of the sample and @xmath266 is the number of spheres .
the radius @xmath263 is related to @xmath262 as @xmath267 for @xmath260 , and @xmath268 for @xmath261 .
the total electric charge inside this cell is zero and the coulomb energy of the system is the sum of the electrostatic energies of these cells .
following ref .
, we obtain the expression for the coulomb energy per site @xmath269 at @xmath260 @xmath270 where @xmath271 is the average permittivity of the sample and @xmath118 is the lattice constant . in the case @xmath261 , we should replace @xmath272 and @xmath273 .
the @xmath27 electrons in the phase - separated state are confined within a restricted volume .
the corresponding size quantization gives rise to the change in the density of states .
the additional contribution to the energy ( per site ) is proportional to the total surface area between f and a phases , and at @xmath260 can be written in the form @xmath274 where surface energy @xmath275 is calculated in appendix [ se ] . in the case
@xmath261 , we should change @xmath273 .
the coulomb and surface contributions to the total energy depend on the size @xmath262 of the inhomogeneities .
minimization of @xmath276 with respect to @xmath262 gives at @xmath260 @xmath277 @xmath278 where @xmath279 .
let us estimate the parameter @xmath280 and the characteristic size of inhomogeneities @xmath262 .
using typical values of parameters for manganites @xmath116 , @xmath281ev , @xmath282 nm , and @xmath283 we find @xmath284ev , and @xmath285 .
the surface energy @xmath286 is calculated in appendix [ se ] .
for example , at @xmath287 the optimization procedure gives @xmath288 , @xmath289 and @xmath290 and from eq .
we find @xmath291 , that is , the inhomogeneity contains @xmath292 unit cells .
thus , we see that @xmath293 for characteristic values of parameters and the wigner - seitz approximation is applicable
. however , this approximation overestimates the coulomb contribution because of a sharp boundary of the inhomogeneities .
therefore , the above values for @xmath262 and @xmath266 could be considered as lower estimates . in the phase - separated state , to find the values of @xmath254 and @xmath255 at given @xmath0 and @xmath83 , it is necessary to minimize the total free energy @xmath294 with respect to @xmath254 and @xmath255 , where @xmath295 . at temperature
@xmath296 ( see fig .
[ fighomphdia ] ) only homogeneous pm state with @xmath246 is possible .
as it follows from the numerical and analytical analyzes of the free energy , below @xmath297 , the function of two variables @xmath298 has two minima if @xmath299 for any phases f and a within the considered hierarchy of parameters ( @xmath300 ) .
the first minimum at the point @xmath301 corresponds to some homogeneous state while the second minimum at the point @xmath302 , @xmath303 corresponds to the phase - separated state . in the limit @xmath304 , the second minimum is the global minimum of the function @xmath298 at @xmath305 and @xmath306 , and the ps state is favorable .
when @xmath280 increases , the range of phase separation in the plane @xmath307 gradually narrows and disappears at some critical value @xmath308 .
note that there is no localized electrons in a more metallic f phase ( @xmath303 ) , and vice versa , in the insulating a phase @xmath309 since @xmath302 , as it was mentioned in connection to fig .
[ figfenergy ] .
since the concentrations @xmath255 and @xmath254 are independent of the doping level , the temperatures of magnetic phase transitions in both phases do not depend on @xmath0 .
the nel temperature of the af phase is @xmath310 , whereas the curie temperature for the fm phase can be found from the equation @xmath311=t_c\,.\ ] ] the nel and curie temperatures in the ps state calculated in such a way correspond to the macroscopic phases with the size of inhomogeneities @xmath312 .
if @xmath313 , the values of @xmath249 and @xmath197 in the ps state can differ from those calculated above .
the region where the ps state is favorable can be found from the comparison of free energy @xmath314 with the free energies of homogeneous states at given @xmath0 and @xmath83 .
the phase diagram of the system in the @xmath315 plane is shown in fig
. [ figpsphdia ] .
the range of existence for the ps state is bounded by the curve @xmath316 . in the ps state ,
the content @xmath252 of the metallic ( @xmath162 ) phase varies with the temperature and the doping level .
hence , the insulator - metal transition is possible when @xmath252 exceeds the percolation threshold .
let us now discuss the transition of the system from the ps to a homogeneous state .
the volume fraction of the f phase in the ps state is @xmath317 .
depending on the relation between the temperatures @xmath318 , @xmath197 , @xmath249 , and @xmath222 , the system can pass from the ps state to the fm ( @xmath319 ) , af ( @xmath320 ) , and pm ( with @xmath321 or @xmath246 ) homogeneous states . in all cases ,
the number of itinerant electrons @xmath73 undergoes a sudden change at the transition to the homogeneous state .
the temperature dependence of @xmath73 is shown in fig .
[ fignbt ] .
in this section , we consider the effect of magnetic field on the properties of the system .
we take into account only the effect of the magnetic field on the local spin .
this corresponds to the limit of classical local spin @xmath322 .
thus , in the presence of external dc magnetic field @xmath323 , we should add the term @xmath324 in hamiltonian , where @xmath325 is the bohr magneton and @xmath46 is the lande factor . as a result
, the magnetic field term modifies only the magnetic hamiltonian , eq . , @xmath326,% \,\,\,\,\,\,\,\mathbf{h}=\mu_{b}gs\mathbf{h}\,.\nonumber\end{aligned}\ ] ] in the fm state , the one - site magnetic hamiltonian corresponding to the mf approximation takes the form @xmath327 where the direction of magnetic field is parallel to @xmath115 axis . the mean value @xmath328 is found by solving the system of equations and with hamiltonian . at @xmath329 , the correction to the free energy in the presence of magnetic field @xmath330 , whereas in paramagnetic phase @xmath331 . in the af or canted states ,
the result depends on the mutual orientation of @xmath332 and the vector @xmath333 .
the minimum of the free energy corresponds to the case @xmath334 .
let the vector @xmath335 be parallel to @xmath115 axis and the magnetic field @xmath332 be parallel to the @xmath0 axis .
the mean values of the directions of local spins in two sublattices , @xmath336 and @xmath337 can be written as @xmath338 , @xmath339 , where @xmath188 is proportional to the magnetization of the system .
the one - site magnetic hamiltonian then has the form @xmath340 where @xmath341 and @xmath342 the values of @xmath26 and @xmath188 are found from the equations @xmath343 expression for the effective bandwidth @xmath344 now takes the form @xmath345 at high magnetic fields @xmath346 , we get from eq . that @xmath347 , and the system passes from the af or canted state to the fm one .
the typical nel temperature in manganites is @xmath348k , and the value of @xmath349 corresponds to the fields of the order of @xmath350 t .
if @xmath351 , the system can be solved perturbatively .
the correction to the free energy of the af state is @xmath352 both below and above @xmath249 , because there is no spontaneous magnetization in the system at @xmath353 .
the external magnetic field favors the fm state in comparison to the af and canted states .
in particular , it reduces the temperature of transition from the af to fm state .
the magnetic field leads to the increase in the effective hopping integral @xmath354 due to the alignment of local spins and thus to the growth in the number of @xmath27 electrons and the value of @xmath222 . at the transition from the ps to pm homogeneous state the magnetic field results in the increase of the transition temperature , and the difference @xmath355 . at the transition from the ps to fm state
, @xmath356 can be both positive and negative depending on the parameters .
as was mentioned above , the number of itinerant electrons @xmath73 differs significantly below and above @xmath318 .
therefore , the shift of the transition temperature @xmath318 with the magnetic field gives rise to a significant change in the number of itinerant electrons .
the temperature dependence of the ratio @xmath357 near the transition temperature @xmath358 is shown in fig .
[ figdnbt ] .
the narrow peak in this ratio is a manifestation of the step in @xmath359 shown in fig .
[ fignbt ] .
since the position @xmath318 of the step in @xmath359 depends on the magnetic field , a small change in @xmath360 causes a significant change in the number of charge carriers at a fixed temperature near @xmath318 .
the number of itinerant charge carriers determines the value of metallic conductivity of the system .
thus , the large change of @xmath73 in magnetic field can be related to the colossal magnetoresistance effect .
we discussed a minimal model dealing with the competition between the localization and metallicity in manganites .
the hamiltonian of the model takes into account the essential physics of strongly correlated electron systems with the jahn - teller ions : it is , in fact , the hubbard model with the strong electron - lattice interaction , the hund s rule intraatomic coupling , and af interatomic exchange between local spins .
such an approach provides a possibility to understand the difference between the number of itinerant charge carriers @xmath73 and the doping level @xcite .
it is shown that @xmath73 can be significantly lower than the number of the charge carriers implied by the doping level .
the models of similar type were discussed in refs . .
however , the possibility of the phase separation was not considered in these papers . here
we demonstrate that in the framework of our model phase separation can exist in a wide range of intermediate doping concentrations disappearing at low and high doping level .
these predictions are in agreement with the general features of the experimentally found phase diagrams of manganites @xcite .
the obtained results suggest the existence of the droplet - type of the electronic phase separation that was widely discussed in literature ( see , e.g. , ref . ) .
we calculated the relative content of different phases in the phase separated states and found the size of such droplets ( ferrons ) .
for the characteristic values of parameters , a droplet includes 1030 unit cells .
these results could , in particular , serve as a key to an adequate description of the transport properties of manganites that could not be done in the framework of the single - band models @xcite . as it was mentioned above , various models related to the strongly correlated electrons exhibit a pronounced tendency toward phase separation .
the effective hamiltonian of our model is , in fact , a generalization of the falicov - kimball model @xcite .
the latter model describes a system with the hybridization of an electronic band and a localized level .
the falicov - kimball model is often used as a toy model in the analysis of heavy - fermion materials and it also leads to the phase separation phenomena @xcite .
so , we believe that our approach could be applicable not only to manganites but also to a wider class of strongly correlated electron systems .
note that the analogy between the falicov - kimball model and the hamiltonian of the @xmath4 type with the jahn - teller interaction was indicated in ref .
but in the case without phase separation . in this paper
, we analyzed the phase diagram of the model in the @xmath1 plane .
the effect of temperature manifests itself mainly in the change of effective hopping integral @xmath78 due to the polaron band narrowing and the entropy term in the free energy due to thermal fluctuations of local spins .
the polaron band narrowing is described by standard formula , see refs . .
the behavior of the local spins was treated using the mean field approximation .
we find that at low temperatures the system is in a state with a long - range magnetic order : af , fm or af fm phase separated state .
we demonstrate that at high temperatures there can exist two types of the paramagnetic state , a usual one with @xmath246 and that with @xmath162 . in the intermediate temperature range ,
the phase diagram includes different kinds of the ps states : af fm , fm pm , and pm with different content of itinerant electrons .
the applied magnetic field leads to the changes in the phase diagram .
it evidently favors the fm ordering and , consequently , to the increase of the number of itinerant electrons .
the effect of the magnetic field was analyzed accounting for the alignment of the local spins in the applied magnetic field .
it is demonstrated that in our model the metal - insulator transition can take place at some characteristic values of the doping @xmath0 corresponding to the crossover between different kinds of the phase separation .
it can be induced by changing temperature or the magnetic field and is of a percolation type .
this transition can be related to the colossal magnetoresistance effect .
note that in the present treatment we assume that the effective parameters @xmath78 and @xmath79 do not depend on the doping level @xmath0 . to verify the applicability of such approximation , we calculated the phase diagram with @xmath78 and @xmath79 depending linearly on @xmath0 .
we found that the phase diagram remains qualitatively the same even if @xmath78 and @xmath79 vary by the factor 23 provided the hierarchy of the model parameters ( @xmath300 ) remains unchanged .
note also that we included to our analysis the long - range coulomb interaction related to the macroscopic charge redistribution in the phase separated state .
it allowed us to estimate the size of the inhomogeneities . at the same time
, we did not take explicitly into account the corresponding terms in the model hamiltonian .
however , if we would like to consider the effect of charge ordering , we have to include at least the nearest - neighbor coulomb repulsion . here
we considered only the minimal model describing the effect of the phase separation .
therefore , we did not include into consideration the possibility of the charge ordering to focus the discussion on the interplay between localized and itinerant electrons .
it is not impossible to include the charge ordering to the model of such type .
the first attempt was made in ref . , but the possibility of phase separation was not considered there . the effect of the charge ordering could change the results for @xmath0 near 0.5 . therefore , it is reasonable to consider only the range @xmath361 , if we want to compare our predictions with the actual situation in the doped magnetic oxides .
the work was supported by the russian foundation for basic research , project no .
05 - 02 - 17600 and by the russian presidential grant no . nsh-1694.2003.2 .
in this section , we calculate the surface energy coming from the size quantization .
the expression for the free energy of @xmath27 and @xmath26 electrons can be written in terms of the density of states in the form ( see eqs . and ) @xmath362 where @xmath363 this expression is valid for @xmath299 and @xmath141 .
the density of states for the system of itinerant electrons in the volume @xmath265 is @xmath364 where momentum @xmath365 varies over a discrete set of values , depending on boundary conditions and geometry of the system .
the function @xmath366 is normalized to unity , that is @xmath367 . in the thermodynamic limit @xmath368 , the sum in eq .
can be replaced by the integral over @xmath365 in the first brillouin zone multiplied by @xmath369 . at finite @xmath265
, we derive an approximate expression for the density of states in the case of cubic lattice , corresponding a small value of @xmath370 , where @xmath371 is the surface area and @xmath118 is the lattice constant . let the sample has the shape of a parallelepiped @xmath372 with sides @xmath373 , @xmath374 , and @xmath375 ( in units of the lattice constant @xmath118 ) .
the dirichlet boundary conditions for the conduction electron wave function @xmath376 is used , that is , @xmath377 for @xmath378 . in this case
, the momentum @xmath379 takes values @xmath380 , where @xmath381 ( @xmath382 ) . for large @xmath383
, we can use the trapezium rule for the sum over @xmath384 @xmath385 as a result , for 3d sum over @xmath31 we obtain @xmath386\right.\nonumber\\% & + & \!\!\!\!\int\!\!\frac{d^2k}{\pi^2l_2}\left[f(\{k^1,0,k^3\})+f(\{k^1,\pi , k^3\})\right]\nonumber\\ & + & \left.\!\!\int\!\!\frac{d^2k}{\pi^2l_3}\left[f(\{k^1,k^2,0\})+f(\{k^1,k^2,\pi\})\right]\right)\,,\nonumber\end{aligned}\ ] ] where 3d and 2d momentum integrations are performed in the range @xmath387 , and @xmath388 is the surface area of the parallelepiped in the units of lattice constant .
using formula with @xmath389 , we can calculate the density of states .
relation can be simplified for the case of cubic symmetry since @xmath390 . in the absence of external fields , we have @xmath391 , and the integration in eq .
can be extended to @xmath392 . as a result
, we obtain for the density of states where @xmath164 is the density of states at @xmath368 .
note , that the density of states in the form depends only on the ratio @xmath394 and does not depend on the shape of the sample .
we believe , that eq .
is applicable for any geometry of the system , provided that the minimum linear dimension @xmath395 is large compared to the lattice constant ( see , for example @xcite ) .
let us calculate now the surface energy for the spectrum in the tight - binding approximation for the simple cubic lattice , @xmath396/3 $ ] . in the limit @xmath368
, the formulas for the density of states @xmath164 , density of itinerant electrons @xmath397 , and their kinetic energy @xmath398 can be written in the following form @xmath399\,,\end{aligned}\ ] ] where @xmath400 is the bessel function .
now , eq . can be rewritten as @xmath401 where @xmath402 is the density of states in the 2d case .
the number @xmath72 and the free energy of electrons ( in the case @xmath403 ) is given by eqs .
and , where instead of @xmath164 we should use the density of states eq . .
note that @xmath72 and @xmath404 are functions of @xmath405 . in the considered limit @xmath406 , we can use the perturbation technique to calculate the surface energy @xmath35 . representing the number of @xmath26 electrons and the free energy @xmath404 in the form @xmath407 , @xmath408 , and expanding eqs . , , and , one obtains @xmath409 @xmath410\right\}%
\frac{\bar w(1-n_l^{(0)})^2}{4},\nonumber\end{aligned}\]]where @xmath411 is determined by eq . at @xmath412 ,
@xmath413 $ ] , @xmath414 and @xmath415 are the number of @xmath27 electrons and their energy in 2d case at @xmath412 , @xmath416\,.\end{aligned}\ ] ] at doping concentration @xmath155 when @xmath138 , we should use the equation @xmath179 for the chemical potential , where @xmath164 is substituted by @xmath417 . as a result
, we obtain the expression for the surface energy @xmath418\right.\nonumber\\% & + & \left.\frac{1}{6}\left[n_0^{(2)}(\mu_0'+\frac{1}{3})-n_0^{(2)}(\mu_0'-\frac{1}{3})\right]\right\},\end{aligned}\ ] ] where @xmath419 is found from the equation @xmath420 at @xmath412 . the dependence @xmath286 is shown in fig .
[ figsigma ] .
the function @xmath286 is discontinuous at @xmath259 .
this singularity stems from the kink in the free energy @xmath421 ( see fig . [ figfenergy ] ) .
s. mori , c.h .
chen , and s .- w .
cheong , nature ( london ) * 392 * , 473 ( 1998 ) ; p.g .
radaelli , d.e .
cox , l. capogna , s .- w .
cheong , and m. marezio , , 14440 ( 1999 ) ; b. raveau , m. hervieu , a. maignan , and c. martin , j. mater
. chem . * 11 * , 29 ( 2001 ) . | a minimal model of the kondo - lattice type is used to describe a competition between the localization and metallicity in doped manganites and related magnetic oxides with jahn - teller ions .
it is shown that the number of itinerant charge carriers can be significantly lower than that implied by the doping level @xmath0 . a strong tendency to the phase separation
is demonstrated for a wide range of intermediate doping concentrations vanishing at low and high doping .
the phase diagram of the model in the @xmath1 plane is constructed . at low temperatures ,
the system is in a state with a long - range magnetic order : antiferromagnetic ( af ) , ferromagnetic ( fm ) , or af fm phase separated ( ps ) state . at high temperatures ,
there can exist two types of the paramagnetic ( pm ) state with zero and nonzero density of the itinerant electrons . in the intermediate temperature range ,
the phase diagram includes different kinds of the ps states : af fm , fm pm , and pm with different content of itinerant electrons .
the applied magnetic field changes the phase diagram favoring the fm ordering .
it is shown that the variation of temperature or magnetic field can induce the metal - insulator transition in a certain range of doping levels . |
in recent years , the study of kaonic atoms and kaonic nuclei has become a hot topic because of the possibility that a high - density nuclear medium is created around antikaons .
experimental data from kaonic hydrogen x - ray spectroscopy at kek@xcite have confirmed the attractive nature of the @xmath2p interaction .
this fact allows us to assume that @xmath3(1405 ) is a @xmath2p bound state due to the strong interaction . on the basis of this assumption , akaishi and
yamazaki predicted meta - stable kaon bound state formation in light nuclei using a variational calculation@xcite .
their prediction has triggered many experimental searches@xcite as well as theoretical studies@xcite .
various candidates for the deeply bound kaonic nuclear state were reported by several international collaborative experiments , but it is difficult to understand all data consistently .
to clarify the situation , a next - generation experimental search for kaonic nuclei by the in - flight @xmath0he(@xmath4 ) reaction ( e15)@xcite is performed at the hadron facility of japan proton accelerator research complex ( j - parc)@xcite . the advantage of the e15 experiment is that it simultaneously performs exclusive measurement by a missing mass study using the primary neutron and invariant mass spectroscopy via the expected decay ( @xmath5 ) . on the other hand , more precise experimental data on kaonic atoms
are necessary to completely understand the @xmath2n interaction .
they also supply extremely important information about the existence of kaonic nuclei .
therefore , precision spectroscopy of kaoinc @xmath0he atom x - rays ( e17 ) is performed at j - parc@xcite .
the purpose is to measure the 2@xmath6-level shift and the width of kaoinc @xmath0he with a precision of a few ev . because these experiments consider kaonic atoms and kaonic nuclei , their physicals motivations are closely related .
a liquid @xmath0he target is an essential component of the experimental setup for both of them .
this paper describes in detail the development and evaluation of the liquid @xmath0he target system for the j - parc e15/e17 experiments . .
the present cryogenic system for the liquid @xmath0he target is shown in figure 1 .
the characteristics and experimental requirements of the target system are as follows : 1 .
@xmath7 to liquefy @xmath0he gas , which boils at 3.2 k at 1 atm , the target system is equipped with a cryostat developed for kek - ps e471/e549 experiments@xcite that uses liquid @xmath1he decompression .
the liquid @xmath1he is transferred from an outside dewar vessel to a separator by a pumping method and from the separator to the evaporator via a needle valve that precisely controls the flow of liquid @xmath1he . in the one - shot operation mode ,
in which the liquid @xmath1he supply to the evaporator is stopped periodically after filling , the achieving temperature of the cryostat achieves a temperature of 1.3 k at a heat load of 0.16 w using a pump at a pumping speed of 120 m@xmath0/h . a heat exchanger for liquefaction of the @xmath0he gas
is placed at the bottom hole of the evaporator ; and its temperature is controlled by the pressure in the evaporator .
the heat exchanger is described in detail in the next section . to reduce the heat load by radiation at room temperature ,
the cryostat is surrounded by a thermal shield cooled by liquid n@xmath8 .
2 . @xmath9 in the e15 experimental setup , the target cell is placed at the center of the cylindrical detector system . to maintain a large acceptance in not only forward detectors but also the cylindrical detector system ,
the target cell is separated from the vertical - type cryostat .
@xmath0he condensed in the heat exchanger moves to the target cell @xmath101.2 m away through a lower pipe by gravity .
the @xmath0he vapor returns to the heat exchanger through the cell s upper pipe .
finally , the heat load at the cell is transported to the heat exchanger by circulation of liquid @xmath0he itself .
3 . @xmath11 the low - temperature target system includes many structures , including the target cell , the radiation shield , and the vacuum vessel between the target material and the detectors .
the amount of material in these structures must be minimized in order to transmit low - energy particles and decrease background events .
@xmath12 because @xmath0he gas is scarce , a gas handling system was produced to reuse it many times during development and the experimental period .
it makes it possible to store , inject , and recover @xmath0he gas without allowing it to escape to the atmosphere owing to damage along the route by sudden changes in the pressure and stress .
the gas circuit , including the gas handling system , is shown in figure 2 .
@xmath0he gas ( 400 l ) at atmospheric pressure is stored in two 200 l tanks mounted on the system . during most of the operation period ,
all the valves placed between the heat exchanger and both gas tanks are open , and they reach pressure equilibrium with vapor pressure depending on the temperature in the heat exchanger .
the gas system is oil - free because it uses a leak - tight scroll pump for recovery . as a safety measure , @xmath0he gas that leaks in the insulation vacuum
is collected once in a 1000 l balloon in the gas system .
afterwards,@xmath0he gas laced with air and oil is returned to the tanks after purification by a liquid n@xmath8 refiner .
a schematic view of the evaporator and heat exchanger is shown in figure 3 .
the cu fin structures , which have a width of 0.5 mm and distance of 0.5 mm are attached to top of the heat exchanger by silver alloy brazing .
high heat - transfer efficiency is obtained by direct contact between the fins and the liquid @xmath1he in the evaporator .
@xmath0he gas is transported from the two tanks of the gas handling system through an inlet pipe thermally anchored on top of the separator and an exhaust duct on the evaporator to the heat exchanger .
liquefied @xmath0he is transported through a pipe of 1/4 `` in diameter from the bottom of the heat exchanger to the target cell @xmath101.2 m away by gravity .
the pipe is removed by a crank structure from the flight path of beam particles and forward neutrons from the ( @xmath4 ) reaction .
the @xmath0he vapor that returns from the cell through a pipe 3/8 '' in diameter is recondensed in the heat exchanger by a siphon method .
finally , the heat load to the cell is transported to the heat exchanger by circulation of liquid @xmath0he itself .
the internal temperature of the heat exchanger and the target cell are each measured by a platinum - cobalt resistance temperature detector ( chino , r800 - 6)@xcite and calibrated ruthenium dioxide resistance temperature detector ( scientific instruments , ro600 ) with an accuracy of @xmath13 0.010 k below 1.5 k@xcite .
the cables of the thermometer placed in the cell and the heat exchanger are connected to the hermetic seal pins at the feed - through port ( kf-16 ) of the heat exchanger . during operation ,
an outlet pipe line isolated from the tanks is used to monitor the pressure in the heat exchanger using an absolute pressure transducer ( mks instruments , baratron 627b)@xcite .
the results of the temperature and pressure measurement are described in the next section .
the target cell must be designed to be pressure tight and leak - tight under a low - temperature environment in order to hold liquid @xmath0he stably .
although the cell is maintained at a pressure of less than 0.1 mpa during operation , it is designed for safe operation at an internal pressure of more than 0.3 mpa .
the allowable leak rate of the cell is less than order of 10@xmath14 pa@xmath15m@xmath0/sec by measurements using superfluid @xmath1he .
if only those requirements are considered , heavy metal such as copper , iron , and stainless steel is advantageous to producing the cell .
however , the e15 experiment requires the amount of materials to be minimized in order to increase the transmission of low - energy particles and decrease the number of background events .
in addition , because the target system is also used for precision x - ray spectroscopy , the experimental requirements were expanded to include the adoption of materials that provide high x - ray transmission ( k@xmath16-@xmath0he 3@xmath17@xmath182@xmath6 : 6.2 kev ) and the exclusion of elements that emit background x - rays ( o , fe , cu , and ti ) for use inside the solid angle of the x - ray detector .
it was quite challenging to satisfy all the requirements . by trial and error including the high - polymer material as well as metal , beryllium and aluminum were adopted as the principal materials of the target cell . a schematic view of the target cell is shown in figure 4 .
because the volume of the cell is 0.48 l , 269 l of @xmath0he gas is necessary to fill it with liquid at a temperature of 1.3 k. the cell consists of a beam window , a cylinder , a ring frame , and end cap . the beam window and end cap are made from an alloy of 62% pure beryllium and 38% pure aluminum ( brush wellman , albemet am162)@xcite .
the window and cap are both 0.6 mm thick in the beam direction . because x - ray detectors are installed around the cell perpendicular to the beam direction , a cylinder having 68 mm inside diameter , 0.3 mm thick , and 100 mm long was produced of 99.40% pure beryllium ( brush wellman , pf-60)@xcite .
the transmissivity of cu - k@xmath19 x - rays ( 8.014 kev ) of 0.305 mm thick pf-60 is 89.5% according to the catalog specifications .
the 1/4 `` pipe is attached to the lower part and the 3/8 '' pipe is attached to the upper part of an aluminum ring frame .
the temperature in the cell is monitored by the ruthenium oxide resistance temperature detector@xcite in the cell s upper pipe .
all components were joined by electron beam welding .
a cooling test for performance evaluation demonstrated that liquid @xmath1he at 1.3 k was maintained stably in the cell for few days .
the leak rate of @xmath1he was stable at less than order of 10@xmath14 pa@xmath15m@xmath0/sec during the entire test period .
thus , good leak - tight performance was confirmed without super - leaks . in the e15 experimental setup ,
the detectors are located not only on the beam line but also on the circumference of the target ; therefore , the amount of material in the radiation shield and vacuum vessel are also minimized .
the base of the shield and the vessel were produced in a tapered form to secure an effective area for the downstream detectors .
figure 5 shows a side view of the radiation shield and vacuum vessel around the target . the shield for reducing the heat load caused by radiation to the target cell and other components having temperatures less than 4.2 k
is cooled to about 80 k by liquid n@xmath8 .
the tip of the shield around the cell is made of 0.3 mm thick aluminum to reduce quantity of material .
the tip , including the cap , is cooled using the good thermal conductivity of highly - pure aluminum .
carbon fiber reinforced plastic ( cfrp ) was adopted for the cylinder around the target region of the vacuum vessel because of its light weight and high strength .
the cylinder must have 150 mm inside diameter , more than 300 mm long , and less than 1 mm thick .
in addition , for safety reasons , it must not be destroyed under an external pressure of 0.3 mpa .
however , it is difficult to design structures using cfrp because many parameters must be considered , such as the strength of the carbon fiber , strength of the plastic , direction of the fiber , and method of knitting the fiber . therefore , by using past development experience@xcite , we determined the effective young s modulus at which the safety factor became 1 when an external pressure of 0.1 mpa acted on a prepreg laminating cylinder having 300 mm inside diameter , 250 mm long , and 0.9 mm thick .
two cfrp cylinders were produced by sankyo manufacturing co , ltd.@xcite .
the specifications of the prepregs used for production , which were made by mitsubishi plastics , inc.@xcite , are given in table 1 .
ts and em represent the tensile strength and elastic modulus of the fiber direction , respectively . for the first cylinder ,
three layers of prepreg with higher rigidity , called 80tonud , were laminated in the circumferential direction ( 90@xmath20 ) and one layer was laminated in the longitudinal direction ( 0@xmath20 ) . because 80tonud has quite high rigidity ,
there was fear that the fiber is broken in the winding process .
therefore , for the other cylinder , a three laminated structure ( two layers at 90@xmath20 , one layer at 0@xmath20 ) made of 60tonud type prepreg , with slightly lower rigidity , was adopted .
both cylinders were 150 mm in inside diameter , 523 mm long , and 1 mm thick .
the design values of the safety factor for withstanding pressure were 3.8 and 3.3 , respectively .
the cylinders were glued to a beam window 0.6 mm thick and a tapered base made of aluminum with stycast1266 .
the safety of both vacuum vessels was confirmed by several pressure tests under a differential pressure of 0.1 mpa .
.specifications of prepregs [ cols="<,^,^,^,^,^,<",options="header " , ]
many cooling tests for performance evaluation were performed at kek and j - parc .
figure 6 shows the typical condensation periods of @xmath0he gas by liquid @xmath1he cooling after precooling by liquid n@xmath8 .
the upper part of the figure shows the decrease in the amount of @xmath0he gas remaining in the tanks according to pressure monitoring . from 2 to 3 h , liquefaction of the gas at a condensation speed of about 3.2 l / min appears .
subsequently , gas condensation stops at an equilibrium vapor pressure of about 0.3 atm at 2.3 k and the pressure in the tanks . to allow the remaining @xmath0he gas to liquefy after storage with liquid @xmath1he in the evaporator ,
the temperature in the heat exchanger was reduced to stop the flow of liquid @xmath1he from the separator to the evaporator .
a needle valve was closed at the time indicated by the dotted line in figure 6 .
@xmath0he gas ( 380 l , 0.95 atm ) was ultimately condensed .
furthermore , the temperature difference between the target cell and the heat exchanger disappeared at ( a ) in figure 6 .
because the temperature sensor was installed in the top of the target cell , this indicates that the cell became filled with liquid @xmath0he .
figure 7(a ) shows a typical temperature achieved without temperature control by heaters attached to the target cell and heat exchanger .
the horizontal axis represents the time normalized to 0 at the moment the needle valve closed .
the temperature of the cell dropped below 1.35 k at 1.3 h after the valve closed .
in addition , a clear difference in temperature between the heat exchanger and the target cell does not appear in figure 7(a ) .
heat transport by the siphon method clearly works well .
the histogram in figure 7(b ) was obtained by expanding the temperature data for the target cell for 8 h ( 2 - 10 h ) in figure 7(a ) to examine the stability .
although the data in figure 7(a ) were trimmed by 1/30 for easier viewing , all data , acquired at 10 s intervals , were used in order to increase the statistical precision of the histogram .
the temperature stability of the target cell was estimated to be @xmath13 0.009 k ( @xmath13 3@xmath21 ) by gaussian fitting .
the density fluctuation of 0.12% , converted from the temperature stability , is acceptable for the e15 experiment .
figure 8 shows the correlation between the temperature and the pressure in the heat exchanger and evaporator .
although the values shift slightly compared with those of the international temperature scale of 1990 ( its-90)@xcite owing to the measurement error of the pressure gauge , their behavior agrees well with that of the its-90 values .
this result strongly suggests that liquid @xmath0he existed in the heat exchanger .
because one month of consecutive operation of the target system is assumed , the consumption of liquid @xmath1he is very important for describing the direct beam time loss by liquid @xmath1he dewar exchange . to minimize the consumption of liquid @xmath1he
, the target system is operated in a one - shot mode .
figure 9 shows the time variation in the fluid volume in the evaporator . because liquid @xmath1he in the evaporator emptied for about 15 h , the evaporator was refilled with liquid @xmath1he transported from the outside dewar via separator . at that time , the valve between the tanks and the heat exchanger was closed in order to stop @xmath0he gas from returning to the tanks owing to the temperature increase in the heat exchanger to 2 k. for the remaining 15 h , the target cell was maintained at a temperature of 1.3 k by isolating the evaporator from the separator .
a heat load of 0.21 w was estimated from the decrease in the speed change in the liquid level in figure 9 .
a total liquid @xmath1he consumption rate of 50 l / day , including transfer loss and natural evaporation in the dewar , is acceptable for the e15 experiment .
lr|c + degree of vacuum & [ mbar ] & @xmath2210@xmath23 + leak rate @xmath0he system @xmath18 vacuum & [ pa@xmath15m@xmath0/sec ] & @xmath2210@xmath14 + amount of condensed @xmath0he gas & [ liter ] & 380/400 + temperature in the target cell & [ k ] & 1.30 + temperature stability in the target cell & [ k ] & @xmath13 0.01 + pressure in the target cell & [ mbar ] & 33 + heat load to the 1 k parts & [ w ] & 0.21 + liq .
@xmath1he consumption & [ liter / day ] & 50 +
we successfully developed a liquid @xmath0he target system for experimental research on deeply bound kaonic nuclear states . in a cooling test ,
380 l of @xmath0he gas were condensed in the cryogenic system . by excellent operation using the siphon method , liquid @xmath0he was maintained at 1.30 @xmath13 0.01 k in the target cell for 8 h. the target system is also superior in cost performance owing to its heat load of 0.21 w and liquid @xmath1he consumption rate of 50 l / day .
a radiation shield of thin pure aluminum and a vacuum vessel using thin cfrp cylinders were produced to minimize the amount of material around the target cell .
the specially developed target cell made of pure be and a be - al composite satisfied all the requirements ; pressure tightness , leak tightness , minimal amount of material , and high x - ray transmission .
experiments using the target system will provide extremely important results in research on kaonic atoms and kaonic nuclei .
we thank the entire staff at kek .
we are specially , grateful to the members of the kek cryogenic science center for their support of the performance evaluation test .
this developmental research was supported by riken , kek , grant - in - aid for scientific research on priority areas ( 17070005 ) , and grant - in - aid for specially promoted research ( 20002003 ) .
m. iwasaki et al . , phys .
78 ( 1997 ) , 3067 .
y. akaishi and t. yamazaki , phys . rev .
c 65 ( 2002 ) 044005 . t. yamazaki and y. akaishi , phys .
b 535 ( 2002 ) 70 . m. agnello et al . , phys . rev .
lett 94 , ( 2005 ) 212303 .
g. bendiscioli et al .
phys . a 789 ( 2007 ) 222 .
t. suzuki et al . , phys .
c 76 ( 2007 ) 068202 .
t. kishimoto et al . , prog .
phys . 118 ( 2007 ) 181 .
m. sato et al . , phys .
b 659(2008 ) 107
. h. yim et al . , phys .
b 688 ( 2010 ) 43 . t. yamazaki et al . , phys . rev .
104 ( 2010 ) 132502 .
n. v. shevchenko , a. gal , j. mare and j. revai , phys .
c 76 ( 2007 ) 044004 . y. ikeda and t. sato , phys .
c 76 ( 2007 ) 035203 .
a. n. ivanov , p. kienle , j. marton and e. widmann , nucl - th/0512037 .
t. nishikawa and y. kondo , phys .
c 77 ( 2008 ) 055202 .
s. wycech and a. m. green , phys .
c 79 ( 2009 ) 014001 .
a. dot , t. hyodo and w. weise , phys .
c 79 ( 2009 ) 014003 .
j. yamagata - sekihara et al .
c 80 ( 2009 ) 045204 .
a. arai , m. oka and s. yasui , prog .
( 2008 ) 103 .
m. iwasaki , et al .
, proposalto the j - parc 50 gev proton synchrotron ( j - parc e15 ) , 2006 .
tanaka et al . , nucl .
phys . a 835 ( 2010 ) 81 .
hayano , et al .
, proposalto the j - parc 50 gev proton synchrotron ( j - parc e17 ) , 2006 . m. sato et al .
instr . and
meth . a 606 ( 2009 )
m.l mcglashan , j. chem .
thermodyn . 22
( 1990 ) , 653 . | a liquid @xmath0he target system was developed for experimental studies of kaonic atoms and kaonic nuclei at j - parc .
@xmath0he gas is liquefied in a heat exchanger cooled below 3.2 k by decompression of liquid @xmath1he . to maintain a large acceptance of the cylindrical detector system for decay particles of kaonic nuclei , efficient heat transport between the separate target cell and the main unit
is realized using circulation of liquid @xmath0he . to minimize the amount of material , a vacuum vessel containing a carbon fiber reinforced plastic cylinder having an inside diameter of 150 mm and a thickness of 1 mm was produced . a target cell made of pure beryllium and beryllium - aluminum alloy
was developed not only to minimize the amount of material but to obtain also high x - ray transmission . during a cooling test ,
the target cell was kept at 1.3 k at a pressure of 33 mbar .
the total estimated heat load to the components including the target cell and heat exchanger cooled by liquid @xmath1he decompression , was 0.21 w , and the liquid @xmath1he consumption rate was 50 l / day .
cryogenic target , liquid @xmath0he , be , cfrp |
from 1990 to october 1995 the lep @xmath2 storage ring was operated at center - of - mass energies close to the z mass ( called lep1 program ) .
lep1 data have been collected and analysed by the 4 lep experiments aleph@xcite , delphi@xcite , l3@xcite and opal@xcite , the electroweak results and their combination are given in ref.@xcite and are still preliminary . since 1992
the sld detector is taking data on the slc @xmath2 storage ring operating at center - of - mass energies also close to @xmath3 with polarised electron beam .
the preliminary results from sld include 1992 to 1997 data @xcite . + the cross sections and the asymmetries of the reactions @xmath4 measured at lep and slc are sensitive , through radiative corrections , to the following standard model parameters : the strong coupling constant @xmath1 , the higgs mass @xmath5 and the top mass @xmath6 .
the value of @xmath1 is mainly determined by the lep total cross section measurements , while the asymmetries measured at lep and slc are most sensitive to @xmath5 .
we will concentrate on the determination of @xmath1 and therefore on the determination of the z lineshape parameters at lep .
at lep , an integrated luminosity of 110 @xmath7 per experiment has been accumulated at the z peak and about 40 @xmath7 off peak , mostly at @xmath8 gev .
about 4 million hadronic and 0.5 million leptonic z decays have been collected by each of the four lep experiments .
this large sample allows a precise determination of the z boson properties : the z mass @xmath3 , the z width @xmath9 , the total hadronic cross section at the pole @xmath10 and the ratio of hadronic to leptonic pole cross sections @xmath11 .
+ these parameters have the advantage of being almost uncorrelated : * @xmath3 is determined by the position of the peak and therefore depends on the absolute energy scale .
its uncertainty is dominated by the uncertainty on the lep energy which is about 1.5 mev@xcite .
* @xmath9 is determined by the width of the z resonance and therefore by peak and off peak relative cross section measurements .
the uncertainty on @xmath9 comes mainly from uncorrelated errors on the off peak energy measurement ( @xmath121.5 mev@xcite ) and from off peak statistics .
* @xmath10 is determined by the height of the resonance and is derived from the measurement of the hadronic cross section .
the statistical and systematic uncertainties of the selections are typically of the order of @xmath13 and @xmath14 respectively , for each experiment@xcite .
the error on @xmath10 is dominated by the theoretical uncertainty on the luminosity which is @xmath15 .
note that a recent study on the theoretical precision of the lep luminosity@xcite will lead to a reduction of this error to @xmath16 .
* @xmath17 is determined by the measurement of the leptonic and hadronic cross sections .
since @xmath17 is a ratio , the uncertainty arising from the luminosity cancels and thus the uncertainty on @xmath17 comes only from the statistical and systematic errors in the @xmath18 event selections which will be discussed in section [ sec : als ] .
c the lowest order diagrams involved in the process @xmath18 are shown in figure [ fig : eeff ] .
the cross section is the sum of the z exchange , the @xmath19 exchange and their interference @xmath20 the cross section due to the z exchange is parametrised with a breit - wigner in a model independent way using @xmath3 , @xmath9 , and the z partial widths @xmath21 @xmath22 the photon contribution is taken from qed @xmath23 where @xmath24 is the number of colours for quarks and 1 for leptons .
the interference term contains combinations of the couplings of the fermions to the z and their charge , which can not be expressed in terms of the z parameters used in equation [ eq : sig_z ] and are included in the @xmath25 parameter @xmath26 since the interference term is expected to be small it is usually set to the standard model value introducing a small model dependence in the fit .
+ the effect of initial state radiation is taken into account in the fitting procedure through its convolution with the born cross section .
.__average line shape parameters from the results of the four lep experiments .
_ _ [ cols="^,^",options="header " , ] @xmath17 and @xmath9 both allow to determine @xmath1 with the same experimental precision should decrease from @xmath27 to about @xmath28 with the decrease of the theoretical error on the luminosity@xcite but @xmath10 will still be less powerful than @xmath17 to determine @xmath1 . ] but @xmath9 varies rapidly with @xmath5 introducing an additional uncertainty if no constraint on @xmath5 is used whereas the @xmath5 dependence of @xmath17 almost cancels in the ratio of the hadronic to leptonic widths .
therefore @xmath17 is the most sensitive to @xmath1 and allows to determine @xmath1 without any assumption on the higgs mass .
+ + one should note that @xmath17 , @xmath9 and @xmath10 all depend on @xmath29 through @xmath30 and that @xmath31 is fixed to its standard model value in order to extract @xmath1 . however , since @xmath32 is very sensitive to new physics through the @xmath33 vertex corrections its value may deviate from standard model expectation .
the experimental determination of @xmath32 is shown in figure [ fig : rb ] and is in agreement within 1.8@xmath34 with the sm prediction for the direct top mass measurement @xmath35 gev/@xmath36@xcite .
this value of @xmath6 is used in the following .
+ + * the measurement of @xmath17 * + as already discussed in section [ sec : zline ] , the uncertainty on @xmath17 is dominated by the dilepton statistical and systematic experimental errors .
the preliminary measurements of the 4 lep experiments @xcite presented in jerusalem 97 are shown in figure [ fig : rl_lep ] .
the statistical uncertainty is @xmath37 for each lep experiment .
the systematic uncertainty from the hadronic channel is of the order of @xmath38 and an effort is being made to reduce the systematic error from the leptonic channel .
+ a new measurement of @xmath17 using aleph data with a global dilepton selection allows to reduce the systematic uncertainty from the leptonic channel to @xmath38 .
the dilepton selection has to be inclusive in order to be compared to the theoretical prediction .
therefore this selection includes also four fermion final state events of the type @xmath39 where @xmath40 is a lepton and v is a low invariant mass pair of fermions .
+ first , dileptons are selected with an efficiency of @xmath41 inside the detector acceptance and background from @xmath42 , @xmath43 and cosmic events is reduced to the level of @xmath44 .
all the systematic uncertainties are at the level of few @xmath45
. then the lepton flavour separation is performed inside the dilepton sample so that the systematic uncertainties are anti - correlated between 2 lepton species and that no additional uncertainty is introduced on @xmath17 .
this separation is needed in order to identify @xmath46 events for which the t - channel contribution has to be subtracted .
this subtraction is performed using theoretical calculation described in @xcite .
the theoretical error assigned to the subtraction is @xmath38 of the @xmath46 s - channel cross section@xcite .
+ the value of @xmath17 obtained with this new selection is @xmath47 the relative error on @xmath17 is reduced from @xmath48 ( 1997 value ) to @xmath49 ( this measurement ) .
+ + * determination of @xmath1 with @xmath17 * + the dependence of @xmath17 on @xmath1 is parametrised with the latest version of zfitter@xcite : @xmath50 in this parametrisation @xmath51 gev / c@xmath52 and @xmath53 gev / c@xmath52 are used .
the small dependence with the higgs and the top masses is also parametrised with zfitter @xmath54 the values of @xmath17 obtained with this parametrisation agree with zfitter prediction at the level of few @xmath55 for any value of @xmath1 in [ 0.100,0.130 ] , @xmath6 in [ 150,200 ] gev / c@xmath52 and @xmath5 in [ 60,1000 ] gev / c@xmath52 .
figure [ fig : als_rl ] shows the new aleph determination of @xmath17 and the standard model expectation as a function of @xmath1 .
the variations arising from @xmath5 and @xmath6 are also shown .
the value obtained for @xmath1 is @xmath56 where the first error comes from the experimental uncertainty on @xmath17 , the second error covers missing higher - order corrections and uncertainties in the interplay of electroweak and qcd corrections@xcite and the last error is obtained by varying the higgs mass from 60 to 1000 gev/@xmath36 .
+ the preliminary lep combined value or @xmath17 ( jerusalem 97 , see ref.@xcite ) is shown in figure [ fig : rl_lep ] and leads to @xmath57 in an overall fit using all the information from lep1 and slc data ( cross sections and asymmetries ) the value of @xmath1 and of @xmath5 are simultaneously constrained ( see section [ sec : mhiggs ] ) , the fitted value of @xmath1 is @xmath58 the experimental error is reduced from 0.004 ( in equation [ eq : fittorl ] ) to 0.003 because @xmath9 and @xmath10 are used in this last fit , and the 0.002 error from the higgs mass disappears since @xmath5 is also constrained .
this value of @xmath1 is lower than in the fit to @xmath17 alone ( equation [ eq : fittorl ] ) because @xmath9 prefers lower values of @xmath1 and the low fitted value of @xmath5 ( 66 gev / c@xmath52 ) brings @xmath1 down by @xmath120.002 ( in the fit to @xmath17 the value @xmath51 gev / c@xmath52 was used ) .
this value is in good agreement with the world average @xmath59@xcite and of comparable precision .
* the determination of @xmath60 * + at the z resonance , the most sensitive parameter to the higgs mass is @xmath60 which is determined through the measurement of the asymmetries * the lepton , b and c quarks forward - backward asymmetries @xmath61 determined at lep1 from @xmath18 angular distributions @xmath62 * the @xmath63 polarisation ( lep1 ) @xmath64 * the left - right asymmetries measured at slc with polarised electron beam @xmath65 where @xmath66 .
all these quantities are expressed in terms of the effective mixing angles of leptons , @xmath60 . the derived values of @xmath60 from the different asymmetry measurements
@xcite are compared in figure [ fig : sineff ] .
sld and lep average differ by 2@xmath34 and sld data prefer lower values of the higgs mass .
since all lep data have been analysed new measurement can only come from slc with increased statistics .
+ + * the global fit * + in the global fit the electroweak measurements are compared to standard model predictions in order to determine simultaneously the values of @xmath0 , @xmath6 and @xmath5 .
the parameters @xmath67 gev@xmath68 from muon decay and @xmath69 from @xcite are also used as input in the fit .
this global fit is performed with three data sets , results are given in table [ tbl : fit ] : 1 .
a first fit to lep data alone including the measurement of the w mass at lep 2 is made ( first column of table [ tbl : fit ] ) .
this fit shows that lep data prefer a light top and a light higgs .
2 . then lep1 and sld data and the measurement of @xmath70 from @xmath71n experiments are used to determine top quark and w masses indirectly ( second column of table [ tbl : fit ] ) .
the good agreement with the direct measurements @xmath72 gev / c@xmath52@xcite and @xmath73 gev / c@xmath52 @xcite provides a test of the standard model .
all electroweak measurements including the direct @xmath6 and @xmath74 measurement are used to obtain the best constraint on @xmath5 ( last column of table [ tbl : fit ] ) .
figure [ fig : mh ] shows the result of this fit using the latest version of zfitter and topaz0 . also shown is the curve obtained when new higher - order corrections are neglected ( ` no @xmath75 ' curve ) .
these corrections lead to a decrease of the higgs mass of @xmath76 gev / c@xmath52 .
the upper limit on @xmath5 is 215 gev / c@xmath52 at 95@xmath77 cl .
the uncertainty on @xmath78 causes an error of 0.2 on log(@xmath5 ) .
the fit is also performed with the new evaluation of @xmath78@xcite . with this value the fitted error on log(@xmath5 ) is reduced by @xmath79 ( from 0.33 to 0.25 ) .
since @xmath5 is mainly constrained by the measurement of @xmath60 , new data from slc will improve this constraint .
moreover , radiative corrections are shared between the top quark and the higgs , therefore the reduction of the error on the direct top mass measurement will improve the constraint on the higgs mass .
the w mass is also sensitive to the higgs mass and its measurement with an error of 30 mev at lep2 will also improve the constraint on @xmath5 . & lep including & all data except & all data + & lep2 @xmath74 & @xmath80 and @xmath74 & + @xmath80 & @xmath81 & @xmath82 & @xmath83 + @xmath84 & @xmath85 & @xmath86 & @xmath87 + @xmath88 & @xmath89 & @xmath90 & @xmath91 + @xmath92 & @xmath93 & @xmath94 & @xmath95 + @xmath96/d.o.f .
& @xmath97 & @xmath98 & @xmath99 + @xmath100 & @xmath101 & @xmath102 & @xmath103 + @xmath104 & @xmath105 & @xmath106 & @xmath107 + @xmath74 & @xmath108 & @xmath109 & @xmath110 +
the large sample of data accumulated at lep1 allows the determination of @xmath1 through radiative corrections , @xmath111 with a precision comparable to the world average ( in which lep1 data have not been included ) .
since all lep1 data have been analysed , this determination may only slightly improve through the reduced error on @xmath10@xcite .
+ the global fit gives for the higgs mass @xmath112 there is still room for improvement in this constraint through better top and w mass direct measurements , new data from slc and the reduction of the error on @xmath78 .
+ a low value of the higgs mass is preferred by the z resonance data : it is still possible to find the higgs at lep2 !
i would like to thank the lep electroweak working group who made the fit results available and in particular c. paus who provided me some plots .
many thanks to the quarks98 organisers for the very pleasant atmosphere at the conference .
it is a pleasure to thank j. lefranois and i. videau for careful reading and comments on this manuscript . 99 aleph collaboration , d. decamp _ et al .
_ , z. phys * c48 * ( 1990 ) 365 ; + aleph collaboration , d. decamp _ et al . _ , z. phys * c53 * ( 1992 ) 1 ; + aleph collaboration , d. buskulic _ et al .
_ , z. phys * c60 * ( 1993 ) 71 ; + aleph collaboration , d. buskulic _ et al .
_ , z. phys * c62 * ( 1994 ) 539 ; + aleph collaboration , _ preliminary results on z production cross section and lepton forward - backward asymmetries using the 1990 - 1995 data _ , contributed paper to ichep96 , warsaw , july 1996 , _ pa07 - 069 _ ; updated for eps - hep-97 , jerusalem .
delphi collaboration , p. aarnio _ et al .
_ , nucl.phys .
* b367 * ( 1991 ) 511 ; + delphi collaboration , p. abreu _ et al .
_ , nucl.phys .
* b417 * ( 1994 ) 3 ; delphi collaboration , p. abreu _ et al .
_ , nucl.phys .
* b418 * ( 1994 ) 403 ; delphi collaboration , delphi note 95 - 62 phys 497 , contributed paper to eps - hep-95 brussels , * eps0404 * ; + delphi collaboration , delphi note 97 - 130 conf 109 , contributed paper to eps - hep-97 jerusalem * eps-463*. l3 collaboration , b. adeva _ et al .
_ , z. phys .
* c51 * ( 1991 ) 179 ; + l3 collaboration , o. adriani _ et al .
_ , phys . rep .
* 236 * ( 1993 ) 1 ; + l3 collaboration , m. acciarri _ et al . _ ,
z. phys .
* c62 * ( 1994 ) 551 ; + l3 collaboration , _
preliminary l3 results on electroweak parameters using 1990 - 96 data _ , l3 note 2065 , march 1997 , available via http://hpl3sn02.cern.ch/note/note-2065.ps.gz opal collaboration , g. alexander _ et al .
_ , z. phys .
* c52 * ( 1991 ) 175 ; + opal collaboration , p.d .
acton _ et al .
_ , z. phys .
* c58 * ( 1993 ) 219 ; + opal collaboration , r. akers _ et al .
_ , z. phys .
* c61 * ( 1994 ) 19 ; + opal collaboration , opal physics notes pn142(1994 ) , pn166(1995 ) , pn242(1996 ) ; + opal collaboration , _ measurements of lepton pair asymmetries using the 1995 data _ , contributed paper to ichep96 , warsaw , 25 - 31 july 1996 * pa07 - 015 * ; + opal collaboration , _ a preliminary update of the z lineshape and lepton asymmetry measurements with the 1995 data _ , opal physics note pn286 , march 1997 .
the lep collaborations aleph , delphi , l3 , opal , the lep electroweak working group and the sld heavy flavour and electroweak groups , _ a combination of preliminary electroweak measurements and constraints on the standard model _ , internal note , lepewwg/98 - 01 , may 1998 .
g. crawford , _ precision electroweak tests recent results from sld _ , xxxiiird rencontres de moriond , les arcs , france , 14 - 21 march 1998 . s. jadach , m. melles , b.f.l .
ward and s.a .
yost , _ new results on the theoretical precision of the lep / slc luminosity _ , uthep-98 - 0501 , may 1998 .
reports of the working group on precision calculations for the z resonance , eds .
d. bardin , w. hollik and g. passarino , cern yellow report 95 - 03 , geneva , march 1995 .
zfitter : d. bardin _ et al .
_ , z. phys .
* c44 * ( 1989 ) 493 ; comp .
. comm . * 59 * ( 1990 ) 303 ; nucl .
* b351 * ( 1991 ) 1 ; phys .
* b255 * ( 1991 ) 290 and cern - th 6443/92 ( may 1992 ) ; + topaz0 : g. montagna , o. nicrosini , g. passarino , f. piccinnii and r. pittau , nucl .
* b401 * ( 1993 ) 3 ; comp .
comm * 76 * ( 1993 ) 328 .
+ these computer codes have been recently updated with results from @xcite , @xcite , @xcite and @xcite .
a. czarnecki and j. k@xmath113hn , phys .
* 77 * ( 1996 ) 3955 ; + r. harlander , t. seidensticker and m. steinhauser hep - ph/9712228 . g. degrassi , p. gambino and a. vicini phys .
* b383 * ( 1996 ) 219 ; + g. degrassi , p. gambino and a. sirlin phys .
* b394 * ( 1997 ) 188 . s. larin , t.v .
ritbergen and j. vermaseren , phys .
* b400 * ( 1997 ) 379 , * b405 * ( 1997 ) 327 ; + k. chetyrkin , b. kniehl and m. steinhauser , phys .
* 79 * ( 1997 ) 2184 , hep - ph/9708255 .
m. jones , _ tevatron top mass measurements _ , xxxiiird rencontres de moriond , les arcs , france , 14 - 21 march 1998 .
w. beenakker , f.a .
berends and s.c .
van der marck , _ large angle bhabha scattering _ , nucl .
* b349 * ( 1991 ) 223 .
w. beenakker and g. passarino , _ large angle bhabha scattering at lep1 _ , phys . lett .
* b425 * ( 1998 ) 199 .
t. hebbeker , m. martinez , g. passarino and g. quast , phys lett .
* b331 * ( 1994 ) 165 ; + p.a .
raczka and szymacha , phys .
* d54 * ( 1996)3073 ; + d.e .
soper and l.r .
surguladze , phys .
rev * d54 * ( 1996 ) 4566 .
barnett _ et al .
_ , phys .
* d54 * ( 1996 ) 1 . s. eidelmann and f. jegerlehner , z. phys * c67 * ( 1995 ) 585 .
kim , talk presented at the lepton photon symposium 1997 , hamburg , 28 july-1 aug , 1997 , to appear in the proceedings .
m. davier and a. h@xmath114cker , _ improved determination of @xmath115 and the anomalous magnetic moment of the muon _ ,
lal 97 - 85 , hep - ph/9711308 , nov 1997 . s. de jong , talk presented at the xxxiiird rencontres de moriond , les arcs , france , 15 - 21 march 1998 . | the determination of the z lineshape parameters at lep1 is presented and the value of @xmath0 is derived from these measurements . the constraint on the higgs mass obtained from a global fit to lep1 and slc data is also given . 2.0 cm lal 98 - 67 + 31st august 1998 1.0 cm * extraction of @xmath1 and constraint on the higgs mass from electroweak fits at the z resonance * 1.0 cm edwige tournefier + laboratoire de lacclrateur linaire + in2p3-cnrs et universit de paris - sud , + bp34 , f-91898 orsay cedex 1.5 cm _ talk given at the 10th international seminar + `` quarks 98 '' + suzdal , russia , may 18 - 24 , 1998 _ |
in a standard scenario , big bang nucleosynthesis ( bbn ) can produce only light elements , up to @xmath0li , and all heavy elements have been synthesized in stars . however , many phase transitions in the early universe could have printed their trace in a non - standard way . for example , some baryogenesis models@xcite predict very high baryon density islands in ordinary low density backgrounds . in the previous paper@xcite
, we studied heavy element production in inhomogeneous bbn from this point of view .
however we limited ourselves to the heavy element abundance and did not discuss about the light element abundance and consistency with observations .
this is because we assumed that high baryon density region is very local and do not affect the global light element abundance . in @xcite , rauscher pointed out that in order not to contradict to observations , the high baryon density region must be very small and can not affect the present heavy element abundance . in this paper , we show that there is a parameter region in which the heavy element can be produced enough to affect observation while keeping the light element abundance consistent with observations .
we consider that the disagreement between rauscher s opinion and our opinion comes from two points .
one is that we are looking at some parameter regions in which neutrons in high baryon density do not diffuse so much as to cause disaster in standard bbn .
we would like to emphasize this point .
the other is that the relevant quantity is not the spatial size of the high baryon density region but the amount of baryon in high density regions .
we will discuss the following issues : in section [ lea ] , we discuss the light element abundance in the homogeneous high baryon density region and after mixing the high and the low baryon density region . in section [ heavy ] , we study the heavy element(ru , mo ) abundance in high and averaged baryon density and show that heavy elements can be produced enough without contradicting the light element observation . in section
[ diff ] , we briefly comment on the diffusion scale of the high baryon density region .
we calculate homogeneous bbn with various values of @xmath1(baryon photon ratio ) . in table.[light - table ] and [ light - normal ] , we show the numerical result of the mass fraction and the number fraction of each light element for @xmath2 and @xmath3 .
ccc + name & mass fraction & number fraction + h & @xmath4 & @xmath5 + @xmath6he & @xmath7 & @xmath8 + @xmath9he & @xmath10 & @xmath11 + @xmath0li+@xmath0be & @xmath12 & @xmath13 + d & @xmath14 & @xmath15 + ccc + name & mass fraction & number fraction + h & @xmath16 & @xmath17 + @xmath6he & @xmath18 & @xmath19 + @xmath9he & @xmath20 & @xmath21 + @xmath0li + @xmath0be & @xmath22 & @xmath23 + d & @xmath24 & @xmath25 + as baryon density becomes higher , more protons and neutrons are bounded to form @xmath6he . at @xmath2 ,
most of the final product of @xmath26 comes from @xmath27 which decays to @xmath26 after bbn .
details on light element production for various @xmath1 can also be found in @xcite . in this paper
we almost concentrate on a case in which the high baryon density region has @xmath2 .
we expect that compared to @xmath28 , the profile of the abundance for @xmath2 is more different from standard bbn because most of the light element abundances change monotonically with respect to @xmath1 and if this case does not contradict to observations , other cases would also be consistent .
briefly , the amount of @xmath29 decreases and @xmath30 increases monotonically as @xmath1 become larger .
the number fraction of @xmath31 is less than @xmath32 for @xmath1 greater than @xmath33 . for @xmath34 ,
the number fraction drastically decreases around @xmath35 down to @xmath36 , and for @xmath26 , the number fraction increases until @xmath37 and drastically decreases for a larger value of @xmath1 . in the following sections , we will see that this non - standard setup does not strongly contradict to the observations . for simplicity
we ignore the diffusion effect before and during bbn , and after bbn both high and low baryon density regions are completely mixed .
detailed analysis such as the case in which the high baryon density region does nt completely mixed , or taking into account diffusion effects are left for future work . in this section ,
we summarize the relations among parameters .
notations : @xmath38 , @xmath39 , @xmath40 are averaged , high , and low baryon number density . @xmath41 , @xmath42 are the volume fractions of the high and the low baryon density region .
@xmath43 , @xmath44 , @xmath45 are the mass fractions of each element ( i ) in averaged- , high- and low - density regions .
the basic relations are @xmath46 under the assumption that the temperature of the universe is homogeneous , the above equation can be written as @xmath47 where @xmath48,@xmath49 conventional parameters for inhomogeneous bbn are @xmath1 , @xmath50 and density ratio @xmath51 . here we use a different combination of parameters .
relevant values for the abundance analysis are products @xmath52 and @xmath53 .
@xmath54 determines the amount of baryon from high- and low- density regions .
@xmath53 determines the mass fraction of each species of nuclei . for convenience ,
we write the ratio of baryon number contribution from high density region as @xmath55 , i.e. , @xmath56 .
there are 5 parameters(@xmath57 and @xmath58 ) and 2 constraints ( eq.([vol ] ) and eq.([density ] ) ) .
we calculate the light element abundance for various values of @xmath53 .
@xmath1 can also take any value , but in order not to contradict observational constraints , we choose @xmath1 from @xmath3 to @xmath59 .
@xmath55 is determined by eq([denseta ] ) .
the aim of the analysis in this section is not to find parameter regions which precisely agree with the observational light element abundance and @xmath1 from cmb .
our model is too simple to determine the constraints to parameters .
for example , we completely ignore the diffusion effect before and during bbn .
instead we see that at least our analysis in previous paper is physically reasonable .
we consider the cases of @xmath60 and @xmath61 .
the mass fractions of and @xmath29 and @xmath34 in the high density region are @xmath62 and @xmath63 , respectively , while those in the low density region are @xmath64 and @xmath65 . from eq.([speceta ] ) , we have @xmath66 @xmath67 we can calculate an averaged value of the abundance ratio of @xmath68he to h as @xmath69 where @xmath55 is related to @xmath1 as @xmath70 here @xmath55 varies from 0 to 0.9 for reasonable values of @xmath1 , or @xmath71 . similarly , for @xmath60 the number fractions are @xmath72 @xmath73 fig.[deps],[heeps ] and [ lieps ] represent the averaged abundance ratio , ( d / h ) , ( @xmath9he / h ) and ( @xmath0li / h ) respectively . , width=340 ] but for ( @xmath68he / h),width=340 ] but for ( @xmath0li / h),width=340 ] we can see that the light element abundance is the same order around @xmath74 as observations @xcite .
@xmath75 we do not discuss detail about diffusion here .
but at least above result suggest that our analysis is not beside the point .
the same analysis can be applied for heavy elements such as @xmath78 , @xmath79 , @xmath80 and @xmath81 .
we are interested in these elements because in many models of supernovae nucleosynthesis , these p - nuclei are less produced .
we will see that some amount of these heavy elements can be synthesized in bbn .
cc + name & mass fraction + h & @xmath82 + @xmath6he & @xmath83 + @xmath84mo & @xmath85 + @xmath86mo & @xmath87 + @xmath88ru & @xmath89 + @xmath90ru & @xmath91 + [ heavy ] from table.[heavy ] , we can derive the expected value of these elements .
@xmath92 @xmath93 @xmath94 @xmath95 we plot expected value of these quantities in fig.[combps ] .
mo / h),(@xmath86mo / h),(@xmath88ru / h ) and ( @xmath90ru / h ) vs @xmath1 .
red , green , blue and pink lines represent the ratio ( @xmath84mo / h),(@xmath86mo / h),(@xmath88ru / h),(@xmath90ru / h ) respectively.,width=340 ] these values should be compared with the solar abundance(table.[solar])@xcite .
.the abundances of @xmath76mo and @xmath77ru in the solar system@xcite [ cols="^,^,^",options="header " , ] compared those observational values with fig.[combps ] , it is clear that the heavy element produced in bbn can affect the solar abundance heavy element . some of them are produced too much .
but this is not a problem of the previous work @xcite , because we assumed that high density regions are very small and do not disturb standard bbn .
the analysis here suggest that even if we assume the density fluctuations are completely mixed , heavy element can have enough affect to the solar abundance .
in the previous analysis , we assumed that the diffusion effect can be ignored during bbn and both high density regions and low density regions are completely mixed after bbn . in this section , we determine the scale of high baryon density island in which the diffusion effect during bbn is very small enough and our assumption is valid .
we do not discuss the diffusion after bbn here . a detail analysis of the comoving diffusion distance of the baryon , the neutron and the proton is in @xcite .
from fig.1 in @xcite , in order to safely ignore the diffusion effect , it is necessary for the high baryon density island to be much larger than @xmath96 cm at t=0.1mev(@xmath97k ) .
notice that @xmath98 , where a is a scale factor
. for scale d now corresponds to @xmath99 at bbn epoch .
present galaxy scale is @xmath100 cm , which corresponds to @xmath101 cm @xmath102 cm at bbn epoch .
cc + temperature & scale + @xmath103k ( bbn ) & d + 3000k ( decouple ) & @xmath104 + 2.725k ( now ) & @xmath105 + the maximum angular resolution of cmb is @xmath1062000 .
the size of universe is @xmath107mpc . in order not to contradict to cmb observation
, the fluctuation of baryon density must be less than @xmath108mpc now .
this corresponds to @xmath109 cm at bbn .
since the density fluctuation size in dolgov and silk s model@xcite is a free parameter , the above brief estimation suggests that we can take the island size large enough to ignore the diffusion effect without contradicting to observations , i.e. , the reasonable size of @xmath96 cm @xmath110@xmath109 cm at the bbn epoch
. we can choose distances between high density islands so that we obtain a suitable value of @xmath50 .
in this paper , we studied the relation between the heavy element production in high baryon density regions during bbn and the light element observation . by averaging the light element abundances in the high and the low density regions we showed that it is possible to produce a relevant amount of heavy element without contradicting to observations .
however we should stress that in this paper we restricted ourselves to some parameter regions where neutrons in high baryon density regions do not destroy the standard bbn .
so our setup is different from the conventional inhomogeneous bbn studies .
we also studied the size of the density fluctuation to show that there is a parameter region in which the neutron diffusion is negligible and which is much smaller than cmb observation scale .
it is worthwhile to investigate further how the produced heavy elements can be related to the detailed observations .
we thank r.h .
cyburt , r. allahverdi and r. nakamura for useful discussions .
this research was supported in part by grants - in - aid for scientific research provided by the ministry of education , science and culture of japan through research grant no.s 14102004 , no.14079202 .
s.m.s work was supported in part by jsps(japan society for the promotion of science ) .
a. dolgov and j. silk , baryon isocurvature fluctuations at small scales and baryonic dark phys .
d * 47 * , 4244 ( 1993 ) .
s. matsuura , s. i. fujimoto , s. nishimura , m. a. hashimoto and k. sato , phys .
d * 72 * , 123505 ( 2005 ) [ arxiv : astro - ph/0507439 ] . jd barrow and
j morgan , mon.not.roy .
. soc . * 203 * , 203 ( 1983 ) ; j.h .
applegate , c.j .
hogan , and r.j .
scherrer , phys . rev .
* d35 * , 1151 ( 1987 ) ; c. alcock , g.m
. fuller , and g.j .
mathews , astrophys .
j. * 320 * , 439 ( 1987 ) ; r.m .
malaney and w.a .
fowler , astrophys .
j * 333 * , 14 ( 1988 ) ; h. kurki - suonio , r. a. matzner , j. m. centrella , t. rothman , and j. r. wilson , phys .
. * d38 * , 1091 ( 1988 ) ; n. terasawa and k. sato , phys . rev .
* d39 * , 2893 ( 1989 ) ; n. terasawa and k. sato , prog .
81 ( 1989 ) , 254 - 259 ; n. terasawa and k. sato , prog .
81 ( 1989 ) , 1085 - 1090 ; h. kurki - suonio and r. matzner , phys.rev .
* d39 * , 1046 ( 1989 ) ; h. kurki - suonio and r. matzner , phys.rev . * d42 * , 1047 ( 1990 ) ; h. reeves , j. richer , k. sato and n. terasawa , astrophys .
j. 355 ( 1990 ) , 18 - 28 ; n. terasawa and k. sato , astrophys .
j. 362 ( 1990 ) l47-l49 ; g. j. mathews , b. s. meyer , c. r. alcock , and g. m. fuller , astrophys .
j. * 358 * , 36 ( 1990 ) ; b. s. meyer , c. r. alcock , g. j. mathews , and g. m. fuller , phys.rev . * d43 * , 1079 ( 1991 ) ; d. thomas , d. n. schramm , k.a .
olive , g. j. mathews , b. s. meyer , and b. d. fields , astrophys .
j. * 430 * , 291 ( 1994 ) ; k. jedamzik , g. m. fuller , g. j. mathews , and t. kajino , astrophys . j. * 422 * , 423 ( 1994 ) ; k. jedamzik , g. m. fuller , and g. j. mathews , astrophys .
j. * 423 * 50 ( 1994 ) ; g. j. mathews , t. kajino , and m. orito , astrophys .
j. * 456 * , 98 ( 1996 ) ; h. kurki - suonio , k. jedamzik , and g. j. mathews , astrophys .
j. * 479 * 31 ( 1997 ) ; in - saeng suh and g. j. mathews , phys . rev . * d58 * , 123002 ( 1998 ) ; k. kainulainen , h. kurki - suonio , and e. sihvola , phys . rev . * d59 * , 083505 ( 1999 ) ; k. jedamzik , and j.b .
rehm , phys .
rev . * d64 * , 023510 ( 2001)[astro - ph/0101292 ] .
t. rauscher , h. applegate , j. cowan , f. thielmann , and m. wiescher , astrophys .
j. * 429 * , 499 ( 1994 ) ; s. matsuura , a. d. dolgov , s. nagataki and k. sato , prog .
* 112 * , 971 ( 2004 ) [ arxiv : astro - ph/0405459 ] .
r. v. wagoner , w. a. fowler and f. hoyle , astrophys .
j. * 148 * , 3 ( 1967 ) .
b. fields and s. sarkar , arxiv : astro - ph/0406663 .
d. kirkman , d. tytler , n. suzuki , j. m. omeara and d. lubin , the cosmological baryon density from the deuterium to hydrogen ratio astrophys .
j. suppl .
* 149 * , 1 ( 2003 ) [ arxiv : astro - ph/0302006 ] . j. m. omeara , d. tytler , d. kirkman , n. suzuki , j. x. prochaska , d. lubin and a. m. wolfe , the deuterium to hydrogen abundance ratio towards a fourth qso : astrophys .
j. * 552 * , 718 ( 2001 ) [ arxiv : astro - ph/0011179 ] .
d. kirkman , d. tytler , s. burles , d. lubin and j. m. omeara , qso 0130 - 4021 : a third qso showing a low deuterium to hydrogen abundance arxiv : astro - ph/9907128 .
j. l. linsky , arxiv : astro - ph/0309099 . s. g. ryan , j. e. norris and t. c. beers , astrophys . j. * 523 * , 654 ( 1999 ) [ arxiv : astro - ph/9903059 ] .
bonifacio _ et al .
_ , arxiv : astro - ph/0204332 . m. h. pinsonneault , g. steigman , t. p. walker and v. k. narayanans , astrophys . j. * 574 * , 398 ( 2002 ) [ arxiv : astro - ph/0105439 ] . j. h. applegate , c. j. hogan and r. j. scherrer , phys . rev .
d * 35 * , 1151 ( 1987 ) .
e. anders and n. grevesse , geochim .
cosmochim .
acta * 53 * , 197 ( 1989 ) . | this is a reply report to @xcite .
we studied heavy element production in the high baryon density region in the early universe@xcite .
however it is claimed by @xcite that a small scale but high baryon density region contradicts observations for the light element abundance or , in order not to contradict to the observations the high density region must be so small that it can not affect the present heavy element abundance . in this paper
we study big bang nucleosynthesis in the high baryon density region and show that in certain parameter spaces it is possible to produce enough amount of the heavy element without contradiction to cosmic microwave background and light element observations . |
long - standing interest in modelling effective elastic properties of composites with microstructure has substantially intensified with the emerging possibility of designing periodic structures in air @xcite and in solids @xcite to form phononic crystals and other exotic metamaterials , which open up exciting application prospects ranging from negative index lenses to small scale multiband phononic devices@xcite .
this new prospective brings about the need for fast and accurate computational schemes to test ideas _ in silico_.
the most common numerical tool is the fourier or plane - wave expansion method ( pwe ) .
it is widely used for calculating various spectral parameters including the effective quasistatic speed of sound in acoustic @xcite and elastic qn phononic crystals . at the same time , the pwe calculation is known to face problems when applied to high - contrast composites @xcite , which are of especial interest for applications .
particularly riveting is the case where a soft ingredient is embedded in a way breaking the connectivity of densely packed regions of stiff ingredient . physically speaking ,
the speed of sound , which is large in a homogeneously stiff medium , should fall dramatically when even a small amount of soft component forms a quasi - insulating network. note that this case , which implies a strong effect of multiple interactions , is also ungainly for the multiple - scattering approach @xcite .
the purpose of present letter is to highlight a new method for evaluating the quasistatic effective sound speed @xmath0 in 2d and 3d phononic crystals .
the idea is to recast the wave equation as a 1st - order ordinary differential system ( ods ) with respect to one coordinate ( say @xmath1 ) and to use a monodromy - matrix operator defined as a multiplicative ( or path ) integral in @xmath1 . by this means , we derive a formula for @xmath0 whose essential advantages are an explicit integration in @xmath1 and an exponentially small error of truncation in other coordinate(s ) .
both these features of the analytical result are shown to significantly improve the efficiency and accuracy of its numerical implementation in comparison with the conventional pwe calculation , which is demonstrated for a 2d steel / epoxy square lattice .
the power of the new approach is especially apparent at high concentration @xmath2 of steel inclusions , where the effective speed @xmath0 displays a steep , near vertical , dependence for @xmath3 , a feature not captured by conventional techniques like pwe .
* a. setup . * consider the scalar wave equation @xmath4for time - harmonic shear displacement @xmath5 in a 2d solid continuum and @xmath6 for solids by @xmath7 and @xmath8 for fluids . ] with @xmath9-periodic density @xmath10 and shear coefficient @xmath11 .
assume a square unit cell @xmath12 ^{2}$ ] with unit translation vectors @xmath13 taken as the basis for @xmath14 imposing the floquet condition @xmath15 where @xmath16 is periodic and @xmath17 ( @xmath18 ) , eq .
( [ 0 ] ) becomes@xmath19regular perturbation theory applied to ( [ 501 ] ) yields the effective speed @xmath20 in the form @xcite@xmath21where @xmath22 , spatial averages are defined by @xmath23and @xmath24 denotes the scalar product in @xmath25 so that @xmath26 ( @xmath27 means complex conjugation ) .
the difficulty with ( [ 6 ] ) is that it involves the inverse of a partial differential operator @xmath28 .
one solution is to apply a double fourier expansion to @xmath29 and @xmath30 in ( [ 6 ] ) .
this leads to the pwe formula for the effective speed@xcite which is expressed via infinite vectors and the inverse of the infinite matrix of fourier coefficients of @xmath11 .
numerical implementation of the pwe formula requires dealing with large dense matrices , especially in the case of high - contrast composites for which the pwe convergence is slow ( see [ n ] ) .
an alternative `` brute force '' procedure of the scaling approach is to numerically solve the partial differential equation @xmath31 for the @xmath32-periodic function @xmath33 ( e.g. via the boundary integral method@xcite ) . the new approach proposed here leads to a more efficient formula for @xmath0 based on direct analytical integration in one coordinate direction .
there are two ways of doing so .
the first proceeds from the ods form of the wave equation ( [ 0 ] ) itself , which means skipping ( [ 6 ] ) .
this is convenient for deriving @xmath34 in the principal directions @xmath35 see iib .
the second method is more closely related to the conventional pwe and scaling approaches in that it also starts from ( [ 6 ] ) but treats it differently , namely , the equation @xmath31 is cast in ods form and analytically integrated in one coordinate .
this is basically equivalent to the former method , but enables an easier derivation of the off - diagonal component @xmath36 for the anisotropic case , see iic . * b. wave speed in the principal directions . * the wave equation ( 0 ) may be recast as @xmath37where @xmath38 stands for @xmath39 .
the solution to eq .
( [ m3 ] ) for initial data @xmath40 at @xmath41 is @xmath42 \pmb{\eta } % ( 0,\cdot ) \
\mathrm{with } \notag \\ \mathcal{m}\left [ a , b\right ] = % \begin{matrix } \widehat{\int } _ { b}^{a}% \end{matrix}% ( \mathcal{i}+\mathcal{q}\mathrm{d}x_{1 } ) .
\label{m5}\end{gathered}\]]the operator @xmath43 $ ] is formally the matricant , or propagator , of ( [ m3 ] ) defined through the multiplicative integral @xmath44 ( with @xmath45 denoting the identity operator ) .
it is assumed for the moment that @xmath10 and @xmath11 are smooth to ensure the existence of @xmath46 .
the matricant over a period , @xmath47 $ ] , is called the monodromy matrix .
assume the floquet condition with the wave vector @xmath48 so that @xmath49 and @xmath50 @xmath51 by ( [ m5])@xmath52 , this implies the eigenproblem @xmath53 \mathbf{w}(k_{1})=e^{ik_{1}}\mathbf{w}(k_{1 } ) .
\label{2}\]]where @xmath47 $ ] depends on @xmath54 .
( [ 2 ] ) defines @xmath55 and hence @xmath56 where @xmath57 is the eigenvalue of ( [ 0 ] ) with @xmath58 .
the effective speed @xmath59 can therefore be determined by applying perturbation theory to ( [ 2 ] ) as @xmath60 . the asymptotic form of @xmath47 $ ] follows from definitions ( [ m3 ] ) and ( [ m5])@xmath61 as @xmath62 = \mathcal{m}_{0}+\omega ^{2}\mathcal{m}% _ { 1}+o(\omega ^{4})\ \mathrm{where : } \notag \\
\mathcal{m}_{0}\left [ 1,0\right ] , \
\mathcal{m}_{0}% \left [ a , b\right ] = % \begin{matrix } \widehat{\int } _ { b}^{a}% \end{matrix}% ( \mathcal{i}+\mathcal{q}_{0}\mathrm{d}x_{1})\ \mathrm{with}\ \notag \\ \mathcal{q}_{0}\equiv \mathcal{q}_{\omega = 0}=% \begin{pmatrix } 0 & \mu ^{-1 } \\
\mathcal{a } & 0% \end{pmatrix}% , \label{m6 } \\ \mathcal{m}_{1}=\int_{0}^{1}\mathcal{m}_{0}\left [ 1,x_{1}\right ] \begin{pmatrix } 0 & 0 \\ -\rho & 0% \end{pmatrix}% \mathcal{m}_{0}\left [ x_{1},0\right ] \mathrm{d}x_{1}. \notag\end{gathered}\]]note the identities @xmath63 and hence@xmath64 \mathbf{w}_{0}=\mathbf{w}_{0},\ \mathcal{m}% _ { 0}^{+}\left [ a , b\right ] \widetilde{\mathbf{w}}_{0}=\widetilde{\mathbf{w}}% _ { 0}\mathbf{\ } \ ( \forall a , b ) \notag \\
\mathrm{for}\ \mathbf{w}_{0}=(1\ 0)^{\mathrm{t}},\ \widetilde{\mathbf{w}}% _ { 0}=(0\ 1)^{\mathrm{t}}. \label{1}\end{gathered}\]]by ( [ 1])@xmath65 @xmath66 is an eigenvector of @xmath67 with the eigenvalue 1 , and it can be shown to be a single eigenvector . therefore @xmath68 and@xmath69 .
insert these expansions along with ( [ m6])@xmath70 in ( [ 2 ] ) and collect the first - order terms in @xmath71 to obtain @xmath72according to ( [ 1 ] ) , @xmath73 has no inverse but is a one - to - one mapping from the subspace orthogonal to @xmath66 onto the subspace orthogonal @xmath74 hence , @xmath75 exists and @xmath76 is uniquely defined .
the terms of second - order in @xmath71 in ( [ 2 ] ) then imply @xmath77scalar multiplication on both sides by @xmath78 leads , with account for ( [ 1 ] ) and ( [ m6])@xmath79 , to @xmath80 , whence by ( [ 3]@xmath61 ) @xmath81where the notation @xmath82 is explained in ( [ 6.1 ] ) .
interchanging variables @xmath83 in the above derivation yields a similar result for @xmath84 as follows @xmath85the result for a rectangular lattice with @xmath86 \times \left [ 0,t_{2}\right ] $ ] is obtained by replacing @xmath87 with @xmath88 * c. the full matrix * @xmath89 .
the anisotropy of the effective speed @xmath90 i.e. its dependence on the wave normal @xmath91 , is determined by the quadratic form @xmath92 ( see eq .
( [ 6 ] ) ) , and represented by the ellipse of ( squared ) slowness @xmath93 . eqs .
( [ 5 ] ) and ( [ f6])@xmath70 , which define @xmath94 and so @xmath95 , suffice for the case where @xmath9 is rectangular and @xmath11 is even in ( at least ) one of @xmath87 so that the effective - slowness ellipse is @xmath96 with the principal axes parallel to @xmath13 .
otherwise @xmath97 for arbitrary @xmath98 requires finding the off - diagonal component @xmath36 . for this purpose , with reference to ( [ 6 ] ) , consider the equation @xmath99for @xmath32-periodic @xmath33 . with the above notations this can be written as @xmath100 or , more conveniently , @xmath101 with @xmath102 .
the latter is equivalent to @xmath103and @xmath104 is given in ( [ m6])@xmath105 .
the general solution to ( [ 7 ] ) is @xmath106 \pmb{\xi } % ( 0,\cdot ) , \label{8}\]]where @xmath107 $ ] is defined in ( [ m6])@xmath61 , and @xmath108 is the initial data at @xmath41 .
the periodicity of @xmath109 implies @xmath110 while @xmath111 by ( [ 8 ] ) .
hence @xmath112 and so ( [ 7.0 ] ) is solved by @xmath106 ( \mathcal{m}% _ { 0}-\mathcal{i})^{-1}\mathbf{w}_{0}. \label{9}\]]substituting ( [ 9 ] ) into the definition of @xmath36 in ( [ 6 ] ) yields @xmath113(\mathcal{m}_{0}-\mathcal{i})^{-1}\mathbf{w}_{0}\right\rangle .
\label{f5}\end{aligned}\ ] ] note that the formula ( [ f5 ] ) for @xmath36 requires more computation than the formulas ( [ 5 ] ) and ( [ f6])@xmath70 for @xmath95 .
interestingly , if the unit cell @xmath9 is square , then , for an arbitrary ( periodic ) @xmath114 , eq .
( [ f5 ] ) can be circumvented by using the identity @xmath115 , where @xmath116 follow from eqs .
( [ 5 ] ) and ( [ f6])@xmath70 applied to the square lattice obtained from the given one by turning it 45@xmath117 .
* d. discussion .
* the two lines of attack outlinedmentioned in ii.a are equivalent in that the formula ( [ 5 ] ) for the effective speed @xmath118 in the principal direction can also be inferred from eq .
( [ 6 ] ) . inserting
the solution ( [ 9 ] ) of ( [ 7.0 ] ) defines the component @xmath119 as @xmath120integrating by parts each term in the last identity and using the periodicity of @xmath11 along with eqs .
( [ m6])@xmath105 , ( [ 1 ] ) , ( [ 7])-([9 ] ) ( see also the notation ( [ 6.1 ] ) ) yields @xmath121thus , @xmath122@xmath123@xmath124 which leads to ( [ 5 ] ) , qed .
note that eq .
( [ f5 ] ) is also obtainable via the monodromy matrix of the wave equation ( [ 0 ] ) ( the approach of iib ) with @xmath15 and @xmath125 but this method of derivation of @xmath36 is lengthier than in iic . as another remark
, it is instructive to recover a known result for the case where @xmath11 is periodic in one coordinate and does not depend on the other , say @xmath126 . using ( [ m6])@xmath61 , ( [ m6])@xmath105 and ( [ f6])@xmath105 gives @xmath127therefore , by ( [ 5 ] ) and ( [ f6])@xmath70 , @xmath128 and @xmath129 while @xmath130 by ( [ f5 ] ) with @xmath131 . finally , we note that , while the above evaluation of quasistatic speed @xmath0 is exact , using the same monodromy - matrix approach also provides a closed - form approximation of @xmath132 for the isotropic case , it is as follows ( see @xcite for more details):@xmath133
the equation for time - harmonic elastic wave motion @xmath134@xmath135@xmath136 is , with repeated suffices summed , @xmath137where density @xmath10 and compliances @xmath138 are @xmath9-periodic in a 3d periodic medium .
assume a cubic unit cell @xmath139 ^{3}$ ] and refer the components @xmath140 and @xmath141 to the orthogonal basis formed by the translation vectors @xmath142 .
impose the condition @xmath143 with periodic @xmath144 and take @xmath145 parallel to one of @xmath146 , e.g. to @xmath147 .
( [ 3d1 ] ) may be rewritten in the form @xmath148where the self - adjoint matrix operators @xmath149 and @xmath150 are @xmath151like in the 2d case , denote the monodromy matrix for ( [ 3d2 ] ) at @xmath152 by @xmath153 where @xmath154 , and also introduce the 6@xmath1553 matrices @xmath156 and @xmath157 .
reasoning similar to that in
ii.c leads us to the conclusion that the effective speeds @xmath158 @xmath159 of the three waves with @xmath160 parallel to @xmath147 are the eigenvalues of the 3@xmath1553 matrix @xmath161
there are several ways to use the above analytical results for calculating the effective speed .
one approach is to transform to fourier space with respect to coordinate(s ) other than the coordinate of integration in the monodromy matrix .
consider the 2d case and apply the fourier expansion @xmath162 in @xmath163 for the functions @xmath164 and @xmath165 .
then the operator of multiplying by the function @xmath166 and the differential operator @xmath167 become matrices@xmath168and eq .
( [ 5 ] ) reduces to following form @xmath169where @xmath170const . for any @xmath98 in the isotropic case .
the above vectors and matrices are , strictly speaking , of infinite dimension , which needs to be truncated for numerical purposes . in this sense there is no loss of generality in assuming a smooth @xmath11 in the course of derivations in [ 2d ] .
implementation of eq .
( [ n3])@xmath171 consists of two steps
. * step 1 . *
calculate the multiplicative integral ( [ n3])@xmath172 defining @xmath173 . for an arbitrary @xmath174 , one way is to use a discretization scheme . divide the segment @xmath175 $ ] into @xmath176 intervals @xmath177 , @xmath178 , of small enough length .
calculate @xmath179 fourier coefficients @xmath180 , @xmath181 and the @xmath182 matrices @xmath183 for each @xmath184 and then use the approximate formula @xmath185 $ ] . recall that @xmath186 satisfies the chain rule and is exactly equal to @xmath187 for @xmath188 if @xmath174 does not depend on @xmath1 within @xmath189 .
therefore the calculation is much simpler in the common case of a piecewise homogeneous unit cell with only a few inclusions of simple shape ( see the example below ) .
* step 2 . *
solve the system @xmath190 for unknown @xmath191 .
first remove one zero row and one zero column in the matrix @xmath192 ( see the remark below ( [ 3 ] ) ) . then the vector @xmath193 is uniquely defined and may be found by any standard method .
note that only a single component of @xmath194 is needed to evaluate @xmath195 finally dividing by @xmath196 yields the desired result ( [ n3])@xmath70 .
as an example , we calculate the effective shear - wave speed @xmath0 versus the volume fraction @xmath2 of square rods periodically embedded in a matrix material forming a 2d square lattice with translations parallel to the inclusion edges . a high - contrast pair of materials is chosen such as steel ( @xmath197 st , with @xmath198kg / m@xmath199 , @xmath200gpa ) and epoxy ( @xmath201 ep , with @xmath202kg / m@xmath199 , @xmath203gpa ) .
we consider two conjugated st / ep and ep / st configurations , where the matrix and rod materials are either st and ep or ep and st , respectively .
the results are displayed in fig . 1 .
the curves @xmath204 are computed by the present monodromy - matrix ( mm ) method , eq .
( [ n3])@xmath70 , they are complemented by the approximation ( [ m18 ] ) .
also shown for comparison are the curves @xmath205 computed from the truncated formula@xcite of the conventional pwe method based on a 2d fourier transform of ( [ 6 ] ) .
calculations are performed for a different fixed number @xmath206 of the 1d fourier coefficients of @xmath207 , which implies @xmath208 monodromy matrix in([n3])@xmath70 and , by contrast , @xmath209 matrix in the pwe formula@xcite .
apart from this advantage of the mm calculation , it is also seen to be remarkably more stable - with a reasonable fit provided already at @xmath210 .
the difference between the mm and pwe numerical curves is especially notable for the case of densely packed steel rods .
interestingly , the mm computation and estimate both predict a steep fall for @xmath211 when a small concentration @xmath212 of epoxy forms a quasi - insulating network. the pwe fails to capture this important physical feature for reasons described next . the far superior stability and accuracy of the mm method observed in fig .
1 can be explained as follows .
the pwe formula@xcite implies calculating @xmath213 with bounded coefficients @xmath214 where @xmath215 are the 2d reciprocal lattice vectors ( we use here that the components of the vector @xmath216 for piecewise constant @xmath11 are of order @xmath217 and that the matrix corresponding to @xmath29 is close to diagonally - dominant and hence its eigenvalues are of order @xmath218 ) . thus the accuracy of the pwe method is expected to be of order @xmath219 in contrast , the accuracy of the mm method , where the 1d fourier expansion is performed inside a multiplicative integral that is close to exponential , is expected to be on the order @xmath220 this can be understood from the mm equation ( [ n3])@xmath171 where the @xmath208 matrix @xmath221 can be replaced by @xmath222 with eigenvalues of order @xmath223 .
9 d. torrent , a. hkansson , f. cervera , and j. snchez - dehesa , phys .
* 96 * , 204302 ( 2006 ) ; d. torrent , j. snchez - dehesa , phys . rev .
b 74 , 224305 ( 2006 ) ; j. phys .
9 , 323 ( 2007 ) ; _ ibid .
_ 10 , 023004 ( 2008 ) .
vasseur , p.a .
deymier , b. djafari - rouhani , y. pennec , a .- c .
hladky - hennion , phys .
b * 77 * , 085415 ( 2008 ) ; y. pennec , j.o .
vasseur , b .
djafari - rouhani , l. dobrzynski , p.a .
deymier , surf .
rep . * 65 * , 229 - 291 ( 2010 ) .
q. ni and j. cheng , j. appl . phys .
* 101 * , 073515 ( 2007 ) .
a. a. kutsenko , a. l. shuvalov , a. n. norris , o. poncelet , submitted .
andrianov , j. awrejcewicz , v.v .
danishevskyy , d. weichert , j. comput .
nonlinear dynam .
* 6 * , 011015 ( 2011 ) | a new formula for the effective quasistatic speed of sound @xmath0 in 2d and 3d periodic materials is reported .
the approach uses a monodromy - matrix operator to enable direct integration in one of the coordinates and exponentially fast convergence in others . as a result , the solution for @xmath0 has a more closed form than previous formulas .
it significantly improves the efficiency and accuracy of evaluating @xmath0 for high - contrast composites as demonstrated by a 2d example with extreme behavior . |
a significant fraction of all active galactic nuclei found in high frequency radio surveys is made up of gigahertz peaked - spectrum ( gps ) and compact steep - spectrum ( css ) sources .
these agn have convex radio spectra that peak at rest - frame frequencies of up to 50 ghz @xcite .
there is a well known correlation between peak frequency ( @xmath6 ) and overall source size in the radio ( @xmath7 ; @xcite ) that is likely an evolutionary effect associated with either free - free @xcite or synchrotron self - absorption @xcite .
the historical division between gps and css classifications has been somewhat arbitrary , with the latter having observed peak frequencies that usually lie below @xmath8 mhz , and are therefore difficult to measure . as a consequence of the @xmath6size correlation ,
css typically have sizes larger than 1 kpc , which is sometimes used as a canonical value for dividing css and gps objects .
there has been much recent work involving a particular sub - class of gps sources that show two - sided radio jets on parsec scales .
their radio morphology is dominated by two `` mini - lobes '' which are typically separating at @xmath9 @xcite .
the core components are usually weak or not detected in vlbi images ( with some exceptions ; see , e.g. , @xcite ) .
their subluminal expansion speeds , relatively stable flux densities , and two - sided morphologies suggest that their jets lie close to the plane of the sky , and that their high apparent luminosities ( @xmath10 ) are not the result of relativistic beaming .
although these agn are usually referred to as `` compact symmetric objects '' ( csos ) , their parsec - scale jets are rarely symmetric , and in some cases are associated with much larger radio sources that extend out to many kiloparsecs @xcite . in such sources ( e.g. , 0108 + 388 ; @xcite ) , it has been speculated that the cso represents a renewed period of activity in the life cycle of the agn , since the kinematic age of the cso ( @xmath11 years ; @xcite ) is much smaller than that inferred for the large - scale structure .
if csos are indeed radio sources that have only recently become ( re)active , it should still be possible to find traces of the triggering events in their host galaxies .
the current paradigm for relativistic jet formation in agn is that a merger disturbs the gas in the inner regions of the host galaxy , sending it toward the central black hole .
an accretion disk forms and then provides the fuel for a relativistic outflow .
although this scheme is now widely accepted , there are relatively few nearby agn in which the predictions can be examined in detail .
one such system is the ultra - luminous infrared galaxy ( ulig ) iras 13451 + 1232 , which contains a double nucleus and appears to have undergone a recent merger .
what makes this agn particularly useful for investigating both the merger and cso `` youth '' scenarios is that it harbors the powerful radio source 4c + 12.50 , which is one of the closest known gps / css sources .
although 4c + 12.50 has a cso - type morphology on parsec - scales , it also has weak , extended radio emission which is possibly a relic of previous nuclear activity .
here we present multi - epoch vlba observations of 4c + 12.50 which reveal for the first time both the existence of superluminal motion and high linear polarization in the jets of a cso .
we find evidence that suggests that the jet nozzle of 4c + 12.50 is slowly precessing with time .
we discuss whether the twisted , symmetric jet morphologies seen in some csos may simply reflect the fact that their black hole spin axes are still precessing , and have not had sufficient time to reach a stable orientation . in
2 we describe our observations and data reduction .
we discuss the overall host galaxy and radio properties of 4c + 12.50 in
3.1 and 3.2 , respectively . in 4 we examine the proper motions in the jet , and discuss the constraints on the viewing angle and flow speed in 4.1 . in
5 we develop a helical precession model that accounts for the dramatic jet curvature in the source , and consider the evidence for 4c + 12.50 being a young radio source in 6 .
we summarize our findings in 7 . in this paper
we assume a standard cosmology with @xmath12 , @xmath13 , and @xmath14 .
the redshift of iras 13451 + 1232 ( @xmath15 ) implies a luminosity distance of 570 mpc , and an angular size of 1 mas on the sky corresponds to 2.20 pc .
in order to study the jet kinematics of 4c + 12.50 , we have used five individual epochs of 2 cm vlba data from three separate observing programs that spanned the time range 1996 to 2001 .
the first epoch consists of archival vlba data observed by @xcite .
the next three epochs are from the vlba 2 cm survey @xcite , and the final epoch is from a full - track observation carried out specifically for this study .
the first and last epochs were recorded in dual circular polarization , and included a single vla antenna in addition to the ten antennas of the vlba ( see table [ journal ] ) .
the data from the individual epochs were all correlated with the vlba correlator in socorro , nm and were processed in identical fashion using aips and difmap .
we applied a - priori amplitude corrections using antenna gains and system temperatures measured during the run , but did not perform any atmospheric opacity corrections , as the antenna gains were adequately calibrated in subsequent self - calibration and imaging iterations in difmap .
we determined the antenna polarization leakage factors for the final epoch by running the aips task lpcal on the calibrator sources oq 208 ( 1404 + 286 ) and pks 1413 + 135 .
these sources were observed at regular intervals during the run in order to obtain a large range of parallactic angles .
the leakage factors ranged up to @xmath16 , with typical values of @xmath17 . by comparing the results obtained from both calibrators
, we estimate that our leakage factor solutions are accurate to within @xmath18 .
our re - analysis of the first epoch data observed by @xcite using the sources oq 208 , 0738 + 313 , 0742 + 103 , and 0743006 gave similar results .
we established the absolute electric vector position angle ( evpa ) on the sky for the last epoch using five scans of the calibrator 3c 279 separated in hour angle throughout the observing run .
the integrated evpa of 3c 279 was measured at 1.3 , 3.6 , and 6 cm on 2000 dec 28 with the vla as part of the vla / vlba polarization monitoring program @xcite . by fitting a rotation measure of 54.7 @xmath19 to these data
, we interpolated an evpa of 51.6@xmath20 at 2 cm for this epoch .
we then rotated the vlba evpas so that in they were in agreement with this value .
based on previous applications of this technique ( e.g. , @xcite ) , we estimate that our evpas are accurate to within @xmath21 .
we calibrated the evpas of the first epoch by comparing single - dish measurements of the compact sources 0552 + 398 and 2134 + 004 made at the university of michigan observatory ( umrao ) to our integrated vlba evpas .
a rotation of @xmath22 degrees was needed to bring the vlba evpas into agreement with the michigan data .
a comparison of our 1996 and 2001 images shows no changes in polarization structure between these epochs , with evpas differing by less than @xmath232 degrees .
in this section we discuss previous observations of 4c + 12.50 , as well as those of its host galaxy iras 13451 + 1232 .
we describe the results of our vlba observations in [ masmorph ] . iras
13451 + 1232 has been imaged extensively by space- and ground - based observatories in the optical and infrared ( e.g. , @xcite ) .
early optical images by @xcite revealed two nuclei separated by 1.6 arcsec ( 3.5 kpc ) , and a v - band image by @xcite showed irregular isophotes , several nearby small companion galaxies , and a strongly curved tidal tail .
the disturbed features of this system and the relatively large nuclear separation are indicative of a recent merger event involving at least one spiral , gas - rich galaxy @xcite .
the east nucleus is optically dull and appears as reddened starlight .
the northwest nucleus , which is coincident with the radio source @xcite , has a seyfert 2 optical spectrum and is highly polarized in the ultraviolet ( @xmath24 ; @xcite ) .
it is highly reddened ( @xmath25 ; @xcite ) with an estimated @xmath26 of 3.5 @xcite , and has broad pa(@xmath27 ) emission in the infrared @xcite .
these observations suggest the presence of a quasar buried behind a foreground dust screen , which is further supported by the detection of hard x - rays from this source by asca ( @xcite ) .
the likely source of fuel for the quasar is a vast quantity of molecular gas that appears to be left over from the merger .
the northwest nucleus marks the peak of strong compact co ( @xmath28 ) emission at 2.7 mm , which @xcite found to be concentrated to within the inner 2 kpc of the galaxy .
in addition to the co emission , @xcite detected hi absorption in front of 4c + 12.50 .
the hi and co profiles span the same velocity range and are likely spatially coincident . @xcite
derive a molecular gas mass of @xmath29 , which is among the highest ever measured for an ulig .
4c + 12.50 is a relatively powerful radio source , with @xmath30 .
it has a curved radio spectrum with a spectral index of approximately @xmath31 at frequencies between 0.4 and 30 ghz .
there is a sharp cutoff at low frequency ( @xmath32 mhz ) characteristic of free - free or synchrotron self absorption .
its radio flux density is relatively stable at 318 and 430 mhz ; @xcite report only very weak ( @xmath33 ) variations over a 14 year time interval .
long term 8 ghz monitoring at umrao showed a slow rise of 300 mjy over several years in the early 1980s ( also seen at 5 ghz by bennett et al .
1984 ) , followed by a period of relatively constant flux density until the end of monitoring in 1992 . during the rise period the @xmath34 ghz spectral index was considerably flatter than at other times , which was likely due to a brightening of the flat spectrum core ( see [ masmorph ] ) . when umrao observations resumed in 2000 ,
the 8 ghz flux density had returned to its original 1980 level .
the radio morphology of 4c + 12.50 is generally compact , with 18 and 6 cm vla a- and b - array observations showing that most of the flux density comes from a region @xmath35 arcsec ( 220 pc ) in extent @xcite .
this size has been confirmed with merlin at 18 cm @xcite .
the total interferometric flux densities at 18 cm are generally consistent with single - dish measurements ( e.g. , @xcite ) , suggesting that the compact component accounts for nearly all of the cm - wave flux density .
however , a vla d - array image at 6 cm by @xcite does show two very weak ( a few mjy ) extensions of diffuse emission , one north of the core , and one to the southwest .
an extracted 20 cm image from the vla first survey also shows slightly extensions in these directions .
the total extent of the diffuse emission is approximately 20 arcsec ( 44 kpc ) . without more sensitive , high - resolution observations
it is difficult to speculate whether or not this represents material emitted during an earlier period of agn activity in the system .
we note that the spatial extent of 4c + 12.50 is similar to that of 0108 + 388 ( 30 kpc ; @xcite ) .
the compact structure in 4c + 12.50 has been imaged with vlbi at wavelengths ranging from 2 to 18 cm @xcite . in the following discussion we will use , as much as possible , the same component nomenclature as @xcite . in figure [ fullmap ]
we show our full - track 2 cm vlba image ( epoch 2001.01 ) , which has a significantly lower noise level ( @xmath36 ) than the epoch 1996.30 image of @xcite , due to the larger bandwidth and longer integration time ( see table [ journal ] ) .
the overall structure consists of a compact core component ( * a1 * ) associated with the active nucleus , a well - defined jet extending 75 mas to the south , a faint extension to the north of the core , and a more diffuse region to the northwest ( * f * ) .
@xcite and @xcite both identified component * a1 * with the core on the basis of its relatively flat spectral index between 2 and 18 cm .
we find additional evidence supporting this identification , in that it is the brightest , most compact feature in our image . in blazars ,
the core is usually appreciably polarized ( e.g. , @xcite ) , and indeed * a1 * is the only polarized feature in this region of the source ( see fig .
[ corezoom ] ) .
the relatively high brightness of the core compared to the lobes is typical for csos found in nearby galaxies with large apparent magnitudes @xcite .
the northern jet is significantly fainter and shorter than the southern jet .
it is well - collimated near the core , but then disappears , only to reappear as a region of diffuse emission further to the northwest .
this emission was also detected at 6 and 18 cm by @xcite and @xcite , respectively , leading to the initial classification of 4c + 12.50 as a cso .
the southern jet is somewhat knotty , and has an unusually high length - to - width ratio compared to the jets of typical agn . for the purposes of measuring apparent motions ,
we have modeled it using 8 distinct emission regions labeled * a1 * through * f * in figure [ fullmap ] .
the jet remains narrow and well - collimated until it undergoes a sharp bend near * d1*. the small opening angle and knottiness prior to * d1 * suggests a high internal mach number that is typical of powerful fr - ii type jets .
past * d1 * the jet widens considerably and becomes rather diffuse , with much of the flux being resolved out in our high resolution image .
a simultaneous 2 cm observation at umrao gave a total flux density of 1420 mjy , compared to only 950 mjy recovered in our vlba observation ( m. and h. aller , private communication ) .
the 6 cm vlbi image of @xcite shows that the southern jet past * d1 * is embedded in a roughly circular , steep - spectrum emission region with diameter @xmath23 30 mas .
their image lacks a similar amount of the total flux at 6 cm ( @xmath23 400 mjy ) .
it is useful to examine whether the increase in apparent opening angle past * d1 * could be a projection effect resulting from bending toward the line of sight . as we will show in 4 , the jet flow prior to the bend is relativistic , and is likely oriented close to the plane of the sky .
the apparent jet opening angle before to the bend is @xmath37 , while after the bend it increases to @xmath38 .
assuming an initial viewing angle of @xmath39 , then the change in viewing angle past the bend is @xmath40 .
this would imply a change of viewing angle of @xmath41 toward the line of sight .
since the jet flow is relativistic , beaming effects would be expected to cause a significant apparent brightening in the jet past * d1 * , yet the surface brightness remains relatively constant before and after the bend .
we therefore conclude that the increase in opening angle past * d1 * is real and not a projection effect .
the integrated linear polarization of 4c + 12.50 is extremely low .
vla - a configuration observations of @xcite give an upper limit ( @xmath42 ) at wavelengths from 3.6 to 18 cm ; similar upper limits were obtained at 2 , 3.6 and 6 cm during the umrao monitoring program .
although @xcite claimed to measure very weak circular polarization at 49 cm ( v = @xmath43 jy ) , there have been no reported detections of circular polarization at other wavelengths .
the generally low polarizations of cso and gps sources have been attributed to faraday depolarization , possibly by a foreground screen associated with the narrow - line region @xcite .
the milliarcsecond - scale linear polarization structure of 4c + 12.50 at epoch 2001.01 is shown in figures [ corezoom ] and [ southjet ] .
as expected , the integrated fractional polarization level over the entire source is very low ( @xmath44 ) , with the fractional polarization at the fitted core position being only ( 0.3 % ) .
the mean level for the surrounding region is @xmath45 .
however , surprisingly high fractional polarization levels are found in the southern region of the jet .
both the total and polarized intensity contours at * d1 * are extended toward the southwest , with the electric vectors being aligned roughly perpendicular to this direction .
the fractional polarization gradually increases toward the southwest , reaching a maximum value of @xmath46 at the outer edge .
the fractional polarization averaged over the entire feature is 11% .
the polarized emission at * d2 * is also concentrated toward the west , and has mean and peak values of 9 and 20% , respectively .
the most interesting polarization feature lies at the extreme southern end of the jet , near component * d3*. along an extension of the line connecting components * d2 * and * d3 * , the fractional polarization steadily increases to a maximum of @xmath47 at the outer edge , where both the i and p contours become extremely steep .
the high degree of field order and perpendicular e vectors of this feature are consistent with a termination shock or `` working surface '' where the jet encounters the interstellar medium and rapidly decelerates . the polarized features at * d1 * and * d2 * are likely associated with shocks that energize relativistic particles in the flow and order the magnetic field . as a jet loses kinetic energy through a series of internal shocks , its mach number will drop , which can eventually lead to subsonic flow and subsequent expansion and decollimation @xcite .
such a situation may be present in the inner kpc - scale jets of fr - i sources such as 3c 31 , in which standing shocks are thought to trigger a sudden flaring in the jet opening angle @xcite .
the jet of m87 also has a relatively constant opening angle until it is disrupted at the site of an oblique shock @xcite .
such shocks can arise from the well - known helical ( @xmath48 ) kelvin - helmholtz instability produced as the jet propagates through the interstellar medium .
we will discuss further evidence for k - h instabilities in the jet of 4c + 12.50 in
[ helicalstreaming ] .
the isolated regions of high polarization seen in 4c+12.50 are highly unusual for a gps / css source , and are somewhat unexpected , given the large quantity of gas present in its host galaxy .
if sufficiently magnetized , such a medium would be expected to efficiently depolarize any synchrotron radiation from the jet .
it is possible , however , that the intervening medium has a low filling factor and is rather clumpy .
recent faraday rotation studies ( e.g. , @xcite ) have revealed highly non - uniform environments in the inner regions of agn .
indeed , there are several isolated regions of the jet ( e.g. , * b1 * , * b2 * , and * c * ) that have comparable brightness to * d2 * and * d3 * , but have low fractional polarization , perhaps as a result of external faraday depolarization .
also , the sharp intensity and polarization gradients at * d3 * suggest that the jet has been disrupted after slamming into a high density region .
we have investigated proper motions within the radio jet by using the `` modelfit '' task in difmap to fit elliptical gaussian components to isolated bright features ( with the exception of component * b1 * where we used a @xmath49function component ) .
we began by fitting a model to a single epoch ( 1999 nov 1 ) .
we then fit the other epochs by allowing the component fluxes and positions to vary , and keeping all other parameters fixed .
we also experimented using combinations of circular and @xmath50-function components , and allowing all parameters to vary , but these methods yielded a much higher degree of scatter in the component positions across epochs .
the formal @xmath51 of our fits were significantly higher than unity due to the large amount of diffuse structure present in the jet , which can not be easily modeled with gaussian components . to obtain error estimates on the component positions
, we used the difwrap package @xcite .
this program varies the component positions by regular increments until the model no longer provides an acceptable fit to the visibilities .
we list the fitted component positions and associated errors for each epoch in table [ cptpos ] . for the bright , isolated components , our positional errors are
typically @xmath52 the width of the restoring beam . in figure [ abcd ]
we show separation versus time plots for the four innermost jet components ( * a2 * , * a3 * , * b1 * , and * b2 * ) with respect to the core component ( * a1 * ) , which we consider to be stationary . with the exception of the first epoch for components * a2 * and * b1 * , the component positions are well - described by linear motion outward from the core . in the case of * a2 * , the fitted position at epoch 1996.30 is likely affected by blending with the core .
the discrepancy in the 1996.30 position of * b1 * is much larger , however , and could be due to a component misidentification at this epoch .
we performed linear regression fits to all of the component positions , the results of which are listed in table [ cptprops ] .
the only components with statistically significant proper motions are * a2 * and * a3*. the remainder are all consistent with zero speed to within the errors .
our time baseline of 4.7 years allows us to set an upper limit of @xmath53 on their speeds .
the mildly superluminal speeds of * a2 * and * a3 * ( @xmath54 and @xmath55 , respectively ) indicate that the inner jet of 4c + 12.50 has only a moderate lorentz factor and/or is viewed at a large angle to the line of sight .
although slow apparent speeds are possible in a highly relativistic jet that is viewed inside the critical angle @xmath56 , where @xmath57 is the bulk lorentz factor , this is not likely to be the case for 4c + 12.50 . its lack of blazar properties ( i.e. , strong variability and polarization ) and its two - sided morphology suggest that the jets lie fairly close to the plane of the sky .
provided that the pattern speed is the same as that of the bulk flow , the apparent speed of component * a3 * limits the viewing angle in that region of the jet to be within 80 degrees of the line of sight , and the bulk flow speed to be greater than @xmath58 .
superluminal motion has so far been detected in only a few css sources : 3c 138 @xcite , 3c 147 @xcite , 3c 216 @xcite , cta 102 @xcite , and 3c 380 @xcite .
all of these sources have one - sided jets on parsec scales . to date there have been no reported superluminal speeds in the jets of csos .
@xcite reported a speed of @xmath59 for one component ( n5 ) of 1946 + 708 , but with a high degree of uncertainty due to positional errors associated with component blending .
the relative flux densities of the jet and counter - jets in 4c + 12.50 can be used to constrain their speed and orientations , provided that they have intrinsically identical properties .
the overall morphology and jet ridgeline of the northern counter - jet region ( * cj * ) differs considerably from the southern jet , even taking into account the possibility of relativistic time - delays .
in particular , there is no diffuse lobe of emission surrounding the northern jet , which suggests that the external environments at the ends of the northern and southern jets are considerably different .
any local differences in the external medium are likely to be less extreme in the small region close to the central engine , so we have chosen to confine our measurements of the jet - to - counter - jet flux density ratio ( @xmath60 ) to within 6 mas of the core .
the integrated flux density of the region @xmath61 mas north of the core is 4 mjy , while the same region south of the core has @xmath62 mjy , giving a ratio of @xmath63 ( we have excluded the region @xmath64 mas in order to eliminate any contribution of the core flux due to smearing of the restoring beam along the jet axis ) . at the position of component * a3 * , the flux ratio @xmath60 is approximately 7 .
we therefore adopt a value of @xmath65 for the inner jet in the following analysis . assuming doppler beaming ( e.g. , @xcite ) , @xmath66 , where @xmath67 is the intrinsic jet speed in units of @xmath68
, @xmath69 is the angle to the line of sight , and @xmath70 for continuous jet emission . here
we assume a typical value of @xmath71 for the jet spectral index .
if we make a further assumption that the pattern speed of component * a3 * ( @xmath72 c ) is representative of the bulk speed responsible for the doppler boosting , then the bulk lorentz factor is @xmath73 , and @xmath74 . the inferred viewing angle at component * a3 * is @xmath75 . in figure [ viewang ]
we show a graphical representation of these constraints in the @xmath76 plane .
the hashed region shows the allowed range of speed and viewing angle for the jet at the position of component *
a3*. the high intrinsic speed of the 4c + 12.50 jet suggests that there are likely to be other csos with end - on orientations that are appreciably beamed .
such a population has not yet been positively identified .
however , their high intrinsic luminosities suggest that they are likely bright enough to already be present in current radio source catalogs . given that the maximum possible doppler factor of 4c + 12.50 ( @xmath77 ) is much less than those inferred for the most highly beamed blazars ( e.g. , @xmath78 ; @xcite ) , the beamed versions of csos would be expected to have less extreme properties .
their integrated fractional polarizations should be much lower than blazars , due to faraday depolarization by dense surrounding gas .
the jet ridgeline in 4c + 12.50 traces out an `` @xmath79 '' shape that is highly reminiscent of those seen in galactic microquasars such as ss 433 @xcite .
the latter have been successfully fit by jet models in which the nozzle axis precesses slowly and material flows out on ballistic trajectories . in these models
the jet ridgeline appears to trace out an apparent three dimensional helical path on a conical surface , but the velocity vectors of the jet material are all directed radially outward from the nozzle . in cases where the velocities are relativistic , the receding jet ridgeline appears compressed , since we are seeing radiation emitted at an earlier time than that of the main ( forward ) jet . on first inspection , the morphology of 4c + 12.50 appears to fit the main predictions of the ballistic model : the counter - jet appears compressed and has a similar bend to the main jet , and all of the jet emission is confined within two conical regions whose apexes are located at the core .
most importantly , the sinusoidal path of the southern jet ridgeline is consistent with that of a helix seen in projection .
a closer examination of the southern jet , however , reveals several features which suggest that the jet material is not moving on simple ballistic trajectories .
the most convincing evidence can be found at the very tip of the southern jet ( component * d3 * in fig .
[ southjet ] ) .
both the * e * vectors and the gradients in total intensity and in fractional polarization run nearly perfectly parallel to the jet ridgeline that joins * d2 * and * d3*. if the material at * d3 * were moving radially outward from the core , one would have to postulate that all of these alignments occurred simply by chance , which is highly unlikely .
also , the fact that the sites of high fractional polarization are located at the major bend and terminus of the southern jet can not be easily explained by the ballistic model .
a more plausible explanation is that the jet material is streaming parallel to the jet ridgeline , and undergoes shocks at * d1 * and * d3*. strong evidence for helical streaming motion has already been found in other agn jets such as 3c 273 @xcite and bl lacertae @xcite .
theoretical studies of kelvin - helmholtz ( k - h ) instabilities in relativistic jets ( e.g. , @xcite ) have shown that they are susceptible to large helical twisting patterns provided there is a suitable perturbation mechanism present at the jet nozzle .
these findings have been confirmed by extensive numerical simulations ( e.g. , @xcite ) that show a variety of k - h surface and displacement modes . according to the k - h model of @xcite , the n = 1 helical fundamental mode displaces the entire jet .
higher modes merely produce helical fluting patterns on the surface , but do not displace the jet ridgeline . for the @xmath80 mode in a conical jet , the dependence of helix wavelength on distance along the jet axis ( @xmath81 ) is of the form @xmath82 , where @xmath83 depends on the external pressure gradient and the adiabatic index of the jet .
the conical boundaries of the helix imply that the pitch angle remains constant ( i.e. , if @xmath84 ) , or decreases with r ( if @xmath85 ) .
the pitch angle of the jet in 4c + 12.50 appears to _ increase _ down the jet , however , which would imply @xmath86 .
such a situation could arise if the instability is driven by precession and the wave speed remains constant down the jet ( p. hardee , private communication ) .
a major difficulty in fitting helical models to relativistic jets is that a minimum of six parameters are needed to describe a helix in three - dimensional space ( see table [ helixparms ] ) . in our case , however , we can narrow the parameter space considerably by using the constraints on the speed and orientation of the inner jet from [ viewingangles ] . in fitting a conical helix to the jets of 4c + 12.50
, we have chosen to adopt the right - hand coordinate system of @xcite , in which the observer s line of sight lies along the * x * axis , and the * y z * plane represents the plane of the sky . the helix axis ( @xmath87 )
lies in the * x z * plane at an angle @xmath69 to the * x * axis .
the helix has a constant wavelength , and lies on the surface of a cone of half angle @xmath88 whose axis is coincident with @xmath87 .
the parametric equation for the helix is @xmath89 where @xmath81 represents the distance along @xmath87 .
the phase angle of the helix varies according to @xmath90 , where @xmath91 , @xmath92 is the wave number , and @xmath93 for a right - handed helix .
the viewing angle to the local velocity vector along the jet is given by @xmath94 substituting equation ( 1 ) into equation ( 2 ) gives @xmath95 \over \left[\sec^2{\psi } + ( h k t\tan{\psi})^2\right]^{1/2}}.\ ] ] for simplicity we adopt a speed of @xmath96 along the entire jet .
the observational constraints on the fitted helix are the position of the ridgeline on the sky , and the speed and viewing angle of the jet at * a3*. the large jet / counter - jet flux ratio at * a3 * implies that the southern jet at this location makes a smaller angle to line of sight than the equivalent segment on the counter - jet side .
the position angle of the ridgeline at * a3 * then implies a right - handed helix ( i.e. , the southern jet would trace a counterclockwise spiral on the sky if it were viewed end - on ) . using the same arguments
, the initial phase angle is constrained to lie in the range @xmath97 . in order to find the best combination of parameters that fit these constraints , we calculated the predicted ridgelines of a set of models with @xmath98 , @xmath99 pc , @xmath100 and @xmath101 , and the above range of @xmath102 . within this set of models
, we considered only those combinations of parameters that gave a viewing angle of @xmath103 for the inner jet .
the combination that gave the best fit to the observed ridgeline of 4c + 12.50 is given in table [ helixparms ] .
although the viewing angle of the inner jet provides a powerful constraint on the uniqueness of this fit , the minimum in parameter space is somewhat shallow due to the fact that the southern jet undergoes barely a complete helical turn .
reasonably good fits to the ridgeline can be obtained with the parameter ranges given in column 4 of table [ helixparms ] .
we show a dot - representation of our best fit model in figure [ dotmodel ] .
this representation assumes a constant jet opening angle of @xmath104 , and a power law emissivity decay with index of @xmath105 along the jet .
the latter value is typically observed in quasar jets ( e.g. , @xcite ) .
the density of the dots is proportional to the beamed total intensity .
we have truncated the northern jet at a distance of 20 mas to match the data .
time - delays may cause a significant difference in the apparent length of the jet and counter - lobe , provided that @xmath106 at the end of the jet . the predicted viewing angle at * d3 * in our model , however , is @xmath107 , which suggests that time delays are unimportant , irrespective of the lobe advance speed .
although the positions of individual features in the remainder of the jets can be affected by time delays , the overall ridgeline will be unaffected due to the fact that material is continually streaming along it .
for these reasons , we did not incorporate any time delay corrections into our helical streaming model .
the model does a fairly good job at reproducing the jet ridgeline , with the exception of the region south of * a3 * , where the jet appears to curve too far to the west .
there also appears to be emission to the east of the fitted ridgeline at this location .
the other major discrepancy with the model occurs at the end of the northern jet , where the emission appears to extend too far in a westerly direction .
the overall fit to the total intensity is somewhat poorer than that of the ridgeline , as the model predicts that the counter - jet should actually increase in flux with distance from the core as the viewing angle twists toward the line of sight .
instead , there appears to be a gap in the counter - jet in this region .
the fact that the jet and counter - jet can not be fit with a symmetric model at distances further than @xmath108 pc from the core suggests that the local properties of the external medium at these radii differ considerably from one side of this source to the other .
asymmetric environments have been invoked to explain faraday asymmetries in other gps and css sources such as 3c 147 @xcite and 3c 216 @xcite .
although the extended emission associated with 4c 12.50 suggests that it may have been active in the past , the compactness of the parsec - scale emission suggests that the jets have only recently restarted after a long period of inactivity .
all of the age determinations carried out for other csos have been based on the relative separation speeds of components located at the extreme end of the jet and counter - jet .
these measurements were simplified somewhat by the fact that the terminal `` hotspots '' in these sources were quite bright and compact , which enabled their positions to be determined with high accuracy . in the case of 4c + 12.50
, the emission at the ends of the jet and counter - jet is too diffuse to obtain a precise lobe advance speed .
our poor positional accuracy on component * d3 * provides only an upper limit of 2 c on the apparent motion at the end of the jet , whereas typical measured advance speeds in csos are on the order of 0.3 c @xcite .
the latter speed would imply an age of only @xmath109 y for the parsec - scale structure in 4c + 12.50 .
a slow lobe advance speed would also imply a significant deceleration of the jet from its initial value of @xmath96 in the inner jet .
this deceleration could conceivably occur either at the strong bend ( component * d2 * ) , or at component * d3*. @xcite have incorporated such jet deceleration in a model of gps sources that can explain their symmetric morphologies and flux outburst behavior .
the overall host galaxy properties of iras 13451 + 1232 are generally consistent with the notion that the parsec - scale outflow is fairly recent .
uligs are thought to represent an early stage of quasar evolution @xcite , and those with relatively `` warm '' infrared colors such as iras 13451 + 1232 may be in a transition from a cold ulig phase to an optical quasar phase @xcite .
@xcite proposes that iras 13451 + 1232 is a young progenitor of a bright radio ir spiral such as arp 220 , and is still confined by large densities of gas and dust .
it was also the most morphologically disturbed object in a sample of nearby luminous radio galaxies studied by @xcite . according to @xcite ,
the disturbed outer isophotes imply that the merger was both violent and recent , since phase mixing should have restored symmetry in about @xmath110 y. although the merger timescales for massive galaxies ( on the order of @xmath111 y ) are much longer than the observed ages of csos , the timescale for black hole mergers which could trigger the radio jets are likely much shorter ( d. meier , private communication ) .
after such a merger , the spin axis of the black hole is predicted to align with the accretion disk in a relatively short time of about @xmath112 y @xcite , which would dampen any precession of the radio jets .
the prevalence of `` s '' shaped morphologies in csos ( e.g. , 0108 + 388 , 3c 186 , 3c 196 , 1946 + 708 , and 2352 + 495 ) may simply reflect the fact that their black hole spin axes are still precessing , and have not had sufficient time to stabilize .
more detailed observations of the extended emission associated with gps and cso sources such as 4c 12.50 could prove useful for constraining the direction of the radio jet axis during periods of previous activity .
we summarize the main results of our multi - epoch vlba study of the kinematics and polarization properties of the compact symmetric object 4c + 12.50 as follows : \1 .
4c + 12.50 has a relatively low integrated linear polarization on parsec - scales ( @xmath44 ) that is typical of other cso and gps sources , and is expected given the large amount of gas present in the host galaxy .
the high - resolution polarization images , however , reveal isolated regions in the southern jet with exceedingly high fractional polarizations that range up to @xmath113 .
these values are highly unusual for a cso , and suggest that the depolarizing medium is somewhat clumpy .
these findings illustrate the need for more high - resolution polarization - sensitive observations of other cso and gps sources in order to better understand their jet magnetic field properties and local environments .
a kinematic analysis of 4c + 12.50 reveals jet component speeds of @xmath114 and @xmath55 c , which represent the first positive detections of superluminal motion in a compact symmetric object .
we are able to fit both the apparent motion and the observed `` @xmath79 '' shape of the source using a conical helical jet .
the helical ridgeline is likely the result of small , periodic perturbations at the jet nozzle that are amplified by kelvin - helmholtz instabilities in the flow .
these perturbations could arise from orbital motion of the black hole and/or precession of the jet nozzle .
the best - fit model suggests that the nozzle is precessing around a cone with half - angle @xmath4 , whose axis lies at an angle of @xmath5 to the line of sight .
the jet has a bulk flow speed of 0.84 c , and follows a right - handed helical path with a wavelength of 280 pc .
the detection of relativistic speeds in the jet of 4c + 12.50 implies the existence of a beamed counterpart population to the csos that has yet to be identified .
although the properties of these objects are not expected to be as extreme as blazars , they should have have high apparent flux densities , and are likely already present in current radio agn catalogs . 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 .
alef , w. , preuss , e. , & kellermann , k. i. 1990 , in compact steep - spectrum & ghz - peaked spectrum radio sources , eds . c. fanti , r. fanti , c. p. odea , & r. schilizzi , ( bologna : consiglio nazionale delle richerche ) , 149 edge , a.c . ,
jones , m. , saunders , r. , pooley , g. , & grainge , k. 1996 , in the second workshop on gigahertz peaked spectrum and compact steep spectrum radio sources , eds .
i. a. g. snellen , r. t. schilizzi , h. j. a. rttgering , & m. n. bremer , ( leiden : leiden observatory ) , 43 paragi , z. , frey , s. , fejes , i. , porcas , r. w. , schilizzi , r. t. , & venturi , t. 2000 , in astrophysical phenomena revealed by space vlbi , ed . h. hirabayashi , p. g. edwards , & d. w. murphy ( sagamihara : institute of space and astronautical science ) , 59 polatidis , a. g. , conway , j. e. , owsianik .
i. , 2002 , in proceedings of the 6th european vlbi network symposium , eds .
e. ros , r. w. porcas , a. p. lobanov , & j. a. zensus , ( bonn : max - planck - institut fuer radioastronomie ) , 139 lllccl 1996 apr 22 & vlba+y1 & dual circular & 16 & 110 & @xcite + 1997 aug 28 & vlba & ll only & 64 & 40 & @xcite + 1998 nov 1 & vlba & ll only & 64 & 40 & @xcite + 1999 nov 6 & vlba & ll only & 64 & 200 & @xcite + 2001 jan 4 & vlba+y1 & dual circular & 32 & 540 & this work + lrrr a1 & & & 203 + a2 & 0.9 @xmath115 & @xmath116 & 59 + a3 & 1.7 @xmath115&@xmath117 & 30 + b1 & 7.6 @xmath118&@xmath116 & 14 + b2 & 9.7 @xmath119&@xmath120 & 224 + c & 20.9 @xmath119&@xmath121 & 87 + d1 & 44.5 @xmath115&@xmath122 & 161 + d2 & 52.2 @xmath119&@xmath123 & 92 + d3 & 65.9 @xmath124&@xmath125 & 139 + a1 & & & 170 + a2 & 0.9 @xmath115&@xmath126 & 74 + a3 & 1.8 @xmath115&@xmath127 & 38 + b1 & 6.7 @xmath118&@xmath126 & 6 + b2 & 9.8 @xmath119&@xmath120 & 225 + c & 19.1 @xmath119&@xmath122 & 91 + d1 & 44.5 @xmath115&@xmath122 & 146 + d2 & 52.6 @xmath119&@xmath120 & 83 + d3 & 66.4 @xmath124&@xmath125 & 89 + a1 & & & 159 + a2 & 1.0 @xmath115&@xmath128 & 64 + a3 & 2.0 @xmath115&@xmath129 & 24 + b1 & 7.1 @xmath118&@xmath130 & 7 + b2 & 9.9 @xmath119&@xmath120 & 196 + c & 20.9 @xmath119&@xmath131 & 80 + d1 & 44.5 @xmath115&@xmath132 & 143 + d2 & 52.3 @xmath119&@xmath123 & 74 + d3 & 65.9 @xmath124&@xmath125 & 85 + a1 & & & 169 + a2 & 1.2 @xmath115&@xmath133 & 61 + a3 & 2.3 @xmath115&@xmath134 & 20 + b1 & 6.9 @xmath118&@xmath130 & 8 + b2 & 9.8 @xmath119&@xmath120 & 194 + c & 20.9 @xmath119&@xmath131 & 91 + d1 & 44.5 @xmath115&@xmath122 & 155 + d2 & 52.4 @xmath119&@xmath123 & 85 + d3 & 66.0 @xmath124&@xmath135 & 97 + a1 & & & 183 + a2 & 1.3 @xmath115&@xmath127 & 41 + a3 & 2.3 @xmath115&@xmath127 & 11 + b1 & 7.1 @xmath118&@xmath136 & 13 + b2 & 9.9 @xmath119&@xmath123 & 155 + c & 21.0 @xmath119&@xmath131 & 80 + d1 & 44.5 @xmath115&@xmath122 & 151 + d2 & 52.4 @xmath119&@xmath123 & 79 + d3 & 66.2 @xmath124&@xmath125 & 124 + lrrrll a1 & 0.47 & 0.67 & @xmath137 & & + a2 & 0.52 & 0.13 & @xmath138 & @xmath139 & @xmath140 + a3 & 0.79 & 0.39 & @xmath141 & @xmath142 & @xmath55 + b1 & 0.00 & & & @xmath143 & @xmath144 + b2 & 2.32 & 0.34 & @xmath145 & @xmath146 & @xmath147 + c & 5.86 & 0.22 & @xmath148 & @xmath149 & @xmath150 + d1 & 6.42 & 0.39 & @xmath151 & @xmath152 & @xmath153 + d2 & 4.73 & 0.61 & @xmath154 & @xmath155 & @xmath156 + d3 & 7.41 & 0.47 & @xmath157 & @xmath158 & @xmath159 + half - opening angle of streaming cone & @xmath88 & 23 & @xmath160 & deg . + sky position angle of cone axis & @xmath161 & 175 & @xmath160 & deg . + viewing angle to cone axis & @xmath69 & 82 & @xmath162 & deg . +
initial phase angle & @xmath102 & 5 & @xmath163 & deg . + helix wavelength & @xmath164 & 280 & @xmath165 & pc + helix handedness & @xmath166 & @xmath167 ( right ) & & + | we present the results of a multi - epoch vlba study of the powerful radio source 4c + 12.50 ( pks 1345 + 125 ) at a wavelength of 2 cm .
this compact radio source is associated with a hidden quasar whose host galaxy shows signs of a recent merger .
it has been classified as a compact symmetric source ( cso ) due to its small overall size ( @xmath0 pc ) and twin - jet morphology , although it also has faint extended emission that may be a relic of previous activity .
we report the detection of exceedingly high linear fractional polarization in isolated features of the southern jet ( up to 60 % ) , which is highly unusual for a cso .
given the large amount of gas present in the host galaxy , we would expect significant faraday depolarization across the whole source , unless the depolarizing gas is fairly clumpy .
the southern jet also contains two features that are moving outward from the core at apparent speeds of @xmath1 and @xmath2 .
these represent the first positive detections of superluminal motion in a cso , and taken together with the jet to counter - jet flux ratio , suggest an intrinsic flow speed of @xmath3 .
the apparent ridgeline of the jet and counter - jet are consistent with a conical helix of wavelength 280 pc that is the result of kelvin - helmholtz instabilities driven by a slow precession of the jet nozzle .
a fit to the data implies that the nozzle is precessing around a cone with half - angle @xmath4 , whose axis lies at an angle of @xmath5 to the line of sight .
we suggest that the `` s''-shaped jet morphologies commonly seen in recent agn outflows such as 4c + 12.50 may simply reflect the fact that their black hole spin axes are still precessing , and have not had sufficient time to align with their accretion disks . |
_ swift _ , a multi - wavelength gamma - ray burst ( grb ) mission ( gehrels et al .
2004 ) , has led to great progress in understanding the nature of the grb phenomenon ( see recent reviews by mszros 2006 ; zhang 2007 ) .
one remarkable advance from _ swift _ is that the on - board x - ray telescope ( xrt ; burrows et al .
2005a ) has established a large sample of x - ray lightcurves from tens of seconds to days , sometimes even months ( e.g. grb 060729 , grupe et al .
2006 ) after the grb triggers , and revealed a canonical x - ray lightcurve that is composed of four successive power - law decaying segments ( zhang et al . 2006
; nousek et al . 2006
; obrien et al .
2006a ) with superimposing erratic flares ( burrows et al .
these segments include a grb tail segment ( segment 1 , with a decay slope , where @xmath4 is the decay slope , @xmath5 is the spectral index , and the subscript of @xmath4 and @xmath5 marks the segment of the lightcurve . ]
@xmath6 ) , a shallow decay segment ( segment 2 , @xmath7 ) , a normal decay segment ( segment 3 , @xmath8 ) , and a jet - like decay segment ( segment 4 , @xmath9 ) .
the grb tail and the shallow decay segment are usually seen in the xrt lightcurves ( obrien et al .
2006a ; liang et al .
2006 ; willingale et al .
2007 ; zhang et al .
2007c , hereafter paper i ; liang et al .
2007 , hereafter paper ii ) .
the jet - like decay segment , however , has occasionally been observed , but only for a small fraction of bursts ( burrows & racusin 2007 ; covino et al .
some bursts were observed with _ swift_/xrt and/or _ chandra _ for weeks and even months after the grb triggers , with no evidence of detecting a jet break in their x - ray lightcurves ( grupe et al .
2006 ; sato 2007 ) .
the jet models had been extensively studied in the pre-_swift _ era ( e.g. , rhoads 1999 , sari et al .
1999 ; huang et al . 2000
; see reviews by mszros 2002 ; zhang & mszros 2004 ; piran 2005 ) . an achromatic break
is expected to be observed in multi - wavelength afterglow lightcurves at a time when the ejecta are decelerated by the ambient medium down to a bulk lorentz factor @xmath10 , where @xmath1 is the jet opening angle ( rhoads 1999 ; sari et al .
most grbs localized in the pre-_swift _ era with deep and long optical monitoring have a jet - like break in their optical afterglow lightcurves ( see frail et al .
2001 ; bloom , frail , & kulkarni 2003 ; liang & zhang 2005 and the references therein ) , but the achromaticity of these breaks was not confirmed outside of the optical band .
panaitescu ( 2007 ) and kocevski & butler ( 2007 ) studied the jet breaks and the jet energy with the xrt data .
however , the lack of detection of a jet - like break in most xrt lightcurves challenges the jet models , if both the optical and x - ray afterglows are radiated by the forward shocks .
multiwavelength observational campaigns raise the concerns that some jet - break candidates may not be achromatic ( burrows & racusin 2007 ; covino et al .
2006 , cf .
dai et al .
2007 ; curran et al . 2007 ) .
issues regarding the nature of previous `` jet breaks '' have been raised ( e.g. zhang 2007 ) .
the observational puzzles require a systematical analysis on both the x - ray and the optical data .
this is the primary goal of this paper .
we analyze the _
swift_/xrt data of 179 grbs ( from 050124 to 070129 ) and the optical afterglow data of 57 pre-_swift _ and _ swift _ grbs , in order to systematically investigate the jet - like breaks in the x - ray and optical afterglow lightcurves ( 2 ) .
we measure a jet break candidate from the data with a uniform method and grade the consistency of these breaks with the forward shock models ( 3 ) , then compare these breaks observed in the x - ray and optical lightcurves ( 4 ) . assuming that these breaks are real jet breaks
, we revisit the grb jet energy budget ( frail et al .
2001 ; bloom et al .
2003 ; berger et al .
2003 ) with the conventional jet models ( 5 ) .
conclusions and discussion are presented in 6 . throughout this paper
the cosmological parameters @xmath11 km s@xmath12 mpc@xmath12 , @xmath13 , and @xmath14 are adopted .
the xrt data are taken from the _ swift _ data archive .
we have developed a script to automatically download and maintain all the xrt data .
the _ heasoft _ packages , including _ xspec _ , _ xselect _ , _ ximage _ , and
_ swift _ data analysis tools , are used for the data reduction .
we have developed an idl code to automatically process the xrt data for a given burst in any user - specified time interval . for details of our code
please see papers i and ii .
we process all the xrt data ( 179 bursts ) observed between 2005 january and 2007 january with our tools .
we are only concerned with the power - law afterglow segments 2 , 3 , & 4 without considering the steep decay segment ( 1 ) and the flares in the lightcurves . since the flares are generally superimposed upon the underlying afterglows ( chincarini et al .
2007 ) and their spectral properties are different from those of the power - law decaying afterglows ( falcone et al . 2007 ) , we do not consider the afterglow phases with significant flares .
first , we inspect the xrt lightcurve of each burst and specify the time interval(s ) that we use to derive the spectral and temporal properties .
then , we fit the lightcurve in this time interval with a power - law - like model as presented below .
we regard that a lightcurve in the specified time interval does not have significant flares , if the reduced @xmath15 of the power law fits is less than 2 .
we obtain a sample of 103 xrt lightcurves that have a good temporal coverage without significant flares .
we fit the lightcurve in the specified time interval to derive the decay slopes of the three segments and the two breaks , @xmath16 ( the shallow to normal transition break , possibly due to cessation of energy injection in the forward shock ; paper ii ) and @xmath17 ( the normal to steep transition break , possibly a jet break ) .
physically , these breaks should be smooth ( e.g. , panaitescu & mszros 1999 ; moderski et al .
2000 ; kumar & panaitescu 2000 ; wei & lu 2000 ) . as shown in paper ii ,
the energy injection break is usually seen in the xrt lightcurves , and a smoothly broken power law ( sbpl ) model fits most xrt lightcurves well , which is defined as @xmath18^{-1/\omega_1},\ ] ] where @xmath19 describes the sharpness of the break at @xmath20 , with a larger value corresponding to a sharper break .
if the jet - like decay segment is also observed , the lightcurve break near @xmath17 evolves as @xmath21^{-1/\omega_1 } \left(\frac{t}{t_{b,2}}\right)^{-\alpha_4}.\ ] ] therefore , a three - segment xrt afterglow lightcurve should be fitted with a smoothed triple power law ( stpl ) model , @xmath22 where @xmath23 is the sharpness factor of the jet break at @xmath17 . at @xmath24 ,
the lightcurve is dominated by the shallow decay phase , @xmath25 , and at @xmath26 , the lightcurve decays as @xmath27 .
as shown in paper ii , @xmath16 and @xmath17 are not significantly affected by @xmath19 and @xmath23 , but @xmath28 is .
the normal decay segment can be smeared by both the pre- and post- segments if @xmath19 and @xmath23 are small ( @xmath29 ) .
we find that @xmath30 can well identify the breaks in the lightcurves .
taking the xrt lightcurve for grb 060814 as an example , in fig.1 we compare the fit curve of the stpl model with a simpler fit by a joint triple power law ( jtpl ) model , which is defined as @xmath31 we find that the breaks at @xmath32 ks and @xmath33 ks are well identified in both models , and the results are consistent with each other . on the other hand , the stpl model is smooth without sharp breaks ( fig.1 ) , coinciding more within the physical context of these breaks .
the fitting result of the jtpl model strongly depends on the initial values of the two breaks .
the results may be misleading , especially when the normal decay phase lasts only a very short time ( in log - scale ) .
therefore , we use the stpl model and fix @xmath30 throughout this analysis .
the jet break signature may not be obvious , therefore we use the following strategy to select the best model among the stpl , sbpl , and single power law ( spl ) models to fit the xrt lightcurves . in the sense of occam s razor ,
the simplest model should be adopted .
on ther other hand , in order to avoid missing a jet break in the lightcurves , we accept a fit model as the best one when the derived breaks are sufficiently constrained by the data ( i.e. @xmath34 , where @xmath35 is the fitting error of @xmath36 , even if the @xmath15 is not significantly improved when compared to a simpler model ) .
we thus first fit the lightcurves with the stpl model ( eq . [ [ stpl ] ] ) .
this model is a reasonable fit to all of the lightcurves . in case of @xmath34
, we suggest that such a lightcurve has three segments and we adopt the stpl model fit .
we find that only 6 lightcurves satisfy this criterion ( see table 1 ) .
we fit the remaining lightcurves with the sbpl model ( eq . [ [ sbpl ] ] ) , and similarly we examine whether or not @xmath16 is sufficiently constrained .
the sbpl fits are adopted for 78 lightcurves .
we fit the remaining lightcurves ( 26 bursts ) with the spl model .
please note that , as shown in paper ii , the sharp breaks in grbs 060413 , 060522 , 060607a , and 070110 are possibly not of external origin ( see also troja et al .
we do not include these sharp breaks in this analysis .
grb 060522 and 070110 have a normal decay segment after an abnormally sharp lightcurve break , therefore we fit this post - break region to a simple power law .
grb 061202 shows significant spectral evolution throughout its lightcurve , so we do not consider this burst either .
our full resulting fits are summarized in table 1 . using the time intervals defined by the fitting results , we extract the spectrum of each segment , and fit it with a simple power law model with absorption by both our galaxy and the host galaxy .
the spectral fitting results are also reported in table 1 . in order to compare the x - ray break candidates with the optical lightcurves
, we also perform an extensive analysis of the optical lightcurves for both pre-_swift _ and _ swift _ bursts .
we search for the optical afterglow data in the literature and compile a sample of 57 optical lightcurves that have a good temporal coverage .
these lightcurves are fit with the same strategy as that for the xrt lightcurves .
the fitting results are reported in table 2 .
a break with @xmath37 is predicted by the forward shock jet models . since it is purely due to dynamic effects , it should be achromatic with no spectral evolution across the break , and both the pre- and post - break segments should also be consistent with the forward shock models . as shown in table 2 , no significant spectral evolution in the segments 3 and 4 is found for most bursts , and the x - ray spectral index is @xmath38 ( see also obrien et al .
2006b ) . assuming that both the optical and the x - ray afterglows are produced by the forward shocks , we select jet break candidates from the results shown in tables 1 and 2 , and grade these candidates as `` bronze '' , `` silver '' , `` gold '' , and `` platinum '' based on the consistency of data with the models .
the definitions of these grades are summarized in table 3 .
a break with a post - break segment being steeper than 1.5 is selected as `` bronze '' .
it is promoted to `` silver '' , if both pre- and post - break segments are consistent with the closure relations of the models .
if multiwavelength data are consistent with an achromatic break with only one band satisfies the jet models , a `` silver '' candidate is elevated to `` gold '' candidate .
if an achromatic break can be established independently at least in two bands with both bands satisfying the jet models , this break is termed as a `` platinum '' jet break candidate .
we first select the `` bronze '' jet break candidates from both the x - ray and optical data shown in tables 1 and 2 . without multiple wavelength modelling , the closure relations between the spectral index ( @xmath39 ) and temporal decay slope of the grb afterglows present an approach to verify whether or not the data satisfy the models ( see table 1 of zhang & mszros 2004 and references therein , in particular sari et al . 1998 ; chevalier & li 2000 ; dai & cheng 2001 ) . as shown in our tables 1 and 2 ,
the observed x - ray and optical spectral indices are larger than 0.5 ( except for the optical data of grb 021004 ) , indicating that the observed x - ray and optical afterglows are usually in the spectral regime of @xmath40 ( regime i ) or @xmath41 ( regime ii ) , where @xmath42 and @xmath43 are the cooling frequency and the typical frequency of the synchrotron radiation , respectively . in the standard
forward shock models , the decay slope of the pre - break segment is @xmath44 for emission in the spectral regime i ( both ism and wind ) and @xmath45 ( ism ) or @xmath46 ( wind ) for emission in regime ii . after the jet break and assuming maximized sideways expansion of jets , the lightcurve evolves as @xmath47 ( spectral regime i ) or @xmath48 ( spectral regime ii ) . if the jet sideways expansion effect can be negligible , the post - break decay index @xmath49 is shallower , i.e. @xmath50 ( ism ) and @xmath51 ( wind ) ( panaitescu 2005 ) .
the observed x - ray spectral indices are greater than 0.5 .
therefore , within the ism forward shock jet model the decay slopes of the pre- and post - break segments of the x - rays in the spectral regime ii should be greater than 0.75 and 1.5 , respectively ( even without significant sideways expansion ) . the wind model ( regime ii ) and the jet model with maximum sideways expansion would make the slopes even steeper .
we therefore pick 1.5 as the critical slope to define the `` bronze '' jet break sample .
as shown in tables 1 and 2 , 42 breaks of the xrt lightcurves and 27 of the optical lightcurves satisfy the `` bronze '' jet break candidate criterion .
these lightcurves are shown in fig .
we summarize the data of these breaks in table 4 .
our `` bronze '' jet break candidate sample is roughly consistent with that reported by panaitescu ( 2007 ) .
the jet breaks in the radio afterglow lightcurve of grbs 970508 ( frail et al.2000 ) and 000418 ( berger et al .
2001 ) are also included in our `` bronze '' sample .
the criterion of the `` bronze '' jet break candidate concerns only the decay slope of the post - break segment .
we notice that the lightcurve of the normal - decaying phase declines as @xmath52 in wind medium , i.e. , @xmath53 for @xmath54 .
as shown in paper ii , some jet - like breaks have a pre - break segment much shallower than that expected from the jet models .
a reasonable possibility would be that they are due to the energy injection effect in the forward shock model in the wind medium .
therefore , some `` bronze '' jet break candidates may be fake energy injection breaks instead .
we promote a `` bronze '' jet break candidate to the `` silver '' sample if both the pre- and post - break segments are consistent with the models in at least one band .
the decay slope of the pre - break segment of a jet break for the bursts in our sample should be steeper than 0.75 .
fifty - two out of the 71 `` bronze '' jet break candidates in table 4 agree with the `` silver '' candidate criterion ( 29 in the x - ray lightcurves and 23 in the optical light curves ) .
the x - ray light curve of grb 970828 has a break at @xmath55 days , with @xmath56 , @xmath57 , and @xmath58 ( djorgovski et al .
the x - ray lightcurve of grb 030329 has a break at @xmath59 days , with @xmath60 , @xmath61 , @xmath62 , and @xmath63 ( willingale et al .
we include these two pre-_swift _ grbs in the x - ray jet break candidate `` silver '' sample .
figure 3 shows the distribution of these bursts in the ( @xmath64)-plane combined with the closure relations for the models ( ism and wind medium ) .
the x - ray data of the `` silver '' jet break candidates are shown in figs .
3(a ) and 3(b ) .
it is found that the x - rays are consistent with the models in the spectral regime i , although the decay slopes of both the pre- and post - break segments are slightly shallower than the model predictions according to the observed spectral indices ( see also willingale et al .
2007 ; paper ii ) . as argued in paper ii
, this may be due to the simplification of the models .
simulations considering more realistic physical effects , such as energy transitions between different epoches ( kobayashi & zhang 2007 ) , evolution of microphysics parameters ( panaitescu et al .
2006 ; ioka et al .
2005 ) , and jet profiles ( zhang et al .
2004 ; yamazaki et al .
2006 ) , could expand the model lines into broad bands , which could accommodate the observational data better . the data for the `` silver '' jet break candidates in the optical band ( 15 pre-_swift _ grbs and 8 _ swift _ grbs ) are shown in figs .
3(c ) and 3(d ) .
since no time - resolved spectral analysis for the optical data is available , we take the same spectral index for both the pre- and post - break segments . differing from the x - rays ,
the optical emission of the post - break segment is consistent with the jet model in the spectral regime ii for most bursts .
however , the pre - break segment is also shallower than that predicted by the models in this spectral regime .
a `` gold '' jet break candidate requires that the break is achromatic at least in two bands , and that the break should satisfy the criteria of a `` silver '' candidate at least in one band . inspecting the data in table 4 and the lightcurves in fig .
2 , one approximately achromatic break is observed in both x - ray and the optical lightcurves of grbs 030329 , 050730 , 050820a , 051109a , and 060605 .
the optical afterglows of grbs 050525a , 060206 , 060526 , and 060614 are bright , and a jet - like break is clearly observed in their optical lightcurves . guided by the optical breaks , some authors argued for achromatic breaks in the xrt lightcurves of these grbs . without the guidance of the optical lightcurves , one can not convincingly argue a break in the xrt lightcurves of these grbs , but the data may be still consistent with the existence of an achromatic break .
both the optical and radio data of grb 990510 are consistent with the jet models .
we inspect the data of these bursts case by case , and finally identify 7 `` gold '' candidates , as discussed below .
grb 990510 : the pre - swift jet break candidate of grb 990510 is an exemplar of a jet break ( harrison et al . 1999 ) .
the break is achromatic in different colors in the optical band .
the radio data post the break is also consistent with the jet models .
however , with the radio data alone , one can not independently claim this break ( d. frail , 2006 , personal communication ) .
therefore , we include this jet candidate in the `` gold '' but not `` platinum '' category .
grb 030329 : its x - ray lightcurve has only five data points . fitting with the sbpl model shows that @xmath65 , @xmath66 , and @xmath67 ks with @xmath68 .
the break is consistent with the closure relations , and the @xmath69 agrees with the @xmath70 within error scope ( see also willingale et al .
however , the achromaticity of this break is somewhat questionable .
the break in the x - ray lightcurve has a great uncertainty since the fit has only one degree of freedom . on the other hand , the fit to the spl model yields @xmath71 and @xmath72 , indicating that the sbpl fit is required by the data .
we cautiously grade this burst to the `` gold '' sample with the caveat of sparse x - ray data in mind .
grb 050525a : blustin et al .
( 2006 ) fitted the xrt data of grb 050525a and derived a break at @xmath73 ks , with @xmath74 and @xmath75 .
the break is consistent with being achromatic with the break identified in the optical band .
the issue with their fitting is that the @xmath15 is too small ( reduced @xmath76 = 0.50 ( 25 dof ) ) .
we increase the signal - to - noise ratio of the data by rebinning the lightcurve , and fit it from 5.94ks to 157.85 ks .
we find that a simple power law is the best fit to the data , with a decay slope @xmath77 ( @xmath78 , 11 dof ) .
the decay slope is much larger than that of the normal decay segment(@xmath38 ) .
a jet - like break is likely embedded in the data .
we thus adopt the fitting by blustin et al .
( 2006 ) and cautiously include this burst in the `` gold '' sample .
grb050820a : its optical lightcurve traces the xrt lightcurves after @xmath79 seconds post - burst .
an achromatic break at @xmath80 days post - burst is observed .
its pre - break segments are well consistent with the models , but the post - break segments are slightly shallower than the prediction of the jet models .
we cautiously promote this break to the `` gold '' sample .
grb 051109a : the break time in both the x - ray and optical lightcurves is @xmath81 ks . although the decay slopes are @xmath82 and @xmath83 , slightly shallower than those predicted by the no - spreading jet models according to the observed spectral index , this burst is included in the `` gold '' candidate sample , similar to grb 050820a .
grb 060526 : the optical lightcurve of grb 060526 has a significant break at @xmath38 day post - burst ( dai et al .
its x - ray flux after @xmath84 ks is very low ( with a significance level of detection being lower than 3 @xmath85 ) .
dai et al .
( 2007 ) suggested a jet - like break in the xrt lightcurve by considering the contamination of a nearby source in the field of view . in all our analyses , we do not try to identify a nearby x - ray contamination source for any grb , so our best fit does not reveal this jet - break within the observational error scope . in view of the analysis of dai et al .
( 2007 ) , we also cautiously grade this break in the `` gold '' category .
grb 060614 : after subtracting the contribution of the host galaxy , the optical lightcurve of grb 060614 shows a clear break at @xmath86 ks ( della valle et al .
mangano et al . ( 2007 ) argue that the xrt lightcurve also has a break at this time .
fitting with our stpl can not reveal this break .
however , we note that @xmath49 in the x - ray lightcurve of this burst is @xmath87 , consistent with a post - jet - break decay slope . although a jet - like break can not be independently claimed at the optical break time with the x - ray data alone , the multi - wavelength data are still consistent with the existence of such a break , with the possibility that the injection break time and the jet break time are close to each other .
therefore , we agree with the suggestion by mangano et al .
( 2007 ) and grade this break as `` gold '' . to be conservative , we do not include grbs 050730 , 060206 , and 060605 in our `` gold '' candidate sample , as discuss below .
grb050730 : the break happens at @xmath88 ks in both the optical and x - ray bands .
the pre - break segment in both the x - ray and optical light curves is much shallower than the forward shock model predictions .
we thus do not grade this break as a `` gold '' candidate .
grb 060605 : a tentative break is observed at @xmath88 ks in both the optical and x - ray afterglow light curves .
however , this break time is uncertain in the optical band because no data point around the break time is available . on the other hand ,
the decay slope of the pre - break segment is only @xmath89 .
similarly to grb 050730 , it is not considered as a `` gold '' candidate .
grb 060206 : curran et al .
( 2007 ) fitted the xrt data of grb 060206 in the range between @xmath90 ks and @xmath91 ks after the grb trigger with the sbpl model , and reported @xmath92 ks , and the decay slopes of pre- and post - break segments are @xmath93 and @xmath94 , respectively .
the reduced @xmath15 of the fit is 0.79 ( 63 dof ) .
fitting with the spl model , they got a slope of @xmath95 with a reduced @xmath96 ( 65 dof ) .
the fitting results with the spl model is more reliable than that of the sbpl model .
also by checking the consistency with the models , we find that the the power law spectral index of the wt mode data @xmath97 after the break is consistent with the `` normal decay '' phase rather than the post - jet - break phase .
for example , for @xmath98 , the model - predicted temporal break index in the normal decay phase is @xmath99 , this is well consistent with the data .
we therefore do not consider this break as a `` gold '' jet break candidate . with our definition , a `` platinum '' jet break should be independently claimed in at least two bands which should be achromatic .
furthermore , the temporal decay slopes and spectral indices in both bands should satisfy those required in the simplest jet break models . since the optical and the x - ray afterglows could be in different spectral regimes ( fig .
3 ) , their lightcurve behaviors may be different ( e.g. sari et al . 1999 ) .
however , none of the seven `` gold '' candidates can be promoted to the `` platinum '' sample due to the various issues discussed above .
for some other `` silver '' candidates in which a prominent `` break '' is observed in one band , the lightcurve in the other band curiously evolves independently without showing a signature of break ( see 4.4 for more discussion ) .
it is fair to conclude that _ we still have not found a textbook version of jet break after many years of intense observational campaigns_.
in this section we compare the statistical characteristics of the jet break candidates in the x - ray and optical lightcurves .
our final graded jet break candidates are shown in table 4 .
the decay slopes of the pre - break segments of those `` bronze '' candidates are much shallower than the prediction of the jet models
. we can not exclude the possibility that some `` bronze '' jet break candidates are due to the energy injection effect in the wind medium ( paper ii ) .
therefore , for the following analysis , we do not include the `` bronze '' jet break candidates . as shown above , within the 103 xrt lightcurves with a good temporal coverage , 27 have `` silver '' or `` gold '' jet break candidates .
this fraction is 23/57 for optical lightcurves .
the detection fraction of jet break candidates in the xrt lightcurves is significantly lower than that in the optical lightcurves .
figure 4 shows the distributions of @xmath100 and @xmath101 in the x - ray and optical lightcurves .
the distributions of @xmath102/s and @xmath103/s peak at @xmath104 and @xmath105 , respectively .
the @xmath106 distribution has a sharp cutoff right at the high edge of the peak , indicating that the peak is possibly not an intrinsic feature . since the histogram depends on the bin size selection , we test the normality of the data set with the shapiro - wilk normality test .
it shows that the probability of a normal distribution for @xmath106 is @xmath107 ( at 0.05 confidence level ) , roughly excluding the normality of the distribution .
therefore , this peak is likely due to an observational selection bias .
by contrast , the @xmath108 distribution is log - normal .
the shapiro - wilk normality test shows @xmath109 ( at confidence level 0.05 ) .
these results suggest that the @xmath108 is systematically smaller than @xmath106 ( see also kocevski & butler 2007 ) .
this raises the possibility that x - ray breaks and optical breaks may not be physically of the same origin . with the closure relations of @xmath110 and assuming sideways expansion
, we derive @xmath111 for the regime - i ism model and all the wind models , and @xmath112 for the regime ii ism model .
therefore we only consider the @xmath110 case . ] . the observed @xmath113 is @xmath38 , hence @xmath114 or @xmath115 .
figure 4 ( right ) shows that the @xmath116 distribution peaks at @xmath38 , which suggests that most x - ray afterglows are consistent with the regime ii models ( i.e. x - ray is above both @xmath43 and @xmath42 ) greater than 1.5 grb 050124 ( @xmath117 ) and grb 051006 ( @xmath118 ) , but they have large errors . ] .
the @xmath119 show a tentative bimodal distribution , with two peaks at @xmath120 and @xmath121 , roughly corresponding to the regime i ( @xmath122 ) and regime ii ( @xmath123 ) ism models , respectively .
being achromatic is the critical criterion to claim a break as a jet break .
as shown above , the distribution of @xmath108 is systematically smaller than @xmath106 , which raises the concern of achromaticity of some of these breaks .
monfardini et al .
( 2006 ) have raised the concern that some jet - like breaks may not be achromatic .
we further check the chromaticity for the jet candidates case by case .
we find 13 bursts that have good temporal coverage in both x - ray and optical bands , with a jet break candidate at least in one band .
the results are the following .
the breaks in the x - ray and optical bands are consistent with being achromatic : grbs 030329 , 050525a , 050820a , 051109a , 060526 , and 060614 .
the x - ray and optical breaks are at different epochs : grbs 060206 and 060210 a `` silver '' jet break candidate in the optical band , but no break in the x - ray band : grbs 051111 and 060729 . a `` silver '' or `` bronze '' jet break candidate in the x - ray band , but no break in the optical band : grbs 050318 ( `` silver '' ) , 050802 ( `` bronze '' ) , and 060124 ( `` silver '' ) .
the ratio of achromatic to chromatic breaks is 6:7 , indicating that the achromaticity is not a common feature of these breaks .
it is a great issue to claim the chromatic breaks as a jet break .
if both the x - ray and optical emissions are from the forward shocks , one can rule out a large fraction ( 7/13 ) of these jet break candidates ( many are `` silver '' candidates ) as a jet break !
we indicate the achromaticity of the jet break candidates in table 4 .
if the above achromatic - to - chromatic ratio is a common value , most of the breaks without multi - wavelength observations ( marked with a `` ? '' in table 4 ) should be also chromatic . a possible way out to still consider these breaks as jet breaks
is to _ assume _ that the band ( either x - ray or optical ) in which the break is detected is from the forward shock , while emission from the other band is either not from the forward shock or some unknown processes have smeared the jet break feature from the forward shock in that band .
such a model does not explicitly exist yet .
we therefore suggest that _ one should be very cautious to claim a jet break , and further infer the grb energetics from a jet break candidate_. we are probably still a long way from understanding grb collimation and energetics .
as shown above , the observed chromatic feature is not consistent with the forward shock models , and it is risky to infer grb collimation and energetics from these data . on the other hand , it may be still illustrative to perform such a study by _ assuming _ that `` silver '' break candidates are jet breaks due to the following reasons . first , most pre-_swift _ works related to jet break and grb energetics ( frail et al .
2001 ; bloom et al . 2003 ; berger et al .
2003 ) were carried out with one - band data only .
if multi - wavelength data were not available for the _ swift _ bursts , one would still confidently take the post-_swift _ `` silver '' breaks as jet breaks .
it is therefore valuable to study this expanded sample and compare the results with the pre-_swift _ sample .
second , we notice that there is no grb that shows a `` silver '' jet break candidate in both the x - ray and optical bands but at different times .
for example , although a chromatic break is observed in both the optical and x - ray lightcurves of grbs 060206 and 060210 , the x - ray break in grb 060206 ( @xmath124 and @xmath125 ) and the optical break in grb 060210 ( @xmath126 and @xmath127 ) are not jet break candidates . on the other hand ,
the optical afterglow lightcurve of grb 060729 show a significant jet - like break , but its xrt lightcurve keeps decaying smoothly without a break .
the lightcurve behaviors in the optical and x - ray bands for most grbs are also enormously different ( see also paper ii for a discussion of achromaticity of the shallow - to - normal decay transition in many bursts ) .
these facts suggest that the jet - break candidates we see may indeed have a genuine origin , but we are probably far from understanding the lightcurve behaviors of most bursts . in this section ,
we _ assume _ that those `` silver '' or `` gold '' jet break candidates are jet breaks , and follow the standard forward shock model to constrain jet collimation and kinetic energy of the grb jets . in the standard afterglow models ,
the isotropic kinetic energy @xmath128 can be derived from the data in the normal decay phase , and the jet kinetic energy @xmath2 can be obtained from the jet break information ( e.g. rhoads 1999 ; sari et al .
1999 ; frail et al . 2001 ) .
the models depend on the power law index @xmath129 of the electron distribution , the spectral regime , and the medium stratification surrounding the bursts ( mszros & rees 1993 ; sari et al .
1998 ; dai & lu 1998 ; chevalier & li 2000 ; dai & cheng 2001 ) . as shown in fig . 3 , most bursts in our sample ( 25 out of 29 ) are consistent with @xmath110 .
we therefore only consider @xmath110 in this analysis .
essentially all the data are consistent with the ism model , although in some bursts the wind model can not be confidently ruled out . on the other hand , interpreting the early afterglow deceleration feature in grb 060418 ( molinari et al .
2007 ) requires that the medium is ism , even at the very early time ( jin & fan 2007 ) .
we therefore consider only the ism case in this paper .
we use the x - ray afterglow data to calculate @xmath3 , following the same procedure presented in our previous paper ( zhang et al .
2007a ) , which gives @xmath130^{4/(p+2)}d_{28}^{8/(p+2)}(1+z)^{-1 } t_d^{(3p-2)/(p+2)}\nonumber \\ & \times & ( 1+y)^{4/(p+2 ) } f_p^{-4/(p+2)}\epsilon_{b,-2}^{(2-p)/(p+2 ) } \epsilon_{e,-1}^{4(1-p)/(p+2 ) } \nu_{18}{^{2(p-2)/(p+2 ) } } \nonumber \\ & & \ \ \ \
( { \rm spectral\ regime\ i})\label{ekiso1}\\ e_{\rm k , iso,52 } & = & \left[\frac{\nu f_\nu ( \nu=10^{18}~{\rm hz})}{6.5\times 10^{-13 } ~{\rm ergs~s^{-1 } ~cm^{-2 } } } \right]^{4/(p+3 ) } d_{28}^{8/(p+3)}(1+z)^{-1 } t_d^{3(p-1)/(p+3)}\nonumber \\ &
\times & f_p^{-4/(p+3 ) } \epsilon_{b,-2}^{-(p+1)/(p+3 ) } \epsilon_{e,-1}^{4(1-p)/(p+3 ) } n^{-2/(p+3 ) } \nu_{18}{^{2(p-3)/(p+3 ) } } \nonumber \\ & & \ \ \ \
( { \rm spectral\ regime\ ii})\label{ekiso2}\end{aligned}\ ] ] where @xmath131 is the energy flux at @xmath132 hz ( in units of @xmath133 ) , @xmath134 the redshift , @xmath135 the luminosity distance , @xmath136 a function of the power law distribution index @xmath129 ( zhang et al . 2007a ) , @xmath137 the density of the ambient medium , @xmath138 the time in the observers frame in days , @xmath139 the inverse compton parameter .
the convention @xmath140 has been adopted .
if the ejecta are conical , the lightcurve shows a break when the bulk lorentz factor declines down to @xmath141 at a time ( rhoads 1999 ; sari et al .
1999 ) @xmath142 the jet opening angle can be derived as @xmath143 the geometrically corrected kinetic energy is then given by @xmath144 thirty _ swift _ grbs in our sample have redshifts available . among them
14 bursts have a jet break candidate detection in the optical or x - ray afterglow lightcurves . for those bursts without jet break detections , we take the time of the last xrt observation as the lower limit of the jet break time .
we calculate @xmath3 and @xmath1 ( or its lower limit ) for these bursts , then derive their @xmath145 ( or lower limits ) .
we use the normal decay phase to identify the spectral regime for each burst using the following method .
we define @xmath146 ^ 2},\end{aligned}\ ] ] where @xmath147 and @xmath148 are the temporal decay slopes ( errors ) from the observations and that predicted from the closure relations using the observed @xmath5 , respectively , for the normal decay phase .
the ratio @xmath149 reflects the nearness of the data point to the model lines within errors . in the case of @xmath150 ,
the data point goes across the corresponding closure relation line .
we derive @xmath151 from the data for both the spectral regimes i and ii . by comparing the two @xmath151 values
, we then assign each burst to the spectral regime with the smaller @xmath151 .
we find that the x - rays of about two - third of the bursts are in the spectral regime i. eq.([ekiso1 ] ) shows that the calculation of @xmath3 is independent of @xmath137 and only weakly depends on @xmath152 and @xmath129 with the data in this spectral regime .
therefore , this spectral regime is ideal to measure @xmath145 .
the x - rays of about one - third of the bursts are in the spectral regime ii .
the inferred @xmath3 in this spectral regime significantly depends on both @xmath152 and @xmath137 .
this makes it complicated to derive @xmath3 . in this case @xmath152 or
@xmath137 must be very small ( e.g. @xmath153 or @xmath154 @xmath155 ) in order to have the cooling frequency above the observed x - rays at @xmath156 day while retaining a reasonable @xmath157 ( zhang et al .
please note that by keeping @xmath158 , the @xmath139 parameter does not increase significantly for a smaller @xmath152 since the klein - nishina correction factor @xmath159 parameter becomes much smaller ( zhang et al .
2007a ) . after identifying the appropriate spectral regime , we derive @xmath129 from the relations between @xmath129 and the spectral index ( table 5 ) .
most derived @xmath129 s are greater than 2 , except for grbs 050820a , 060912 , and 060926 , and we assign @xmath160 for these bursts . in our calculation , we fix @xmath161 @xmath155 ( frail et al .
2001 ) and take initial values of @xmath152 and @xmath139 as @xmath162 and @xmath163 , respectively .
we iteratively search for the maximum value of @xmath152 that ensures the x - rays are in the proper spectral regime .
previous broadband fits and statistical analyses suggest that @xmath164 is typically around 0.1 ( wijers & galama 1999 ; panaitescu & kumar 2002 ; yost et al .
2003 ; liang et al . 2004
; wu et al . 2004 ) . therefore we take @xmath165 for all the bursts .
the @xmath3 is calculated with the observed energy flux at a given time .
after the energy injection is over , @xmath3 is a constant in the scenario of an adiabatic decelerating fireball . in principle
, one can derive @xmath157 at any time @xmath138 with eqs.([ekiso1 ] ) and ( [ ekiso2 ] ) .
we take the flux at a time @xmath166 .
our results are reported in table 4 .
we calculate @xmath1 and @xmath145 for the pre-_swift _ grbs with the same method .
we collect the x - ray afterglow data of the pre-_swift _ grbs from the literature .
the results are shown in table 5 .
eight bursts in table 5 are included in the sample presented by frail et al .
( 2001 ) . assuming @xmath161 @xmath155 and grb efficiency @xmath167 , frail et al .
( 2001 ) derived the jet opening angles @xmath1 of these bursts with the observed gamma - ray energy .
we compare our results with theirs ( @xmath168 ) in fig .
they are generally consistent with each other .
since our calculations derive @xmath3 directly rather than assuming an @xmath169 value , this result indicates that the derivation of @xmath1 is insensitive to @xmath169 , as suggested by the @xmath170 dependence of @xmath3 in eq.([theta ] ) .
the distributions of @xmath3 and @xmath129 are displayed in fig .
no significant differences between the pre-_swift _ and the _ swift _ samples are found for these parameters .
the kolmogorov - smirnov test shows that @xmath171 for the @xmath3 distribution and @xmath172 for the @xmath129 distribution .
as mentioned above , since we only consider @xmath110 , the sharp cutoff at @xmath173 is an artifact .
a small fraction of bursts might have @xmath174 ( such as grbs 050820a , 060912 , and 060926 ) , which would extend the @xmath129-distribution to smaller values .
no evidence for @xmath129-clustering among bursts is found ( see also shen et al .
2006 ; paper ii ) .
the @xmath3 distribution spans almost 3 orders of magnitude , ranging from @xmath175 to @xmath176 ergs with a log - normal peak at @xmath177 ergs .
the probability of the normality is @xmath178 at 0.05 confidence level .
the @xmath1 and @xmath2 distributions are shown in fig.7 . a sharp cutoff at @xmath179 is observed .
the @xmath1 of the _ swift _ grbs derived from xrt observations tends to be smaller than that of the pre-_swift _ grbs .
the @xmath2 of the pre-_swift _ grbs log - normally distribute around @xmath180 with a dispersion of 0.44 dex ( at @xmath181 confidence level ) . however , the @xmath2 of the _ swift _ grbs randomly distribute in the range of @xmath182 ergs ( see also kocevski & butler 2007 ) .
we examine the correlation between @xmath3 and @xmath1 in fig .
a tentative anti - correlation is found , but it has a large scatter . the best fit yields @xmath183 , with a linear correlation coefficient @xmath184 and a chance probability of @xmath185 ( n=28 ) .
this suggests that although @xmath2 has a much larger scatter than the pre-_swift _ sample , it is still quasi - universal among bursts .
we have presented a systematic analysis on the _ swift_/xrt data of 179 grbs observed between jan . , 2005 and jan . ,
2007 and the optical afterglow lightcurves of 57 grbs detected before jan .
2007 , in order to systematically investigate the jet - like breaks in the x - ray and optical afterglow lightcurves . among the 179 xrt lightcurves ,
103 have good temporal coverage and have no significant flares in the afterglow phase .
the 103 xrt lightcurves are fitted with the stpl , sbpl , or spl model , and the spectral index of each segment of the lightcurves is derived by fitting the spectrum with a simple absorbed power law model .
the same fitting is also made for the 57 optical light curves .
we grade the jet break candidates through examining the data with the forward shock models with `` bronze '' , `` silver '' , `` gold '' , or `` platinum '' .
we show that among the 103 well - sampled xrt lightcurves with a break , 42 are `` bronze '' , and 27 are `` silver '' .
twenty - seven out of 57 optical breaks are `` bronze '' , and 23 `` silver '' .
thirteen bursts have well - sampled lightcurves of both the x - ray and optical bands , but only 6 cases are consistent with being achromatic . together with the grb 990510 (
in which an achromatic break in optical and radio bands can be claimed , harrison et al .
1999 ) , we have 7 `` gold '' jet break candidates . however , none of them can be classified as `` platinum '' , i.e. a textbook version of a jet break . curiously , 7 out of the 13 jet - break candidates with multi - wavelength data suggest a chromatic break at the `` jet break '' , in contrary to the expectation of the jet models . the detection fraction of a jet break candidate in the xrt lightcurves is lower than that of the optical lightcurves , and the break time is also statistically earlier .
these facts suggest that one should be very cautious in claiming a jet break and using the break information to infer grb collimation and energetics . on the other hand , the possibility that some of these breaks are jet breaks is not ruled out .
the `` silver '' and `` gold '' jet break candidates have both the pre- and post - break temporal decay segments satisfying the simplest jet models , suggesting that these break are likely indeed jet breaks . in order to compare with the previous work on jet breaks , we then cautiously assume that the breaks in discussion are indeed jet breaks and proceed to constrain the @xmath1 and @xmath2 by using the x - ray afterglow data using the conventional jet models .
we show that the geometrically corrected afterglow kinetic energy @xmath2 has a broader distribution than the pre-_swift _ sample , disfavoring the standard energy reservoir argument . on the other hand , a tentative anti - correlation between @xmath1 and @xmath3
is found for both the pre-_swift _ and _ swift _ grbs , indicating that the @xmath2 could still be quasi - universal .
the grb jet models had been extensively studied in the pre-_swift _ era ( e.g. , rhoads 1999 ; sari et al .
1999 ; panaitescu & mszros 1999 ; moderski et al .
2000 ; huang et al .
2000 ; wei & lu 2000 ; see reviews by mszros 2002 ; zhang & mszros 2004 ; piran 2005 ) . the results of this paper suggest that for most bursts the x - ray and optical afterglows can not be simultaneously explained within the simplest jet models .
data suggest that we may be missing some basic ingredients to understand grb afterglows .
there have been skepticism about the jet break interpretations before ( e.g. dai & lu 1999 ; wei & lu 2002a , b ) .
the current data call for more open - minded thoughts on the origin of lightcurve breaks ( zhang 2007 ) .
observationally , at the epoch when the jet - like breaks show up the flux level is typically low .
source contaminations ( e.g. grb 060526 ; dai
2007 ) would complicate the picture .
careful analyses are needed to claim the breaks . on the other hand , most of the curious late afterglow break behaviors are likely not caused by these observational uncertainties .
for example , even if the contamination source is removed , the broad band afterglow lightcurves of grb 060526 ( dai et al .
2007 ) can not be incorporated within any simplest jet models . as cosmic beacons extending to high redshift universe ( e.g. lamb 2000 ; bromm & loeb 2002 ; gou et al .
2004 ; lin et al . 2004 ) , grbs have the potential to probe the high-@xmath134 universe . using the pre-_swift _ jet break sample , ghirlanda et al .
( 2004a ) discovered a tight correlation between the cosmic rest - frame peak energy ( @xmath186 ) of the grb @xmath187 spectrum and the geometrically - corrected grb jet energy ( @xmath188 ) .
this correlation was taken as a potential standard candle to perform cosmography studies ( e.g. dai et al .
2004 ; ghirlanda et al .
liang & zhang ( 2005 ) proceed with a model - independent approach , and derived a tight correlation among three observables , @xmath189 , @xmath190 , and @xmath191 , with the later being the cosmic rest frame optical break time only .
this correlation was also used to constrain cosmological parameters ( liang & zhang 2005 ; wang & dai 2006 ) . as shown in this paper , it is difficult to accommodate both the x - ray and optical afterglow data within a unified jet model , so that the ghirlanda relation is not longer supported by the _
swift _ data .
in fact , even with the optical data only , the _ swift _ bursts make the ghirlanda relation more dispersed than the pre-_swift _ sample ( campana et al .
2007 ) . as shown in figs . 4 and 6 ,
the break times in the xrt lightcurves are significantly smaller than that in the optical lightcurves ( most are pre - swift bursts ) , but no significant difference is observed in the @xmath189 distributions of the pre-_swift _ and _ swift _ grbs .
these results tend to suggest that the jet break candidates in the xrt lightcurves do not share the same liang - zhang relation derived from the pre-_swift _ optical data .
since the energy band of _ swift _ bat is too narrow to reliably derive @xmath186 and @xmath189 for most grbs , it is non - trivial to test the liang - zhang relation rigorously .
we plan to explore this interesting question in the future .
grbs fall into short - hard and long - soft categories ( kouveliotou et al .
1993 ) or more generally type i and type ii categories ( zhang et al . 2007b
; zhang 2006 ) .
the progenitors of the two classes are distinctly different : type ii grbs are related to deaths of massive stars ( woosley & bloom 2006 and references therein ) , and type i grbs are likely related to mergers of compact objects ( gehrels et al .
2005 ; fox et al .
2005 ; barthelmy et al . 2005 ; berger et al .
2005a ) . inspecting our sample of grbs with known redshifts ,
there are two type i grbs : 051221a and 060614ni ejection ( fynbo et al.2006 ; della valle et al .
2006 ; gal - yam et al .
2006 ; king et al . 2007 ; tominaga et al . 2007 ) .
their x - ray afterglows are very bright , and the derived @xmath2 from the xrt data are @xmath192 ergs and @xmath193 ergs , respectively , roughly about 1 order of magnitude smaller than that of the typical type ii grbs .
the @xmath1 of the two bursts are @xmath194 and @xmath195 , respectively .
they are wider than those of the other ( type ii ) _ swift _ grbs in our sample . combining our results with the fact that the @xmath1 of another short grb 050724 is @xmath196 ( grupe et al.2006 ; malesani et al .
2007 ) , we cautiously suggest that the short grbs might be less collimation , if the breaks are explained as a jet break .
we thank the referee for helpful suggestions , and z. g. dai , dirk grupe and goro sato for valuable comments .
we acknowledge the use of the public data from the swift data archive .
this work is supported by nasa under grants nng06gh62 g ,
nng05gb67 g , nnx07aj64 g , nnx07aj66 g , and the national natural science foundation of china under grant no . 10463001 ( ewl ) and 10640420144 .
aoki , k. , hattori , t.,kawabata , k. s. , & kawai , n. 2006 , gcn , 4703 , 1 barthelmy , s. d. , et al . 2005 , , 438 , 994 berger , e. , & gladders , m. 2006 , gcn , 5170 , 1 berger , e. , & mulchaey , j. 2005 , gcn , 3122 , 1 berger , e. , & soderberg , a. m. 2005 , gcn , 4384 , 1 berger , e. , et al .
2001 , , 556 , 556 berger , e. , kulkarni , s. r. , & frail , d. a. 2003 , , 590 , 379 berger , e. et al . 2005a , nature , 438 , 988 berger , e. et al .
2005b , apj , 634 , 501 berger , e. , cenko , s. b. , steidel , c. , reddy , n. , & fox , d. b. 2005c , gcn , 3368 , 1 bloom , j. s. , frail , d. a. , & kulkarni , s. r. 2003 , , 594 , 674 bloom , j. s. , foley , r .
j. , koceveki , d. , & perley , d. 2006a , gcn , 5217 , 1 bloom , j. s. , perley , d .
a. , & chen , h. w. 2006b , gcn , 5826 , 1 blustin , a. j. , et al .
2006 , , 637 , 901 bromm , v. , & loeb , a. 2002 , , 575 , 111 burrows , d. n. , & racusin , j. 2007 , preprint(astro - ph/200702633 ) burrows , d. n. , et al .
2005a , space science reviews , 120 , 165 burrows , d. n. , et al . 2005b , science , 309 , 1833 burrows , d. n. , et al .
2006 , , 653 , 468 campana , s. , guidorzi , c. , tagliaferri , g. , chincarini , g. , moretti , a. , rizzuto , d. , & romano , p. 2007
, , 472 , 395 castro - tirado , a .
j. , amado , p. , negueruela , i. , gorosabel , j. , jelinek , m. , & de ugarte postigo , a. 2006 , gcn , 5218 , 1 cenko , s. b. , et al .
2005 , gcn , 3542 , 1 cenko , s. b. , berger , e. , & cohen , j. 2006a , gcn , 4592 , 1 cenko , s. b. , et al . 2006b , gcn , 5155 , 1 chevalier , r. a. , & li , z .- y .
2000 , , 536 , 195 chincarini , g. , et al .
2007 , apj , submitted ( arxiv : astro - ph/0702371 ) covino , s. , et al . 2006 , arxiv : astro - ph/0612643 cucchiara , a. , fox , d. b. , & berger , e. 2006 , gcn , 4729 , 1 cucchiara , a. , price , p. a. , fox , d. b. , cenko , s. b. , & schmidt , b. p. 2006 , gcn , 5052 , 1 curran , p. a. , et al .
2007 , mnras , in press(arxiv:0706.1188 ) dai , z. g. , & cheng , k. s. 2001 , , 558 , l109 dai , z. g. & lu , t. 1999 , apj , 519 , l155 dai , z. g. , & lu , t. 1998 , , 333 , l87 dai , z. g. , liang , e. w. , & xu , d. 2004 , , 612 , l101 dai , x. , halpern , j. p. , morgan , n. d. , armstrong , e. , mirabal , n. , haislip , j. b. , reichart , d. e. , & stanek , k. z. 2007 , , 658 , 509 delia , v. , et al .
2005 , gcn , 4044 , 1 delia , v. , et al .
2006 , gcn , 5637 , 1 della valle , m. , et al . 2006 , , 444 , 1050 djorgovski , s. g. , frail , d. a. , kulkarni , s. r. , bloom , j. s. , odewahn , s. c. , & diercks , a. 2001 , , 562 , 654 falcone , a. d. , et al .
2007 , apj , submitted ( arxiv:0706.1564 ) fox , d. b. , et al .
2005 , , 437 , 845 frail , d. a. , waxman , e. , & kulkarni , s. r. 2000 , , 537 , 191 frail , d. a. , et al .
2001 , , 562 , l55 fugazza , d. , et al .
2006 , gcn , 5276 , 1 fynbo , j. p. u. , hjorth , j . ,
jensen , b. l. , jakobsson , p. , moller , p. , & naranen , j. 2005a , gcn , 3136 , 1 fynbo , j. p. u. , et al .
2005b , gcn , 3176 , 1 fynbo , j. p. u. , et al . 2005c , gcn , 3749 , 1 fynbo , j. p. u. , et al .
2006 , , 444 , 1047 gal - yam , a. , et al .
2006 , , 444 , 1053 gehrels , n. , et al .
2004 , , 611 , 1005 gehrels , n. , et al .
2006 , , 444 , 1044 ghirlanda , g. , ghisellini , g. , lazzati , d. , & firmani , c. 2004a , , 613 , l13 ghirlanda , g. , ghisellini , g. , & lazzati , d. 2004b , , 616 , 331 gou , l. j. , mszros , p. , abel , t. , zhang , b. 2002 , apj , 604 , 508 grupe , d. , burrows , d. n. , patel , s. k. , kouveliotou , c. , zhang , b. , mszros , p. , wijers , r. a. m. , & gehrels , n. 2006 , , 653 , 462 harrison , f. a. , et al . 1999 , , 523
, l121 huang , y. f.,gou , l. j. , dai , z. g. , & lu , t. 2000 , , 543 , 90 ioka , k. , kobayashi , s. , & zhang , b. 2005 , , 631 , 429 ioka , k. , toma , k. , yamazaki , r. , & nakamura , t. 2006 , , 458 , 7 jakobsson , p. , et al .
2006a , , 460 , l13 jakobsson , p. , vreeswijk , p. , fynbo , j. p. u. , hjorth , j. , starling , r. , kann , d. a. , & hartmann , d. 2006a , gcn , 5320 , 1 jakobsson , p. , levan , a. , chapman , r. , rol , e. , tanvir , n. , vreeswijk , p. , & watson , d. 2006b , gcn , 5617 , 1 jakobsson , p. , et al .
2006e , , 447 , 897 jaunsen , a. 0 . , et al . 2006 ,
gcn , 6010 , 1 jin , z. p. , & fan , y. z. 2007 , , 378 , 1043 kelson , d. , & berger , e. 2005 , gcn , 3101 , 1 king , a. , olsson , e. , & davies , m. b. 2007 , , 374 , l34 kobayashi , s. , & zhang , b. 2007 , , 655 , 973 kocevski , d. , & butler , n. 2007 , arxiv:0707.4478 kouveliotou , c. , meegan , c. a. , fishman , g. j. , bhat , n. p. , briggs , m. s. , koshut , t. m. , paciesas , w. s. , & pendleton , g. n. 1993 , , 413 , l101 kumar , p. , & panaitescu , a. 2000 , , 541 , l9 lamb , d. q. 2000 , , 333 , 505 ledoux , c. , et al .
2005 , gcn , 3860 , 1 liang , e. , & zhang , b. 2005 , , 633 , 611 liang , e. w. , dai , z. g. , wu , x. f. 2004 , apj , 606 , l29 liang , e. w. , et al .
2006 , , 646 , 351 liang , e .- w . , zhang ,
, & zhang , b. 2007 , apj , in press(arxiv:0705.1373 ) ( paper ii ) lin , j. r. , zhang , s. n. , & li , t. p. 2004
, , 605 , 819 malesani , d. , et al .
2007 , , 473 , 77 mszros , p. 2006
, reports of progress in physics , 69 , 2259 mszros , p. 2002
, , 40 , 137 mangano , v. , et al . 2007 , arxiv e - prints , 704 , arxiv:0704.2235 mszros , p. , & rees , m. j. 1993 , , 405 , 278 moderski , r. , sikora , m. , & bulik , t. 2000 , , 529 , 151 molinari , e. , et al . 2007 , , 469 , l13 monfardini , a. et al . 2006 ,
apj , 648 , 1125 nousek , j. a. , et al .
2006 , , 642 , 389 obrien , p. t. , et al . 2006a , , 647 , 1213 obrien , p. t. , willingale , r. , osborne , j. p. , & goad , m. r. 2006b , new journal of physics , 8 , 121 panaitescu , a. , & kumar , p. 2002
, , 571 , 779 panaitescu , a. , & mszros , p. 1999
, , 526 , 707 panaitescu , a. , mszros , p. , burrows , d. , nousek , j. , gehrels , n. , obrien , p. , & willingale , r. 2006 , , 369 , 2059 panaitescu , a. 2005 , , 363 , 1409 panaitescu , a. 2007 , , 380 , 374 piran , t. 2005 , reviews of modern physics , 76 , 1143 rhoads , j. e. 1999 , , 525,737 rol , e. , jakobsson , p. , tanvir , n. , & levan , a. 2006 , gcn , 5555 , 1 sako , m. , harrison , f. a. , & rutledge , r. e. 2005 , , 623 , 973 sari , r. , piran , t. , & narayan , r. 1998 , , 497 , l17 sari , r. , piran , t. , & halpern , j. p. 1999
, , 519 , l17 sato , g. , et al .
2007 , , 657 , 359 shen , r. , kumar , p. , & robinson , e. l. 2006 , , 371 , 1441 soderberg , a. m. , berger , e. , & ofek , e. 2005 , gcn , 4186 , 1 still , a. , et al .
2006 , gcn , 5226 , 1 thoene , c. c. , et al .
2006 , gcn , 5373 , 1 thoene , c. c. , perley , d. a. , & bloom , j. s. 2007 , gcn 6663 tominaga , n. , et al .
2007 , , 657 , l77 troja , e. , et al .
2007 , , 665 , 599 wang , f. y. , & dai , z. g. 2006 , , 368 , 371 wei , d. m. , & lu , t. 2000 , , 541 , 203 wei , d. m. , & lu , t. 2002a , , 332,994 wei , d. m. & lu , t. 2002 , a&a , 381 , 731 wijers , r. a. m. j. & galama , t. j. 1999 , apj , 523 , 177 willingale , r. , osborne , j. p. , obrien , p. t. , ward , m. j. , levan , a. , & page , k. l. 2004 , , 349 , 31 willingale , r. , et al . 2007 , , 662 , 1093 woosley , s. e. , & bloom , j. s. 2006 , , 44 , 507 wu , x. f. , dai , z. g. , liang , e. w. 2004 , apj , 615 , 359 yamazaki , r. , toma , k. , ioka , k. , & nakamura , t. 2006 , , 369 , 311 yost , s. a. , harrison , f. a. , sari , r. , & frail , d. a. 2003 , , 597 , 459 zhang , b. 2006 , , 444 , 1010 zhang , b. , & mszros , p. 2004
, international journal of modern physics a , 19 , 2385 zhang , b. 2007 , chinese journal of astronomy and astrophysics , 7 , 1 zhang , b. , dai , x. , lloyd - ronning , n. m. , & mszros , p. 2004
, , 601 , l119 zhang , b. , fan , y. z. , dyks , j. , kobayashi , s. , mszros , p. , burrows , d. n. , nousek , j. a. , & gehrels , n. 2006 , , 642 , 354 zhang , b. , et al . 2007a , , 655 , 989 zhang , b. , zhang , b .- b . , liang , e .- w . , gehrels , n. , burrows , d. n. , & mszros , p. 2007b , , 655 , l25 zhang , b .- b . ,
liang , e .- w .
, & zhang , b. 2007c , , 666 , 1002 & stpl + 050128 & 0.25 & 70.72 & 1.13(0.74 ) & 30.67(14.19 ) & 0.34(0.15 ) & 1.00(0.13 ) & 1.98(0.39 ) & 27(46 ) & 1.76(0.07 ) & 2.05(0.08 ) & 1.95(0.15 ) + 060210 & 3.90 & 861.94 & 5.51(0.86 ) & 186.65(76.48 ) & -0.20(0.39 ) & 1.00(0.05 ) & 1.85(0.27 ) & 134(131 ) & & 2.12(0.08 ) & 2.11(0.33 ) + 060510a & 0.16 & 343.41 & 2.89(1.87 ) & 47.65(16.75 ) & 0.01(0.09 ) & 0.87(0.17 ) & 1.74(0.12 ) & 84(140 ) & 1.91(0.07 ) & 2.04(0.14 ) & 2.06(0.14 ) + 060807 & 0.28 & 166.22 & 3.80(1.15 ) & 14.89(5.88 ) & -0.22(0.13 ) & 0.96(0.24 ) & 1.92(0.12 ) & 42(34 ) & 2.19(0.16 ) & 2.18(0.09 ) & 2.40(0.20 ) + 060813 & 0.09 & 74.25 & 0.19(0.04 ) & 15.24(3.88 ) & -0.01(0.19 ) & 0.87(0.03 ) & 1.63(0.13 ) & 56(73 ) & 2.05(0.09 ) & 1.99(0.05 ) & 2.10(0.07 ) + 060814 & 0.87 & 203.31 & 5.92(2.88 ) & 68.58(23.27 ) & 0.32(0.13 ) & 1.06(0.12 ) & 2.38(0.40 ) & 44(48 ) & 2.21(0.05 ) & & 2.30(0.05 ) + & sbpl + 050124 & 11.37 & 58.66 & & 29.37(12.61 ) & & 0.62(0.56 ) & 2.53(0.78 ) & 6(11 ) & & 2.05(0.29 ) & 1.93(0.21 ) + 050315 & 5.40 & 450.87 & & 224.64(38.68 ) & & 0.66(0.03 ) & 1.90(0.28 ) & 42(52 ) & & 2.31(0.12 ) & 2.17(0.07 ) + 050318 & 3.34 & 45.19 & & 10.64(4.97 ) & & 0.90(0.23 ) & 1.84(0.19 ) & 27(20 ) & & 2.01(0.08 ) & 2.02(0.06 ) + 050319 & 6.11 & 84.79 & 11.20(13.26 ) & & 0.23(0.59 ) & 0.99(0.25 ) & & 9(9 ) & 2.00(0.06 ) & 2.04(0.07 ) & + 050401 & 0.14 & 801.04 & 5.86(0.78 ) & & 0.58(0.02 ) & 1.39(0.06)& & 107(92 ) & 2.06(0.06 ) & 2.03(0.04 ) & + 050416a & 0.25 & 261.69 & 1.74(1.12 ) & & 0.43(0.12 ) & 0.90(0.04)& & 36(38 ) & 2.19(0.20 ) & 2.15(0.10 ) & + 050505 & 3.07 & 97.19 & 7.87(1.57 ) & & 0.15(0.19 ) & 1.30(0.06 ) & & 26(45 ) & 2.00(0.07 ) & 2.03(0.04 ) & + 050713a & 4.61 & 1600.08 & 5.86(1.24 ) & & -0.27(1.05 ) & 1.16(0.03 ) & & 28(17 ) & 2.25(0.05 ) & 2.21(0.17 ) & + 050713b & 0.79 & 478.50 & 10.80(1.59 ) & & -0.00(0.07 ) & 0.94(0.04 ) & & 40(63 ) & 1.83(0.11 ) & 1.94(0.09 ) & + 050716 & 0.64 & 74.40 & 7.53(9.02 ) & & 0.76(0.16 ) & 1.35(0.24 ) & & 31(36 ) & 1.60(0.08 ) & 2.01(0.13 ) & + 050717 & 0.32 & 11.23 & & 1.84(0.95 ) & & 0.57(0.21 ) & 1.65(0.12 ) & 28(56 ) & & 1.61(0.08 ) & 1.89(0.12 ) + 050726 & 0.42 & 17.05 & &1.17(0.33 ) & & 0.80(0.03 ) & 2.32(0.22 ) & 27(34 ) & & 2.06(0.08 ) & 2.14(0.09 ) + 050730 & 3.93 & 108.75 & & 6.66(0.29 ) & & -0.37(0.25 ) & 2.49(0.04 ) & 203(215 ) & & 1.65(0.03 ) & 1.70(0.03 ) + 050801 & 0.07 & 46.10 & 0.25(fixed ) & & 0(fixed ) & 1.10(0.03 ) & & 44(45 ) & & 1.91(0.12 ) & + 050802 & 0.51 & 83.83 & & 4.09(0.61 ) & & 0.32(0.10 ) & 1.61(0.04 ) & 58(72 ) & & 1.92(0.05 ) & 1.89(0.07 ) + 050803 & 0.50 & 368.89 & &13.71(0.90 ) & & 0.25(0.03 ) & 2.01(0.07 ) & 94(57 ) & & 1.78(0.10 ) & 2.00(0.08 ) + 050820a & 4.92 & 1510.14 & & 420.78(179.33 ) & & 1.11(0.02 ) & 1.68(0.21 ) & 246(292 ) & & 1.63(0.05 ) & 1.87(0.04 ) + 050822 & 6.41 & 523.32 & 66.99(44.38 ) & & 0.60(0.10 ) & 1.25(0.19 ) & & 29(44 ) & 2.29(0.23 ) & 2.36(0.11 ) & + 050824 & 6.31 & 330.49 & 11.52(4.25 ) & & -0.40(0.52 ) & 0.61(0.06 ) & & 45(41 ) & 2.00(0.16 ) & 2.01(0.09 ) & + 050908 & 3.97 & 33.36 & &7.81(5.33 ) & & 0.13(0.96 ) & 1.58(0.46 ) & 0(1 ) & & - & 2.09(0.25 ) + 050915a & 0.32 & 88.77 & 1.94(1.11 ) & & 0.39(0.27 ) & 1.24(0.09 ) & & 7(6 ) & 2.32(0.17 ) & 2.42(0.20 ) & + 051006 & 0.23 & 13.13 & & 0.93(0.71 ) & & 0.57(0.26 ) & 2.23(0.56 ) & 15(19 ) & & 1.61(0.14 ) & 1.84(0.20 ) + 051008 & 3.09 & 43.77 & 14.67(3.82 ) & & & 0.86(0.09 ) & 2.01(0.19 ) & 52(49 ) & & 2.15(0.32 ) & 2.11(0.10 ) + 051016a & 0.37 & 37.41 & 0.63(0.40 ) & & -0.41(1.18 ) & 0.91(0.12 ) & & 0(7 ) & 2.40(0.26 ) & & + 051016b & 4.78 & 150.47 & &66.40(23.09 ) & & 0.71(0.08 ) & 1.84(0.46 ) & 15(16 ) & & - & 2.19(0.13 ) + 051109a & 3.73 & 639.16 & &27.28(7.90 ) & & 0.79(0.07 ) & 1.53(0.08 ) & 39(48 ) & & 1.91(0.07 ) & 1.90(0.07 ) + 051109b & 0.39 & 87.63 & 5.11(4.73 ) & & 0.56(0.17 ) & 1.22(0.17 ) & & 15(17 ) & 2.73(0.44 ) & 2.35(0.24 ) &
+ 051117a & 18.19 & 970.14 & 104.23(151.17 ) & & 0.51(0.25 ) & 1.07(0.24 ) & & 21(19 ) & 2.25(0.04 ) & 2.39(0.15 ) &
+ 051221a & 6.87 & 118.64 & &40.74(15.89 ) & & 0.46(0.16 ) & 1.75(0.41 ) & 11(14 ) & & 2.08(0.09 ) & 2.02(0.19 ) + 060105 & 0.10 & 360.83 & &2.31(0.14 ) & & 0.84(0.01 ) & 1.72(0.02 ) & 653(754 ) & & 2.23(0.05 ) & 2.15(0.03 ) + 060108 & 0.77 & 165.26 & &22.08(7.38 ) & & 0.26(0.09 ) & 1.43(0.17 ) & 7(7 ) & & 2.17(0.32 ) & 1.75(0.15 ) + 060109 & 0.74 & 48.01 & 4.89(1.10 ) & & -0.17(0.14 ) & 1.32(0.09 ) & & 19(13 ) & 2.32(0.15 ) & 2.34(0.14 ) & + 060124 & 13.30 & 664.01 & &52.65(10.33 ) & & 0.78(0.10 ) & 1.65(0.05 ) & 165(132 ) & & 2.10(0.06 ) & 2.06(0.08 ) + 060202 & 1.03 & 96.23 & 3.50(6.95 ) & & 0.68(0.37 ) & 1.14(0.13 ) & & 51(31 ) & 2.96(0.19 ) & 3.41(0.14 ) & + 060203 & 3.80 & 32.95 & &12.95(6.69 ) & & 0.40(0.30 ) & 1.65(0.47 ) & 4(7 ) & & 2.08(0.19 ) & 2.25(0.13 ) + 060204b & 4.06 & 98.80 & &5.55(0.66 ) & & -0.49(0.65 ) & 1.47(0.07 ) & 21(34 ) & & 2.54(0.14 ) & 2.64(0.16 ) + 060206 & 0.11 & 621.77 & 8.06(1.46 ) & & 0.40(0.05 ) & 1.26(0.04 ) & &43(44 ) & 2.31(0.12 ) & 2.33(0.32)& + 060211a & 5.40 & 527.10 & &267.24(165.67 ) & & 0.38(0.08 ) & 1.63(1.27 ) & 10(9 ) & & 2.15(0.06 ) & 2.11(0.26 ) + 060306 & 0.25 & 124.39 & 4.67(2.91 ) & & 0.40(0.11 ) & 1.05(0.07 ) & & 30(32 ) & 2.10(0.11 ) & 2.21(0.10 ) & + 060313 & 0.09 & 93.22 & &11.18(2.89 ) & & 0.82(0.03 ) & 1.76(0.18 ) & 95(128 ) & & 1.84(0.34 ) & 1.78(0.09 ) + 060319 & 0.33 & 304.52 & &99.70(26.78 ) & & 0.84(0.02 ) & 1.92(0.30 ) & 72(93 ) & & 1.93(0.22 ) & 2.25(0.11 ) + 060323 & 0.33 & 16.28 & &1.29(0.32 ) & & -0.11(0.23 ) & 1.55(0.16 ) & 4(7 ) & & 1.99(0.16 ) & 2.02(0.13 ) + 060428a & 0.23 & 271.10 & &125.31(47.19 ) & & 0.48(0.03 ) & 1.46(0.37 ) & 26(21 ) & & 2.11(0.24 ) & 1.97(0.10 ) + 060428b & 0.96 & 200.36 & 3.95(5.55 ) & & 0.53(0.41 ) & 1.16(0.13 ) & & 19(21 ) & 2.41(0.24 ) & 2.10(0.33 ) &
+ 060502a & 0.24 & 593.06 & &72.57(15.05 ) & & 0.53(0.03 ) & 1.68(0.15 ) & 11(26 ) & & 2.11(0.29 ) & 2.15(0.13 ) + 060507 & 3.00 & 86.09 & 6.95(1.68 ) & & -0.06(0.55 ) & 1.12(0.07 ) & & 13(24 ) & 2.06(0.23 ) & 2.15(0.14 ) & + 060510b & 4.40 & 77.71 & &67.90(29.88 ) & & 0.44(0.18 ) & 2.40(0.00 ) & 4(8 ) & & 1.71(0.04 ) & + 060526 & 1.09 & 45.20 & &11.60(6.39 ) & & 0.42(0.12 ) & 1.58(0.34 ) & 5(9 ) & & 2.07(0.09 ) & 2.08(0.16 ) + 060604 & 4.14 & 403.81 & 11.51(9.81 ) & & 0.20(0.77 ) & 1.17(0.09 ) & & 32(36 ) & 2.44(0.15 ) & 2.43(0.17 ) & + 060605 & 0.25 & 39.85 & & 7.14(0.93 ) & & 0.45(0.04 ) & 1.80(0.13 ) & 22(34 ) & & 1.62(0.17 ) & 1.83(0.09 ) + 060614 & 5.03 & 451.71 & &49.84(3.62 ) & & 0.18(0.06 ) & 1.90(0.07 ) & 70(54 ) & & 2.02(0.02 ) & 1.93(0.06 ) + 060707 & 5.32 & 813.53 & 22.21(54.08 ) & & 0.37(0.96 ) & 1.09(0.17 ) & & 8(11 ) & 1.88(0.08 ) & 2.06(0.20 ) & + 060708 & 0.25 & 439.09 & 7.28(2.34 ) & & 0.57(0.08 ) & 1.32(0.07 ) & & 39(35 ) & 2.30(0.20 ) & 2.36(0.11 ) & + 060712 & 0.56 & 317.56 & 7.89(2.67 ) & & 0.12(0.16 ) & 1.15(0.10 ) & & 15(14 ) & 3.21(0.38 ) & 2.94(0.28 ) & + 060714 & 0.32 & 331.97 & 3.70(0.97 ) & & 0.34(0.10 ) & 1.27(0.05 ) & & 53(73 ) & 2.15(0.08 ) & 2.04(0.11 ) & + 060719 & 0.28 & 182.15 & 9.57(2.70 ) & & 0.40(0.06 ) & 1.31(0.10 ) & & 19(26 ) & 2.35(0.13 ) & 2.28(0.26 ) & + 060729 & 0.42 & 2221.24 & 72.97(3.02 ) & & 0.21(0.01 ) & 1.42(0.02 ) & & 459(459 ) & 2.33(0.08 ) & 2.29(0.07 ) & + 060804 & 0.18 & 122.07 & 0.86(0.22 ) & & -0.09(0.15 ) & 1.12(0.07 ) & & 18(24 ) & 2.04(0.23 ) & 2.14(0.15 ) &
+ 060805a & 0.23 & 75.91 & 1.30(0.70 ) & & -0.17(0.41 ) & 0.97(0.13 ) & & 11(17 ) & & 1.97(0.37 ) & + 060906 & 1.32 & 36.69 & &13.66(3.29 ) & & 0.35(0.10 ) & 1.97(0.36 ) & 3(7 ) & & 2.28(0.37 ) & 2.12(0.17 ) + 060908 & 0.08 & 363.07 & & 0.95(0.34 ) & & 0.70(0.07 ) & 1.49(0.09 ) & 98(59 ) & & 2.01(0.22 ) & 2.00(0.08 ) + 060912 & 0.12 & 86.80 & 2.92(2.77 ) & & 0.65(0.12 ) & 1.24(0.11 ) & & 31(56 ) & & 2.03(0.12 ) &
+ 060923a & 0.22 & 280.62 & 3.33(1.03 ) & & -0.16(0.22 ) & 1.30(0.06 ) & & 34(21 ) & 2.05(0.25 ) & 1.86(0.18 ) &
+ 060923b & 0.16 & 6.03 & 0.42(0.64 ) & & -0.73(0.99 ) & 1.08(0.82 ) & & 2(10 ) & 2.47(0.53 ) & 2.25(0.31 ) & + 060926 & 0.09 & 5.96 & 1.13(0.92 ) & & 0.04(0.14 ) & 1.23(0.52 ) & & 11(9 ) & 1.93(0.16 ) & 1.88(0.14 ) & + 060927 & 0.11 & 5.64 & &4.24(8.22 ) & & 0.73(0.32 ) & 1.82(2.60 ) & 4(7 ) & & 1.65(0.19 ) & 1.92(0.15 ) + 061004 & 0.39 & 69.99 & 1.50(0.52 ) & & -0.08(0.29 ) & 1.04(0.09 ) & & 13(17 ) & 1.84(0.34 ) & 3.04(0.34 ) & + 061019 & 9.07 & 287.03 & 10.84(2.15 ) & & -1.38(2.88 ) & 1.15(0.08 ) & & 6(10 ) & 2.32(0.20 ) & 1.93(0.28 ) & + 061021 & 0.30 & 594.16 & 9.59(2.17 ) & & 0.52(0.03 ) & 1.08(0.03 ) & & 94(87 ) & 1.90(0.06 ) & 1.72(0.05 ) & + 061121 & 4.89 & 353.10 & &24.32(4.38 ) & & 0.75(0.06 ) & 1.63(0.05 ) & 121(147 ) & & 1.71(0.03 ) & 1.96(0.07 ) + 061201 & 0.10 & 15.42 & &2.09(0.75 ) & & 0.57(0.07 ) & 1.61(0.23 ) & 20(29 ) & & 1.30(0.09)& + 061222a & 10.94 & 724.64 & &60.51(8.89 ) & & 0.81(0.07 ) & 1.86(0.06 ) & 144(95 ) & & 2.45(0.06 ) & 2.22(0.12 ) + 070103 & 0.11 & 143.98 & &2.88(0.48 ) & & 0.20(0.10 ) & 1.63(0.08 ) & 43(30 ) & & 2.32(0.25 ) & 2.52(0.21 ) + 070129 & 1.32 & 546.36 & 20.12(3.14 ) & & 0.15(0.07 ) & 1.31(0.06 ) & &42(70 ) & 2.25(0.07 ) & 2.30(0.10 ) & + & spl + 050219b & 3.21 & 85.26 & & & & 1.14(0.03 ) & & 24(32 ) & & 2.27(0.14 ) & + 050326 & 3.34 & 142.24 & & & & &1.63(0.04 ) & 45(34 ) & & &2.15(0.14 ) + 050408 & 2.60 & 3223.36 & & & & 0.78(0.01 ) & & 52(44 ) & & 2.01(0.18 ) &
+ 050525a & 5.94 & 157.85 & & & & 1.40(0.05 ) & & 11(11 ) & & 2.17(0.18 ) & + 050603 & 39.72 & 166.22 & & & & &1.71(0.10 ) & 8(10 ) & & & 1.84(0.09 ) + 050721 & 0.30 & 257.24 & & & & 1.18(0.02 ) & & 80(98 ) & & 1.77(0.10 ) & + 050814 & 2.17 & 87.85 & & & & 0.65(0.05 ) & & 21(16 ) & & 1.91(0.07 ) & + 050826 & 0.13 & 61.93 & & & & 1.02(0.03 ) & & 23(21 ) & & 2.19(0.19 ) & + 050827 & 65.95 & 246.35 & & & & 1.24(0.15 ) & & 12(15 ) & & 1.88(0.15 ) & + 051001 & 6.71 & 273.86 & & & & 0.70(0.06 ) & & 30(25 ) & & 1.93(0.19 ) & + 051111 & 10.98 & 34.24 & & & & 1.09(0.17 ) & & 1(6 ) & & & + 051117b & 0.22 & 0.62 & & & & &1.68(0.27 ) & 0(2 ) & & & + 060115 & 5.44 & 326.04 & & & & 0.88(0.04 ) & & 12(12 ) & & 2.50(0.38 ) & + 060116 & 0.21 & 6.87 & & & & 0.88(0.06 ) & & 3(6 ) & & 2.33(0.39 ) & + 060403 & 0.05 & 79.82 & & & & &1.67(0.07 ) & 70(57 ) & & &1.58(0.13 ) + 060418 & 0.20 & 201.65 & & & & 1.45(0.02 ) & & 272(283 ) & & 2.24(0.05 ) & + 060421 & 0.12 & 6.52 & & & & 0.93(0.05 ) & & 11(7 ) & & 1.60(0.35 ) & + 060512 & 0.11 & 104.01 & & & & 1.39(0.02 ) & & 76(58 ) & & 3.60(0.19 ) & + 060522 & 5.50 & 432.75 & & & & 1.07(0.10 ) & & 7(13 ) & & & + 060825 & 0.23 & 63.15 & & & & 1.08(0.04 ) & & 4(6 ) & & 1.64(0.29 ) & + 061007 & 0.09 & 97.82 & & & & & 1.68(0.01 ) & 2153(1880 ) & & & 2.08(0.05 ) + 061019 & 2.90 & 287.03 & & & & 0.95(0.03 ) & & 28(20 ) & & 2.12(0.21 ) & + 070110 & 43.70 & 439.51 & & & & 1.05(0.14 ) & & 9(5 ) & & 2.36(0.24 ) & + 970508 & 30.00 & 7421.93 & 139.67(3.16 ) & -2.73 & 1.21(0.02 ) & 29(21 ) & + 980703 & 81.26 & 343.92 & 214.92(10.15 ) & 1.11 & 2.83 & 7(7 ) & + 990123 & 13.31 & 1907.45 & 155.13(78.79 ) & 0.98(0.10 ) & 1.71(0.10 ) & 12(8 ) & + 990510 & 12.44 & 340.24 & 101.91(12.48 ) & 0.86(0.03 ) & 1.95(0.14 ) & 17(17 ) & + 990712 & 15.25 & 2991.47 & 2000.00(fixed ) & 0.97 & 2.32 & 15(11 ) & + 991216 & 41.17 & 1100.60 & 248.71(67.63 ) & 1.22(0.04 ) & 2.17 & 27(13 ) & + 000301 & 134.00 & 4198.10 & 562.87(18.70 ) & 1.04 & 2.97 & 25(24 ) & + 000926 & 74.48 & 591.61 & 175.18(4.62 ) & 1.48 & 2.49 & 35(24 ) & + 010222 & 13.09 & 2124.75 & 32.12(3.62 ) & 0.43(0.08 ) & 1.29(0.02 ) & 29(48 ) & + 011211 & 34.40 & 2755.47 & 198.66(16.68 ) & 0.85(0.05 ) & 2.36 & 26(33 ) & + 020124 & 5.77 & 2787.67 & 8.47(7.39 ) & 0.76(1.19 ) & 1.85(0.11 ) & 8(9 ) & + 020405 & 85.04 & 882.60 & 236.88(15.90 ) & 1.21 & 2.48 & 6(10 ) & + 020813 & 14.18 & 362.83 & 40.03(0.21 ) & 0.63 & 1.42 & 69(43 ) & + 021004 & 21.12 & 2030.14 & 300.30(fixed ) & 0.82(0.02 ) & 1.39(0.05 ) & 82(90 ) & + 030226 & 17.34 & 609.12 & 88.83(16.30 ) & 0.88(0.12 ) & 2.41(0.12 ) & 10(12 ) & + 030323 & 34.68 & 895.74 & 400.00(fixed ) & 1.29 & 2.11 & 10(10 ) & + 030328 & 4.90 & 227.46 & 18.50(4.32 ) & 0.52(0.09 ) & 1.25(0.05 ) & 52(70 ) & + 030329 & 4.60 & 100.00 & 41.00(0.42 ) & 0.84 & 1.89(0.01 ) & 870(956 ) & + 030429 & 12.53 & 574.04 & 158.73(fixed ) & 0.72(0.03 ) & 2.72 & 30(10 ) & + 030723 & 15.00 & 800.00 & 103.22(5.02 ) & 0.05(0.06 ) & 2.01(0.05 ) & 20(15 ) & + 040924 & 0.95 & 134.12 & 1.49(0.96 ) & 0.34(0.64 ) & 1.11(0.06 ) & 19(10 ) & + 041006 & 0.23 & 550.00 & 14.24(1.15 ) & 0.44(0.02 ) & 1.27(0.01 ) & 97(69 ) & + 050319 & 0.03 & 3.00 & 0.61(0.25 ) & 0.38(0.06 ) & 1.02(0.12 ) & 29(29 ) & + 050525 & 2.83 & 91.80 & 40.72(8.18 ) & 1.02(0.12 ) & 3.00(0.57 ) & 28(5 ) & + 050730 & 0.07 & 358.90 & 11.61(1.95 ) & 0.26(0.08 ) & 1.67(0.09 ) & 58(16 ) & + 050801 & 0.02 & 9.49 & 0.20(0.01 ) & 0.00(0.02 ) & 1.11(0.01 ) & 140(42 ) & + 050820a & 0.12 & 663.30 & 344.98(32.78 ) & 0.88(0.01 ) & 1.48 & 439(25 ) & + 050922c & 0.25 & 69.60 & 3.13(2.75 ) & 0.63(0.13 ) & 1.14(0.10 ) & 14(17 ) & + 051109a & 0.04 & 265.20 & 36.02(8.28 ) & 0.68(0.01 ) & 1.42(0.12 ) & 116(40 ) & + 051111 & 0.03 & 20.00 & 2.61(0.25 ) & 0.79(0.01 ) & 1.70(0.14 ) & 107(84 ) & + 060206 & 20.00 & 201.58 & 71.21(3.65 ) & 1.07(0.02 ) & 1.96 & 25(50 ) & + 060210 & 0.09 & 7.19 & 0.72(0.17 ) & 0.04(0.22 ) & 1.21(0.05 ) & 13(12 ) & + 060526 & 0.06 & 893.55 & 84.45(5.88 ) & 0.67(0.02 ) & 1.80(0.04 ) & 116(56 ) & + 060605a & 0.43 & 111.96 & 8.83(1.21 ) & 0.41 & 2.33(0.16 ) & 2(1 ) & + 060607a & 0.07 & 13.73 & 0.16(fixed ) & -3.07(0.25 ) & 1.18(0.02 ) & 92(35 ) & + 060614 & 20.00 & 934.36 & 112.35(8.53 ) & 0.77(0.10 ) & 2.70(0.07 ) & 16(16 ) & + 060714 & 3.86 & 285.87 & 10.00(fixed ) & 0.01 & 1.41(0.03 ) & 35(11 ) & + 060729 & 70.00 & 662.39 & 297.49(69.62 ) & 1.09(0.10 ) & 2.13(0.44 ) & 18(19 ) & + 061121 & 0.26 & 334.65 & 1.70(0.73 ) & 0.17 & 0.99(0.05 ) & 18(23 ) & + 980326 & 36.46 & 117.68 & & 2.14(0.09 ) & & 15(6 ) & + 991208 & 179.52 & 613.24 & - & 2.30(0.12 ) & & 17(9 ) & + 000131 & 357.44 & 699.06 & - & 2.55(0.29 ) & - & 0(1 ) & + 000418 & 214.27 & 2000.00 & - & 0.81(0.03 ) & - & 13(9 ) & + 000911 & 123.35 & 1466.26 & - & 1.36(0.06 ) & - & 9(2 ) & + 011121 & 33.36 & 1000.00 & - & 1.98(0.06 ) & - & 7(5 ) & + 021211 & 0.13 & 1865.64 & - & 1.18(0.01 ) & - & 78(50 ) & + 050318 & 3.23 & 22.83 & - & 0.84(0.22 ) & - & 0(1 ) & + 050401 & 0.06 & 1231.18 & - & 0.80(0.01 ) & - & 43(12 ) & + 050408 & 8.64 & 434.81 & - & 0.72(0.04 ) & - & 9(15 ) & + 050502 & 6.12 & 29.22 & - & 1.42(0.02 ) & - & 31(19 ) & + 050603 & 34.09 & 219.71 & - & 1.75(0.20 ) & - & 16(7 ) & + 050802 & 0.34 & 127.68 & - & 0.85(0.02 ) & - & 50(10 ) & + 050908 & 1.32 & 57.81 & - & 0.71(0.09 ) & - & 11(10 ) & + 060124 & 3.34 & 1979.30 & - & 0.85(0.02 ) & - & 11(19 ) & + 060418 & 3.92 & 69.53 & - & 1.36(0.04 ) & - & 8(11 ) & + 060904b & 0.50 & 163.13 & - & 0.86(0.02 ) & - & 60(19 ) & + 070110 & 0.66 & 34.76 & - & 0.43(0.08 ) & - & 1(4 ) &
+ radio & & & & + 970508&&&&&@xmath81 ( days)&&bronze & ?
+ 000418&&&&&@xmath197 ( days)&&bronze & ?
+ optical & & & & + 980703 & 1.01(0.02 ) & & 1.11 & 2.83 & 214.92(10.15 ) & 1.71 & silver & ? + 990123 & 0.80(0.10 ) & & 0.98(0.10 ) & 1.71(0.10 ) & 155.13(78.79 ) & 0.73(0.14 ) & silver & ?
+ 990510 & 0.75(0.07 ) & & 0.86(0.03 ) & 1.95(0.14 ) & 101.91(12.48 ) & 1.09(0.14 ) & gold&@xmath198 + 990712 & 0.99(0.02 ) & & 0.97 & 2.32 & 2000 & 1.35 & silver & ? + 991216 & 0.74(0.05 ) & & 1.22(0.04 ) & 2.17 & 248.71(67.63 ) & 0.95(0.04 ) & silver & ? + 000301c & 0.90(0.02 ) & & 1.04 & 2.82 & 562.87(18.70 ) & 1.78 & silver & ? + 000926 & 1.00(0.20 ) & & 1.48 & 2.49 & 175.18(4.62 ) & 1.01 & silver & ? + 011211 & 0.74(0.05 ) & & 0.85(0.05 ) & 2.36 & 198.66(16.68 ) & 1.52(0.05 ) & silver & ?
+ 020124 & 0.91(0.14 ) & & 0.76(1.19 ) & 1.85(0.11 ) & 8.47(7.39 ) & 1.09(1.19 ) & silver & ? + 020405 & 1.23(0.12 ) & & 1.21 & 2.48 & 236.88(15.90 ) & 1.27 & silver & ? + 020813 & 0.85(0.07 ) & & 0.63 & 1.42 & 40.03(0.21 ) & 0.79 & silver & ? + 021004 & 0.39(0.12 ) & & 0.65(0.02 ) & 1.57(0.05 ) & 300.30 & 0.92(0.05 ) & silver & ? + 030226 & 0.70(0.03 ) & & 0.88(0.12 ) & 2.41(0.12 ) & 88.83(16.30 ) & 1.53(0.17 ) & silver & ?
+ 030323 & 0.89(0.04 ) & & 1.29 & 2.11 & 400 & 0.82 & silver & ?
+ 030329 & 0.66 & & 0.84 & 1.89(0.01 ) & 41.00(0.42 ) & 1.05(0.01 ) & gold&@xmath198 + 030429 & 1.22(0.04 ) & & 0.72(0.03 ) & 2.72 & 158.73 & 2.00(0.03 ) & silver & ?
+ 030723 & 1 & & 0.05(0.06 ) & 2.01(0.05 ) & 103.22(5.02)&1.96(0.08 ) & bronze & ?
+ 050525 & 0.97(0.10 ) & & 1.02(0.12 ) & 3.00(0.57 ) & 40.72(8.18 ) & 1.98(0.58 ) & gold&@xmath198 + 050730 & 0.75 &
& 0.26(0.08 ) & 3.00(0.57 ) & 1.67(0.09 ) & 2.74(0.58 ) & bronze & ?
+ 050820a & 0.57(0.06 ) & & 0.88(0.01 ) & 1.48 & 344.98(32.78 ) & 0.60 & gold&@xmath198 + 051109a & 0.65(0.15 ) & & 0.68(0.01 ) & 1.42(0.12 ) & 36.02(8.28 ) & 0.74(0.12 ) & gold&@xmath198 + 051111 & 0.84(0.02 ) & & 0.79(0.01 ) & 1.70(0.14 ) & 2.61(0.25 ) & 0.91(0.14 ) & silver&x + 060206 & 0.70 & & 1.07(0.02 ) & 2.00(0.26 ) & 71.21(3.65 ) & 0.93(0.26 ) & silver&x + 060605 & 0.8 & & 0.41 & 2.33(0.16 ) & 8.83(1.21 ) & 1.92 & bronze & @xmath198 + 060526 & 1.69(0.53 ) & & 0.67(0.02 ) & 1.80(0.04 ) & 84.45(5.88 ) & 1.13(0.04 ) & gold&@xmath198 + 060614 & 0.94(0.08 ) & & 0.77(0.10 ) & 2.70(0.07 ) & 112.35(8.53 ) & 1.93(0.12 ) & gold&@xmath198 + 060729 & 0.74(0.07 ) & & 1.09(0.10 ) & 2.13(0.44 ) & 297.49(69.62 ) & 1.03(0.45 ) & silver&x + x - ray & & & + 980828 & @xmath38 & & 1.44 & 2.6 & 190 & 1.16 & silver &
? + 030329 & 1.17 & 0.8(0.3)&0.87(0.05 ) & 1.84(0.07)&44.93(4.32 ) & 0.97(0.09)&gold&@xmath198 + 050124 & 1.05(0.29 ) & 0.93(0.21 ) & 0.62(0.56 ) & 2.53(0.78 ) & 29.37(12.61 ) & 1.91(0.96 ) & silver & ? + 050128 & 1.05(0.08 ) & 0.95(0.15 ) & 1.00(0.13 ) & 1.98(0.39 ) & 30.70(14.20 ) & 0.98(0.41 ) & silver & ? + 050315 & 1.31(0.12 ) & 1.17(0.07 ) & 0.66(0.03 ) & 1.90(0.23 ) & 224.64(38.68 ) & 1.24(0.23 ) & silver & ? + 050318 & 1.01(0.08 ) & 1.02(0.06 ) & 0.90(0.23 ) & 1.84(0.19 ) & 10.60(4.97 ) & 0.94(0.30 ) & silver&x + 050525a & 1.17(0.18 ) & 1.17(0.18 ) & 1.20(0.03 ) & 1.62(0.16 ) & 13.73(7.47 ) & 0.42(0.16 ) & gold&@xmath198 + 050717 & 0.61(0.08 ) & 0.89(0.12 ) & 0.57(0.21 ) & 1.65(0.12 ) & 1.84(0.95 ) & 1.08(0.24 ) & silver & ? + 050726 & 1.06(0.08 ) & 1.14(0.09 ) & 0.79(0.03 ) & 2.32(0.22 ) & 8.78(1.11 ) & 1.53(0.22 ) & silver & ?
+ 050730 & 0.65(0.03 ) & 0.70(0.03 ) & -0.37(0.25 ) & 2.49(0.04 ) & 6.66(0.29 ) & 2.86(0.25 ) & bronze&@xmath198 + 050802 & 0.92(0.05 ) & 0.89(0.07 ) & 0.32(0.10 ) & 1.61(0.04 ) &
4.09(0.61 ) & 1.29(0.11 ) & bronze&x + 050803 & 0.78(0.10 ) & 1.00(0.08 ) & 0.25(0.03 ) & 2.01(0.07 ) & 13.71(0.90 ) & 1.76(0.08 ) & bronze & ?
+ 050820a & 0.63(0.05 ) & 0.87(0.04 ) & 1.11(0.02 ) & 1.68(0.21 ) & 421.00(179.00 ) & 0.57(0.21 ) & gold&@xmath198 + 050908 & 2.09(0.25 ) & 1.09(0.25 ) & 0.13(0.96 ) & 1.58(0.46 ) & 7.81(5.33 ) & 1.45(1.06 ) & bronze&x + 051006 & 0.61(0.14 ) & 0.84(0.20 ) & 0.57(0.26 ) & 2.23(0.56 ) & 0.93(0.71 ) & 1.66(0.62 ) & silver & ? + 051008 & 1.15(0.32 ) & 1.11(0.10 ) & 0.86(0.09 ) & 2.01(0.19 ) & 14.67(3.82 ) & 1.15(0.21 ) & silver & ? + 051016b & 1.19(0.13 ) & 1.19(0.13 ) & 0.71(0.08 ) & 1.84(0.46 ) & 66.40(23.09 ) & 1.13(0.47 ) & silver & ?
+ 051109a & 0.91(0.07 ) & 0.90(0.07 ) & 0.79(0.07 ) & 1.53(0.08 ) & 27.28(7.90 ) & 0.74(0.11 ) & gold&@xmath198 + 051221a & 1.07(0.36 ) & 1.02(0.19 ) & 1.20(0.06 ) & 1.92(0.52 ) & 354.00(103.00 ) & 0.72(0.52 ) & silver&@xmath198 + 060105 & 1.23(0.05 ) & 1.15(0.03 ) & 0.84(0.01 ) & 1.72(0.02 ) & 2.31(0.14 ) & 0.88(0.02 ) & silver & ? + 060108 & 1.17(0.32 ) & 0.75(0.15 ) & 0.26(0.09 ) & 1.43(0.17 ) & 22.08(7.38 ) & 1.17(0.19 ) & bronze & ?
+ 060124 & 1.10(0.06 ) & 1.06(0.08 ) & 0.81(0.09 ) & 1.66(0.05 ) & 52.60(10.30 ) & 0.85(0.10 ) & silver&x + 060203 & 1.08(0.19 ) & 1.25(0.13 ) & 0.40(0.30 ) & 1.65(0.47 ) & 12.95(6.69 ) & 1.25(0.56 ) & bronze & ? + 060204b & 1.54(0.14 ) & 1.64(0.16 ) & -0.49(0.65 ) & 1.47(0.07 ) & 5.55(0.66 ) & 1.96(0.65 ) & bronze & ?
+ 060210 & 1.12(0.08 ) & 1.11(0.33 ) & 1.00(0.05 ) & 1.85(0.27 ) & 187.00(76.50 ) & 0.85(0.27 ) & silver&x + 060211a & 1.15(0.06 ) & 1.11(0.26 ) & 0.38(0.08 ) & 1.63(1.27 ) & 267.24(165.67 ) & 1.25(1.27 ) & bronze & ? + 060313 & 0.84(0.34 ) & 0.78(0.09 ) & 0.82(0.03 ) & 1.76(0.18 ) & 11.18(2.89 ) & 0.94(0.18 ) &
+ 060319 & 0.93(0.22 ) & 1.25(0.11 ) & 0.84(0.02 ) & 1.92(0.30 ) & 99.70(26.78 ) & 1.08(0.30 ) & silver & ?
+ 060323 & 0.99(0.16 ) & 1.02(0.13 ) & -0.11(0.23 ) & 1.55(0.16 ) & 1.29(0.32 ) & 1.66(0.28 ) & bronze & ?
+ 060428a & 1.11(0.24 ) & 0.97(0.10 ) & 0.48(0.03 ) & 1.46(0.37 ) & 125.31(47.19 ) & 0.98(0.37 ) & bronze & ?
+ 060502a & 1.11(0.29 ) & 1.15(0.13 ) & 0.53(0.03 ) & 1.68(0.15 ) & 72.57(15.05 ) & 1.15(0.15 ) & bronze & ?
+ 060510a & 1.04(0.05 ) & 1.06(0.14 ) & 0.93(0.14 ) & 1.77(0.10 ) & 47.70(16.70 ) & 0.84(0.17 ) & silver & ? + 060526 & 1.07(0.09 ) & 1.08(0.16 ) & 0.42(0.12 ) & 1.58(0.34 ) & 11.60(6.39 ) & 1.16(0.36 ) & gold&@xmath198 + 060605 & 0.62(0.17 ) & 0.83(0.09 ) & 0.45(0.04 ) & 1.80(0.13 ) & 7.14(0.93 ) & 1.35(0.14 ) & bronze&@xmath198 + 060614 & 0.96(0.16 ) & 0.93(0.06 ) & 1.03(0.02 ) & 2.13(0.07 ) & 36.60(2.40 ) & 1.10(0.07 ) & gold&@xmath198 + 060807 & 1.18(0.09 ) & 1.40(0.20 ) & 0.96(0.24 ) & 1.92(0.12 ) & 14.90(5.88 ) & 0.96(0.27 ) & silver & ? + 060813 & 0.99(0.05 ) & 1.10(0.07 ) & 0.87(0.03 ) & 1.63(0.13 ) & 15.20(3.88 ) & 0.76(0.13 ) & silver & ? + 060814 & 1.30(0.05 ) & 1.30(0.05 ) & 1.06(0.12 ) & 2.38(0.40 ) & 68.60(23.30 ) & 1.32(0.42 ) & silver & ? + 060906 & 1.28(0.37 ) & 1.12(0.17 ) & 0.35(0.10 ) & 1.97(0.36 ) & 13.66(3.29 ) & 1.62(0.37 ) & bronze & ?
+ 060908 & 1.01(0.22 ) & 1.00(0.08 ) & 0.70(0.07 ) & 1.49(0.09 ) & 0.95(0.34 ) & 0.79(0.11 ) & bronze & ? + 060927 & 0.65(0.19 ) & 0.92(0.15 ) & 0.73(0.32 ) & 1.82(2.60 ) &
4.24(8.22 ) & 1.09(2.62 ) & silver & ? + 061121 & 0.71(0.03 ) & 0.96(0.07 ) & 0.75(0.06 ) & 1.63(0.05 ) & 24.32(4.38 ) & 0.88(0.08 ) & silver & ? + 061201 & 0.30(0.15 ) & 0.30(0.15 ) & 0.57(0.07 ) & 1.61(0.23 ) & 2.09(0.75 ) & 1.04(0.24 ) & bronze & ? + 061222a & 1.45(0.06 ) & 1.22(0.12 ) & 0.81(0.07 ) & 1.86(0.06 ) & 60.50(8.89 ) & 1.05(0.09 ) & silver & ? + 070103 & 1.32(0.25 ) & 1.52(0.21 ) & 0.20(0.10 ) & 1.63(0.08 ) & 2.88(0.48 ) & 1.43(0.13 ) & bronze & ? + 050315&1.95&i&2.76&1.00&2.45&2.9&55.06&52.17(0.05)&11.71&16.61&1 + 050318&1.44&i&2.08&1.01&6.91&1.6&53.30&49.91(0.16)&11.86&18.00&2 + 050319&3.24&i&2.16&1.00&4.93&@xmath1992.5&53.76&@xmath19950.75&11.69&17.70&3 + 050401&2.9&ii&2.98&0.20&2.59&@xmath1994.2&54.97&@xmath19952.40&12.51&@xmath19918.00&4 + 050416a&0.65&i&2.32&10.78&0.72&@xmath1999.3&51.94&@xmath19950.06&11.80&@xmath19918.00&5 + 050505&4.27&i&2.1&1.00&6.28&@xmath1992.5&54.03&@xmath19951.00&11.56&17.53&6 + 050525a&0.606&ii&3.34&0.99&0.10&2.6&53.98&50.99(0.06)&12.79&@xmath19918.69&7 + 050820a&2.61&i&2.01&1.00&11.05&3.6&54.88&52.17(0.14)&9.65&17.08&8 + 050922c&2.2&ii&2.44&1.49&3.16&@xmath1992.9&53.24&@xmath19950.35&13.15&@xmath19918.00&9 + 051016b&0.94&i&2.18&3.30&1.31&4.82&52.24&49.79(0.12)&10.31&18.00&10 + 051221a&0.5465&i&2.14&9.76&0.79&12.06&51.53&49.87(0.10)&10.19&18 & 11 + 060124&2.3&ii&3.12&0.23&1.06&1.5&55.41&51.91(0.06)&12.11&@xmath19918.00&12 + 060206&4.05&i&2.62&1.00&2.65&1.8&54.48&51.18(0.02)&11.74&16.90&13 + 060210&3.91&i&2.24&1.00&5.75&2.7&54.33&51.39(0.14)&12.25&17.37&14 + 060502a&1.51&ii&3.3&0.79&0.15&2.1&54.88&51.71(0.07)&12.04&18.00&15 + 060512&0.44&ii&3.36&1.00&0.12&@xmath1996.1&52.38&@xmath19950.14&13.41&@xmath19919.09&16 + 060522&5.11&i&2.26&1.00&2.91&@xmath1994.8&53.17&@xmath19950.73&11.50&17.81&17 + 060526&3.21&i&2.14&1.01&5.26&2.82&53.41&50.49(0.02)&11.58&17.65&18 + 060604&2.68&i&2.54&1.00&2.38&@xmath1994.6&53.75&@xmath19951.26&12.02&17.59&19 + 060605&3.7&ii&2.98&0.60&2.38&@xmath1991.6&54.21&@xmath19950.81&13.32&@xmath19918.00&20 + 060614&0.13&ii&2.72&1.00&0.43&6.8&52.45&50.30(0.02)&10.74&@xmath19918.32&21 + 060714&2.71&i&2.12&1.00&4.26&@xmath1995.1&53.32&@xmath19950.91&11.17&17.94&22 + 060729&0.54&i&2.26&1.00&2.42&6.6&53.39&51.21(0.08)&9.93&17.54&23 + 060814&0.84&i&2.60&2.33&1.41&3.63&53.34&50.64(0.12)&12.6&18 0&24 + 060908&2.43&ii&2.5&1.00&1.62&@xmath1998.4&52.78&@xmath19950.81&11.99&@xmath19918.26&25 + 060912&0.94&i&2.01&1.02&7.06&@xmath1994.4&52.88&@xmath19950.35&9.46&17.98&26 + 060926&3.2&i&2.01&1.01&15.71&@xmath1990.9&54.21&@xmath19950.26&11.21&17.21&27 + 061007&1.26&ii&3.16&1.00&1.93&@xmath1997.6&53.99&@xmath19951.94&14.35&@xmath19918.22&28 + 061121&1.31&ii&2.7&0.95&1.01&1.93&53.88&50.63(0.06)&11.57&18.01&29 + 070110&2.35&i&2.72&1.00&1.70&@xmath1997.8&54.31&@xmath19952.27&11.38&16.98&30 + 970508&0.835&i&47160&7.13&1.1&@xmath200 & 2160.00(432.00 ) & 16.7&0.391&2.28&3.4&1.34&52.53&51.15(0.07)&11.26&18.00 + 970828&0.958&ii&14400&118&1.44(0.07)&@xmath201&190.08(34.56)&3.9&0.128&3.2&0.99&0.80&54.31&51.68(0.06)&13.46&18.20 + 980703&0.966&i&122400&4(1)&1.24(0.18)&@xmath202&214.92(10.15)&6.1&0.2&2.1&1.01&3.52&52.91&50.67(0.02)&9.81&17.87 + 990123&1.6&i&84240&19.11(2.2)&1.41(0.05)&@xmath203&155.13(787.86 ) & 2.9&0.089&2.98&0.40&1.51&54.77&51.87(0.18)&12.32&18.00 + 990510&1.619&i&42120&32.8(1.4)&1.41(0.18)&@xmath204&101.91(124.81)&3.1&0.054&2.38&1.00&3.63&53.98&51.14(0.04)&12.07&17.62 + 990705&0.84&ii&52200&1.9(0.6)&&1.05&86.40(17.28)&3.8&0.096&3.1&0.99&0.22&53.52&50.85(0.06)&12.18&18.67 + 991216&1.02&ii&39240&250(10)&1.61(0.07)&@xmath205&248.71(67.63)&3.7&0.051&2.01&1.0&11/11&54.79&52.12(0.11)&9.57&17.29 + 000926&2.307&i&197640&2.23(0.77)&&@xmath206&175.18(4.62)&3.3&0.14&2.01&1.00&7.11&54.12&51.34(0.01)&8.29&17.16 + 010222&1.477&i&117720&1.87(0.18)&1.33(0.04)&@xmath207&80.35(12.96)&2.7&0.08&2.02&1.00&7.59&54.21&51.25(0.05)&9.21&17.29 + 011211&2.14&ii&29600&0.248&0.95(0.02)&@xmath208&198.66(16.68)&3.5&-&3.32&0.99&0.21&54.11&51.39(0.02)&12.84&18.31 + 020405&0.689&i&147600&13.6(2.5)&1.15(0.95)&@xmath209&236.88(15.90)&5.7&0.285&2.02&1.00&5.55&53.53&51.21(0.02)&8.63&17.67 + 020813&1.254&ii&114840&22&1.42(0.05)&@xmath210&397.44(0.864)&2.2&0.066&2.6&0.55&1.86&54.08&50.95(0.01)&11.57&18.00 + 021004&2.323&i&113040&4.3(0.7)&1(0.2)&@xmath211&300.30(8.64)&4.7&0.24&2.2&1.00&3.69&53.61&51.13(0.01)&10.85&17.54 + 030329&0.1678&i&22377&157.0(8.7 ) & 1.2(0.1)&@xmath212 & 40.95(0.43 ) & 3.8&0.052&2.34&2.15&2.23&53.09&50.43(0.01)&11.95&18.00 + | the _ swift_/xrt data of 179 grbs ( from 050124 to 070129 ) and the optical afterglow data of 57 pre- and post-_swift _ grbs are analyzed , in order to investigate jet - like breaks in the afterglow lightcurves . using progressively rigorous definitions of jet breaks , we explore whether the observed breaks in the x - ray and optical lightcurves can be interpreted as a jet break , and their implications in understanding grb energetics if these breaks are jet breaks .
we find that not a single burst can be included in the `` platinum '' sample , in which the data satisfy all the criteria needed to define a jet break , i.e. a clear achromatic break observed in both the x - ray and optical bands , and that the pre- and post - break decay segments satisfy the closure relations in the same , simplest jet model .
however , by releasing one or more requirements to define a jet break , some jet - break candidates of various degrees could be identified . in the x - ray band ,
42 out of the 103 well - sampled x - ray lightcurves have a decay slope of the post - break segment @xmath0 ( `` bronze '' sample ) , and 27 of them also satisfy the closure relations of the forward shock models for both the pre- and post- break segments ( `` silver '' sample ) .
the numbers of the `` bronze '' and `` silver '' candidates in the optical lightcurves are 27 and 23 , respectively .
thirteen bursts have well - sampled optical and x - ray lightcurves , but only seven cases are consistent with an achromatic break , but even in these cases only one band satisfies the closure relations ( `` gold '' sample ) .
the breaks in other grbs are all chromatic .
the observed break time in the xrt lightcurves is statistically earlier than that in the optical bands .
all these raise great concerns in interpreting jet - like breaks as jet breaks and further inferring grb energetics from these breaks . on the other hand ,
if one _ assumes _ that these breaks are jet breaks , one can proceed to perform a similar analysis as previous work to study grb collimation and energetics .
we have performed such an analysis with the `` silver '' and `` gold '' jet break candidates .
we calculate the jet opening angle ( @xmath1 ) and kinetic energy ( @xmath2 ) or their lower limits with the ism forward shock models using the x - ray afterglow data .
the derived @xmath2 distribution reveals a much larger scatter than the pre-_swift _ sample . a tentative anti - correlation between @xmath1 and @xmath3
is found for both the pre-_swift _ and _ swift _ grbs , indicating that the @xmath2 could still be quasi - universal , if the breaks in discussion are indeed jet breaks . |
membrane proteins are fundamental for life , and their structures and dynamics are essential for their biological functions . about 30 % of proteins encoded in genomes are estimated to be of membrane proteins by bioinformatics.@xcite although membrane - protein folding has been studied extensively by experiments,@xcite only about 2 % in whole known structures in pdb are membrane proteins , because biomembrane environment makes crystallization very difficult.@xcite thus , simulation studies are getting more important ( for previous simulations , see , for instance , @xcite ) .
however , simulations often suffer from sampling insufficiency and the efficient sampling methods like generalized - ensemble algorithms and/or the reduction of the size of systems are required . in particular , replica - exchange method ( rem ) @xcite and its extensions are often used in generalized ensemble algorithms due to their efficiency , parallelization ease , and usability ( for reviews , see , e.g. , refs.@xcite ) .
one of useful approaches to reduce system sizes is to employ an implicit membrane model , which mimics some elements of membrane properties such as dielectric profile , chain order , pressure profile , and intrinsic curvature by parameters for electrostatic solvent free energy.@xcite while these methods are mainly based on the free energy difference between solvent and solute , simpler implicit membrane model was introduced previously , where transmembrane helices keep a helix structure and are always restricted within membrane regions during folding , which greatly reduces the effort for the search in the conformational space during folding processes.@xcite this model assumed that the native structure of membrane proteins can be predicted by helix - helix interactions between transmembrane helices with fixed helix structures , and that the membrane environment constraints the regions where helices can exist ( namely , within membranes ) and stabilizes transmembrane helix structures .
this model is supported by many experimental data such as those leading to the two - stage model , in which each helix structure is formed first , and they aggregate each other by helix - helix packing in membrane protein folding to reach the native conformation ( for a review , see ref .
@xcite ) . the previous method @xcite could predict the native structures by the rem simulation using known native helix structures ( for a review , see ref .
@xcite ) . however ,
if the native structures consist of distorted helix structures , the previous prediction method will not work because the method treated helix structures as rigid bodies .
it is actually known from experimental structures in pdb that transmembrane helices are distorted or bent in about 25 % of all transmembrane helix structures .
@xcite therefore ,
in this article , we propose a new treatment of helix structures by taking into account helix distortions and kinks instead of treating them as rigid bodies .
we tested our new prediction method for native structures .
our test systems consist of the case with only ideal helix structures and that with a distorted helix structure .
this article is organized as follows . in section 2 , we explain the details of our methods .
the potential energy function used for our new models and the method to introduce the helix kinks are described . in section 3 , we show the results of the rem simulation applied to glycophorin a and phospholamban .
after we check that rem simulation are properly performed , the free energy minimum states are identified by the principal component analysis .
finally , section 4 is devoted to the conclusions .
we first review our previous method@xcite .
only the transmembrane helices are used in our simulations , and loop regions of membrane proteins as well as lipid and water molecules are neglected .
our assumptions are that a role of water is to push the hydrophobic transmembrane regions of membrane proteins into the lipid bilayer and that a major role of lipid molecules is to prepare a hydrophobic environment and construct helix structures in the transmembrane regions .
loop regions of membrane proteins are often outside the membrane and we assume that they do not directly affect the structure of transmembrane regions . due to the difference in surface shapes of helices and lipids , the stabilization energy for
helix - helix packing will be larger than that for helix - lipid packing .
therefore , water , lipids , and loop - region of proteins are not treated explicitly in our simulations , although the features of membrane boundaries are taken into account by the constraint conditions below .
we update configurations with a rigid translation and rotation of each @xmath0-helix and torsion rotation of side - chains by monte carlo ( mc ) simulations .
we use mc method although we can also use molecular dynamics in principle .
there are 2@xmath1 + @xmath2 kinds of mc move sets , where @xmath1 is the total number of transmembrane helices in the protein , and @xmath2 is the total number of dihedral angles in the side - chain of @xmath1 helices .
we add the following three elementary harmonic constraints to the original potential energy function .
the constraint function is given by @xmath3 where each term on the right - hand side is defined as follows : @xmath4 ^ 2 , \label{const - ene1}\end{aligned}\ ] ] @xmath5 ^ 2\right .
\nonumber \\ & + \left.k_2~ \theta \left ( \left| z^{\rm u}_{i}-z^{\rm u}_{0 } \right| -d^{\rm u } \right ) \left [ \left| z^{\rm u}_{i}-z^{\rm u}_{0 } \right| -d^{\rm u } \right]^2 \right\ } , \label{const - ene2}\end{aligned}\ ] ] @xmath6 ^ 2 .
\label{const - ene3}\end{aligned}\ ] ] @xmath7 is the energy that constrains pairs of adjacent helices along the amino - acid chain not to be apart from each other too much ( loop constraints ) .
@xmath8 is the distance between the c atom of the c - terminus of the @xmath9-th helix and the c@xmath10 atom of the n - terminus of the @xmath11-th helix , and @xmath12 and @xmath13 are the force constant and the central value constant of the harmonic constraints , respectively , and @xmath14 is the step function : @xmath15 this term has a non - zero value only when the distance @xmath8 becomes longer than @xmath13 .
only the structures in which the distance between neighboring helices in the amino - acid sequence is short are searched because of this constraint term .
@xmath16 is the energy that constrains helix n - terminus and c - terminus to be located near membrane boundary planes . here
, the z - axis is defined to be the direction perpendicular to the membrane boundary planes .
@xmath17 is the force constant of the harmonic constraints . @xmath18 and @xmath19 are the z - coordinate values of the c@xmath10 atom of the n - terminus or c - terminus of the @xmath9-th helix near the fixed lower membrane boundary and the upper membrane boundary , respectively . @xmath20 and @xmath21 are the fixed lower boundary z - coordinate value and the upper boundary z - coordinate value of the membrane planes , respectively . @xmath22 and @xmath23 are the corresponding central value constants of the harmonic constraints .
this term has a non - zero value only when the c@xmath10 atom of the n - terminus or c - terminus of the @xmath9-th helix are apart more than @xmath24 ( or @xmath25 ) .
this constraint energy was introduced so that the helix ends are not too much apart from the membrane boundary planes .
@xmath26 is the energy that constrains all c@xmath10 atoms within the sphere ( centered at the origin ) of radius @xmath27 .
@xmath28 is the distance of c@xmath10 atoms from the origin , and @xmath29 and @xmath27 are the force constant and the central value constant of the harmonic constraints , respectively .
this term has a non - zero value only when c@xmath10 atoms go out of this sphere and is introduced so that the center of mass of the molecule stays near the origin .
the radius of the sphere @xmath27 is set to a large value in order to guarantee that a wide conformational space is sampled .
these constraints are considered to be a simple implicit membrane model which mimics membrane environment during membrane protein folding .
moreover , all constraints limit the conformational space of proteins to improve sampling and are useful when we use limited computational resources . in summary , this procedure is consistent with the two - stage model , and it assumes that side - chain flexibility is essential in their folding . because backbone structures of main chain are treated as rigid bodies in the previous method , the method can not be applied if transmembrane helices are distorted . however , most of transmembrane helix structures in pdb have distorted or bent helix structures .
we , therefore , need to treat the deformations of backbone helix structures during simulations .
namely , the @xmath30 and @xmath31 torsion rotations and concerted rotation of backbone are used to reproduce the distorted helix structures of experimental structures from the initial ideal helix structures in monte carlo move sets . here , we also update configurations with a rotation of torsion angles of backbones by directional manipulation and concerted rotation.@xcite there are 2@xmath1 + @xmath2 + @xmath32 + @xmath33 kinds of mc moves now , where @xmath32 is the total number of @xmath34 torsion angles in the helix backbones , and @xmath33 is the total number of the combination of seven successive backbone torsion angle by the concerted rotation in the helix backbone .
one mc step in this article is defined to be an update of one of these degrees of freedom , which is accepted or rejected according to the metropolis criterion . in order to keep helix conformations of the distortions ,
we introduce the fourth constraint term as follows : @xmath35 @xmath36 where @xmath37 is the newly - introduced energy term which constrains dihedral angles of main chains within bending or kinked helix structures from ideal helix structures and prevent them from bending and distortions too much . @xmath38 and @xmath39 are the main - chain torsion angles of the @xmath40-th residue . @xmath41 and @xmath42 are the fixed reference values of the harmonic constraint , @xmath43 and @xmath44 are the force constants , and @xmath45 are the central values of the harmonic constraint .
we now explain the replica - exchange method briefly .
this method prepares @xmath46 non - interacting replicas at @xmath46 different temperatures .
while conventional canonical mc simulation is performed for each replica , temperature exchange between pairs of replicas corresponding to temperatures is attempted at a fixed interval based on the following metropolis criterion .
let the label @xmath9 ( = 1 , @xmath47 , @xmath46 ) correspond to the replica index and label @xmath48 ( = 1 , @xmath47 , @xmath46 ) to the temperature index .
we represent the state of the entire system of @xmath46 replicas by @xmath49 } , \cdots , x_{m(m)}^{[m ] } \right\}$ ] , where @xmath50 } = \left\{q^{[i]}\right\}$ ] are the set of coordinates of replica @xmath9 ( at temperature @xmath51 ) , and @xmath52 is the permutation of @xmath9 . the boltzmann - like probability distribution for state @xmath53 is given by @xmath54})]}.\ ] ] we consider exchanging a pair of temperatures @xmath51 and @xmath55 , corresponding to replicas @xmath9 and @xmath40 : @xmath56 } , \cdots , x_n^{[j ] } , \cdots \right\ } \rightarrow \nonumber \\ & x^\prime = \left\ { \cdots , x_m^{[j ] } , \cdots , x_n^{[i ] } , \cdots \right\ } .\end{aligned}\ ] ] the transition probability @xmath57 of metropolis criterion is given by @xmath58 } \mid x_n^{[j ] } ) \nonumber \\ & = { \rm min}\left(1 , \frac{w_{{\rm rem } } ( x^\prime)}{w_{{\rm rem } } ( x)}\right ) \nonumber \\ & = { \rm min}(1,\exp(- \delta ) ) , \end{aligned}\ ] ] where @xmath59 } ) - e(q^{[i ] } ) ) $ ] . because each replica reaches various temperatures followed by replica exchange , the rem method performs a random - walk in temperature space during the simulation .
expectation values of physical quantities are given as functions of temperatures by solving the multiple - histogram reweighting equations.@xcite the density of states @xmath60 and dimensionless helmholtz free energy are obtained by solving the following equations iteratively : @xmath61 and @xmath62 where @xmath63 and @xmath64 be the energy histogram and the total number of samples obtained of temperature @xmath51 , respectively .
after we obtained @xmath65 at each temperature , the expectation value of a physical quantity @xmath66 at any temperature @xmath67 is given by @xcite @xmath68 where @xmath69 are the set of coordinates at temperature @xmath51 obtained from the trajectories of the simulation .
we analyze the simulation data by the principal component analysis ( pca).@xcite the structures are superimposed on an arbitrary reference structure , for example , the native structure from pdb .
the variance - covariance matrix is defined by @xmath70 where @xmath71 and @xmath72 .
@xmath73 are cartesian coordinates of the @xmath9-th atom , and @xmath74 is the total number of atoms .
this symmetric 3@xmath74 @xmath75 3@xmath74 matrix is diagonalized , and the eigenvectors and eigenvalues are obtained . for this calculation , we used the r program package.@xcite the first superposition is performed to remove large eigenvalues from the translations and rotations of the system , because we want to analyze the internal differences of structures .
therefore , this manipulation results in the smallest value close to zero for the six eigenvalues corresponding to translations and rotations of the center of geometry .
the eigenvalues are ordered in the decreasing order of magnitude .
thus , the first , second , and third principal component axes are defined as the eigenvectors corresponding to the largest , second largest , and third largest eigenvalues , respectively .
the @xmath9-th principal component of each sampled structure @xmath76 is defined by the following inner product : @xmath77 where @xmath78 is the ( normalized ) @xmath9-th eigenvector .
the mc program is based on charmm macromolecular mechanics program,@xcite and replica - exchange monte carlo method was implemented in it . in this work ,
we studied two membrane proteins : glycophorin a and phospholamban .
both proteins are registered in orientation of proteins in membrane ( opm).@xcite the former has a dimer of an almost ideal helix structure in pdb ( pdb code : 1afo ) .
the number of amino - acid residues in the helix is 18 , and the sequence is identical and tliifgvmagvigtilli .
the other has a single transmembrane helix structure in pdb ( pdb code : 1fjk ) .
the number of amino - acid residues in the helix is 25 , and the sequence is lqnlfinfclilifllliciivmll .
the n - terminus and the c - terminus of each helix were blocked with the acetyl group and the n - methyl group , respectively . in the previous works , a 13-replica rem mc simulation of glycophorin a was performed with 13 replicas with the following temperatures : 200 , 239 , 286 , 342 , 404 , 489 , 585 , 700 , 853 , 1041 , 1270 , 1548 , and 1888 k. @xcite
although this simulation predicted the structures close to the native one successfully , the backbones structures were fixed to the ideal helix structures . in the present simulation , the flexibility of backbone helix structures is newly taken into account , and 16 replicas were used with the following temperatures : 300 , 333 , 371 , 413 , 460 , 512 , 571 , 635 , 707 , 787 , 877 , 976 , 1087 , 1210 , 1347 , and 1499 k. the total number of mc steps was 60,000,000 . for phospholamban , 16 replicas were also used with the following temperatures : 300 , 340 , 386 , 438 , 497 , 564 , 640 , 727 , 825 , 936 , 1062 , 1205 , 1368 , 1553 , 1762 , and 2000 k. the total number of mc steps was 100,000,000 .
the above temperatures were chosen so that all acceptance ratios of replica exchange are almost uniform and sufficiently large for computational efficiency .
the highest temperature was chosen sufficiency high so that no trapping in local - minimum - energy states occurs in both simulations .
replica exchange was attempted once at every 1000 mc steps for glycophorin a and 100 mc steps for phospholamban , respectively .
we used the charmm19 parameter set ( polar hydrogen model ) for the original potential energy of the system.@xcite no cutoff was introduced to the non - bonded terms .
each structure was first minimized subjected to harmonic restraint on all the heavy atoms .
the value of the dielectric constant was set as @xmath79 = 1.0 , as in the previous works.@xcite because previous studies showed that this value was better for the predictions of transmembrane helix structures than that of @xmath79 = 4.0 , although @xmath79 = 4.0 is close to the lipid environment of electrostatic potential effects .
this may be due to the fact that few lipid molecules lie between helices in native transmembrane structures . for concerted rotation we selected the backbone atoms except for those in cysteine residues .
we selected 6 or 7 continuous bonds from the first atom along backbone for the driver torsion .
third bond and fifth bond were allowed to rotate following the driver bonds .
the number of degrees of freedom in total was equal to 190 in glycophorin a and 132 in phospholamban .
we set @xmath80 for glycophorin a and @xmath81 for phospholamban @xmath82 ( kcal / mol)/@xmath83 , @xmath84 , @xmath85 ( kcal / mol)/@xmath83 , @xmath86 ( kcal / mol)/@xmath83 , @xmath87 , @xmath88 ( kcal / mol)/degrees@xmath83 , @xmath89 degrees , @xmath90 degrees , @xmath91 degrees , and @xmath92 degrees for our simulations . for membrane thickness parameters
, we set @xmath93 , @xmath94 , and @xmath95 for glycophorin a , and @xmath96 , @xmath97 , and @xmath95 for phospholamban .
for pca analyses , 60,000 and 100,000 conformational data were chosen in a fixed interval at each temperature from the rem simulation for glycophorin a and phospholamban , respectively .
we used the pdb structures ( pdb codes : 1afo for glycophorin a and 1fjk for phospholamban ) as the reference structures to judge the prediction ability .
we first examine how the replica - exchange simulation performed . fig .
[ 1afo - integrate](a ) shows the time series of the replica index at the lowest temperature of 300 k. we see that the minimum temperature visited different replicas many times during the rem simulation , and we observe a random walk in the replica space .
the complementary picture is the temperature exchange for each replica .
[ 1afo - integrate](b ) shows the time series of temperatures for one of the replicas ( replica 11 ) .
we see that replica 11 visited various temperatures during the rem simulation .
we observe random walks in the temperature space between the lowest and highest temperatures .
other replicas behaved similarly .
[ 1afo - integrate](c ) shows the corresponding time series of the total potential energy for replica 11 .
we see a strong correlation between time series of temperatures ( fig .
[ 1afo - integrate](b ) ) and that of potential energy ( fig .
[ 1afo - integrate](c ) ) , as is expected .
we next examine how widely the conformational space was sampled during the rem simulation .
we plot the time series of the root mean - square deviation ( rmsd ) of all the c@xmath10 atoms from the experimental structure ( pdb code : 1afo ) for replica 11 in fig .
[ 1afo - integrate](d ) .
when the temperature becomes high , the rmsd takes large values , and when the temperature becomes low , the rmsd takes small values . by comparing figs .
[ 1afo - integrate](b ) and [ 1afo - integrate](d ) , we see that there is a positive correlation between the temperature and the rmsd values .
the fact that the rmsds at high temperatures are large implies that our simulations did not get trapped in local - minimum potential - energy states .
these results confirm that the rem simulation was properly performed .
time series of various quantities for the rem simulation of glycophorin a. ( a ) time series of replica index at temperature 300 k. ( b ) time series of temperature change for replica 11 .
( c ) time series of total potential energy change for replica 11 .
( d ) time series of the rms deviation ( in ) of all the c@xmath10 from the pdb structures for replica 11 . ]
table [ 1afo accept ] lists the acceptance ratios of replica exchange between all pairs of nearest neighboring temperatures .
we find that the acceptance ratio is high enough ( @xmath98 0.1 ) in all temperature pairs .
[ 1afo - wham](a ) shows the canonical probability distributions of the potential energy obtained from the rem simulation at 16 temperatures .
we see that the distributions have enough overlaps between the neighboring temperature pairs .
this ensures that the number of replicas was sufficient . in fig .
[ 1afo - wham](b ) , the average potential energy and its components , namely , the electrostatic energy @xmath99 , van der waals energy @xmath100 , torsion energy @xmath101 , and constraint energy @xmath102 , are shown as functions of temperature , which were calculated by eq .
( [ wham ] ) . because the helices are generally far apart from each other at high temperatures , the energy components , especially electrostatic energy and van der waals energy , are higher at high temperatures . at low temperatures , on the other hand
, the side - chain packing among helices is expected .
we see that as the temperature becomes lower , @xmath100 , @xmath101 , and @xmath99 decrease almost linearly up to @xmath103 1200 k , and as a result @xmath104 is also almost linearly decreasing up to @xmath103 1200 k. on the other hand , when the temperature becomes @xmath105 1200 k , @xmath100 contributes more to the decrease of @xmath104 .
this is reasonable , because @xmath100 decreases as a result of side - chain packing and the stability of the conformation increases .
note that we used only transmembrane regions in the rem simulation .
transmembrane helices are generally considered to be hydrophobic , and helix - helix association is sometimes considered only by vdw packing ( lock - and - key model ) .
however , fig .
[ 1afo - wham](b ) shows that @xmath99 also changes much as a function of temperature .
this implies that electrostatic effects also contribute to the formation of the native protein conformation .
.acceptance ratios of replica exchange corresponding to pairs of neighboring temperatures from the rem simulation of glycophorin a.[1afo accept ] [ cols="<,^,<,^ " , ] the following abbreviations are used : str : the number of structures , tote : total potential energy , elec : electrostatic energy , vdw : van der waals energy , dih : dihedral energy , geo : constraint energy ( all in kcal / mol ) , rmsd : root - mean - square deviation of all c@xmath106 atoms ( in ) .
( color online ) typical structures of phospholamban in each cluster selected by free energy local minimum state .
the purple structure is native structure .
the rmsd from the native conformation with respect to the backbone atoms is 1.27 and 2.89 for cluster 1 and cluster 2 , respectively . ]
table [ 1fjk allcluster ] lists average quantities of two clusters of similar structures .
the rows of cluster 1 and cluster 2 represent various average values for the structures that belong to each cluster .
we see that rmsd is as small as 2.06 for cluster 1 , while it is 2.93 for cluster 2 .
hence , cluster 1 has very similar structures to the native one .
however , it is not the global - minimum free energy state but a local - minimum one , comparing the number of conformations ( str entries in table [ 1fjk allcluster ] ) in both clusters . in fig .
[ 1fjk - minimumenestr ] , representative structures of each cluster in table [ 1fjk allcluster ] and the structure obtained by solution nmr experiments ( pdb code : 1fjk ) are shown .
we confirm that cluster 1 is very similar to the native structure .
it is bent at the same position and in the same direction , although the amount of bent is not as much as the native one .
cluster 2 is also bent at the same position and about the same amount as the native one , but it has a bend in the opposite direction .
hence , the present simulation can predict the position of bend , but it gives both directons of bend as local - minimum free energy states and cluster 2 as the global - minimum one .
the present system is a helix monomar , and without interactions with other helices , it seems very difficult to decide the direction of distorsions within the approximation of the present method .
we remark that a preliminary rem simulation of bacteriorhodopsin with seven helices predicts correct directions of helix bending ( manuscript in preparation ) .
in this article , we introduced deformations of helix structures to the replica - exchange monte carlo simulation for membrane protein structure predictions .
the membrane bilayer environment was approximated by restraining the conformational space in virtual membrane region .
the sampled helix structures were limited so that helix structures by introducing the restraints on the backbone @xmath30 and @xmath31 angles are not completely destroyed . in order to check the effectiveness of the method , we first applied it to the prediction of a dimer membrane protein , glycophorin a. we successfully reproduced the native - like structure as the global - minimum free energy state .
we next applied the method to phospholamban , which has one distorted transmembrane helix structure in the pdb structure .
the results implied that a native - like structure was obtained as a local - minimum free energy state .
two local - minimum free energy states were found with the same bend position as the native one , but the global - minimum free energy state had an opposite direction of helix bend .
therefore , our results seem to imply that the location of bends of helix structures in transmembrane helices are determined by their amino - acid sequence , but the direction and amount of distortion of helices are dependent on the interactions with surrounding lipid molecules , which we represented only implicitly .
our next targets will be more complicated membrane proteins with multiple transmembrane helices such as g protein coupled receptors . our preliminary results for bacteriorhodopsin show that native - like structures with the correctly bent helices can be predicted by our method .
some of the computations were performed on the supercomputers at the institute for molecular science , at the supercomputer center , institute for solid state physics , university of tokyo , and center for computational sciences , university of tsukuba .
this work was supported , in part , grants - in - aid for scientific research ( a ) ( no .
25247071 ) , for scientific research on innovative areas ( dynamical ordering & integrated functions ) , program for leading graduate schools integrative graduate education and research in green natural sciences , and for the computational materials science initiative , and for high performance computing infrastructure from the ministry of education , culture , sports , science and technology ( mext ) , japan . | we propose an improved prediction method of the tertiary structures of @xmath0-helical membrane proteins based on the replica - exchange method by taking into account helix deformations .
our method allows wide applications because transmembrane helices of native membrane proteins are often distorted . in order to test the effectiveness of the present method
, we applied it to the structure predictions of glycophorin a and phospholamban .
the results were in accord with experiments . |
the structure of the @xmath3 resonance , which will be referred to as `` the roper '' in this article , has been a long - standing mystery .
its mass and large width are difficult to understood in a framework of quark models @xcite .
the decay branches imply that it is a mixture of @xmath5 and @xmath6 states ( dominated by the mixture of @xmath7 and @xmath8 channels ) .
recently , the precision data from clas at jlab @xcite ( single and double pion electroproduction ) , and a global analysis from maid @xcite in the region @xmath9 gev@xmath10 have become available .
the extracted @xmath2 helicity amplitudes can give hints on the internal structure of the roper .
indeed , the amplitudes ( and the transition form factors ) extracted from the experiments , turned out to show a surprising @xmath4 dependence . in a simple quark model picture ,
the roper can be regarded as the first radial excitation of the nucleon .
recent experimental data at large @xmath4 @xcite support this picture @xcite . in the high @xmath4 region
( @xmath11 gev@xmath10 ) the data is consistent with the predicted @xmath4 dependence of relativistic constituent and light - front quark models @xcite .
these models , however , fail to explain the lower @xmath4 data , particularly at the photon point , @xmath12 .
this discrepancy has been interpreted as a manifestation of the missing meson cloud effects in the models @xcite .
unfortunately , the magnitude of the meson cloud contributions is less constrained , and they are more model dependent than the valence quark contributions , which are dominant at larger @xmath4 .
( in general , the meson cloud contributions are expected to fall as @xmath4 increases . ) in this study , we estimate the meson cloud contributions using a covariant spectator constituent quark model which is consistent with the large @xmath4 region data of the clas .
the present model requires no extra parameters except for those already fixed by the previous studies for the nucleon @xcite .
the @xmath2 transition amplitudes have been studied in several ways : treating the roper as a quark - gluon system @xcite , a pure valence quark system @xcite , and a valence quark system dressed by @xmath13 pairs @xcite .
relativity has proven to be very important , not only for a large @xmath4 , but also for the region @xmath14 .
it may even decide the sign of the amplitudes near @xmath12 @xcite .
the nucleon to roper transition amplitudes have been also studied using dynamical baryon - meson coupled - channel models .
they usually treat the baryons and mesons as effective degrees of freedom , and the baryons are dressed with a meson cloud nonperturbatively .
( for details , see e.g. , refs .
@xcite and a review @xcite . )
the nonvalence quark degrees of freedom was also introduced by coupling the pion and other meson fields to the valence quarks @xcite .
the @xmath2 transition was also studied in lattice qcd @xcite . in this study
we compute the @xmath2 transition form factors using a valence quark model based on the covariant spectator formalism @xcite .
relativity is implemented consistently . in this model
a baryon is described as a quark - diquark system , where the diquark has all possible polarizations ( spin state 0 or 1 ) and acts as a spectator .
the isolated quark in the baryon interacts with the photon in the impulse approximation .
the model has been applied successfully for the studies of spin 1/2 and spin 3/2 low - lying baryons @xcite .
we assume that the roper is the first radial excitation of the nucleon , which also ensures that it is orthogonal to the nucleon ( valence quark ) state . using a model with no extra new adjustable parameters ,
we compute the @xmath4 dependence of the transition form factors . we find an excellent agreement with the data @xcite in the high @xmath4 region .
furthermore , we extend the model to a lattice qcd regime and compare with the lattice data for a large pion mass ( @xmath15 mev ) , and find also a good agreement , particularly in the low @xmath4 region , @xmath16 gev@xmath10 . note that under the two conditions , high @xmath4 and lattice simulations with heavy pions , meson cloud effects are expected to become small or negligible @xcite .
the two agreements mentioned above , support that the present model can describe well the valence quark contributions .
encouraged by the successful features of the model , we estimate the meson cloud contributions focusing on the low @xmath4 region .
this article is organized as follows : in sec .
[ secamp ] general relations among the current , helicity amplitudes , and transition form factors are given .
formalisms , wave functions , and explicit expressions for the transition form factors are presented in sec .
[ secff ] . in sec .
[ secresults ] numerical results are presented , and a comparison is made with the lattice simulation data .
furthermore , meson cloud contributions are estimated in the present approach . finally , discussions and conclusions are given in sec .
[ secconclusions ] .
the electromagnetic transition between a nucleon ( mass @xmath17 ) and a spin 1/2 positive parity nucleon resonance @xmath18 ( mass @xmath19 ) can be written using the dirac @xmath20 and pauli @xmath21 type form factors , defined by the current @xcite , & & j^= |u_r ( p_+ ) \ { ( ^- ) f_1^+ f_2^ } u(p_- ) , + & & [ eqjgen ] where , @xmath22 ( @xmath23 ) is the @xmath18 ( nucleon ) dirac spinor , @xmath24 ( @xmath25 ) is the final ( initial ) momentum and @xmath26 .
spin projection indices are suppressed for simplicity .
usually , experimental data measured for hadron electromagnetic structure are reported in terms of the helicity amplitudes in a particular frame .
the most popular choice is the rest frame of the final state , projecting the current on the photon polarization states , @xmath27 , where @xmath28 is the photon spin projection . in the @xmath18 rest frame ,
the helicity amplitudes , @xmath29 and @xmath30 , are given by @xcite : & & a_1/2(q^2)= k n^,+| _ + j |n , - , [ eqa1 ] + & & s_1/2(q^2)= k n^,+| _ 0 j |n , + . [ eqs0 ] here , @xmath31 is the magnitude of the electron charge with @xmath32 , and = , with @xmath33 .
@xmath34 is the photon momentum in the @xmath18 rest frame , |*q*|= , [ eqq2 ] where @xmath35 , with @xmath36 .
the helicity amplitudes , @xmath29 and @xmath30 , can be related with the form factors @xmath20 and @xmath21 via eqs .
( [ eqjgen])-([eqs0 ] ) @xcite : & & a_1/2(q^2)= r \ { f_1^(q^2 ) + f_2^(q^2 ) } , [ eqa12 ] + & & s_1/2(q^2)= |*q*| \ { f_1^(q^2 ) -f_2^(q^2 ) } , [ eqs12 ] where @xmath37 , and = .
note that the amplitude @xmath30 is determined by virtual photons and not specified at @xmath12 , but obtained only in the limit @xmath38 . in this case , one has & & a_1/2(0 ) = r f_2^(0 ) , [ eqa0 ] + & & s_1/2(0 ) = ( m_r+m ) k \ { ( 0 ) - } .
+ & & the inverse relations for @xmath39 in terms of the helicity amplitudes are : & & f_1^(q^2 ) = \ { a_1/2(q^2 ) + s_1/2(q^2 ) } , + & & [ eqf1 ] + & & f_2^(q^2)= \ { a_1/2 ( q^2 ) - s_1/2 ( q^2 ) } . + & & [ eqf2 ]
if the amplitudes @xmath29 and @xmath30 are finite at @xmath12 , eq . ( [ eqf1 ] ) implies that f_1^(0)=0 . as for @xmath40 , there is no particular condition . for the roper
a fit to the data suggests a small value for @xmath41 @xcite .
thus , at low @xmath4 , @xmath20 and @xmath21 are determined essentially by @xmath42 .
so far , a covariant spectator quark model has been developed and applied successfully to the spin 1/2 @xcite and 3/2 @xcite ground states with no radial excitations . in the model ,
a three - quark baryon is described as an isolated quark which interacts with a photon , and a spectator diquark .
the wave function of a baryon is represented in terms of the flavor and spin states of the quark and diquark combined with the relative angular momentum . in the spirit of the covariant spectator theory
the quark pair degrees of freedom are integrated out to form an on - shell diquark with a certain mass @xmath43 . to describe the momentum distribution of the quark - diquark system in a baryon @xmath44
, we introduce a scalar wave function @xmath45 , which depends on the relative angular momentum and the radial excitation of the system . as the baryon and
the diquark are on - shell the scalar wave function @xmath45 can be written as a function of @xmath46 @xcite , where @xmath47 is the baryon total momentum and @xmath48 the diquark momentum .
we represent that dependence in term of the dimensionless variable @xcite , _
b= , [ eqchi ] where @xmath44 is the baryon index ( @xmath49 , etc ) and @xmath50 the mass .
the nucleon wave function , @xmath51 , can be written in the simplest s - wave model @xcite : _
n(p , k)= _ n(p , k ) , [ eqpsin ] where @xmath52 represents isospin ( @xmath53 ) or spin ( @xmath54 ) states corresponding to the total magnitude of either 0 or 1 in the diquark configuration @xcite .
( see appendix [ appsqm ] for the detail . )
the wave function represented by eq .
( [ eqpsin ] ) , satisfies the dirac equation @xcite . the scalar wave function @xmath55 ( s state )
is given by _
n(p , k ) = , [ eqpsin ] where @xmath56 is obtained by inserting @xmath57 in eq .
( [ eqchi ] ) , and @xmath58 the normalization constant . in a parameterization
where @xmath59 , @xmath60 is associated with the long range physics , while @xmath61 the short range physics . in the present approach ,
the spin and isospin of the roper are the same as those of the nucleon . assuming the roper to be the first radial excitation of the nucleon , we can write the roper scalar wave function @xmath62 as _ r(p , k ) & = & n_r + & & , [ eqpsir ] where @xmath63 is a new parameter which will be discussed later , and @xmath64 the normalization constant .
the sign of the normalization constant can be fixed by the experimental data @xcite .
this particular form with the order one polynomial in @xmath65 ( as @xmath66 ) , is motivated by the harmonic - oscillator potential model for the three - quark system .
a similar dependence was adopted in ref .
note that the wave function represented by eq .
( [ eqpsir ] ) preserves the short range behavior as presented in the nucleon scalar wave function , and that it simultaneously modifies the long range properties through the factor @xmath67 .
the roper wave function @xmath68 also satisfies the dirac equation with mass @xmath19 .
the parameter @xmath63 in the roper scalar wave function in eq .
( [ eqpsir ] ) , will be fixed by the orthogonality condition between the nucleon and the roper states .
@xmath64 will be fixed by @xmath69 at @xmath12 , the same as that for the nucleon @xcite .
thus , the description of the roper requires no extra parameters to be fixed for the present purpose , since the parameters @xmath70 and @xmath61 have already been fixed @xcite . in the covariant spectator quark model
the transition current for the @xmath2 reaction can be written in a relativistic impulse approximation @xcite , j^= 3 _ _ k _ r(p_+,k ) j_i^_n(p_-,k ) , [ eqjsp ] where @xmath71 with @xmath72 the diquark on - shell energy , @xmath73 ( @xmath36 ) , and @xmath74 is the quark current parameterized as j_i^= j_1(q^2 ) ^+ j_2(q^2 ) .
[ eqji ] the dirac ( @xmath75 ) and pauli ( @xmath76 ) quark form factors in the above are also decomposed as j_i(q^2)= f_i+(q^2 ) + f_i-(q^2 ) _ 3 , ( i=1,2 ) .
[ eqji2 ] the quark form factors @xmath77 are normalized as @xmath78 and @xmath79 ( isoscalar and isovector quark anomalous moments ) .
their explicit expressions can be found in refs .
@xcite and in appendix [ appsqm ] .
using the expressions for the nucleon and roper wave functions , eqs .
( [ eqpsin])-([eqpsir ] ) , and the hadronic current , eq .
( [ eqjsp ] ) , one gets explicit expressions for @xmath80 and @xmath81 : f_1^(q^2)&= & j_1 i + j_3 i + & & - j_4 i , [ eqf1b ] + f_2^(q^2)&= & j_2 i - j_3 i + & & + j_4 i , [ eqf2b ] where @xmath82 is the overlap integral for the roper and nucleon scalar wave functions : ( q^2)= _ k _ r(p_+,k ) _ n ( p_-,k ) , and j_(i+2)&= & _ 3 j_i_3 + & = & f_i+(q^2 ) - f_i-(q^2 ) _ 3 , ( i=1,2 ) . [ eqjii ] equations ( [ eqf1b])-([eqf2b ] ) are the main expressions of the present model for the transition form factors . at @xmath12 ,
one has f_1^(0)&= & ( j_1+j_3 ) i ( 0 ) , + & = & ( 1 + _ 3 ) i(0 ) . in this case , the desired result , @xmath83 , is ensured only if @xmath84 .
but this is indeed fulfilled by the orthogonality condition between the nucleon and the roper wave functions .
more detail will be discussed in next section .
the orthogonality condition between the nucleon and the roper states in the present approach is : ( 0)= _ k _ r(|p_+,k ) _ n(|p_-,k)=0 .
[ eqi0 ] this can be regarded as a generalization of the nonrelativistic orthogonality condition when @xmath12 in the final roper rest frame , namely , @xmath85 and @xmath86 . in appendix
[ apporth ] , we discuss more on eq .
( [ eqi0 ] ) and the nonrelativistic orthogonality condition .
note that the condition eq .
( [ eqi0 ] ) is not automatically verified for the nucleon and the roper wave functions in eqs .
( [ eqpsin])-([eqpsir ] ) .
the orthogonality condition is satisfied only for a particular value of @xmath63 that sets @xmath84 . a direct consequence of the orthogonality condition eq .
( [ eqi0 ] ) , is that @xmath83 , and also that @xmath87 .
@xmath83 is consistent with the general properties of the @xmath2 reaction .
@xmath87 is a prediction of our model as a consequence of the orthogonality condition eq .
( [ eqi0 ] ) , but it also corresponds to an approximation , since the sea quark contributions are ignored in the present approach
. a more accurate treatment would give a small value for @xmath88 .
the nucleon and the roper wave functions are described by eq .
( [ eqpsin ] ) with their scalar functions , eqs .
( [ eqpsin])-([eqpsir ] ) .
the parameters for the roper wave function are determined by those of the nucleon @xcite : @xmath89 and @xmath90 , and by the orthogonality condition eq .
( [ eqi0 ] ) to give @xmath91 and @xmath92 . before presenting the results ,
we emphasize again that there are no new free parameters to be fixed in the model .
thus , for a given @xmath4 , the form factors can be calculated using eqs .
( [ eqf1b])-([eqf2b ] ) .
the valence quark contributions calculated in the present model for the @xmath20 and @xmath21 form factors are shown in fig .
[ figf1f2 ] by the solid lines . in fig .
[ figf2 ] also a part magnified from fig .
[ figf1f2 ] for @xmath21 is shown .
the magnitudes of the present results are consistent with constituent and light - front quark models @xcite . for convenience ,
we present in fig .
[ figa12s12 ] the helicity amplitudes calculated in the roper rest frame , using the relations eqs .
( [ eqa12])-([eqs12 ] ) . in fig .
[ figf1f2 ] we also compare the valence quark contributions with the clas data for @xmath20 and @xmath21 .
the clas data were extracted combining dispersion relations with a unitary isobar model analysis @xcite .
as one can see our result is very close to the data in the region @xmath93 gev@xmath10 ( see especially fig . [ figf2 ] for @xmath21 ) , which supports the idea that the meson cloud contributions are suppressed in the high @xmath4 region , and the assumption that the roper is the first radial excitation of the nucleon .
this achievement may be impressive since we have introduced no extra new parameters as already mentioned .
one only may identify the long range behavior of the roper wave function with that of the nucleon , and may ensure the orthogonality between the roper and the nucleon wave functions . in fig .
[ figf1f2 ] we also show a fit to the data from ref .
@xcite , using the code maid2007 ( abbreviated maid ) @xcite .
both the clas data and the maid fit are similar for @xmath94 gev@xmath10 region , although some differences may be noticeable , in particular for @xmath20 .
this discrepancy in @xmath20 can be also a consequence of the different data sets used in the analysis between the clas data and the maid fit .
our predictions for @xmath20 are closer to the clas analysis than to the maid fit , although the differences are comparable with the error bars of the clas data .
next , we discuss the asymptotic behavior for @xmath20 and @xmath21 as @xmath95 .
perturbative qcd ( pqcd ) @xcite predicts @xmath96 and @xmath97 as @xmath95 , apart from the @xmath98 corrections .
the predictions of the present model from eqs .
( [ eqf1b])-([eqf2b ] ) , are consistent with these results : @xmath99 and @xmath100 , with @xmath101 .
( see appendix g of ref .
@xcite for details . ) for the helicity amplitudes , our results are also consistent with the pqcd predictions and @xmath102 .
note that the extra power in @xmath103 in contrast to @xmath104 .
this takes account of changing the helicity one unit between the initial and final states @xcite .
thus , one gets , @xmath105 and @xmath106 . ] : @xmath107 @xcite , again apart from the logarithmic corrections .
note , however , that based on what observed for the nucleon @xcite , and especially for the @xmath108 transition form factors @xcite , the predicted scaling behaviors may not be observed in such a small @xmath4 region .
also it is not realistic to expect that our calibration of the quark current based on the reactions in the regime @xmath109 gev@xmath10 , can be naively extended to the high @xmath4 region where pqcd is valid ( such as @xmath110 gev@xmath10 ) . extending the model to the lattice simulation regime
, we compare our results with the lattice qcd results .
hereafter , the results obtained in the lattice regime will be referred to as `` lattice regime '' , which will be explained below .
one can expect that , the heavier the pion mass becomes in the lattice simulations , the closer our results become to the lattice one , since the meson cloud effects become smaller .
a comparison is made in fig .
[ figf1f2lat ] for the lattice data corresponding to @xmath111 mev @xcite . to extend the model to the lattice regime
, we include an implicit dependence on the pion mass for the hadron masses , following the procedure proposed in refs .
this is done by replacing the hadron masses in the model by the respective lattice masses in the baryon wave function and the quark electromagnetic current [ eq . ( [ eqji ] ) ] . in the lattice regime ( the dashed lines in fig .
[ figf1f2lat ] ) we have used the nucleon and the roper masses , @xmath112 gev and @xmath113 gev , respectively , corresponding to the pion mass @xmath114 mev @xcite .
we have also used the value , 1.083 gev for the @xmath115 meson mass , according to the parameterization in ref .
@xcite .
one can see in fig .
[ figf1f2lat ] that the lattice regime results are fairly consistent with the lattice data for @xmath21 . as for @xmath20 ,
the lattice results overestimate our predictions for @xmath116 gev@xmath10 , as well as the covariant spectator model results `` spectator ( valence ) '' [ the same as those shown in figs .
[ figf1f2 ] and [ figf2 ] ] .
however , for @xmath117 gev@xmath10 , the agreement is excellent .
more lattice qcd simulation data with smaller pion masses are desired to constrain better the valence quark contributions in the low @xmath4 region .
the agreement between the lattice regime results and the lattice qcd data , supports that the present estimate of the valence quark contributions is reasonable . to estimate the meson cloud effects
, we take the valence quark contributions estimated in sec .
[ secphysical ] as a reference .
the valence quark contributions will be denoted by @xmath118 .
we need to know a full contribution to estimate the meson cloud contributions . as a first step approach
, we regard the maid fit results @xcite , which will be denoted by @xmath119 , as the total contribution for the form factors .
then , the meson cloud contributions @xmath120 , may be estimated by f_i^mc(q^2)= f_i^maid(q^2)-f_i^bare(q^2 ) , ( i=1,2 ) .
the results for @xmath120 are presented in fig .
[ figmc ] .
to have an idea for the uncertainties in this estimate we use an analytical form for the each form factor upper limit of the clas data error bars [ denoted by @xmath121 below ] , and calculate the bands with one - standard deviation . specifically , we use @xmath122 and @xmath123 .
first , we analyze @xmath20 . a situation for @xmath20
is a bit subtle .
the magnitude of the data is smaller than that of the @xmath21 .
the upper limit for @xmath124 is around 0.1 , which is about half of that for @xmath125 .
the meson cloud contributions are negative , and they can be between 0.01 to 0.03 in the high @xmath4 region . probably , the contributions of 0.03 are overestimated , since they are expected to become small in the high @xmath4 region .
we note however , the meson cloud contributions here are estimated by subtracting the model results from the maid fit .
there are differences in the clas data and the maid fit in the high @xmath4 region , since the maid analysis @xcite uses a different data set from that of the clas @xcite .
the clarification between the differences in the two data sets , and more precision data in this high @xmath4 region ( less contaminations from meson cloud ) , are desired to constrain better the meson cloud contributions for @xmath20 .
next we discuss @xmath21 . because of the larger magnitude of @xmath21
, the contributions from the meson cloud may be easier to estimate .
the meson cloud contributions for @xmath21 are negative as well as those for @xmath20 . as one can see in fig .
[ figmc ] the meson cloud gives main contributions in the low @xmath4 region .
( at @xmath12 , they are the only contributions . )
however , as expected , the meson cloud contributions become small for @xmath116 gev@xmath10 , and the valence quark contributions dominate . in higher @xmath4 region ,
the magnitude of the meson cloud contributions are of order of the error bars ( @xmath126 ) of the data .
overall , our estimate of the meson cloud contributions qualitatively agree with those of the recent works @xcite , particularly for @xmath20 . in terms of the helicity
amplitudes the meson cloud contributions at @xmath12 are estimated by @xmath127 gev@xmath128 and @xmath129 gev@xmath128 . here
the meson cloud is the only contribution for @xmath130 . as for @xmath41 ,
the meson cloud contributions cancel significantly with those of the valence quark .
this feature can be understood better by observing fig .
[ figa12s12 ] .
thus , we predict negative contributions from the meson cloud for @xmath29 and @xmath30 in the low @xmath4 region .
this is consistent with the estimate made in ref .
@xcite , where meson cloud contributions are at least 70% in @xmath130 ( 100% in our case ) , although in their case the bare contributions differ in sign from our results .
a similar difference exists also for @xmath30 , in contradiction to the clas data in the high @xmath4 region .
also ref .
@xcite predicts significant negative contributions from meson cloud for @xmath30 in the low @xmath4 region , implying large negative values for @xmath30 in that region , contrary to what is suggested by the clas data @xcite .
we also attempt to compare our results with those of the recent calculations based on the dynamical meson - baryon coupled - channel model of sato and lee @xcite . there , the bare contributions are very small for @xmath130 @xcite . in the region ,
@xmath131 gev@xmath10 , the absolute magnitudes of the meson cloud contributions ( their amplitudes are complex numbers ) are dominant in both amplitudes , although the meson cloud contributions decrease in the high @xmath4 region for @xmath29 @xcite .
unfortunately , their analysis @xcite leads to the amplitudes different from those of the clas .
thus , it is expected that their meson cloud contributions and ours differ quantitatively .
in this work we have studied the transition form factors for @xmath2 reaction using a covariant spectator quark model .
the model is fully relativistic , and has no new adjustable parameters .
this is fulfilled by the orthogonality condition between the nucleon and the roper states .
the model can describe the reaction well in the high @xmath4 region ( 1.5 gev@xmath10 @xmath132 gev@xmath10 ) , supporting the idea that the roper is the first radial excitation of the nucleon . in the low @xmath4 region ( @xmath133 gev@xmath10 )
the model predictions deviate from the experimental data , and this fact suggests that the meson cloud contributions ( missing in the model ) are significant in this @xmath4 region . to check whether or not the description of our estimate for the valence quark contributions is realistic , we have extended the model to the lattice regime to compare with the heavy - pion lattice qcd data .
our results are in an excellent agreement with them in the low @xmath4 region .
this fact supports that the valence quark degrees of freedom is well described in our model . having a good confidence in the description of the valence quark sector
, we have estimated the meson cloud contributions for the dirac ( @xmath20 ) and pauli ( @xmath21 ) type transition form factors .
a characteristic feature of the model , @xmath87 , is a direct consequence of the orthogonality condition between the nucleon and the roper states .
however , this may not have significant consequences in the estimate of the meson cloud contributions , since they appear to be dominant at low @xmath4 , or conversely , the valence quark contributions are expected to be small at @xmath12 @xcite . in this work
we have shown that the transition form factors @xmath20 and @xmath21 for the @xmath2 reaction may be more suitable quantities to study than the helicity amplitudes in order to disentangle the meson cloud contributions .
for example , @xmath124 is small for both in experimental data and the valence quark model because of the cancellation between the @xmath29 and @xmath30 amplitudes .
this implies also the meson cloud contributions are small . as for @xmath21 ,
it is mainly determined by @xmath42 in the very low @xmath4 region , which measures the meson cloud contributions in our approach ( for @xmath38 ) .
we admit that our estimate of the meson cloud contributions has some limitations due to the parameterization of the amplitudes @xmath29 and @xmath30 based on the maid analysis .
furthermore , we have faced that there are some discrepancies between the two data sets of the clas and maid in the high @xmath4 region .
more precision experimental data in this high @xmath4 region are desired to constrain better the meson cloud contributions . for a possible improvement
, there is a room to add one extra free parameter to the model .
the parameter @xmath60 ( fixed by the nucleon wave function ) in the factor @xmath134 in the roper wave function , can be replaced by a new free parameter @xmath135 to introduce a new long range scale , and may be adjusted by more precise , high @xmath4 data .
needless to say , more lattice qcd data are also useful to constrain the valence quark contributions better .
we plan to extend the present model for higher mass resonance region , like for instance , @xmath136 . in the past dynamical coupled - channel models
are very successful in the description of the meson - baryon electro- and photoreaction , involving resonances such as @xmath137 , @xmath3 and @xmath136 .
however , such models require a parameterization in the interaction with the quark core . in general
, it is based on the baryon - meson phenomenology and not based on the quark ( and gluon ) degrees of freedom .
the approach developed in this work is very promising to study the valence quark contributions in the meson - baryon systems
. it can be tested , or compared with the bare parameters used in the dynamical coupled - channel models .
an independent test may be to compare with the bare contributions ( no meson cloud ) determined by dynamical meson - baryon coupled - channel models by fitting e.g. , to the cross section data at each @xmath4 @xcite .
this kind of an extraction of the bare contributions was done for the @xmath108 reaction @xcite , as well as other resonances including the roper @xcite .
however , often the meson cloud contributions become larger than the quark core contributions , and an estimate of the bare contributions are sometimes difficult to get as a smooth function of @xmath4 .
another possible method is to use the lattice qcd data in the heavy - pion regime to fix the valence quark contributions .
this method was successfully applied in ref .
@xcite for the nucleon to @xmath138 electromagnetic transition .
* acknowledgments : * the authors thank b. juli - daz , h. kamano , m. t. pea , a. sibirtsev and a. stadler for helpful discussions , and a. w. thomas for suggestions in the manuscript .
this work was support by jefferson science associates , llc under u.s .
doe contract no .
de - ac05 - 06or23177 , and in part by the european union ( hadronphysics2 project `` study of strongly interacting matter '' ) .
g. r. was also supported by the portuguese fundao para a cincia e tecnologia ( fct ) under the grant no .
sfrh / bpd/26886/2006 .
k. t. acknowledges the cssm ( adelaide , australia ) for hospitality , where this work was completed .
below , we present some details of the covariant spectator quark model . in the covariant spectator quark model the s - state wave function for the nucleon with the ( initial ) momentum @xmath25 is given by @xcite , _
n(p_-,k)= _ n(p_-,k ) , where @xmath48 is the diquark momentum and @xmath139 is the combination of the quark flavors associated with the isospin 0 or 1 diquark .
the spin-1 polarization vector is represented by @xmath140 in a fixed - axis base @xcite , and @xmath23 and @xmath141 are the spinors related by @xcite , u^(p)= _ 5 ( ^- ) u(p ) .
the roper wave function with the ( final ) momentum @xmath24 is defined by _ r(p_+,k)= _
r(p_+,k ) , where @xmath22 and @xmath142 are spin states associated with a spin 1/2 particle and mass @xmath19 , and @xmath143 is the polarization vector . in both cases the scalar functions ,
@xmath144 and @xmath145 , are the functions of @xmath146 and @xmath147 , respectively .
the quark current associated with eqs .
( [ eqji ] ) and ( [ eqji2 ] ) is expressed in terms of the quark form factors @xmath148 , inspired by a vector meson dominance form : & & f_1(q^2)= + ( 1- ) + c _ , + & & f_2 ( q^2)= _ \ { d_+ ( 1-d _ ) } . in the above , @xmath149 defines the quark charge in deep inelastic scattering , @xmath150 are the isoscalar and isovector quark anomalous magnetic moments .
the mass @xmath151 ( @xmath152 ) corresponds to the light ( heavy ) vector meson , and @xmath153 , @xmath154 are the mixture coefficients . in the present model we set @xmath155 ( @xmath156 ) for the light vectorial meson and @xmath157 ( twice the nucleon mass ) to represent the short range physics .
the values of the parameters were previously fixed by the nucleon elastic form factors @xcite , and the values are presented in table [ tabparam ] .
note that the present model uses @xmath158 .
.quark current parameters . [
cols="^,^,^,^,^,^,^",options="header " , ] in the present model , the asymptotic expressions for @xmath95 , associated with the quark current of eqs .
( [ eqji2 ] ) and ( [ eqjii ] ) are given by & & j_1 , [ eqaj1 ] + & & j_3 ( c_+-c_- ) , + & & j_2 ( _ + + 3_- ) , [ eqj2 ] + & & j_4 ( _ + - _ - ) , [ eqaj4 ] with = d_+ m_v^2 + ( 1-d_+ ) m_h^2 , where , we have used the relation , @xmath158 , corresponding to the parameterization of the model ii in ref .
in particular , @xmath159 gev@xmath10 should be noted .
using eqs .
( [ eqaj1])-([eqaj4 ] ) we can derive the expressions for @xmath20 and @xmath21 for @xmath95 .
in a static , nonrelativistic formalism the wave function is a function of the particle s three - momentum , @xmath160 . in the limit where there is no momentum transfer , the arguments in the initial ( @xmath161 ) and final ( @xmath162 ) wave functions are the same .
this corresponds to @xmath163 , or @xmath164 .
then , the overlap integral between the orthogonal states , @xmath165 , is given by _
f^*(*k * ) _ i(*k*)=0 .
[ eqorth ] the equivalent relation for the covariant spectator theory for eq .
( [ eqorth ] ) is . _
k _ f^*(p_+,k ) _ i ( p_-,k ) |_q^2=0=0 , [ eqint1 ] where the subindex @xmath12 indicates that @xmath166 the simplest case is in the initial ( or final ) baryon s rest frame . for the equal mass case , @xmath167 .
next , consider the inelastic case with the masses of the initial and final states , @xmath17 and @xmath19 , respectively .
the generalization of the condition eq .
( [ eqint1 ] ) would correspond to @xmath168 and @xmath169 , with @xmath170 when @xmath171 .
( @xmath34 is given by eq .
( [ eqq2 ] ) . )
this gives also q^2q^2= -(m_r - m)^2 , which will be denoted by the pseudo - threshold point , the point where both initial and final state are at rest .
this situation is , however , unphysical for @xmath172 because @xmath173 .
the generalization of eq .
( [ eqint1 ] ) for unequal mass case would be , .
_ k _ f^*(p_+,k ) _ i ( p_-,k ) |_q^2=q^2=0 .
[ eqint2 ] since the physical reaction are restricted to @xmath174 , we need to redefine the orthogonality condition in the unequal mass case .
to do so we impose the condition , _ k _ f^*(|p_+,k ) _ i ( |p_-,k)=0 , [ eqint3 ] where the four - momenta @xmath175 and @xmath176 are respectively defined by , & & |p_-= ( , 0,0,- ) , + & & |p_+=(m_r,0,0,0 ) , which correspond to @xmath12 , but @xmath177 . the use of eq .
( [ eqint3 ] ) may be regarded as an approximation , or the simplest relativistic extension for the orthogonality condition .
the exact treatment needs to impose the overlap integral to vanish at the pseudo - threshold as in eq .
( [ eqint2 ] ) .
however , it would require an extension of the wave functions to the unphysical region , and beyond a scope of the present study .
instead , we use eq .
( [ eqint3 ] ) .
both prescriptions should give similar results when @xmath178 is small enough .
h. clement _ et al .
_ , arxiv : nucl - ex/0612015 .
a. v. sarantsev _ et al .
_ , phys .
b * 659 * , 94 ( 2008 ) [ arxiv:0707.3591 [ hep - ph ] ] .
n. suzuki , b. julia - diaz , h. kamano , t. s. lee , a. matsuyama and t. sato , arxiv:0909.1356 [ nucl - th ] .
i. g. aznauryan , phys .
c * 76 * , 025212 ( 2007 ) [ arxiv : nucl - th/0701012 ] .
i. g. aznauryan _ et al . _
[ clas collaboration ] , phys .
c * 80 * , 055203 ( 2009 ) [ arxiv:0909.2349 [ nucl - ex ] ] .
d. drechsel , s. s. kamalov and l. tiator , eur .
j. a * 34 * , 69 ( 2007 ) [ arxiv:0710.0306 [ nucl - th ] ] . l. tiator and m. vanderhaeghen , phys .
b * 672 * , 344 ( 2009 ) [ arxiv:0811.2285 [ hep - ph ] ] .
i. g. aznauryan , v. d. burkert and t. s. lee , arxiv:0810.0997 [ nucl - th ]
. v. d. burkert , aip conf .
* 1056 * , 348 ( 2008 ) [ arxiv:0808.2326 [ nucl - ex ] ] . b. julia - diaz , t. s. h. lee , a. matsuyama , t. sato and l. c. smith , phys .
c * 77 * , 045205 ( 2008 ) [ arxiv:0712.2283 [ nucl - th ] ] .
f. gross , g. ramalho and m. t. pea , phys .
c * 77 * , 015202 ( 2008 ) [ arxiv : nucl - th/0606029 ] .
z. p. li , phys .
d * 44 * , 2841 ( 1991 ) ; z. li , v. burkert and z. li , phys .
d * 46 * , 70 ( 1992 ) .
m. warns , w. pfeil and h. rollnik , phys .
d * 42 * , 2215 ( 1990 ) ; m. warns , h. schroder , w. pfeil and h. rollnik , z. phys .
c * 45 * , 627 ( 1990 ) .
s. capstick and b. d. keister , phys .
d * 51 * , 3598 ( 1995 ) [ arxiv : nucl - th/9411016 ] .
f. cardarelli , e. pace , g. salme and s. simula , phys .
b * 397 * , 13 ( 1997 ) [ arxiv : nucl - th/9609047 ] .
m. m. giannini , e. santopinto and a. vassallo , eur .
j. a * 12 * , 447 ( 2001 ) [ arxiv : nucl - th/0111073 ] . b. julia - diaz , d. o. riska and f. coester , phys .
c * 69 * , 035212 ( 2004 ) [ erratum - ibid .
c * 75 * , 069902 ( 2007 ) ] [ arxiv : hep - ph/0312169 ] .
h. j. weber , phys .
c * 41 * , 2783 ( 1990 ) .
f. cano and p. gonzalez , phys .
b * 431 * , 270 ( 1998 ) [ arxiv : nucl - th/9804071 ] .
q. b. li and d. o. riska , phys .
c * 74 * , 015202 ( 2006 ) [ arxiv : nucl - th/0605076 ] .
e. santopinto and r. bijker , few body syst .
* 44 * , 95 ( 2008 ) .
m. dillig and m. schott , phys .
c * 75 * , 067001 ( 2007 ) [ erratum - ibid .
c * 76 * , 019903 ( 2007 ) ] [ arxiv : nucl - th/0405063 ] .
b. julia - diaz and d. o. riska , nucl .
a * 780 * , 175 ( 2006 ) [ arxiv : nucl - th/0609064 ] .
o. krehl , c. hanhart , c. krewald and j. speth , phys .
c * 62 * , 025207 ( 2000 ) [ arxiv : nucl - th/9911080 ] .
b. julia - diaz , t. s. lee , a. matsuyama and t. sato , phys .
c * 76 * , 065201 ( 2007 ) [ arxiv:0704.1615 [ nucl - th ] ] .
b. julia - diaz , t. s. h. lee , t. sato and l. c. smith , phys .
c * 75 * , 015205 ( 2007 ) [ arxiv : nucl - th/0611033 ] . b. julia - diaz , h. kamano , t. s. h. lee , a. matsuyama , t. sato and n. suzuki , phys .
c * 80 * , 025207 ( 2009 ) [ arxiv:0904.1918 [ nucl - th ] ] .
n. suzuki , t. sato and t. s. lee , arxiv:0910.1742 [ nucl - th ] .
b. golli , s. sirca and m. fiolhais , eur .
j. a * 42 * , 185 ( 2009 ) [ arxiv:0906.2066 [ nucl - th ] ] . b. golli and s. sirca , few body syst . * 44 * , 157 ( 2008 ) .
s. schneider , s. krewald and u. g. meissner , eur .
j. a * 28 * , 107 ( 2006 ) [ arxiv : nucl - th/0603040 ] .
s. s. kamalov , s. n. yang , d. drechsel , o. hanstein and l. tiator , phys .
c * 64 * , 032201(r ) ( 2001 ) [ arxiv : nucl - th/0006068 ] .
v. d. burkert and t. s. h. lee , int .
j. mod .
e * 13 * , 1035 ( 2004 ) [ arxiv : nucl - ex/0407020 ] .
k. bermuth , d. drechsel , l. tiator and j. b. seaborn , phys .
d * 37 * , 89 ( 1988 ) .
y. b. dong , k. shimizu , a. faessler and a. j. buchmann , phys .
c * 60 * , 035203 ( 1999 ) .
d. y. chen and y. b. dong , commun .
phys . *
50 * , 142 ( 2008 ) .
h. w. lin , s. d. cohen , r. g. edwards and d. g. richards , phys .
d * 78 * , 114508 ( 2008 ) [ arxiv:0803.3020 [ hep - lat ] ] .
f. gross , phys .
* 186 * , 1448 ( 1969 ) ; f. gross , j. w. van orden and k. holinde , phys .
c * 45 * , 2094 ( 1992 ) .
g. ramalho , m. t. pea and f. gross , eur .
phys . j. a * 36 * , 329 ( 2008 ) [ arxiv:0803.3034 [ hep - ph ] ] .
g. ramalho , m. t. pea and f. gross , phys . rev .
d * 78 * , 114017 ( 2008 ) [ arxiv:0810.4126 [ hep - ph ] ] . g. ramalho and m. t. pea , j. phys .
g * 36 * , 085004 ( 2009 ) [ arxiv:0807.2922 [ hep - ph ] ] ; g. ramalho , m. t. pea and f. gross , phys .
b * 678 * , 355 ( 2009 ) [ arxiv:0902.4212 [ hep - ph ] ] .
g. ramalho , m. t. pea and f. gross , arxiv:1002.4170 [ hep - ph ] .
g. ramalho , k. tsushima and f. gross , phys .
d * 80 * , 033004 ( 2009 ) [ arxiv:0907.1060 [ hep - ph ] ] .
f. gross , g. ramalho and k. tsushima , arxiv:0910.2171 [ hep - ph ] .
p. wang , d. b. leinweber , a. w. thomas and r. d. young , phys .
d * 75 * , 073012 ( 2007 ) [ arxiv : hep - ph/0701082 ] . c. amsler
_ et al . _
[ particle data group ] , phys .
b * 667 * , 1 ( 2008 ) . c. e. carlson , phys .
d * 34 * , 2704 ( 1986 ) ; g. sterman and p. stoler , ann .
nucl . part .
sci . * 47 * , 193 ( 1997 ) [ arxiv : hep - ph/9708370 ] .
g. ramalho and m. t. pea , j. phys .
g * 36 * , 115011 ( 2009 ) [ arxiv:0812.0187 [ hep - ph ] ] .
g. ramalho and m. t. pea , phys .
d * 80 * , 013008 ( 2009 ) [ arxiv:0901.4310 [ hep - ph ] ] .
d. b. leinweber , a. w. thomas , k. tsushima and s. v. wright , phys .
d * 64 * , 094502 ( 2001 ) [ arxiv : hep - lat/0104013 ] .
f. gross , g. ramalho and m. t. pea , phys .
c * 77 * , 035203 ( 2008 ) . | using a covariant spectator quark model we estimate valence quark contributions to the @xmath0 and @xmath1 transition form factors for the @xmath2 reaction .
the roper resonance , @xmath3 , is assumed to be the first radial excitation of the nucleon .
the present model requires no extra parameters except for those already fixed by previous studies of the nucleon .
our results are consistent with the experimental data in the high @xmath4 region , and those from lattice qcd . we also estimate the meson cloud contributions , focusing on the low @xmath4 region , where they are expected to be dominant . |
entanglement is one of the most striking phenomena in quantum physics .
composite systems exhibiting genuine quantum correlations defies our intuition , in the sense that they are not interpreted by classical or semiclassical means @xcite .
quantum correlations are at the heart of many quantum information tasks , such as quantum teleportation and quantum communication , as well as at the core of a variety of many - body physics phenomena @xcite . _
quantum thermodynamics _ seeks the understanding of the emergence of thermodynamics laws from those of quantum dynamics @xcite . in this sense
, one may wonder about the role of quantum correlations and coherence in different phenomena of interest in quantum thermodynamics , for instance in the thermodynamics of quantum thermal machines @xcite or , more generally , in thermal non - equilibrium systems @xcite .
quantum correlations have different fates depending on the environmental influence @xcite .
although entanglement is fragile with respect to thermal fluctuations and decoherence , stationary entanglement still could remain in a system subjected to dissipation @xcite .
quantum discord @xcite on the other hand seems more stable to environmental noise @xcite .
recently , efforts have been made to study how dissipation may precisely drive the system onto preferred states , e.g. onto a genuine entangled state @xcite , by engineering the interaction with environments @xcite . according to the non - equilibrium theory , the analysis of the ( _ linear _ ) response of many - body systems to macroscopic thermodynamic forces , such as those induced by temperature or chemical potential gradients , and to ( weak ) external fields provides an opportunity to test some predictions from condensed matter theory and statistical physics . as an illustration , multipartite entanglement in spin chains
has been explored through precise measurements of the magnetic susceptibility @xcite and the heat capacity @xcite .
also theoretical studies of these two magnitudes seem to provide observable signatures of entanglement in spin chains at thermal equilibrium @xcite . nowadays
, the energy transport through systems involving spatial continuous variables , such as chains of trapped ions , can be experimentally measured @xcite . given the increasing interest in systems under non - equilibrium thermal conditions in the quantum regime @xcite
, one may naturally ask whether the stationary response of a system to a temperature gradient may be influenced by the presence of pure quantum correlations , and in particular by _ genuine _ multipartite entanglement .
the present work tries to elucidate whether the _ average _ properties of the stationary energy current across a harmonic chain , such the mean values and fluctuations , are sensitive to the presence of two - mode and _ genuine _ tripartite entanglement in the system , and more generically to quantum correlations as measured by discord .
significant advances in the context of quantum spin networks indicate that the presence of bipartite entanglement does not play an important role on excitation transport @xcite , whereas a strong correlation between quantum coherence and transport efficiency can be present @xcite . although quantum correlations tend to disappear in systems subjected to a temperature gradient , it has been shown that entanglement and discord can still survive in systems under such conditions @xcite .
much less is known about the influence of genuine multipartite entanglement and the structure of discord on the stationary energy current in strongly dissipated harmonic chains at low temperatures .
this work focuses on stationary quantum correlations in a continuous - variable system within such domain , and analyzes their possible relation to non - equilibrium conditions .
we consider an open system model composed of a linear arrange of three harmonic oscillators , each of them interacting with its own independent heat bath .
we assume that the heat baths are in an initial squeezed thermal state @xcite .
this set - up is particularly interesting in the study of the generation of entanglement between distant modes of a quantum network @xcite , and also a convenient model to analyze many issues concerning the quantum thermodynamics of continuous variable systems @xcite .
we employ the open - system formalism based on the generalized langevin equation ( gle ) @xcite to carry out an extensive numerical study of the stationary properties .
we focus on the two- and three - mode entanglement and the discord in the presence of an energy current through the harmonic chain , for a large range of system parameters . we will analyze whether the average and the fluctuations of the energy current exhibit any evidence of the quantum correlations emerging under non - equilibrium thermal conditions .
the paper is organized as follows . in section [ sec:1 ]
we describe the model of the system and introduce the covariance matrix , which fully characterizes the stationary state of the system .
section [ sla ] reviews the generalized langevin equation approach considered to obtain this state . in section [ sec:3 ] we derive the expressions giving the average and the fluctuations of the energy current in terms of two - time correlation functions , and introduce the quatum correlations , characterized by means of the two- and three - mode entanglement and the ( right ) discord .
the numerical results are presented in section [ sres ] , the corresponding discussion is given in section [ sdis ] , and the main conclusions are put together in section [ scon ] .
we consider an open one - dimensional chain composed of three harmonic oscillators , see figure [ system ] , labeled as @xmath0 ( left ) , @xmath1 ( center ) , and @xmath2 ( right ) , with identical mass @xmath3 , natural frequencies @xmath4 , and position and momentum operators @xmath5 .
we assume bilinear interactions between first - neighbor oscillators , @xmath6 and @xmath7 , with strength given by a single parameter @xmath8 .
each @xmath9 oscillator is coupled with an independent heat bath composed of @xmath10 independent harmonic oscillators , with masses @xmath11 , frequencies @xmath12 and position and momentum operators @xmath13 .
eventually we will consider the quasi - continuum limit @xmath14 .
the hamiltonian of the global system can be written as @xmath15 where @xmath16 with @xmath17 corresponds to the isolated chain , and @xmath18 describes the three independent baths and their interactions with the oscillators , which are assumed bilinear with coupling constants @xmath19 . ,
@xmath20 and @xmath21 . @xmath22 indicates the energy current from the heat bath to the @xmath9 oscillator , and @xmath23 the energy current from the @xmath24 to the @xmath9 oscillator , in the case of @xmath25 .
@xmath8 is the coupling constant between first - neighbor oscillators , and @xmath19 the coupling constant between the @xmath9 chain oscillator and the @xmath26 oscillator of the bath . ]
the interaction term in the microscopic model given by @xmath27 includes the renormalization terms @xmath28 which ensures that the frequency @xmath29 is maintained as the bare frequency of the @xmath9 oscillator @xcite , and the complete positivity of the total hamiltonian ( [ mm ] ) . in general
, a system under the influence of dissipative effects will evolve in the long time limit toward a stationary state in which any trace of its initial state has been wiped out .
the initial condition is only relevant in determining the transient dynamics previous to this asymptotic state .
we fix the initial state at @xmath30 , and assume a barely chance of interaction between the system and the environment at to this point .
then it is reasonable to consider that the system and the environment are initially uncorrelated . as our analysis
is based on quantum properties in the asymptotic stationary state , without loss of generality we will assume an initial product state given by @xmath31 @xcite , where @xmath32 is the initial state of the isolated chain and @xmath33 initial gaussian quantum states corresponding to the baths , which are not necessarily at thermal equilibrium states . assuming that initially the baths are in squeezed thermal states with zero first - moments @xcite , the following averages over the initial state @xmath34 are satisfied @xmath35\,\right ] , \nonumber \\
\frac{1}{2}\left\langle \left\lbrace { \hat p}_{i\nu}(t_{0 } ) , { \hat p}_{i\mu}(t_{0})\right\rbrace\right\rangle_{\hat{\rho}_{0 } } & = & \delta_{\nu\mu}\,\frac{\hbar m_{i\mu}\omega_{i\mu}}{2}\left[\,1 + 2\,n\left(\omega_{i\mu}\right)\,-\,2\,\text{re}\left[\,m\left(\omega_{i\mu}\right)\,\right]\,\right]\,,\nonumber \\
\frac{1}{2}\left\langle \left\lbrace { \hat q}_{i\nu}(t_{0 } ) , { \hat p}_{i\mu}(t_{0})\right\rbrace\right\rangle_{\hat{\rho}_{0 } } & = & \delta_{\nu\mu}\,\hbar\,\text{im}\left[\,m\left(\omega_{i\mu } \right)\,\right ] , \label{cmhb}\end{aligned}\ ] ] where @xmath36 $ ] and @xmath37 $ ] denotes the real and imaginary part of @xmath38 , and @xmath39 which satisfy the relation @xmath40 .
we have considered the same squeeze @xmath41 for all the oscillators of the @xmath9 bath .
@xmath42 is a global arbitrary rotation of the bath state @xmath43 , and @xmath44 is the average occupation number of the @xmath45 oscillator in the @xmath9 bath in a thermal equilibrium state . to induce an energy current across the chain ,
see figure [ system ] , we fix the left and right heat baths at different temperatures , @xmath46 and @xmath47 respectively , with @xmath48 low enough to ensure that the system remains within the quantum regime .
then we modify the temperature of the central bath , @xmath49 , by considering different values of @xmath50 .
this setup is particularly interesting as it makes possible to establish a quasiclassical regime in the central oscillator while maintaining the lateral oscillators in the quantum regime .
below we will show that the asymptotic stationary state derived from this manipulation of the central oscillator can exhibit a rich variety of quantum correlations , such as two - mode and bipartite three - mode and genuine tripartite entanglement .
since the total hamiltonian ( [ mm ] ) is quadratic in both positions and momenta , and we have considered gaussian initial bath states , the asymptotic stationary state , denoted by @xmath51 , will be gaussian for any initial state of the oscillators @xcite . then , the stationary quantum properties will be determined by just the first and second moments of the positions @xmath52 and the momenta @xmath53 .
the former can be made arbitrarily close to zero by unitary local transformations that do not affect the non - local properties such as entanglement . whereas the second moments determining all the correlation properties required in our analysis are given in terms of the covariance matrix @xmath54 with @xmath55 and @xmath56 , and the two - point ( symmetrical ) correlation functions @xmath57 the second moments of the energy currents also involve the imaginary part of the two - point correlation @xmath58 $ ] , given by @xmath59 \,\right)\ , .
\label{cv2}\\\ ] ] the covariance matrix of the state @xmath60 corresponding to the subsystem defined by the @xmath9 and @xmath24 oscillators can be obtained from ( [ cvmatrix ] ) by just taking the elements associated with these two oscillators .
it should be emphasized that at stationary conditions the correlation functions ( [ cv ] ) and ( [ cv2 ] ) only depend on the time difference @xmath61 , and a particular initial time @xmath62 is irrelevant to obtain them .
this will be important in what follows , when computing these stationary correlations by using the generalized langevin equation approach .
within the langevin approach , the equations of motion that govern the evolution of the stationary correlations are derived from the microscopic model ( [ mm ] ) by writing the heisenberg equations for the oscillator positions and tracing out the degrees of freedom of the heat baths .
this leads to the so - called generalized langevin equation , @xmath63 where we have introduced the potential @xmath64 the susceptibilities @xmath65 and the fluctuating forces @xmath66 with @xmath67 the heaviside step function , and the spectral density of the environment given by @xmath68 as long as the stationary solution of the generalized langevin equation is guaranteed , one may take the limit @xmath30 in eq .
( [ leq ] ) , and then , use the fourier transform @xmath69 to obtain the stationary solution of the position and momentum operators @xcite . by replacing these solutions into the correlation elements ( [ cv ] ) and averaging over the initial state @xmath34 , it follows @xmath70 with @xmath71 and @xmath72 where @xmath73 .
the two - points correlation functions @xmath74 ( [ cv2 ] ) satisfy an expression identical to ( [ sleq1 ] ) , but replacing in @xmath75 the anticommutator of the fluctuating force by the commutator .
expression ( [ greenf ] ) is nothing but the fourier transform of the ( matrix ) green function @xcite for the generalized langevin equation ( [ leq ] ) .
the real and imaginary parts of the fourier transform @xmath76 of the susceptibility ( [ sus ] ) are given by ( see appendix [ asusc ] for further details ) @xmath77&=&\hbar\left[\,\theta(\omega)\,j_{i}(\omega)-\theta(-\omega)\,j_{i}(-\omega)\,\right ] , \label{imx } \\
\text{re}\left[\,\tilde{\chi}_{i}(\omega)\,\right]&=&\frac{1}{\pi}\,p\int\frac{\text{im}\left[\,\tilde{\chi}_{i}(\omega')\,\right]}{\omega'-\omega}\,d\omega ' , \label{rex}\end{aligned}\ ] ] where @xmath78 denotes the cauchy principal value .
the second expression is the well - known kramers - kronig relation @xcite arising from the causal nature of the susceptibility . in order
the three - mode system can reach a stationary state , the function @xmath79 must approach a combination of decaying exponentials in the long time limit @xcite . according to a previous study of the equations of motion of the system - plus - environment complex in terms of normal modes @xcite
, the existence of a well defined stationary solution entails that @xmath80 has no any real root @xmath81 corresponding to the frequency of a bound normal mode , which implies that @xmath82\neq 0 $ ] . from eq .
( [ imx ] ) , the latter condition means that @xmath81 must be contained within the domain of the bath spectral density @xmath83 @xcite . in general , the heat baths can be considered as composed of a large number of degrees of freedom with finite broad band spectrum , in which the most energetic environmental - degree is roughly determined by a cut - off frequency @xmath84 .
this ensures that the natural frequencies of the system are well embedded in the environmental spectrum , and consequently it makes possible an irreversible energy transfer from the system to the environment , at least in a finite time much larger than the natural time scale of the system .
we shall impose @xmath85 in order to ensure an irreversible evolution of the three - oscillator chain toward a well defined asymptotic stationary state .
the covariance matrix of this stationary state is completely determined by the correlation functions ( [ sleq1 ] ) evaluated at equal time , once the correlation functions of the fluctuating forces ( [ ff ] ) have been obtained .
these correlations depend only on the initial environmental state .
below we show the relation between the fluctuating forces and the initial covariance matrix ( [ cmhb ] ) of the heat baths .
our choice of the initial environmental state implies the statistical independence of the fluctuating forces corresponding to different heat baths , _
i.e. _ @xmath86 for all @xmath87 . whereas , according to eqs .
( [ cmhb ] ) , the symmetrical two - time correlation function of the fluctuating forces associated with a given @xmath88 bath is given by ( see appendix [ aff ] for further details ) @xmath89\cos\left(\omega_{l\mu}(t+t'-2t_{0})\right)+\text{im}\left[\,m_{l}(\omega_{l\mu})\,\right]\sin\left(\omega_{l\mu}(t+t'-2t_{0})\right)\bigg]\ , .
\label{ffe}\end{aligned}\ ] ] the average of the corresponding commutator can be expressed as @xmath90\right\rangle_{\hat{\rho}_{0}}= i\,\delta_{lm}\,\frac{\hbar}{2}\sum_{\mu=1}^n \frac{g_{l\mu}^2}{m_{l\mu}\omega_{l\mu}}\,\sin\left(\omega_{l\mu}(t'-t)\right)\ , .
\label{aff2a}\end{aligned}\ ] ] the dependence of the symmetrical two - time correlation functions on the initial time @xmath91 is eliminated in the case of an initial thermal equilibrium state of the @xmath88 bath , in which @xmath92 and @xmath93 for all @xmath94 values .
although in the previous section we have already fixed the time limit @xmath95 in order to obtain the stationary solution , we shall maintain the notation @xmath96 for convenience in order to make more clear the following discussion .
as shown in appendix [ aff ] , the non - stationary terms in eq.([ffe ] ) come from the average of factors involving @xmath97 and @xmath98 , with @xmath99 the annihilation ( creation ) operator of the @xmath45 mode in the @xmath100th reservoir .
these terms describe non - conservative energy processes that take place in the heat bath at the initial time @xmath96 , and therefore , they may influence the transient dynamics of the three - oscillator chain .
however , they become highly oscillatory in the long time limit @xmath101 and their contribution to the stationary properties may be disregarded @xcite . when taking the quasi - continuum limit @xmath102 in the environment spectral density , only the stationary term in eq.([ffe ] ) remains .
this assertion holds for an environment with a broad spectrum limited by @xmath103 , and a finite interaction between the reservoir modes and the system oscillators .
mathematically , the latter translates into that the spectral densities @xmath104 are finite continuous functions , and the corresponding coupling strengths should decay at least as @xmath105 at high frequencies . under these conditions ,
the long time limit of the symmetrical two - time correlation function ( [ ffe ] ) reduces to the following expression in the frequency domain ( see appendix [ aff ] for details ) @xmath106\coth\left(\frac{\hbar\,\omega}{2\,k_{b}\,t_{l}}\right)\cosh\left(2r_{l}\right)\ , .
\label{cffdal}\ ] ] the average of the corresponding commutator reduces to @xmath107\right\rangle_{\hat{\rho}_{0}}= 2\pi\,\delta(\omega+\omega')\,\text{im}\left[\,\chi_{l}(\omega)\,\right]\ , .
\label{cffdal2}\ ] ] similar results have been previously obtained within the path integral formalism @xcite , see also @xcite .
we point out that an initially squeezed state of the environment makes the reduced system to _ notice _ an effective temperature above the temperature @xmath108 of the heat bath at thermal equilibrium .
this effect has interesting consequences in the efficiency of thermal machines within the quantum regimen @xcite .
now we can replace the autocorrelations ( [ cffdal ] ) into the expressions ( [ gij ] ) , and perform the integral in the frequency @xmath109 to obtain a closed - form expression for the two - time correlation functions @xmath110 .
notice that @xmath111 due to the stationary condition of the fluctuating force correlation . a similar procedure is followed for the functions @xmath74 ( [ cv2 ] ) . in general , there are not analytic expressions giving the integrals involved in the correlation functions in terms of the system parameters , such as the bath temperatures , the oscillator frequencies and the coupling strengths .
we will compute them by means of numerical methods .
the non equilibrium conditions imposed by the different bath temperatures drive an energy current through the system , see figure [ system ] .
a discrete definition of the energy currents associated with each chain oscillator can be derived from its local energy @xcite @xmath112 with @xmath113 for @xmath114 , and @xmath115/4 $ ] .
the time derivative of @xmath116 leads to the discrete continuity equations @xmath117 with @xmath118 for @xmath114 , and @xmath119 .
the term @xmath120 \label{j1}\end{aligned}\ ] ] can be identified as the energy current from the @xmath24 oscillator to the @xmath9 oscillator , whereas @xmath121 corresponds to the energy current from the @xmath9 heat bath into the @xmath9 oscillator . at stationary conditions
the total current coming from the baths into the system becomes zero .
here we will focus on the analysis of the total current flowing from the @xmath0- to the @xmath2-oscillator , defined as @xmath122 our study is based on the stationary properties of the total energy current , which are basically determined by its first- and second - moments , or equivalently , by its average and fluctuations . the steady state average of the total energy current @xmath123
can be obtained by tracing ( [ j1 ] ) over the the stationary state and using the stationary solutions of the two - time correlation functions ( [ sleq1 ] ) , which leads to @xmath124\ , \label{ef1 } \end{aligned}\ ] ] for the local currents @xmath125 and @xmath126 .
since the quantum correlations shared by the oscillators , in particular entanglement , are partially encoded on the correlation terms indicated in ( [ j1 ] ) , one might expect that the energy current could be sensitive to these correlations .
notice that the total current involves the correlations between the central and the side oscillators , while it does not depend on the crossed correlation function between the two side oscillators .
we shall further analyze this issue in section ( [ sdis ] ) . to have a better understanding of the system behavior under non - equilibrium conditions we also study the two - time correlation functions of the energy currents ( [ j1 ] ) .
the fluctuations can be obtained from the symmetrical version of the classical two - time correlations , and expressed as @xmath127 theoretically , the response of a system under external perturbations can be studied in terms of these correlations functions @xcite .
notice that we evaluate the fluctuations in the non - equilibrium stationary state , so that we might expect that eq.([cj1 ] ) can elucidate some properties of stationary non - equilibrium , rather than the equilibrium quantum correlations @xcite .
furthermore , it has been shown that eq.([cj1 ] ) is related to the fluctuations of the stationary energy current across the chain @xcite .
as the stationary state obeys a gaussian distribution , the four - time correlation terms implicit in @xmath128 can be decomposed into terms involving the product of two - time correlations , in the form @xmath129\right\rangle \left\langle \left [ \hat p_{j}(\tau),\hat p_{m } ( 0)\right]\right\rangle\,\right ] \nonumber \\ & + & 2\,\left[\left\langle \left\lbrace \hat x_{i}(\tau),\hat p_{m}(0 ) \right\rbrace\right\rangle \left\langle \left\lbrace \hat x_{l } ( 0),\hat p_{j}(\tau)\right\rbrace\right\rangle - \left\langle \left [ \hat x_{i}(\tau),\hat p_{m}(0 ) \right]\right\rangle \left\langle \left [ \hat x_{l } ( 0 ) , \hat p_{j}(\tau)\right]\right\rangle\right ] \ , . \label{ftc}\end{aligned}\ ] ] then the current - current response function ( [ cj1 ] ) can be expressed as @xmath130\right .
\\ \nonumber \left.\,+\,s_{\alpha,\beta'}\left[\,c_{x_\alpha p_\beta}(\tau,0)\,c_{x_{\beta'}p_{\alpha'}}(-\tau,0)- y_{x_\alpha p_\beta}(\tau,0)\,y_{x_{\beta'}p_{\alpha'}}(-\tau,0)\right ] \,\right]\,,\hspace*{1cm}\end{aligned}\ ] ] where @xmath131 is the sign of the cofactor of the element @xmath132 in the @xmath133 array defined by the indexes @xmath134 . finally , according to eq.([gef1 ] ) , the autocorrelation function of the total current flowing from the @xmath0- to the @xmath2-oscillator is given by @xmath135 in contrast to the average energy current , the correlation function @xmath136 involves crossed terms between the @xmath2- and the @xmath137- oscillators .
in addition , while @xmath138 ( [ tec ] ) is given by a linear combination of two - time correlation terms , @xmath136 has a nonlinear dependence on such terms .
these two aspects will be useful in the subsequent discussion .
alternatively , the behavior of @xmath138 and @xmath136 will help us to gauge whether the average properties of the energy current are sensitive to quantum correlations , such as genuine tripartite entanglement .
we shall analyze the two - mode quantum correlations between the @xmath9 and @xmath24 oscillators by means of the discord measure on the right @xcite , denoted by @xmath139 .
the entanglement between both modes can be quantified by the well - known logarithmic negativity @xmath140 @xcite . in particular
we devote special attention to the entanglement @xmath141 and the discord @xmath142 between the side oscillators .
we use a recent criterion in the realm of continuous - variable systems @xcite to study tripartite entanglement , which is a good estimator for @xmath143-partite entanglement @xcite in @xmath144-mode gaussian as well as non - gaussian states .
a tripartite harmonic system may develop _ bipartite _ three - mode entanglement , which means that there is at least a bipartition of the three - mode system that is entangled , or _
genuine tripartite _
entanglement , which corresponds to the case in which all the bipartitions are entangled and the state @xmath51 can not be written as a convex combination of bipartite separable states . here , the criterion reduces to evaluate a figure of merit @xmath145 , such as a positive value of @xmath146 ( @xmath147 ) indicates that the state @xmath51 is genuine tripartite entangled ( bipartite three - mode entangled ) @xcite . as we are dealing with stationary gaussian states , all the previously mentioned indicators of quantum correlations can be directly computed from the covariance matrix @xmath148 given by eq.([cvmatrix ] ) .
the logarithmic negativity can be expressed as @xcite , @xmath149 where @xmath150 stands for the lowest symplectic eigenvalue of the partial transpose covariance matrix @xmath151 , corresponding to the reduced density matrix @xmath60 .
the ( right ) discord is given by @xcite , @xmath152 with the total correlations @xmath153 and the classical correlations @xmath154 which are given in terms of the von newman entropy @xmath155 .
closed form expressions for the quantum discord as a function of the covariance matrix @xmath156 have been derived in @xcite .
it is important to realize that these indicators of quantum correlations involve a non - linear dependence on the density operator and the two - time correlation functions .
see appendix [ asc ] for further details of the logarithmic negativity , the quantum discord and the separability criteria @xmath145 .
we now investigate the average properties of the total current @xmath157 when the three oscillators share two - mode and tripartite entanglement . in many realistic situations , e.g. quantum brownian motion
, the interaction with the environment leads to an ohmic dissipation . in a first approach , for a nanomechanical setup
, one may think that the thermal relaxation is mainly due to the coupling with the acoustic phonons of the substrate , which may lead to linear spectral density at low oscillator frequencies .
however , in some cases , the dimensionality of the environment may induce super - ohmic dissipation . here
we analyze both the ohmic and super - ohmic dissipations , which are characterized by the spectral densities @xmath158 and @xmath159 respectively , with @xmath160 the dissipative rate for the @xmath9 oscillator and @xmath84 the frequency cut - off of the environmental spectrum .
as argued in section [ sla ] , the stationary state is reached in a time scale larger than any of the natural time scales implicit in the dynamics of the open chain ; namely @xmath161 .
from now on we set the environmental parameters @xmath162 , @xmath163 , @xmath164 , and the typical values for nanomechanical oscillators @xmath165 and @xmath166 . with this configuration
the system begins to exhibit quantumness at temperatures in the range of mili - kelvin .
we also assume off resonance oscillators with frequencies , @xmath167 , @xmath168 , and @xmath169 , given in terms of a detuning parameter @xmath170 .
we start by analyzing the behavior of two - mode entanglements and the total energy current with the temperature gradient @xmath50 . as figure [ fig1 ] ) shows ,
the three oscillators become two - mode entangled at low temperature gradients , with this entanglement exhibiting a plateau for negative gradients .
interestingly the total current flowing through the oscillator chain presents a similar plateau .
this can be related to the proximity of the central oscillator to its ground state at very low temperatures @xmath171 , which is effectively reached for @xmath172 .
a similar result for entanglement has been obtained in the study of the influence of heat transport on the two - mode entanglement between oscillators that are embedded in a disordered harmonic chain connected to heat baths at both ends @xcite .
it has been shown that a plateau emerges when the energy spectrum is bounded from below since each site of the chain suffers an harmonic potential .
as the central oscillator gets closer to the ground state for negative values of @xmath50 , the energy flowing across this oscillator becomes bounded as the temperature gradient decreases . in the absence of the harmonic confinement
the logarithmic negativity would continue growing up to a maximum value , as the heat transport decreases @xcite .
moreover , the plateau in the entanglement remains even when the average energy current across the chain becomes zero , though the temperature gradient @xmath50 is not zero .
this occurs when @xmath173 and the left and right oscillators are identical @xmath174 , and therefore @xmath175 .
this last point also underlines that the appearance of entanglement is mainly attributed to proximity of the system to its ground state , rather than to the presence of an energy flow induced by non - equilibrium conditions .
the two - mode entanglement rapidly decreases for positive @xmath50 , while the energy current grows monotonically .
this is expected as the temperature of the whole three - mode system is increased on average , which is generally harmful for entanglement .
previous results have suggested this behavior , in fact it has been shown that in a harmonic chain an increasing @xmath176 is detrimental to build up bipartite or tripartite entanglement due to the rise in the thermal noise @xcite .
in addition , it can be shown from ( [ cffdal ] ) that an initially squeezed environmental state effectively increases the temperature .
hence , in the present setting an initial squeezed bath state does not favor the appearance of stationary entanglement .
moreover , one may expect that non - markovian effects , which are more relevant for super - ohmic dissipation , would substantially degrade the entanglement with respect to the ohmic situation . according to figure ( [ fig1 ] ) ,
the two - mode entanglement is essentially the same for the chain suffering ohmic or super - ohmic dissipation ; namely , @xmath141 , @xmath177 and @xmath178 practically coincide for both situations .
this result is in contrast with the observed transient evolution of the two - mode entanglements under different environmental spectral densities , in which the super - ohmic dissipation induces stronger disentanglement effects @xcite .
the coincidence of the stationary two - mode entanglements also differs from the emergence of entanglement in a situation in which the oscillators are affected by the same bath @xcite . in the case of the energy current
, we observe that it is strongly affected by the interaction with the heat baths , determined by the fixed spectral density .
we have performed an extensive analysis of the two - mode entanglements and the energy currents involving the central oscillator , in terms of both the temperature gradient @xmath50 and the coupling strength @xmath8 .
as figure ( [ fig2 ] ) shows , a similar behavior to that illustrated in the figure ( [ fig1 ] ) is reproduced for different values of @xmath8 .
the entanglements @xmath177 and @xmath179 increase for lower temperature gradients and stronger couplings .
whereas the energy currents @xmath180 and @xmath181 exhibit a relatively weak dependence on the coupling strength , and the expected increase with the temperature gradient .
also the plateau of small energy currents arising in the proximity of the ground state of the central oscillator can be clearly observed .
hence , our results indicate that the energy currents across the system are insensitive to the emerge of two - mode entanglements between the oscillators , both under ohmic and super - ohmic dissipation , and irrespective of the coupling strength with the heat baths .
the two - mode entanglement @xmath141 and total energy current @xmath182 remain nearly unchanged provided that the central oscillator is close enough to the ground state , at temperatures between @xmath183 and @xmath184 .
an increase in the temperature of this oscillator results in a deterioration of the entanglement , and an increase in the energy current . and @xmath185 as a function of the temperature gradient for ohmic dissipation .
right : the time evolution of the fluctuations of the total current under ohmic dissipation , in a system that is genuine entangled @xmath186 ( black solid line ) , bipartite three - mode entangled @xmath187 ( blue dashed line ) , and likely separable in the three possible bipartitions @xmath188 ( red dot - dashed line ) .
the remaining parameters are @xmath189 , @xmath190 , @xmath191 , and @xmath192.,title="fig : " ] and @xmath185 as a function of the temperature gradient for ohmic dissipation . right : the time evolution of the fluctuations of the total current under ohmic dissipation , in a system that is genuine entangled @xmath186 ( black solid line ) , bipartite three - mode entangled @xmath187 ( blue dashed line ) , and likely separable in the three possible bipartitions @xmath188 ( red dot - dashed line ) .
the remaining parameters are @xmath189 , @xmath190 , @xmath191 , and @xmath192.,title="fig : " ] in this section we analyze the three - mode entanglement and the energy current correlations between the left and right oscillators , which includes correlation terms involving the three oscillators , see eq.([ccr ] ) .
figure ( [ fig3 ] ) shows the bipartite three - mode ( @xmath193 ) and the genuine tripartite ( @xmath194 ) entanglements measured by the corresponding criteria @xmath195 @xcite .
the results for both the ohmic and super - ohmic dissipations are quite similar . in the low temperature and strong coupling regime ,
the three - mode system exhibits genuine tripartite entanglement , though this feature rapidly disappears for positive values of @xmath50 , such as occurs with the two - mode entanglement .
strikingly , the system still remains bipartite three - mode entangled at relatively high temperature gradients ( @xmath196 ) .
hence the tripartite entanglement is more robust to temperature changes than the two - mode entanglement between the side oscillators .
figure ( [ fig3 ] ) also shows the initial time evolution of the energy current correlations for three different three - mode entanglement configurations .
as expected , the fluctuations of the energy current exhibit an oscillatory behavior , which should be progressively attenuated at larger time intervals . according to a previously reported exponential time decay of the two - time correlation functions ( [ sleq1 ] ) in a damped harmonic oscillator at low temperature @xmath48
, such oscillations should be effectively suppressed at time @xmath197 @xcite . as evidenced by figure ( [ fig3 ] ) , the energy current correlations exhibit similar oscillations as the system evolves from genuine tripartite to bipartite three - mode entanglement .
the most significant discrepancy between these two configurations is an increase in the oscillation amplitude , which can be attributed to the thermal fluctuations that arise with increasing the temperature gradient .
indeed , a similar oscillating behavior in the fluctuations is still observed at relatively large temperature gradients @xmath198 , when the system is expected to be separable in the three possible bipartitions .
the results we have obtained from an analysis considering an extensive set of parameters @xmath199 corresponding to different multipartite entanglement configurations , for both ohmic and super - ohmic dissipations , also indicate that the energy currents correlations across the harmonic chain are insensitive to the emerge of tripartite genuine or bipartite three - mode entanglement . to conclude this section we focus on the energy current correlations evaluated at equal time , see figure ( [ fig31 ] ) , which will be useful in the subsequent discussion .
both the stationary fluctuations and the average of the energy current , see figure ( [ fig2 ] ) , grow with increasing the temperature gradient .
but , in contrast to the average energy current , the plateau at small values of the fluctuations is observed above a given value of the coupling strength , which is larger in the case of ohmic dissipation . once again , the fluctuations of the energy current are insensitive to whether the system experiences bipartite three - mode or genuine tripartite entanglement .
similar results are obtained for the current - current response involving the central and the side oscillators .
one might expect that a scenario similar to the one previously described for the two- and three - mode entanglement would be repeated in the presence of other non - classical correlations , such as discord . in this section
we analyze a possible connection between the energy current and the quantum correlations measured by the right - discord @xmath200 .
although not shown in this work , similar results are obtained from the analysis of the discord @xmath201 measured from the left .
we also point out that the two - mode discord contains the contribution of the two - mode entanglement studied in previous sections .
we have found that discord exhibits a strong dependence on both the temperature gradient and the frequencies of the oscillators .
as shown in figure ( [ fig4 ] ) , its presents a sharp peak centered at the resonance condition @xmath202 , and with an amplitude that grows with increasing the temperature gradient .
interestingly , the discord appears strongly correlated with the mean interaction energy between the oscillators , @xmath203\ , , \label{mei}\ ] ] which has maximum strength also at resonance , see figure ( [ fig4 ] ) . this resonant interaction becomes stronger for higher temperature gradients , which also increase the discord .
as all the negative contribution to @xmath204 comes from the crossed correlation terms , it becomes evident that the maximum interaction strength occurs when these correlations take the highest values , which also turns into the optimal conditions for discord .
this result evidences an underlying connection between the interaction energy and the discord , as could be anticipated considering that the discord would be zero in the absence of interaction .
figure ( [ fig4 ] ) also shows that the discord in the resonant system begins to grow almost linearly with the temperature gradient , and then approaches a constant value at higher gradients .
an initially squeezed central reservoir enhances the creation of discord , both for ohmic and super - ohmic dissipation .
this is in agreement with the foregoing results , as the squeeze of the initial bath state effectively increases the temperature perceived by the oscillators , see eq.([cffdal ] ) .
then an increase of the stationary discord between the side oscillators may be induced either by initially squeezing the central reservoir or increasing its temperature .
it has been shown that discord may be additionally created by local noisy operations , such as dissipation @xcite .
hence , it may happen that discord would be generated by an energy current induced by a temperature gradient , as this current would make each oscillator to dissipate .
considering that the discord contains all the quantum correlations , it would be interesting to analyze whether the entanglement available in the system contributes to its increase with the temperature gradient . at this respect ,
since entanglement can be only created by non - local manipulations and it becomes zero at high temperature gradients , see section ( [ steec ] ) , we may conclude that such increase of the discord must be mainly due to local operations .
the paramount role of the local manipulations in the creation of quantum correlations at resonance conditions is correlated with a maximum average interaction strength between oscillators of similar frequencies , see figure ( [ fig4 ] ) and eq . ( [ mei ] ) .
notice that the discord grows with @xmath50 even in the absence of an energy current between the side oscillators ( @xmath173 ) .
the plateau of maximum discord at high temperature gradients can be attributed to the very low temperatures of the side oscillators @xmath205 , which guarantees the `` coherence '' of the local manipulations .
indeed , the increase in the discord gradually disappear as the mean temperature @xmath48 increases , and therefore the three - mode system approaches to a classic state .
we have observed that the discord @xmath200 has almost disappeared at temperature @xmath206 .
as expected , the two - mode discord between the central and the side oscillators is enhanced by increasing the interaction strength , see figure ( [ fig5 ] ) . whereas , the trend of the central oscillator towards a classical state , by increasing its temperature through higher values of @xmath50 , causes a progressive deterioration of the discord . in the case of the energy currents between the central and the side oscillators
, they exhibit an almost linear increase with the temperature gradient , which is barely modified by the strength of the coupling interaction .
once again , the energy currents do not seem to be related to the significant non - classical correlations shared by the oscillators .
( left panel ) and the energy current @xmath207 ( right panel ) between the central and right oscillators in the resonant system under ohmic dissipation .
similar results are obtained for the discord and the energy current between the @xmath1- and @xmath0- oscillators , both under ohmic and super - ohmic dissipations .
the parameters are the same as figure ( [ fig4]).[fig5],title="fig : " ] ( left panel ) and the energy current @xmath207 ( right panel ) between the central and right oscillators in the resonant system under ohmic dissipation .
similar results are obtained for the discord and the energy current between the @xmath1- and @xmath0- oscillators , both under ohmic and super - ohmic dissipations .
the parameters are the same as figure ( [ fig4]).[fig5],title="fig : " ]
considering that the energy current between two oscillators has an explicit dependence on crossed correlations between them , one could expect that the emerge of entanglement in the system should have detectable effects on such current .
therefore , it would be interesting to determine whether a formal connection between entanglement and the average properties of the energy current can be formally established .
the results presented in section [ steec ] indicate that the behavior of the energy current is not modified by the presence of two - mode entanglement . a direct comparison between the two - mode entanglement between the side oscillators @xmath208 and the total energy current @xmath182 suggests an elusive correlation between them , see figure [ fig1 ] ) .
although the most important contribution to the local currents @xmath209 ( [ ef1 ] ) comes from crossed correlation terms that encode part of the quantum correlations shared by the oscillators , the total current @xmath182 ( [ tec ] ) does not have an explicit dependence on the correlation terms involving the @xmath0- and @xmath2- oscillators . then the two - mode entanglement @xmath208 should not necessarily affect the total current .
this reasoning does not exclude , however , the possibility that the total current could be sensitive to the two - mode entanglement shared by the central and the side oscillators .
the conjecture that entanglement and energy current are intimately related would lead to the natural question whether the current @xmath210 could serve as an useful witness of the entanglement between the @xmath9 and @xmath24 oscillators . according to the theory of entanglement ,
an entanglement witness @xmath211 based on a ( bounded ) hermitian operator @xmath212 may be constructed as @xmath213 , where the last term is the infimum value of @xmath212 among all the product states @xmath214 @xcite . for @xmath215 ( [ j1 ] )
it follows @xmath216 where @xmath217 is obtained from ( [ sleq1 ] ) by considering the average over product states .
@xmath218 is a good candidate to unveil the two - mode entanglement @xmath140 provide that @xmath219 takes a negative value for at least one entangled state .
unfortunately such a rigorous proof requires a closed form expression for @xmath217 , which is currently out of scope as we are dealing with a system under the non - equilibrium conditions induced by two different temperature gradients . though we can not guarantee whether eq.([ewj ] ) constitutes a good estimator of entanglement , the results of section [ steec ] evidence the difficulty of assessing entanglement through @xmath218 , mainly due to the apparent insensitivity of the energy current @xmath220 to the two - mode entanglement @xmath140 , see figure ( [ fig2 ] ) .
in addition , though the correlation terms in eq .
( [ j1 ] ) partially carry the quantum correlations shared by the chain oscillators , they themselves do not necessarily manifest entanglement .
indeed , the so - called peres - hodorecki - simon inequality @xcite @xmath221 which provides a criterion to detect the two - mode entanglement between the @xmath9 and @xmath24 oscillators , already displays a non - linear relation between entanglement and the elements of the covariance matrix . a common feature of the entanglement measures ( or entanglement monotones ) is their non - linear functional dependence on the density operator @xcite , as occurs in the case of the logarithmic negativity @xcite . in some sense ,
the reliable observation of entanglement relies on the ability to measure non - linear properties of the quantum state @xcite .
according to the expressions for the total energy current @xmath138 ( [ ef1 ] ) and the criterion for entanglement ( [ sc ] ) , the first - moment of the energy current depends linearly on the crossed correlation terms , whereas the entanglement exhibits a non - linear dependence on such terms .
hence , the energy current between the @xmath9 and @xmath24 is not expected to manifest the emergence of two - mode entanglement . following the previous argument
, the fluctuations of the total current ( [ ccr ] ) could manifest the emergence of entanglement , as it involves correlation terms between all the oscillators . in section [ sccr ] we focused on the tripartite entanglement , and showed that @xmath222 seems to be insensitive to the inseparability properties of the three - mode chain .
a similar conclusion was drawn from the comparison of the time evolution of the fluctuations @xmath136 at different entanglement configurations of the stationary state , see figure ( [ fig3 ] ) .
we remark that this result is not in contradiction with the previous argument based on criterion ( [ sc ] ) , as it provides a necessary , but not sufficient , condition for the existence of entanglement .
considering that , in contrast to entanglement , almost any quantum state has a non - negative discord @xcite , a distinct behavior of these two quantum correlations might be expected @xcite .
the results of section ( [ snc ] ) indicate that in the proximity of a resonance condition , a finite energy current induced by the temperature gradient @xmath50 may generate non - classical correlations between the left and right oscillators , even when the total energy current between them becomes zero .
this behavior has been correlated with a maximum strength of the average interaction between the harmonic oscillators , see figure ( [ fig4 ] ) .
the same results also show that the energy current is not modified by the emerge of discord , see figure ( [ fig5 ] ) .
the average properties ( mean value and fluctuations ) of the energy current as a whole exhibit a ` linear ' behavior ruled by the temperature gradients , irrespective of the significant two - mode quantum correlations that may be present in the system . as a measure of such correlations
we have analyzed the logarithmic negativity , the tripartite entanglement , characterized by the criteria @xmath223 and @xmath224 , and the quantum discord . finally ,
although we have focused on a specific system configuration in which each chain oscillator is connected to an independent heat bath , similar results might be expected for other arrangements , since the emergence of quantum correlations in linear harmonic chains or lattices essentially stems from the proximity of the system to the ground state .
the study of the energy current and quantum correlations in systems exhibiting non - linearities deserves further attention .
we have performed an extensive analysis of the quantum correlations , and the mean value and fluctuations of the energy current across a three - oscillator linear chain at the stationary state , both under ohmic and super - ohmic dissipation .
we have considered initially squeezed reservoir thermal states , and applied the gle formalism to determine the correlation functions between the position and momentum operators of the chain oscillators , which completely characterize the stationary properties .
interestingly , the results obtained for the quantum correlations are quite similar for both ohmic and super - ohmic dissipation .
this suggests that the non - markovian effects do not significantly modify the markovian results for the _ stationary _ properties of the quantum correlations in a system of oscillators in contact with independent heat baths .
moreover , the initial squeezing of a heat bath effectively increases the temperature that the oscillator chain perceives from this bath , and eventually becomes detrimental for the build - up of stationary entanglement .
a different behavior is observed in the case of discord , which can be created by local noisy operations .
these quantum correlations highly depend on both the interaction strength between the oscillators and the detuning of their natural frequencies .
in particular , the two - mode discord between the side oscillators in the presence of temperature gradients is enhanced at resonance , which indicates that quantum coherence may be favored by thermal non - equilibrium conditions . according to our results , both the average and the fluctuations of the stationary energy current across the oscillator chain
are mainly determined by the two imposed temperature gradients , and do not seem to be related to the appearance of a rich variety of quantum correlations in the system , comprising two - mode discord and entanglement , bipartite three - mode and genuine tripartite entanglement .
the absence of quantum correlation effects in the average energy current can be partially understood in terms of its linear dependence on the correlation terms between the oscillators . in the case of the fluctuations ,
the more intricate dependence on such terms makes more complex to elucidate their possible connection with quantum correlations . nowadays
the quantum correlations under thermal non - equilibrium conditions have become a topic of great interest in the fields of quantum information , quantum thermodynamics and the theory of open quantum system .
generally the non - classical correlations , such as entanglement , exhibit a non - linear dependence on the density operator that make difficult to establish any formal connection between them and the response of the system to non - equilibrium constrains .
we hope that this work may contribute to stimulate further research in this direction .
the authors acknowledge fruitful discussions with l.a .
correa , j.p .
palao and b. witt .
this project was funded by the spanish micinn ( grant nos . fis2010 - 19998 and fis2013 - 41352-p ) and by the european union ( feder ) .
a.a.v . acknowledges financial support by the government of the canary islands through an aciisi fellowship ( 85% co - financed by the european social fund ) .
in this appendix we derive the real and imaginary parts of the fourier transform of the susceptibility @xmath225 ( [ sus ] ) . considering the spectral density ( [ jden ] ) , the imaginary part of the fourier transform @xmath226
can be expressed as @xmath227=\frac{\pi \hbar}{2}\sum_{\mu=1}^{n}\frac{g_{i\mu}^2}{m_{i\mu } \omega_{i\mu}}\,\left[\,\delta\left(\omega-\omega_{i\mu}\right)-\delta\left(\omega+\omega_{i\mu}\right)\,\right]=\hbar\left[\,\theta\left(\omega\right)j_{i}\left(\omega\right)-\theta\left(-\omega\right)j_{i}\left(-\omega\right)\,\right]\ , .
\label{appa1}\ ] ] then the real part can be obtained from the causality of @xmath225 , according to @xmath228=\mathcal{h}\left[\,\text{im}\left(\chi_{i}(\omega')\right)\,\right](\omega)\doteqdot \frac{1}{\pi}\,p\int\frac{\text{im}\left[\,\chi_{i}(\omega')\,\right]}{\omega'-\omega}\,d\omega'\,,\ ] ] where @xmath229(\omega)$ ] denotes the hilbert transform of @xmath38 . using eq .
( [ appa1 ] ) it follows that @xmath230=\mathcal{h}\left[\,\theta(\omega')j_{i}(\omega')\,\right](\omega)+\mathcal{h}\left[\,\theta(\omega')j_{i}(\omega')\,\right](-\omega)\ , .
\label{appa2}\ ] ] below are given the expressions of the susceptibility for ohmic and super - ohmic spectral densities .
assuming the ohmic spectral density ( [ sd1 ] ) , and considering the well - known properties of the hilbert transform , the expression ( [ appa2 ] ) leads to @xmath231&=&\frac{\hbar}{2}\,\pi m \gamma_{i}\left[\,\mathcal{h}\left[\,\theta(\omega')\,\omega'e^{-\frac{\omega'}{\omega_{c}}}\,\right](\omega)+\mathcal{h}\left[\,\theta(\omega')\,\omega'e^{-\frac{\omega'}{\omega_{c}}}\,\right](-\omega)\,\right]\nonumber \\ & = & \frac{\hbar}{2}\,\pi m \gamma_{i}\,\omega\left[\,\mathcal{h}\left [ \theta(\omega')\,e^{-\frac{\omega'}{\omega_{c}}}\,\right](\omega)-\mathcal{h}\left[\,\theta(\omega')\,e^{-\frac{\omega'}{\omega_{c}}}\,\right ] ( -\omega)\,\right ] + \hbar\,m \gamma_{i}\,\omega_{c } \nonumber \\ & = & \frac{\hbar}{2}\,m \gamma_{i}\,\omega\left[\,e^{-\frac{\omega}{\omega_{c}}}\,\gamma(0,-\omega/\omega_{c } ) -e^{\frac{\omega}{\omega_{c}}}\,\gamma(0,\omega/\omega_{c})\,\right ] + \hbar\,m\gamma_{i}\,\omega_{c}\,,\end{aligned}\ ] ] where @xmath232 denotes the incomplete gamma function .
following the same procedure as in the previous section for the super - ohmic spectral density ( [ sd2 ] ) , one obtains @xmath231&=&\frac{\hbar}{2\omega_{c}}\,\pi m \gamma_{i}\left[\,\mathcal{h}\left[\,\theta(\omega')\,(\omega')^2\,e^{-\frac{\omega'}{\omega_{c}}}\,\right](\omega)+\mathcal{h}\left[\,\theta(\omega')\,(\omega')^2\,e^{-\frac{\omega'}{\omega_{c}}}\,\right ] ( -\omega)\,\right]\nonumber \\ & = & \frac{\hbar}{2\omega_{c}}\,\pi m\gamma_{i}\,\omega^2\left[\,\mathcal{h}\left[\,\theta(\omega')\,e^{-\frac{\omega'}{\omega_{c}}}\,\right ] ( \omega)+\mathcal{h}\left[\,\theta(\omega')\,e^{-\frac{\omega'}{\omega_{c}}}\,\right](-\omega)\,\right ] + \hbar m \gamma_{i}\,\omega_{c } \nonumber \\ & = & \frac{\hbar}{2\omega{c}}\,m \gamma_{i}\,\omega^2\left[\,e^{-\frac{\omega}{\omega_{c}}}\gamma(0,-\omega/\omega_{c})+e^{\frac{\omega}{\omega_{c}}}\gamma(0,\omega/\omega_{c})\,\right ] + \hbar m \gamma_{i}\,\omega_{c}\,.\end{aligned}\ ] ]
in this appendix we derivate the correlation function of the fluctuation forces associated with the heat baths given in eq.([cffdal ] ) . considering the time dependence of these forces
, it follows @xmath233 replacing the identities ( [ cmhb ] ) in eq.([aff1 ] ) , and applying the fourier transform , one obtains @xmath234 \int\!\!\int dt\ , dt ' e^{i\omega t}\,e^{i\omega ' t ' } \cos(\omega_{i\mu}(t+t'-2t_{0 } ) ) \nonumber \\ & + & \text{im}\left[m(\omega_{i\mu})\right]\int\!\!\int dt\,dt ' e^{i\omega t}\,e^{i\omega ' t'}\sin(\omega_{i\mu}(t+t'-2t_{0}))\big)\ , .
\nonumber \end{aligned}\ ] ] to compute the previous integrals we express the trigonometric functions as complex exponentials , introduce the change of variable @xmath235 in the first integral , and @xmath236 in the second and third ones .
secondly , we use the definition of the delta function @xmath237 and the identity @xmath238 , which lead to @xmath239\coth\left ( \frac{\hbar \omega}{2k_{b } t_{l}}\right)\cosh(2r_{l } ) \nonumber \\ & -&2\pi\hbar\delta(\omega-\omega')\coth\left ( \frac{\hbar \omega}{2k_{b } t_{l}}\right)\sinh(2r_{l})\text{re}\left[e^{i \theta_{l}}\right]\bigg(\int_{0}^{\infty}d{\bar\omega } { \bar j}_l(\omega-{\bar\omega})e^{2i{\bar\omega } t_{0 } } - \int_{0}^{\infty}d{\bar\omega } { \bar j_l}(\omega+{\bar\omega})e^{-2i{\bar\omega } t_{0 } } \bigg ) \nonumber \\ & + & 2i \pi\hbar\delta(\omega-\omega ' ) \coth\left ( \frac{\hbar \omega}{2k_{b } t_{l}}\right)\sinh(2r_{l})\text{im}\left[e^{i \theta_{l}}\right ] \bigg(\int_{0}^{\infty}d{\bar\omega } { \bar j}_l(\omega-{\bar\omega})e^{2i{\bar\omega } t_{0 } } + \int_{0}^{\infty}d{\bar\omega } { \bar j}_l(\omega+{\bar\omega})e^{-2i{\bar\omega } t_{0}}\bigg)\ , , \nonumber \\ \label{cffd}\end{aligned}\ ] ] with @xmath240 . before to continue , we pay attention to the four integrals having an explicit dependence on the initial time @xmath96 . for both the ohmic and super - ohmic spectral densities , ( [ sd1 ] ) and ( [ sd2 ] )
, the corresponding @xmath241 decays more rapidly than @xmath242 at high frequencies .
this allows us to use the riemann - lebesgue lemma which states @xcite @xmath243 for @xmath244 an absolutely integrable function in the interval @xmath245 . as a consequence , only the first term in eq.([cffd ] ) survives after taking the long - time limit @xmath30 .
the riemann - lebesgue lemma has been successfully employed in the study of the properties of the stationary state for the damped harmonic oscillator @xcite .
in the following we briefly describe the computation of the logarithmic negativity , quantum discord , and the separability criteria @xmath145 .
the evaluation of the logarithmic negativity ( [ logneg ] ) requires the symplectic eigenvalues of the partial transpose @xmath246 .
these are given by the positive square roots of the eigenvalues of @xmath247 @xcite , which are given in terms of the so - called symplectic matrix @xmath248 where @xmath249 is the @xmath144-dimensional unit matrix .
the corresponding partial transpose matrix is obtained from @xmath250 , with @xmath251 the evaluation of the quantum discord ( [ disc ] ) on the state @xmath60 involves an optimization procedure over all positive operator - valued measurements ( povms ) on the @xmath252-mode , denoted by @xmath253 . in eq .
( [ disc2 ] ) , @xmath254 is the probability associated with the @xmath88 measurement outcome , and @xmath255 is the corresponding post - measurement reduced state of the @xmath100-mode .
an explicit formula providing @xmath256 for any input gaussian state @xmath257 is given in @xcite .
@xmath258 can be obtained from a similar optimization procedure on the povmss in the subsystem @xmath100 . in general , both evaluations of discord may return different values ( i.e. @xmath259 ) .
now we describe a hierarchy of separability criteria recently proposed to characterize from genuine multipartite to bipartite entanglement @xcite . according to this proposal , the state @xmath260 of a @xmath144-partite system is @xmath143-partite entangled , _
i.e. _ there is at least a entangled subsystem composed of @xmath143 parties , provide that a given function @xmath261 takes a positive value .
the evaluation of the function @xmath262 involves the selection of a set of @xmath263 pure states that allows to assess multipartite entanglement .
the important point is that a reliable characterization of entanglement requires an appropriate choice of such probe states .
however , a priori there is no information about the ` optimal ' probe states enable to unveil the entanglement encapsulated by an arbitrary density operator @xmath260 .
one may circumvent this difficulty by carrying out an optimization procedure over all possible selections in order to obtain the maximum of @xmath262 , whose positive value would reveal the entanglement in the system .
we denote @xmath145 such maximum . in continuous - variable states
a gaussian selection of the probe states provides a readable expression of @xmath262 @xcite , which can be optimized with standard procedures , and which is strong enough to detect entanglement for a broad class of gaussian and non - gaussian states . in the gaussian case
this expression reads @xcite , @xmath264 where @xmath265 are constant values @xcite , @xmath266 is a real @xmath263-vector , @xmath267 is the standard form of the so - called symplectic matrix , and @xmath268 and @xmath269 are @xmath270 real matrices .
the objects @xmath266 and @xmath269 denote a compact form of the first and second moments of the probe states @xcite .
hence , the detection of entanglement consists basically in optimizing eq.([stau ] ) over the variables @xmath266 and @xmath269 , _ i.e. _
@xmath271 69 ifxundefined [ 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 | we present a study that addresses both the stationary properties of the energy current and quantum correlations in a three - mode chain subjected to ohmic and super - ohmic dissipations .
an extensive numerical analysis shows that the mean value and the fluctuations of the energy current remain insensitive to the emergence of a rich variety of quantum correlations : such as , two - mode discord and entanglement , bipartite three - mode and genuine tripartite entanglement .
the discussion of the numerical results is based on the derived expressions for the stationary properties in terms of the two - time correlation functions of the oscillator operators , which carry quantum correlations .
interestingly , we show that quantum discord can be enhanced by both considering initially squeezed thermal bath states and imposing temperature gradients . |
is one of fourteen known transneptunian binaries ( tnbs ) ( noll 2003 ) .
in addition to , three others of these fourteen have orbits that have been determined sufficiently to derive masses , , , and ( veillet et al . 2002 ; osip et al . 2003 ; noll et al .
the opportunity to directly measure the mass makes tnbs among the most attractive objects for study among the @xmath71000 known transneptunian objects ( tnos ) .
was discovered at the canada - france - hawaii telescope ( chen et al .
1997 ) and was found to be a binary object from observations made with the hubble space telescope in november 2001 ( noll and stephens 2002 ) .
follow - up observations made in june and july 2002 showed a wider separation than at initial discovery and a very small rate of angular motion .
our initial conclusions were that an eccentric orbit was likely ( noll et al .
2002 ) . with additional epochs we are now able to determine an orbit for this binary system yielding a direct measurement of its mass .
photometry of the individual components can be used to constrain the relative masses and radii of the components .
albedo as a function of density is also constrained . in this work we detail our observations and explore the physical implications of the mass determined for this binary system .
we have been able to constrain the orbit of by combining data obtained with the hubble space telescope ( hst ) and with the magellan project baade and clay 6.5 m telescopes .
the observations are summarized in table 1 and described in detail separately below .
hst observations were obtained using the wide field planetary camera 2 ( wfpc2 ) and near infrared camera multi - object spectrograph ( nicmos ) in three separate observing programs . all observations employed moving target tracking with hst to follow the apparent motion of .
the observations made with the wfpc2 camera on four separate dates have been described in detail by noll et al .
the first epoch of wfpc2 observation was obtained in the wf3 detector with 0.1 pixels ; the remaining pointings centered on the higher spatial resolution , 0.045-per - pixel pc detector .
three different filters were used in the wf3 observations , the f555w , f675w , and f814w ; only the f814w was used with the pc .
two or more undithered exposures were recorded in each filter to facilitate the removal of cosmic rays .
three additional observations of have been made using hst s near infrared camera and multi - object spectrometer ( nicmos ) on 8 may and 16 may 1998 and 23 april 2003 .
the may 1998 data were exposed as target location images for attempted grism spectroscopy of several bright tnos ( kern et al .
1999 ) . on each date three
32 second exposures of the target using the step32 multiaccum sequence were obtained at dithered positions in the nic3 camera with the f150w filter in place .
these relatively short exposures yielded low s / n images of .
we were able to reliably detect the secondary only on 16 may 1998 , as indicated in table 1 .
the april 2003 observation of consisted of four exposures , two each in the f110w and f160w filters ( fig . 1 ) .
exposures were 256 sec and 512 sec in the two filters at each of two dithered positions on the detector .
a step256 sample sequence was used with 11 and 12 samples respectively .
hst tracked the apparent motion of the binary during all nicmos ( and wfpc2 ) observations and , importantly , simultaneously corrected for the parallax induced by the telescope s orbital motion .
these data yielded high s / n detections of both components of the binary system .
nicmos data were reduced using the standard on - the - fly pipeline reduction with the most up - to - date dark and flat reference files .
magellan data were obtained with the raymond and beverly sackler magellan instant camera ( magic ) on eight separate nights in 2002 - 2004 . of these , we have considered three epochs not already sampled by high quality hst observations , ut 28 - 29 january 2003 , 15 december 2003 , and 08 - 09 march 2004 which offer the best combination of optimal observing conditions and potential to add unique constraints to the orbit ( fig . 2 ) .
other dates either overlapped in time with hst observations or were obtained on nights with poor seeing .
magic is a cryogenic ccd camera with a 2k@xmath82k site detector .
permanently mounted at a folded port of the magellan project clay 6.5 m telescope , magic provides a pixel scale of 0.069 .
all observations reported here were obtained with the sloan @xmath9 filter in place .
standard overscan subtraction and flat fielding were applied to all frames . on each date two series of at least three separate 150 - 300 second exposures
were obtained . during the integration ,
the pointing was fixed ( sidereal tracking ) so that moved . in january 2003 during 300 second exposures , the target moved 0.195 at a position angle of 294.7 degrees . in december 2003 ,
300 second exposures correspond to only 0.07 of sky motion for the target .
finally , in march 2004 , shorter integrations of 150 seconds resulted in target image motion of 0.127 at a position angle of 293.1 degrees .
this motion further complicates accurate measurement of the separation of the two components .
in order to obtain the most accurate positional information from our images , we applied a detailed point - spread - function ( psf ) fitting routine to each of the individual images .
model psfs were generated using the tinytim software package ( krist and hook 2003 ) . following the method described in noll et al .
( 2002 ) we iteratively scaled and shifted synthetic psf pairs , subtracted from the image , and computed residuals .
models with the smallest residuals were used to calculate the most likely separation and position angle for the two components .
errors in centroiding each component are typically of order @xmath1 0.1 pixel . the uncertainty in separation and position angle is then calculated directly from the combined centroiding uncertainties of each component .
results are listed in table 1 . the positional information derived from our images
are the raw data that we employed in fitting orbits . in an ideal case where there are high quality astrometric measurements at several points along an orbit , including measurements near pericenter and apocenter , orbit analysis can result in a precise determination of the system mass .
the measurements of are , however , far from ideal .
all but one epoch of our high precision hst data are clustered near position angles of 340 degrees .
the sparseness of the sampling ( table 1 , fig .
2 ) limits our ability to determine the orbit and the mass with the hst data alone .
the hst data robustly determine the orbital period but do not fully constrain the remaining orbital elements . to break this impasse , we sought additional constraints from ground - based observations . was observed at magellan as part of an ongoing program of tno observations that resulted in the discovery and characterization of ( osip et al .
because the ground - based data are considerably lower in resolution than hst data , we chose to include only two epochs in our analysis based on their ability to add constructively to our orbit fitting . seeing on the nights of ut 28 january 2003 , 15 december 2003 , and 09 march 2004 was 0.65 , 0.55 , and 0.30 respectively . because of the small 0.069 pixels , the point - spread - function ( psf ) is oversampled in all images . while this leads to a reduction in the signal - to - noise per pixel for any given source , it is ideal for identifying partially resolved binaries through comparison of the tno psf with stellar psfs in the same frame .
a 2-d truncated gaussian is fit to the images of stars to determine the best - fit fwhm of the psf for each image .
this psf is then compared to the image of the binary .
the images of the binary on the night of 09 march 2004 show two clearly resolved peaks ( fig . 2 ) .
the separation is measured by fitting psfs to each peak .
the data from 28 january 2003 do not resolve the binary components , but do indicate elongation of the stellar psf corresponding to the target motion vector .
for a two - component model with a magnitude difference consistent with that observed at other epochs we can place an upper limit on the separation at 0.3 , or slightly less than half the fwhm of the seeing on that night .
data from 15 december 2003 suffer contamination by a similar magnitude background object within 0.5 - 0.7 .
nevertheless , adopting a two - component model , these data do constrain the separation of the primary and secondary components to be no greater than 0.3 .
orbit fitting was accomplished by iteratively minimizing differences between observed and model sky - plane offsets between primary and secondary , projected for the time of each observation .
this procedure has the advantage of properly accounting for the time variable projection of the orbit as seen from earth .
the parallax motion of the earth and the seasonally changing aspect of the tnb s orbit both produce appreciable variations in the projected shape of the orbit . with our observations spread over several years
, it was essential to include these effects .
two separate solutions to the orbit are found , each comprising equally good fits to the observed data .
the orbital elements of both possible solutions are listed in table 2 .
the orbits differ primarily in the orientation of the orbit plane and in the timing of mutual events but are otherwise very similar .
we have plotted the solution with the lower @xmath10 in fig . 3 , though we note that the difference is insignificant .
because of the sparseness of the sampling and the clustering of the highest quality data at a position angle near 340 degrees , the orbit is dependent on the lowest quality data points we have included in the fit , the upper limits from 28 january and 15 december 2003 .
as described above , we have attempted to apply conservative criteria to determining this upper limit in order to insulate the orbit fit from an error in this single datum .
it is worth noting that even for one - sigma upper limits to the separation as high as 0.5 , the orbit fit remains essentially unchanged , although with greater uncertainties in the orbit parameters .
apparently , even very weak , fractional - sigma constraints are all that are needed to collapse the uncertainties in the orbit fits .
it is also possible to consider the extreme case where we simply eliminate the upper - limit points from our fits . in that case
the orbit fitting routine again finds essentially the same orbit as above , but with uncertainties large enough that the eccentricity and semi - major axis of the orbit become indeterminate , though the orbit period remains well - determined .
a single orbit fit is replaced by a family of orbits with fixed @xmath3 that are able to equally - well fit the remaining data .
the constrained orbit we find when using all the data is near the low - mass , low - eccentricity , high probability end of this family of possible orbits .
higher masses correspond to increasingly unlikely orbits having high eccentricity and an orbit plane inclination approaching an edge - on view .
further observation , particularly at times when position angles are away from our cluster of data at 340 degrees , will be able to confirm the validity of our derived orbit .
photometry of the individual components was performed using the same psf - fitting technique used to determine relative positions .
the optimally - scaled psf is used to determine the flux including appropriate corrections for finite aperture size .
we have corrected photometry for phase angle effects using a phase function of 0.15 mags / deg at all wavelengths ( sheppard and jewitt 2002 ) .
measured photometry is listed in table 3 .
photometric data from five filters can be combined to produce a low resolution optical - near - infrared spectrum as we have done in fig .
4 . as is typical for tno albedo spectra ( e.g. noll et al .
2000 , mcbride et al .
2003 ) there is a red slope from the optical to approximately 1 micron after which the slope flattens .
the spectral slope in the optical has been observed to be highly correlated independent of filter over a range from b through i ( 430 - 800 nm ; jewitt and luu 2001 ) .
the spectral gradient , @xmath11 , defined as the fractional change in reflectance per 100 @xmath12 ( boehnhardt et al .
2001 ) is @xmath13 and @xmath14 for components a and b respectively measured over the range from f555w to f814w which approximates v and i. this falls in the mid - range of v - i gradients we have measured for 41 non - resonant tnos ( stephens et al .
the gradients we measure for the individual components are marginally within the error of the v - i gradient for the combined pair , @xmath15 , measured from the ground ( boehnhardt et al .
2001 ) .
variability is an important issue that must be addressed when combining photometric data from different epochs .
tnos are known to have intrinsic lightcurves , some with significant amplitudes ( need a reference and quantitative number ) . as listed in table 1 , the data obtained with hst at multiple epochs is ill - suited for an evaluation of possible lightcurve variations .
the f814w filter is the only one used on multiple dates , a total of four different epochs . as noted in detail in noll et al .
( 2002 ) , there is evidence for possible large amplitude variations in component a ( normally the brighter of the pair ) with variations in the i band of as much as a magnitude .
however , with limited sampling , this observation remains unverified .
ground - based observations of the lightcurve of this object could resolve the question of whether large albedo or shape asymmetries exist on either or both of the components of this binary . once the orbit is sufficiently well determined that an orbital period and semimajor axis can be determined , it is possible to retrieve the total mass of the system using kepler s third law : @xmath16 .
both orbit solutions result in a system mass of 0.42@xmath17 .
this mass is notable in that it is an order of magnitude lower than the system masses of the three other binary tnos that have been measured so far .
it is possible to use measured photometry to determine the relative size and mass of the components of .
the mass ratio is given by @xmath18 conditioned by the assumption that both components have the same albedo in the relevant photometric band .
we computed this ratio from a weighted average of the flux ratios measured in the f555w , f675w , and f814w filters with wfpc2 and the f110w and f160w filters with nicmos .
individual masses can be calculated using this ratio and the derived system mass .
we find @xmath19 and @xmath20 .
similarly , the ratio of radii is @xmath21 .
if we assume a bulk density , it is possible to derive radii from the masses of the components . for @xmath22
we find @xmath23 and @xmath24 .
for different assumed densities the derived radii scale by a factor of @xmath25 .
a direct consequence of the low system mass is the conclusion that must have a high albedo for any plausible assumed density . as shown in fig . 5 , even at an extremely low density of @xmath26 we find a red albedo of @xmath27 that is nearly six times higher than has typically been assumed for tnos .
as is the case for a single object of measured flux and unknown mass , the mass of the kuiper belt as a whole is proportional to the assumed albedo to the minus three halves power , independent of assumptions about the size distribution .
typically , estimates of the mass of the belt as a whole have relied on the assumption that @xmath6 = 0.04 . if the average albedo is five times higher , @xmath6 = 0.2 , the mass calculation overestimates the mass of the belt by more than an order of magnitude .
for example , bernstein et al . ( 2004 ) estimate the total mass of the non - resonant , non - scattered population of the kuiper belt to be 0.010 m@xmath28 using the standard albedo assumption and their latest estimate of the size distribution .
this already low mass estimate could be significantly too large if the average albedo of tnos is greater than the standard @xmath6=0.04 .
unless the density of is far lower than any other solar system object known , it indicates that at least some objects in the kuiper belt have significantly higher albedos .
observations of a significant number of objects at thermal wavelengths using the spitzer space telescope will constrain this important physical parameter . the period , @xmath0 , is generally relatively easy to determine , given observations spanning a substantial fraction , or more , of an orbital period . the determination of @xmath3 , however , can be more problematic , particularly in cases where the orbit is highly eccentric and/or the normal to the orbit plane is highly inclined . in these situations
the simple strategy of observing at apocenter and pericenter to directly determine the quantity @xmath29 may be impractical or impossible and , in any case , requires knowledge of the orbit plane orientation . for sparsely sampled orbits ,
the uncertainty in the orbital eccentricity , @xmath2 , is also a significant barrier to uniquely determining @xmath3 .
our analysis of the orbit of provides a clear example of the difficulties involved .
determination of other tnb orbits will inevitably face similar complications .
there is growing evidence that high eccentricity is a common property of tnbs . at the present time , four tnbs ( excluding pluto / charon which has been circularized due to tidal interactions ) have orbits sufficiently well - determined that their eccentricity is constrained .
the first tnb to have its orbit measured , , has an eccentricity of @xmath2 = 0.82 @xmath10.05 ( veillet et al .
osip et al .
( 2003 ) have determined that has an eccentricity of @xmath2 = [email protected] .
we have recently measured the orbit of and find an eccentricity of @xmath2 = [email protected] ( noll et al .
adding to the mix , with @xmath2 = [email protected] appears to indicate that moderately high eccentricity may be the norm for tnbs , though extremely high eccentricities are not .
this result suggests that the formation of binaries through the exchange reaction mechanism proposed by funato et al .
( 2004 ) , which results almost exclusively in eccentricities @xmath30 0.8 is not the primary mechanism for the formation of tnbs .
however , the continuing uncertainty in the formation of stellar binaries ( e.g. mathieu 1994 ) , despite more than 200 years of study , is a sobering reminder of the difficulty of making genetic inferences from orbital data alone .
the radius of the hill sphere , @xmath31 , in a three - body system is given by the equation @xmath32 where @xmath3 is the semimajor axis of the primary s heliocentric orbit .
orbits within the hill sphere are stable relative to the sun , though perturbations from a fourth body ( e.g. neptune in the case of tnos ) can alter the zone of stability . for binary systems where the mass is known
, it is possible to compute the separation of binary components in terms of the hill radius .
the size of the hill radius is @xmath77000 times the radius of the largest component of the binary in the four tnbs with measured mass .
the derived orbits of these systems have semimajor axes that are 0.8% to 6% of the hill radius ( table 4 ) .
pluto has a similarly large hill radius , but differs in that charon orbits at only 0.25% of this distance .
tnbs with comparably tight orbits may exist but be undetected with current searches . indeed
, the large amplitude lightcurve of 2001 qg@xmath33 may be the first recognized contact binary in the kuiper belt and suggests that 10% to 20% of the transneptunian population may consist of similarly close pairs ( sheppard and jewitt 2004 ) .
the majority of known binaries in the main asteroid belt orbit much closer to their primaries in terms of absolute distance and also in terms of the semimajor axis relative to the size of the primary , @xmath34 ( merline et al . 2003 ) than known tnbs .
this is partially due to observational selection effects limiting the discovery of close tnbs .
however , it is interesting to note that , in terms of the hill radius , the semimajor axes of main - belt asteroids are quite comparable to tnbs , falling in a range of one to a few percent ( table 4 ) .
we can also compare tnbs to planetary satellite systems .
the four galilean satellites orbit in a range from 0.7% to 3% of jupiter s hill sphere radius while the most distant satellites , the retrograde irregulars , orbit at nearly half the hill radius ( sheppard and jewitt 2003 ) . in terms of their separations in units of the hill radius , tnbs as well as main - belt binaries are more similar to regular satellites systems than to the irregular captured objects .
the significance of this similarity awaits a comprehensive theory of planetesimal binary formation in the early solar system .
the orbit of the binary system has been determined with a combination of data from the hubble space telescope and the magellan observatory .
the derived orbital parameters reveal a system of nearly equal - sized bodies in a moderately eccentric orbit .
the mass derived from the orbit is an order of magnitude lower than the three other tnbs that have been measured to date , and is only @xmath71/35,000 of the mass of the pluto / charon system . with probable diameters of less than 80 km ,
the components of are approaching the size range of the largest cometary nuclei .
for any reasonable assumed density , these objects must be remarkably reflective with red albedos greater than 20% .
if objects this reflective are common in the kuiper belt , current mass estimates may be too high by as much as an order of magnitude .
the separations of transneptunian binaries in terms of the hill radius are comparable to the separations of main belt binaries suggesting possible parallels in their formation .
[ merline , w. j. , weidenschilling , s. j. , durda , d. d. , margot , j - l . , pravec , p. , storrs , a. d. 2002 , in _ asteroids _ , eds . w. f. bottke , a. cellino , p. paolicchi , r. p. binzel , university of arizona press , tucson , pp .
289 - 312 ] = 1000 kg m@xmath5
. the albedo scales with density to the 2/3 power .
the solid curve corresponds to the larger member of the binary , component a , and the dotted curve to the smaller component b. the dotted bars at the bottom of the figure are show the bandpasses of the five filters used .
the discrepant f814w photometry from 18 june 2002 was not included in computing the average flux at this wavelength .
[ fig4 ] ] = 1000 kg m@xmath5 , the average albedo is @xmath39 = 0.39 , far higher than generally assumed for transneptunian objects .
if albedos as high as this are typical , the mass of the kuiper belt may be currently overestimated by an order of magnitude . | we have measured the separations and position angles of the two components of the binary transneptunian object at eight epochs . from these data
we are able to constrain the orbit and mass of this binary system .
the best fitting orbit has an orbital period of @xmath0 = 312@xmath13 days .
the orbital eccentricity is @xmath2 = [email protected] and the semimajor axis is @xmath3 = 8,010@xmath180 km .
the mass of the system is surprisingly low at [email protected] @xmath4 kg . to account for the observed brightness consistent with the low mass and a range of plausible densities
, the geometric albedo must be very high ; for an assumed density of 1000 kg m@xmath5 we find a red geometric albedo of @xmath6 = 0.37 , an order of magnitude higher than has been generally assumed for transneptunian objects .
if objects with comparably high albedos are common in the kuiper belt , estimates of the mass of the belt must be significantly reduced .
the semimajor axis of s orbit is 2.8% of the hill radius of the primary , a value strikingly similar to this same ratio in other transneptunian binaries , main - belt binaries , and regular satellite systems .
[ 0.5in=1 ] |
the paper is devoted to the numerical solution of small - strain quasi - static elastoplastic problems .
such a problem consists of the constitutive initial value problem ( civp ) and the balance equation representing the principle of virtual work .
a broadly exploited and universal numerical / computational concept includes the following steps : * time - discretization of civp leading to an incremental constitutive problem ; * derivation of the constitutive and consistent tangent operators ; * substitution of the constitutive ( stress - strain ) operator into the balance equation leading to the incremental boundary value problem in terms of displacements ; * finite element discretization and derivation of a system of nonlinear equations ; * solving the system using a nonsmooth variant of the newton method .
civp satisfies thermodynamical laws and usually involves internal variables such as plastic strains or hardening parameters .
several integration schemes for numerical solution of civp were suggested . for their overview , we refer , e.g. , @xcite and references introduced therein . if the implicit or trapezoidal euler method is used then the incremental constitutive problem is solved by the elastic predictor / plastic corrector method .
the plastic correction leads to the return - mapping scheme .
we distinguish , e.g. , implicit , trapezoidal or midpoint return - mappings depending on a chosen time - discretization ( * ? ? ?
* chapter 7 ) . in this paper
, we assume that the plastic flow direction is generated by the plastic potential , @xmath3 .
if @xmath3 is smooth then the corresponding plastic flow direction is uniquely determined by the derivative of @xmath3 and consequently , the plastic flow rule reads as follows , e.g. ( * ? ? ?
* chapter 8) : @xmath4 here , @xmath5 , @xmath6 , @xmath7 , and @xmath8 denotes the plastic strain rate , the plastic multiplier rate , the stress tensor and the hardening thermodynamical forces , respectively .
the corresponding return - mapping scheme is relatively straightforward and leads to solving a system of nonlinear equations .
a difficulty arises when @xmath3 is _
nonsmooth_. mostly , it happens if the yield surface contains singular points , such as apices or edges .
then the function @xmath3 is rather pseudo - potential than potential and its derivative need not exist everywhere . in such a case , the rule ( [ eqn_flow_rule ] ) is usually completed by some additive formulas depending on particular cases of @xmath3 and @xmath7 in an ad - hoc manner .
for example , the implementation of the mohr - coulomb model reported in ( * ? ? ?
* chapter 6 , 8) employs one , two , or six plastic multipliers @xmath9 , depending on the location of @xmath7 on the yield surface . since the stress tensor @xmath7 is unknown in civp one must blindly guess its right location .
moreover , for each tested location , one must usually solve an auxilliary system of nonlinear equations whose solvability is not guaranteed in general .
_ these facts are evident drawbacks of the current return - mapping schemes . _ in associative plasticity
, it is well - known that the plastic flow rule ( [ eqn_flow_rule ] ) together with a hardening law and loading / unloading conditions can be equivalently replaced by the principle of maximum plastic dissipation within the constitutive model .
this alternative formulation of civp does not require special treatment for nonsmooth @xmath3 and enables to solve civp by techniques based on mathematical programming @xcite . in particular , if the implicit or trapezoidal euler method is used then the incremental constitutive problem can be interpreted by a certain kind of the closest - point projection @xcite . for some nonassociative models ,
civp can be re - formulated using a theory of bipotentials that leads to new numerical schemes @xcite .
these alternative definitions of the flow rule enable a variational re - formulation of the initial boundary value elastoplastic problem .
consequently , solvability of this problem can be investigated ( see , e.g. , @xcite ) .
therefore , the corresponding numerical techniques are usually also correct from the mathematical point of view . on the other hand ,
such a numerical treatment is not so universal and its implementation is more involved / too complex in comparison with standard procedures of computational inelasticity .
the approach pursued in this paper builds on the subdifferental formulation of the plastic flow rule , e.g. ( * ? ? ?
* section 6.3.9 ) , @xmath10 for nonsmooth @xmath3 . here
, @xmath11 denotes the subdifferential of @xmath3 at @xmath12 with respect to the stress variable .
if @xmath3 is convex at least in vicinity of the yield surface then this definition is justified , e.g. , by ( * ? ? ?
* corollary 23.7.1 ) and is valid even when @xmath3 is not smooth at @xmath7 . on the first sight , it seems that ( [ inclusion_flow_rule ] ) is not convenient for numerical treatment due to the presence of the multivalued flow direction .
the main goal of this paper is to show that the _ opposite is true _ , by demonstrating that the implicit return - mapping scheme based on ( [ inclusion_flow_rule ] ) leads to solving a just one system of nonlinear equations regardless whether the unknown stress tensor lies on the smooth portion of the yield surface or not at least for a wide class of models with nonsmooth plastic pseudo - potentials . using this technique , we eliminate the blind guessing and
thus considerably simplify the solution scheme .
moreover , the new technique enables to investigate some useful properties of the constitutive operator , like uniqueness or semismoothness , that are not obvious for the current technique .
first of all , we illustrate the new technique on a simple 2d projective problem that mimics the structure of an incremental elastoplastic constitutive problem .
consider the convex set @xmath13 and define the projection @xmath14 of a point @xmath15 as follows : @xmath16 the scheme of the projection is depicted in figure [ fig_projection ] . clearly , the function @xmath17 is convex in @xmath18 , nondifferentiable at @xmath19 and @xmath20 if @xmath21 then @xmath22 .
conversely , if @xmath23 it follows from the karush - kuhn - tucker conditions and ( [ motivation0 ] ) that the projective problem can be written as follows : _
find @xmath24 and the lagrange multiplier @xmath25 : @xmath26 where _
@xmath27 [ -1,1 ] , & w_2^*=0 , \end{array } \right.\ ] ] to find a solution to ( [ motivationb2 ] ) , it is crucial to rewrite the inclusion ( [ motivationb2])@xmath28 as an equation .
observe that @xmath29 where @xmath30 denotes a positive part of a function .
this crucial transformation will be derived in detail in section [ sec_dp ] on an analogous elastoplastic example .
thus ( [ motivationb2 ] ) leads to the following system of equations : @xmath31 since ( [ motivationb3])@xmath28 implies @xmath32 , the system of three nonlinear equations reduces to a single one @xmath33 consequently , @xmath9 can be found in the closed form as @xmath34 from which one can easily compute @xmath35 by ( [ motivationb3])@xmath36 and ( [ motivationb3])@xmath28 .
the presented idea is systematically extended on some elastoplastic models .
this paper , part i , is focused on isotropic models containing : @xmath0 yield surfaces with one or two apices ( singular points ) laying on the hydrostatic axis ; @xmath1 plastic pseudo - potentials that are independent of the lode angle ; @xmath2 nonlinear isotropic hardening ( optionally ) .
such models are usually formulated by the haigh - westergaard coordinates .
further , the implicit euler discretization of civp is considered and thus two types of return on the yield surface within the plastic correction are distinguished : @xmath37 return to the smooth portion of the yield surface ; @xmath38 return to the apex ( apices ) .
the paper is organized as follows .
section [ sec_preliminaries ] contains some preliminaries related to invariants of the stress tensor and semismooth functions .
section [ sec_dp ] is devoted to the drucker - prager model including the nonlinear isotropic hardening .
although the plastic corrector can not be found in closed form , the new technique enables to a priori decide about the return type and prove existence , uniqueness and semismoothness of the implicit constitutive operator
. the consistent tangent operator is also introduced . in section [ sec_jg ]
, we derive similar results for the perfect plastic part of the jirsek - grassl model @xcite . in section [ sec_general ] ,
the new technique is extended on an abstract model written by the haigh - westergaard coordinates . in particular , within the plastic correction , we formulate a unique system of nonlinear equations which is common for the both type of the return .
it can lead to a more correct and/or simpler solution scheme in comparison with the current technique .
section [ sec_realization ] is devoted to numerical realization of the incremental boundary value elastoplastic problem using the semismooth newton method . in section [ sec_experiments ] , illustrative numerical examples related to a slope stability benchmark are considered .
here , limit load is analyzed by an incremental method depending on a mesh type and mesh density for the drucker - prager and jirsek - grassl models . within this paper , second order tensors , matrices and vectors
are denoted by bold letters . as usual ,
small letters are used for vectors and capitals for matrices ( see section [ sec_realization ] ) .
further , the fourth order tensors are denoted by capital blackboard letters , e.g. , @xmath39 or @xmath40 .
the symbol @xmath41 means the tensor product @xcite .
we also use the following notation : @xmath42 and @xmath43 for the space of symmetric , second order tensors .
consider a stress tensor @xmath44 and its splitting into the volumetric and deviatoric parts : @xmath45 here , @xmath46 , @xmath47 , @xmath48 and @xmath49 denote the identity second order tensor , the fourth order deviatoric projection tensor , the hydrostatic pressure , and the deviatoric stress , respectively . _
the haigh - westergaard coordinates _ are created by the invariants @xmath48 , @xmath50 and @xmath51 , where @xmath52 @xmath53 clearly , @xmath54 and @xmath55 $ ] . since @xmath51 is not well - defined when @xmath56 , the lode angle is included in elastoplastic models indirectly .
usually , it is considered another invariant in the form @xmath57 where @xmath58 is assumed to be a smooth function such that @xmath59 is at least continuous . however , as we will see below , it is more convenient to assume _ strong semismoothness _ of @xmath59 . as a particular case , it will be considered the invariant @xmath60 where @xmath61 the function @xmath62 was proposed in @xcite and contains the excentricity parameter @xmath63 $ ] .
it holds : @xmath0 @xmath64 is a bounded and smooth function for any @xmath44 , @xmath65 ; @xmath1 @xmath66 when @xmath56 ; @xmath2 @xmath67 , @xmath68 when @xmath69 .
we will also use the following derivatives : @xmath70 @xmath71,\quad \frac{\partial r_e}{\partial{\mbox{\boldmath$\sigma$}}}=-r_e'(\cos\theta)\sin\theta\frac{\partial \theta}{\partial{\mbox{\boldmath$\sigma$}}}. \label{inv_der2}\ ] ] notice that the derivatives of @xmath50 , @xmath72 , @xmath51 and @xmath62 do not exist when @xmath56 .
further , @xmath51 is not differentiable when @xmath7 satisfies either @xmath73 or @xmath74 . on the other hand
, @xmath62 has derivatives for such stresses @xcite . for purposes of this paper , it is crucial to derive the subdifferential of @xmath50 at @xmath7 when @xmath75 : @xmath76 if @xmath65 then @xmath77 by ( [ inv_der1])@xmath78 .
it is readily seen that @xmath79 regardless @xmath80 or not .
semismoothness was originally introduced by mifflin @xcite for functionals .
qi and j. sun @xcite extended the definition of semismoothness to vector - valued functions to investigate the superlinear convergence of the newton method .
we introduce a definition of _ strongly semismooth _ functions @xcite . to this end , consider finite dimensional spaces @xmath81 and @xmath82 with the norms @xmath83 and @xmath84 , respectively . in the context of this paper ,
the abstract spaces @xmath85 represent either subspaces of @xmath86 or the space @xmath87 .
let @xmath88 be locally lipschitz function in a neighborhood of some @xmath89 and @xmath90 denote the generalized jacobian in the clarke sense @xcite .
we say that @xmath91 is strongly semismooth at @xmath92 if 1 .
@xmath91 is directionally differentiable at @xmath92 , 2 . for any @xmath93 , @xmath94 , and any @xmath95 , @xmath96 [ def_semi ]
notice that the estimate ( [ semi_1 ] ) is called the quadratic approximate property in @xcite or the strong @xmath97-semismoothness in @xcite . in literature
, there exists several equivalent definitions of strongly semismooth functions , see @xcite .
for example the condition ( [ semi_1 ] ) can be replaced with @xmath98 where @xmath99 is the subset of @xmath81 , @xmath91 is frchet differentiable and @xmath100 denotes the directional derivative of @xmath91 at a point @xmath92 and a direction @xmath101 .
we say that @xmath88 is strongly semismooth on an open set @xmath102 if @xmath91 is strongly semismooth at every point of @xmath103 . since it is difficult to straightforwardly prove ( [ semi_1 ] ) or ( [ semi_3 ] ) , we summarize several auxilliary results .
firstly , piecewise smooth ( @xmath104 ) functions with lipschitz continuous derivatives of selected functions belong among strongly semismooth functions @xcite . especially , we mention the max - function , @xmath105 , @xmath106 .
further , scalar product , sum , compositions of strongly semismooth functions are strongly semismooth . finally , we will use the following version of the implicit function theorem @xcite .
let @xmath107 be a locally lipschitz function in a neighborhood of @xmath108 , which solve @xmath109 .
let @xmath110 denote the generalized derivatives of @xmath111 at @xmath112 with respect to the variables @xmath92 .
if @xmath113 is of maximal rank , i.e. the following implication holds , @xmath114 then there exists an open neighborhood @xmath115 of @xmath116 and a function @xmath117 such that @xmath91 is locally lipschitz continuous in @xmath115 , @xmath118 and for every @xmath119 in @xmath115 , @xmath120 .
moreover , if @xmath111 is strongly semismooth at @xmath108 , then @xmath91 strongly semismooth at @xmath116 .
[ implicit ] the semismoothness of constitutive operators in elastoplasticity has been studied e.g. in @xcite .
namely in @xcite , one can find an abstract framework how to investigate it for operators in an implicit form .
however , the framework can not be straightforwardly used for models investigated in this paper .
therefore , we introduce the following auxilliary result .
let @xmath121 , @xmath122 , @xmath123 , @xmath124 , @xmath125 , @xmath126 and @xmath127 be the functions introduced in section [ sec_invariants ] .
further , ler @xmath128 be strongly semismooth functions and assume that @xmath129 vanishes for any @xmath44 , @xmath80 .
define , @xmath130 then the functions @xmath50 , @xmath131 , @xmath132 and @xmath133 are strongly semismooth in @xmath87 .
[ prop_semi ] since the functions @xmath125 and @xmath134 are bounded and have lipschitz continuous derivatives in @xmath135 , it is easy to see that the functions @xmath50 , @xmath131 , @xmath132 and @xmath133 are locally lipschitz continuous in @xmath87 and strongly semismooth for any @xmath44 , @xmath136 . therefore , it remains to show strong semismoothness at @xmath44 , @xmath80 . to this end
, we show that ( [ semi_3 ] ) holds for @xmath50 , @xmath131 , @xmath132 and @xmath133 at such @xmath7 .
let @xmath137 be such that @xmath138 .
then @xmath139 hence @xmath140 , @xmath141 and @xmath142{\mbox{\boldmath$r$}}_e(\cos\theta({\mbox{\boldmath$\tau$}})){=}o(\|{\mbox{\boldmath$\tau$}}\|^2),\\ { \mbox{\boldmath$s$}}'({\mbox{\boldmath$\sigma$}}+{\mbox{\boldmath$\tau$}};{\mbox{\boldmath$\tau$}})-{\mbox{\boldmath$s$}}'({\mbox{\boldmath$\sigma$}};{\mbox{\boldmath$\tau$}})&=&[\hat p'({\mbox{\boldmath$\sigma$}}+{\mbox{\boldmath$\tau$}};{\mbox{\boldmath$\tau$}})-\hat p'({\mbox{\boldmath$\sigma$}};{\mbox{\boldmath$\tau$}})]{\mbox{\boldmath$i$}}+[\hat\varrho'({\mbox{\boldmath$\sigma$}}+{\mbox{\boldmath$\tau$}};{\mbox{\boldmath$\tau$}})-\hat\varrho'({\mbox{\boldmath$\sigma$}};{\mbox{\boldmath$\tau$}})]{\mbox{\boldmath$n$}}({\mbox{\boldmath$\tau$}}){=}o(\|{\mbox{\boldmath$\tau$}}\|^2),\end{aligned}\ ] ] since the functions @xmath143 , @xmath144 satisfy ( [ semi_3 ] ) by the assumption .
the function @xmath133 introduced in proposition [ prop_semi ] has the same scheme as a mapping between trial and unknown stress tensors for models introduced in section [ sec_dp][sec_general ] . here
, the trial stress is represented by @xmath7 and the unknown stress is in the form @xmath145 .
therefore , it is sufficient to prove only semismoothness of the scalar functions @xmath143 , @xmath144 representing invariants of the unknown stress tensor .
the semismoothness of @xmath146 has been derived to prove theorem [ th_semi_jg ] .
we consider the elastoplastic problem containing the drucker - prager criterion , a nonassociative plastic flow rule and a nonlinear isotropic hardening . within a thermodynamical framework with internal variables ,
we introduce the corresponding constitutive initial value problem , see @xcite : 1 .
_ additive decomposition of the infinitesimal strain tensor @xmath147 on elastic and plastic parts _ :
@xmath148 2 .
_ linear isotropic elastic law between the stress and the elastic strain _ : @xmath149 where @xmath150 denotes the bulk , and shear moduli , respectively .
3 . _ non - linear isotropic hardening _ : @xmath151 here @xmath152 denotes an isotropic ( scalar ) hardening variable , @xmath153 is the corresponding thermodynamical force and @xmath154 is a nondecreasing , strongly semismooth function satisfying @xmath155 .
drucker - prager yield function _ : @xmath156 here , the parameters @xmath157 are usually calculated from the friction angle using a sufficient approximation of the mohr - coulomb yield surface and @xmath158 denotes the initial cohesion .
_ plastic pseudo - potential_. @xmath159 here @xmath160 denotes a parameter depending on the dilatancy angle .
_ nonassociative plastic flow rule _
@xmath161 where @xmath162 is a multiplier and @xmath163 denotes the subdifferential of the convex function @xmath3 at @xmath7 . using ( [ inv_der1 ] ) , ( [ rho_subgrad ] ) and ( [ potential_function ] ) , the flow rule ( [ flow_rule ] ) can be written as @xmath164 consequently by ( [ split ] ) , ( [ elastic_law ] ) and ( [ rho_subgrad_prop ] ) , @xmath165 7 .
_ associative hardening law : _
. _ loading / unloding criterion : _
@xmath167 then the elastoplastic constitutive initial value problem reads as follows : _ given the history of the strain tensor @xmath168 , @xmath169 $ ] , and the initial values @xmath170 find @xmath171 such that @xmath172 hold for each instant @xmath173$]_. we discretize civp using the implicit euler method . to this end
we assume a partition @xmath174 of the pseudo - time interval and fix a step @xmath175 .
for the sake of brevity , we omit the index @xmath175 and write @xmath176 , @xmath177 , @xmath178 and @xmath179 .
further , we define the following trial variables : @xmath180 , @xmath181 and @xmath182 .
then the discrete elastoplastic constitutive problem for the @xmath175-step reads as follows : _ given @xmath147 , @xmath183 and @xmath184 .
find @xmath7 , @xmath185 and @xmath186 satisfying : _
@xmath187 notice that the remaining input parameter for the next step , @xmath188 , can be computed using the formula @xmath189 after finding a solution to problem ( [ k - step_problem ] ) .
we standardly use the elastic predictor / plastic corrector method for solving ( [ k - step_problem ] ) . *
elastic predictor * applies when @xmath190 then the triplet @xmath191 is the solution to ( [ k - step_problem ] ) .
* plastic corrector * applies when ( [ trial_admissibility ] ) does not hold . then @xmath192 and ( [ k - step_problem ] ) reduces into @xmath193 since the functions @xmath17 and @xmath3 depend on @xmath7 only through the variables @xmath50 and @xmath48 , it is natural to reduce a number of uknowns in problem ( [ k - step_problem_plast ] ) . to this end , we split ( [ k - step_problem_plast])@xmath36 into the deviatoric and volumetric parts : @xmath194 where @xmath195 , @xmath196 denotes the deviatoric stress , and the hydrostatic stress related to @xmath183 , respectively . using ( [ inv_der1])@xmath78 , the equality ( [ flow_dev ] ) yields @xmath197 \triangle\lambda g\sqrt{2}\hat{{\mbox{\boldmath$n$}}}&\mbox{if } & \varrho= 0
. \end{array}\right .
\label{flow_s_k}\ ] ] denote @xmath198 and recall that @xmath199 for @xmath56 by ( [ rho_subgrad ] )
. then from ( [ flow_s_k ] ) we obtain @xmath200 following the arguments developed in section 1.1 , we now rewrite ( [ flow_rho ] ) as follows : @xmath201 notice that ( [ flow_rho ] ) and ( [ flow_rho_improve ] ) are equivalent .
further from ( [ flow_s_k])@xmath36 , we standardly have : @xmath202 the following theorem summarizes and completes the derived results .
let @xmath203 .
if @xmath204 is a solution to problem ( [ k - step_problem_plast ] ) and @xmath205 , @xmath206 , @xmath207 , then @xmath208 is a solution to the following system : @xmath209 conversely , if @xmath208 is a solution to ( [ k - step_problem_plast_red ] ) then @xmath204 is the solution to ( [ k - step_problem_plast ] ) where @xmath210 [ lem_auxilliary_problem ] notice that the knowledge of the subdifferential of @xmath50 enables us to formulate the plastic corrector problem as a unique system of nonlinear equations in comparison to the current technique introduced in @xcite .
moreover , one can eliminate the unknowns @xmath211 similarly as for the current return - mapping scheme of this model .
inserting of ( [ k - step_problem_plast_red])@xmath212 into ( [ k - step_problem_plast_red])@xmath213 leads to the nonlinear equation @xmath214 where @xmath215 using ( [ yield_function ] ) .
we have the following solvability result .
let @xmath203 .
then there exists a unique solution , @xmath192 , of the equation @xmath214 .
furthermore , problems ( [ k - step_problem_plast_red ] ) , ( [ k - step_problem_plast ] ) and ( [ k - step_problem ] ) have also unique solutions .
in addition , if @xmath216 then @xmath217 and @xmath218 .
conversely , if @xmath219 @xmath220 then @xmath221 and @xmath56 .
[ th_solvability_dp ] from ( [ q_dp ] ) and the assumptions on @xmath222 , it is readily seen that @xmath223 is a continuous and decreasing function .
further , @xmath224 as @xmath225 and @xmath226 . therefore , the equation @xmath214 has just one solution in @xmath227 .
if @xmath216 then @xmath217 .
otherwise , @xmath221 .
the rest of the proof follows from theorem [ lem_auxilliary_problem ] and the elastic prediction .
the second part of theorem [ th_solvability_dp ] is very useful from the computational point of view : one can a priori decide whether return to the smooth portion of the yield surface happens or not .
this is the main difference in comparison with the current return - mapping scheme .
the improved return - mapping scheme reads as follows .
* return to the smooth portion * 1 .
_ necessary and sufficient condition _ : @xmath228 and @xmath216 .
_ find _ @xmath217 : @xmath229 3 . _
* return to the apex * 1 .
_ necessary and sufficient condition _ : @xmath231 .
find _ @xmath221 : @xmath232 3 .
@xmath233 nonlinear equations ( [ lambda_problem1 ] ) and ( [ lambda_problem2 ] ) can be solved by the newton method .
then it is natural to use the initial choice @xmath234 , @xmath235 for ( [ lambda_problem1 ] ) , and ( [ lambda_problem2 ] ) , respectively . in case of perfect plasticity , @xmath236 , or linear hardening , @xmath237 , @xmath238 , equations ( [ lambda_problem1 ] ) and ( [ lambda_problem2 ] )
are linear and thus @xmath186 can be found in the closed form . solving the problem ( [ k - step_problem ] )
, we obtain a nonlinear and implicit operator between the stress tensor , @xmath239 , and the strain tensor , @xmath240 .
the stress - strain operator , @xmath241 , also depends on @xmath242 and @xmath243 through the trial variables . to emphasize this fact we write @xmath244 . from the results introduced in section [ subsec_const_sol ] , we have @xmath245 , where @xmath246 where @xmath186 is the solution to ( [ lambda_problem1 ] ) , ( [ lambda_problem2 ] ) in ( [ s_def])@xmath28 , and ( [ s_def])@xmath78 , respectively , i.e. @xmath214 . the function @xmath241 is strongly semismooth in @xmath247 with respect to @xmath147 .
[ th_semi_dp ] we use the framework introduced in section [ sub_semi ] . consider the function @xmath248 satisfying @xmath249 if @xmath250 , otherwise @xmath214 . applying theorem [ implicit ] on the implicit function @xmath223
, one can easily find that the function @xmath186 is strongly semismooth .
consequently , the functions @xmath251 and @xmath252 are strongly semismooth . since @xmath253
, we obtain strong semismoothness of the functions @xmath133 and @xmath241 using proposition [ prop_semi ] . notice that @xmath241 is not smooth if @xmath254 or @xmath255 or if @xmath222 has not derivative at @xmath256 .
we introduce the derivative @xmath257 under the assumption that any of these conditions does not hold .
set @xmath258 . using ( [ inv_der1 ] ) , ( [ inv_der2 ] ) ,
( [ elastic_law ] ) and the chain rule , we obtain the following auxilliary derivatives : @xmath259 we distinguish three possible cases : 1 .
let @xmath260 ( _ elastic response _ ) . then clearly , @xmath261 2 .
let @xmath228 and @xmath262 ( _ return to the smooth surface _ ) .
then the derivative of ( [ sigma_smooth ] ) reads @xmath263 applying the implicit function theorem on ( [ lambda_problem1 ] ) , we obtain @xmath264 hence , @xmath265 3 .
let @xmath266 ( _ return to the apex _ ) .
then the derivative of ( [ sigma_apex ] ) yields @xmath267 applying the implicit function theorem on ( [ lambda_problem2 ] ) , we obtain @xmath268 hence , @xmath269 the derivatives ( [ deriv_elast])([deriv_apex ] ) define the consistent tangent operator , @xmath270 .
it is readily seen that the tangent operator is symmetric if @xmath271 , i.e. for the associative plasticity . for purposes of section [ sec_realization ] ,
it is useful to extend the definition of @xmath270 for nondifferential points .
for example , one can write @xmath272 ( \ref{deriv_smooth } ) & \mbox{if}&q_{tr}(0)>0,\;q_{tr}\left({\varrho^{tr}}/{g\sqrt{2}}\right)<0,\\[2pt ] ( \ref{deriv_apex } ) & \mbox{if } & q_{tr}\left({\varrho^{tr}}/{g\sqrt{2}}\right)\geq0 , \end{array } \right.\ ] ] where @xmath273 in ( [ deriv_smooth ] ) , ( [ deriv_apex ] ) is the derivative from left of @xmath222 at @xmath256 .
notice that @xmath274
the jirsek - grassl model was introduced in @xcite .
it is a plastic - damage model proposed for complex modelling of concrete failure .
the model has been further developed .
for example , unteregger and hofstetter @xcite have improved a hardening law and used the model in rock mechanics . for the sake of simplicity
, we only consider a perfect plastic part of this model to illustrate the suggested idea and improve the implicit return - mapping scheme .
the whole plastic part of the jirsek - grassl model can be included to an abstract model studied in the next section .
the perfect plastic model contains the yield function proposed in @xcite : @xmath275 where @xmath276 is the friction parameter and @xmath277 is the uniaxial compressive strength .
the invariants @xmath48 , @xmath50 and @xmath278 were introduced in section [ sec_preliminaries ] .
notice that the couple @xmath279 defines the apex of the yields surface generated by the function @xmath280 .
the yield surface is not smooth only at this apex .
scheme of the yield surface can be found in @xcite .
further , the following plastic pseudo - potential is considered @xcite : @xmath281 where @xmath282 the subdifferential of @xmath3 consists of the following directions : @xmath283 where @xmath284 is defined by ( [ rho_subgrad ] ) and @xmath285 the @xmath175-step of the incremental constitutive problem received by the implicit euler method reads as follows .
_ given @xmath177 and @xmath286 .
find @xmath239 and @xmath186 satisfying : @xmath287,\quad \hat{{\mbox{\boldmath$n$}}}\in\partial\varrho({\mbox{\boldmath$\sigma$}}),\\[5pt ] \triangle\lambda\geq0,\quad \hat f(p({\mbox{\boldmath$\sigma$}}),\varrho({\mbox{\boldmath$\sigma$}}),\varrho_e({\mbox{\boldmath$\sigma$}}))\leq0,\quad \triangle\lambda\hat f(p({\mbox{\boldmath$\sigma$}}),\varrho({\mbox{\boldmath$\sigma$}}),\varrho_e({\mbox{\boldmath$\sigma$}}))=0 .
\end{array } \right\ } \label{k - step_problem_jg}\ ] ] _
we solve this problem again by the elastic predictor / plastic corrector method . within the plastic correction
, we define the trial variables @xmath195 , @xmath288 , @xmath289 , @xmath290 , @xmath291 , @xmath292 and @xmath293 associated with @xmath183 and obtain the following result .
let @xmath294 .
if @xmath295 is a solution to problem ( [ k - step_problem_jg ] ) and @xmath205 , @xmath206 , @xmath207 , then @xmath296 is a solution to the following system : @xmath297 \varrho=\left[\varrho^{tr}-\triangle\lambda 2g\left(\frac{3\varrho}{\bar f_c^2}+\frac{m_0}{\sqrt{6}\bar f_c}\right)\right]^+,\\[5pt ] \hat f(p,\varrho,\varrho r_e^{tr})=0 .
\end{array } \right\ } \label{k - step_problem_plast_jg}\ ] ] conversely , if @xmath296 is a solution to ( [ k - step_problem_plast_jg ] ) then @xmath295 is the solution to ( [ k - step_problem_jg ] ) where @xmath298 p{\mbox{\boldmath$i$ } } & \mbox{if } & \varrho^{tr}\leq\triangle\lambda 2g\left(\frac{3\varrho}{\bar f_c^2}+\frac{m_0}{\sqrt{6}\bar f_c}\right).\\ \end{array } \right.\ ] ] [ lem_auxilliary_problem_jg ] to prove theorem [ lem_auxilliary_problem_jg ] we use the same technique as in section [ subsec_const_sol ] .
it is based on the splitting the stress tensor on the deviatoric and volumetric parts , and on using linear dependence between @xmath49 and @xmath195 to reduce a number of unknowns .
in particular , we have @xmath299 for @xmath218 .
consequently , we obtain ( [ k - step_problem_plast_jg])@xmath28 , @xmath300 and also @xmath301 for @xmath218 using ( [ theta ] ) . finally , notice that @xmath302 as @xmath303 .
indeed , @xmath304 as @xmath303 and the function @xmath305 is bounded .
analogously to the drucker - prager model , one can analyze existence and uniqueness of a solution to problem ( [ k - step_problem_plast_jg ] ) , and a priori decide whether the return to the smooth portion of the yield surface happens or not . to this end , we define implicit functions @xmath306 and @xmath307 such that @xmath308^+=0,\ ] ] respectively , for any @xmath309 .
the following lemma is a consequence of the implicit function theorem .
the functions @xmath310 and @xmath311 are well - defined in @xmath227 .
further , @xmath310 is smooth and decreasing in @xmath227 , @xmath311 is decreasing in the interval @xmath312 and its closed form reads as follows : @xmath313 [ lem_jg ] now , consider the function @xmath314 , @xmath315 let @xmath294 .
then there exists a unique solution , @xmath192 , of the equation @xmath214 .
furthermore , problems ( [ k - step_problem_plast_jg ] ) and ( [ k - step_problem_jg ] ) have also unique solutions .
in addition , if @xmath316 then @xmath317 and @xmath218 .
conversely , if @xmath318 then @xmath319 and @xmath56 .
[ th_solvability_jg ] since @xmath320 and @xmath321 in @xmath312 , the functions @xmath311 , @xmath322 are decreasing in this interval . for @xmath323 ,
these functions vanish .
therefore , from ( [ q_jg ] ) and lemma [ lem_jg ] , it is follows that @xmath223 is a continuous and decreasing function in @xmath227 .
furthermore , @xmath224 as @xmath225 and @xmath324 .
hence , the equation @xmath214 has a unique solution in @xmath227 . if @xmath316 then @xmath317
otherwise , @xmath319 .
the rest of the proof follows from theorem [ lem_auxilliary_problem_jg ] and the elastic prediction .
although the function @xmath223 is implicit the decision criterion introduced in theorem [ th_solvability_jg ] can be found in closed form .
@xmath325 since @xmath326 , @xmath327 hence , @xmath328 . using the definitions of @xmath310 and @xmath329 , we have @xmath330 by theorem [ lem_auxilliary_problem_jg ] and theorem [ th_solvability_jg ] , the return - mapping scheme reads as follows
. * return to the smooth portion * 1 .
_ necessary and sufficient condition _ : @xmath228 and @xmath316 .
2 . _ find _ @xmath331 , @xmath218 and @xmath317 : @xmath332 \varrho+\triangle\lambda 2g\left(\frac{3\varrho}{\bar f_c^2}+\frac{m_0}{\sqrt{6}\bar f_c}\right)-\varrho^{tr}=0,\\[5pt ] \frac{3}{2}\left(\frac{\varrho}{\bar f_c}\right)^2+m_0\left(\frac{\varrho r_e^{tr}}{\sqrt{6}\bar f_c}+\frac{p}{\bar f_c}\right)-1=0 .
\end{array } \right\ } \label{lambda_problem1_jg}\ ] ] 3 .
* return to the apex * 1 .
_ necessary and sufficient condition _ : @xmath318 .
set _ @xmath334
the system ( [ lambda_problem1_jg ] ) of nonlinear equations can be solved by the newton method with the initial choice @xmath335 , @xmath336 , @xmath234 .
it was shown that the system has a unique solution subject to @xmath228 and @xmath316 . without these conditions
, one can not guarantee existence and uniqueness of the solution to ( [ lambda_problem1_jg ] ) .
solving the problem ( [ k - step_problem_jg ] ) , we obtain a nonlinear and implicit operator between the stress tensor , @xmath239 , and the strain tensor , @xmath240 .
the stress - strain operator , @xmath241 , also depends on @xmath242 through the trial stress . to emphasize this fact we write @xmath337 .
we have @xmath338 where the function @xmath223 is defined by ( [ q_jg ] ) and @xmath48 , @xmath50 are components of the solution to ( [ lambda_problem1_jg ] ) .
the function @xmath241 is strongly semismooth in @xmath247 with respect to @xmath147 .
[ th_semi_jg ] consider the function @xmath339 satisfying @xmath249 if @xmath250 , otherwise @xmath214 . to apply theorem [ implicit ] on the implicit function @xmath314 ,
it is necessary to show that @xmath340 is strongly semismooth w.r.t .
the variables @xmath341 .
this follows from ( [ hat_varrho ] ) and proposition [ prop_semi ] .
the rest of the proof coincides with the proof of theorem [ th_semi_dp ] . if @xmath254 or @xmath342 then @xmath241 is not smooth .
we derive the derivative @xmath257 under the assumption that any of these conditions does not hold . if @xmath260 ( elastic response ) then @xmath343 . if @xmath344 ( return to the apex ) then @xmath345 ( vanishes ) .
let @xmath228 and @xmath316 , i.e. , return to the smooth portion happens .
then the derivative @xmath257 can be found as follows . 1 .
find the solution @xmath296 to ( [ lambda_problem1_jg ] ) .
2 . use ( [ inv_der1 ] ) , ( [ inv_der2 ] ) , ( [ elastic_law ] ) and the chain rule and compute : @xmath346 @xmath347,\quad \frac{\partial r_e^{tr}}{\partial{\mbox{\boldmath$\varepsilon$}}}=-r_e'(\cos\theta^{tr})\sin\theta^{tr}\frac{\partial \theta^{tr}}{\partial{\mbox{\boldmath$\varepsilon$}}}.\ ] ] 3 . compute : @xmath348 \frac{\partial \varrho}{\partial{\mbox{\boldmath$\varepsilon$}}}\\[5pt ] \frac{\partial \triangle\lambda}{\partial{\mbox{\boldmath$\varepsilon$ } } } \end{array } \right)=\left ( \begin{array}{c c c } 1+\triangle\lambda k\frac{m_g''(p)}{\bar f_c } & 0 & k\frac{m_g'(p)}{\bar f_c}\\ 0 & 1+\triangle\lambda \frac{6g}{\bar f_c^2 } & 2g\left(\frac{3\varrho}{\bar f_c^2}+\frac{m_0}{\sqrt{6}\bar f_c}\right)\\ \frac{m_0}{\bar f_c } & \frac{3\varrho}{\bar f_c^2}+\frac{m_0r_e^{tr}}{\sqrt{6}\bar f_c } & 0 \end{array } \right)^{-1}\left(\begin{array}{c } \frac{\partial p^{tr}}{\partial{\mbox{\boldmath$\varepsilon$}}}\\[5pt ] \frac{\partial \varrho^{tr}}{\partial{\mbox{\boldmath$\varepsilon$}}}\\[5pt ] -\frac{m_0\varrho}{\sqrt{6}\bar f_c}\frac{\partial r_e^{tr}}{\partial{\mbox{\boldmath$\varepsilon$ } } } \end{array}\right ) .
\label{linearized_system}\ ] ] notice that the matrix in ( [ linearized_system ] ) arises from linearization of ( [ lambda_problem1_jg ] ) around the solution @xmath296 .
the matrix is invertible since its determinant is negative .
4 . compute @xmath349 for numerical purposes , we use the following generalized consistent tangent operator : @xmath350 ( \ref{deriv_smooth_jg } ) & \mbox{if}&q_{tr}(0)>0,\;q_{tr}\left(\sqrt{6}\bar f_c\varrho^{tr}/2gm_0\right)<0,\\[2pt ] \mathbb o & \mbox{if } & \qquad \qquad \quad q_{tr}\left(\sqrt{6}\bar f_c\varrho^{tr}/2gm_0\right)\geq0 . \end{array } \right.\ ] ]
the aim of this section is an extension of theorem [ lem_auxilliary_problem ] and [ lem_auxilliary_problem_jg ] on a specific class of elastoplastic models that are usually formulated in the haigh - westergaard coordinates .
we consider an abstract model containing the isotropic hardening and the plastic flow pseudo - potential .
_ given the history of the strain tensor @xmath168 , @xmath169 $ ] , and the initial values @xmath351 find the generalized stress @xmath352 and the generalized strain @xmath353 such that @xmath354 hold for each instant @xmath173$]_. further , we have the following assumptions on ingredients of the model : 1 .
@xmath355 where @xmath280 is increasing with respect to @xmath50 and @xmath356 , convex and continuously differentiable at least in vicinity of the yield surface .
2 . @xmath357 where @xmath358 is an increasing function with respect to @xmath50 , convex and twice continuously differentiable at least in vicinity of the yield surface .
3 . @xmath222 is a nondecreasing , continuous and strongly semismooth function satisfying @xmath155 .
4 . @xmath359 is a positive function .
invariants @xmath48 , @xmath50 , @xmath131 and @xmath356 are the same as in section [ sec_preliminaries ] .
notice that the assumptions on @xmath280 and @xmath358 guarantee convexity of @xmath17 and @xmath3 using properties of @xmath62 introduced in @xcite .
let @xmath360 then one can write the plastic flow rule as follows : @xmath361,\quad \hat{{\mbox{\boldmath$n$}}}\in\partial\varrho({\mbox{\boldmath$\sigma$ } } ) .
\label{flow_rule2_abstract}\ ] ] the @xmath175-th step of the incremental constitutive problem received by the implicit euler method reads as follows . _ given @xmath177 , @xmath286 and @xmath180 .
find @xmath239 , @xmath362 and @xmath186 satisfying : @xmath363,\quad \hat{{\mbox{\boldmath$n$}}}\in\partial\varrho({\mbox{\boldmath$\sigma$}}),\\[3pt ] \bar{\varepsilon}^{p}=\bar{\varepsilon}^{p , tr}+\triangle\lambda \ell({\mbox{\boldmath$\sigma$}},\kappa),\\[3pt ] \triangle\lambda\geq0,\quad f({\mbox{\boldmath$\sigma$}},h(\bar{\varepsilon}^{p}))\leq0,\quad \triangle\lambda f({\mbox{\boldmath$\sigma$}},h(\bar{\varepsilon}^{p}))=0 .
\end{array } \right\ } \label{k - step_problem_abstract}\ ] ]
_ if we use the elastic predictor / plastic corrector method then we derive the following straightforward extension of theorem [ lem_auxilliary_problem ] and theorem [ lem_auxilliary_problem_jg ] within the plastic correction .
let @xmath203 .
if @xmath364 is a solution to problem ( [ k - step_problem_abstract ] ) then @xmath365 , @xmath192 , is a solution to the following system : @xmath366 \varrho=\left[\varrho^{tr}-\triangle\lambda2g\hat g_\varrho(p,\varrho)\right]^+,\\[3pt ] \bar{\varepsilon}^{p}=\bar{\varepsilon}^{p , tr}+\triangle\lambda
\hat\ell\left(p,\varrho,\varrho\tilde r(\cos \theta^{tr})\right),\\[3pt ] \hat f\left(p,\varrho,\varrho r_e(\cos \theta^{tr}),h(\bar{\varepsilon}^{p})\right)=0 .
\end{array } \right\ } \label{k - step_problem_abstract_red}\ ] ] conversely , if @xmath367 is a solution to ( [ k - step_problem_abstract_red ] ) then @xmath368 solves ( [ k - step_problem_abstract ] ) , where @xmath369 p{\mbox{\boldmath$i$ } } & \mbox{if } & \varrho=0 . \end{array } \right .
\label{sigma_k_abstract}\ ] ] [ lem_auxilliary_problem_abstract ] notice that it is generally impossible to a priori decide about the type of the return as in the models introduced above . to be in accordance with the current approach introduced e.g. in @xcite
one can split ( [ k - step_problem_abstract_red ] ) into the following two systems : @xmath370 @xmath371 and guess which of these systems provides an admissible solution . beside the blind guessing , the current approach has another drawback : it can happen that ( [ k - step_problem_abstract_red ] ) has a unique solution and mutually one of the systems ( [ f_equality_apex ] ) , ( [ f_equality_smooth ] ) does not have any solution or have more than one solution .
therefore , we recommend to solve ( [ k - step_problem_abstract_red ] ) directly by a nonsmooth version of the newton method with the standard initial choice @xmath335 , @xmath336 , @xmath372 and @xmath234 .
consider an elasto - plastic body occupying a bounded domain @xmath373 with the lipschitz continuous boundary @xmath374 .
it is assumed that @xmath375 , where @xmath376 and @xmath377 are open and disjoint sets . on @xmath376 ,
the homogeneous dirichlet boundary condition is prescribed .
surface tractions of density @xmath378 are applied on @xmath377 and the body is subject to a volume force @xmath379 .
notice that the above defined stress , strain and hardening variables depend on the spatial variable @xmath380 , i.e. @xmath381 , etc .
let @xmath382 ^ 3\ |\;{\mbox{\boldmath$v$ } } = { \mbox{\boldmath$0$}}\ \mbox{on } \gamma_d \right\}$ ] denote the space of kinematically admissible displacements . under the infinitesimal small strain assumption
, we have @xmath383 substitution of the stress - strain operator @xmath241 into the principle of the virtual work leads to the following problem at the @xmath175-th step : @xmath384 where @xmath385 and @xmath386 are the prescribed volume , and surface forces at @xmath387 , respectively . after finding a solution @xmath388 ,
the remaining unknown fields @xmath389 important for the next step can be computed at the level of integration points .
problem @xmath390 can be standardly written as the operator equation in the dual space @xmath391 to @xmath392 : @xmath393 , where @xmath394 since we plan to use the semismooth newton method , we also introduce the operator @xmath395 as follows : @xmath396 to discretize the problem in space we use the finite element method
. then the space @xmath392 is approximated by a finite dimensional one , @xmath397 .
if linear simplicial elements are not used then it is also necessary to consider a suitable numerical quadrature on each element .
let @xmath398 , @xmath399 , @xmath400 denote the approximation of operators @xmath401 , @xmath402 , @xmath403 , respectively , and @xmath404 , @xmath405 , @xmath406 be their algebraic counterparts
. then the discretization of problem @xmath390 leads to the system of nonlinear equations , @xmath407 , and the semismooth newton method reads as follows : 1.2 initialization : @xmath408 find @xmath409 : @xmath410 compute @xmath411 set @xmath412 .
if @xmath241 is strongly semismooth in @xmath87 then @xmath413 is strongly semismooth in @xmath86 .
notice that the strong semismoothness is an essential assumption for local quadratic convergence of this algorithm .
in numerical examples introduced below , we observe local quadratic convergence when the tolerance is sufficiently small .
in particular , we set @xmath414 .
the improved return - mapping schemes in combination with the semismooth newton method have been partially implemented in codes sifel @xcite and matsol @xcite . here , for the sake of simplicity
, we consider the slope stability benchmark ( * ? ? ?
* page 351 ) for the presented models , the drucker - prager ( dp ) and the jirasek - grassl ( jg ) ones .
the benchmark is formulated as a plane strain problem .
we focus on : a ) incremental limit analysis and b ) dependence of loading paths on element types and mesh density . for purposes of such an experiment ,
special matlab codes have been prepared to be transparent .
these experimental codes are available in @xcite together with selected graphical outputs .
the geometry of the body is depicted in figure [ fig.mesh_p1 ] or [ fig.mesh_q2 ] .
the slope height is 10 m and its inclination is @xmath415 . on the bottom , we assume that the body is fixed and , on the left and right sides , zero normal displacements are prescribed . the body is subjected to self - weight .
we set the specific weight @xmath416kn / m@xmath417 with @xmath418 being the mass density and @xmath3 the gravitational acceleration .
such a volume force is multiplied by a scalar factor , @xmath419 .
the loading process starts from @xmath420 .
the gravity load factor , @xmath419 , is then increased gradually until collapse occurs .
the initial increment of the factor is set to 0.1 . to illustrate loading responses we compute settlement at the corner point @xmath8 on the top of the slope depending on @xmath419 .
as in ( * ? ? ?
* page 351 ) , we set @xmath421kpa , @xmath422 , @xmath423 and @xmath424kpa , where @xmath425 denotes the cohesion for the perfect plastic model . hence , @xmath426kpa and @xmath427kpa . in comparison to @xcite , we use the presented models instead of the mohr - coulomb one . the remaining parameters for these models will be introduced below .
we analyze the problem for linear triangular ( @xmath428 ) elements and eight - pointed quadrilateral ( @xmath429 ) elements . in the latter case ,
@xmath430-point gauss quadrature is used . for each element type
, a hierarchy of four meshes with different densities is considered .
the @xmath428-meshes contain 3210 , 12298 , 48126 , and 190121 nodal points , respectively .
the @xmath429-meshes consist of 627 , 2405 , 9417 , and 37265 nodal points , respectively .
the coarsest meshes for @xmath428 and @xmath429 elements are depicted in figure [ fig.mesh_p1 ] and [ fig.mesh_q2 ] .
let us complete that the mesh in figure [ fig.mesh_p1 ] is uniform in vicinity of the slope and consists of right isoscales triangles with the same diagonal orientation .
further , it is worth mentioning that the @xmath428-meshes are chosen much more finer in vicinity of the slope than their @xmath429-counterparts within the same level .
the drucker - prager parameters @xmath431 , @xmath432 and @xmath433 are computed from the friction angle , @xmath434 , and the dilatancy angle , @xmath435 , as follows @xcite : @xmath436 at first , we introduce results obtained for the model with associative perfect plasticity . in such a case , @xmath437 , @xmath438 and @xmath439kpa .
the received loading curves for the investigated meshes and elements are depicted in figure [ fig.load_path_h_p1 ] and [ fig.load_path_h_q2 ] .
although @xmath428-meshes are much finer , we observe more significant dependence of the curves on the mesh density for @xmath428-elements than for @xmath429-elements .
also computed limit load factors are greater and tend more slowly to a certain value as @xmath440 for @xmath428-meshes than for @xmath429-meshes .
the expected limit value is 4.045 as follows from considerations introduced in @xcite . using the finest @xmath428 and @xmath429 meshes
, we receive the values 4.230 , and 4.056 , respectively . in general , higher order elements are recommended when a locking effect is expected . in this example
, it can be caused due to the presence of the limit load and/or @xmath441 .
on the other hand , the strong dependence on mesh density is influenced by other factors like mesh structure or choice of a model .
for example , this dependence is not so significant for the jirasek - grassl model ( see the next subsection ) .
further , in @xcite , there is theoretically justified and illustrated that the dependence of the limit load on the mesh density is minimal for bounded yield surfaces and that an approximation of unbounded yield surfaces by bounded ones ( the truncation ) leads to a lower bound of the limit load . for illustration , we add figure [ fig.multiplier_perf_plas ] and [ fig.displacement_perf_plas ] with plastic multipliers and total displacements at collapse , respectively .
the figures are in accordance with literature .
to compare the current return - mapping scheme with improved one , we have also considered the nonassociative model with nonlinear hardening where @xmath442 here , @xmath443 represents the initial slope of the hardening function and the material response is perfect plastic for sufficiently large values of the hardening variable .
this nonassociative model yields a slightly lower values of the limit load factors and also the other results are very similar to the associative model .
the related graphical outputs are available in ( * ? ? ?
* ss - dp - nh ) .
further , in vicinity of the limit load , we have observed lower rounding errors for the improved return - mapping scheme and thus lower number of newton steps is necessary to receive the prescribe tolerance than for the current scheme
. however , the computational time for both schemes are practically the same since return to the apex happens only on a few elements lying in vicinity of the yield surface . to be the simplified jirasek - grassl ( jg ) model applicable for the investigated soil material we fit its parameters using the associative perfect plastic drucker - prager ( dp ) model as follows : @xmath69 , @xmath444 , @xmath445 , @xmath446 , @xmath447 , @xmath448 and @xmath449 .
recall that @xmath69 implies @xmath67 .
further the value of @xmath277 corresponds to the uniaxial compressive strength computed from the drucker - prager model . to eliminate the influence of the exponential term in the function @xmath329
, the value of @xmath450 is chosen sufficiently large .
then the model is insensitive on @xmath451 and one can vanish it .
finally , we require the same flow direction for both the models under the uniaxial compressive strength .
since the yield function in the jg model is normalized in comparison to the dp model it is convenient to introduce the following relation between the plastic multipliers : @xmath452 , where @xmath453 is a scale factor .
then the values of @xmath276 and @xmath454 are determined from the following equations : @xmath455 to be @xmath276 positive , @xmath453 must be greater than @xmath456 . to be in accordance with results of the dp model , we set @xmath457 . , the limit load factor is underestimated and for greater values of @xmath453 the limit load factor is overestimated .
] comparison of yield surfaces ( in the meridean plane ) and flow directions for the dp and jg models is illustrated in figure [ fig_model_comparison ] . here , the fixed value @xmath458 is used for vectors representing the flow directions .
loading curves for the investigated @xmath428 , @xmath429 meshes and the jg model are depicted in figure [ fig.load_path_h_jg ] and [ fig.load_path_h_q2_jg ] .
we observe much faster convergence of the @xmath428-loading curves than for the dp model .
moreover , the results for @xmath428 and @xmath429 elements are comparable .
the computed values of the limit load factor on the finest @xmath428 and @xmath429 meshes are 4.124 , and 4.107 , respectively .
the main idea of this paper is that the subdifferential formulation of the plastic flow rule is also useful for computational purposes and numerical analysis .
namely , it has been shown that such an approach improves the implicit return - mapping scheme for non - smooth plastic pseudo - potentials as follows . *
the unique system of nonlinear equations is solved regardless on a type of the return .
* it can be a priori determined the type of the return from a given trial state for some models ( without knowledge of the solution ) .
* the scheme can be more correct than the current one , and its form enables to study properties of constitutive operators like existence , uniqueness and semismoothness . in this paper ( part i )
, the new technique has been systematically built on a specific class of models containing singularities only along the hydrostatic axis . beside an abstract model , two particular models
have been studied : the drucker - prager and the simplified jirasek - grassl model
. however , the presented idea seems to be more universal .
for example , it has been successfully used for the mohr - coulomb model in `` part ii '' @xcite .
the authors would like to thank to pavel marlek for generating the quadrilateral meshes with midpoints .
this work has been supported by the project 13 - 18652s ( ga cr ) and the european regional development fund in the it4innovations centre of excellence project ( cz.1.05/1.1.00/02.0070 ) .
armero f , prez - foguet a. on the formulation of closest - point projection algorithms in elastoplasticity - part i : the variational structure .
_ international journal for numerical methods in engineering _ 2002 ; * 53 * : 297 - 329 . de angelis f , taylor rl .
an efficient return mapping algorithm for elastoplasticity with exact closed form solution of the local constitutive problem , _ engineering computations _ 2015 ; * 32 * : 2259 - 2291 .
de saxc g. the biponential method , a new variational and numerical treatment of the dissipative laws of materials . in : _ 10th .
int . conf . on mathematical and computer modelling and scientific computing _ ,
pp . 1 - 6 , 1995 . hjiaj m , fortin j , de saxc g. a complete stress update algorithm for the non - associated drucker - prager model including treatment of the apex .
_ international journal of engineering science _
2003 * 41 * : 1109 - 1143 . prez - foguet a , armero f. on the formulation of closest - point projection algorithms in elastoplasticity - part ii : globally convergent schemes . _
international journal for numerical methods in engineering _ 2002 ; * 53 * : 331 - 374
. willam kj , warnke ep .
constitutive model for the triaxial behavior of concrete .
i. : concrete structure subjected to triaxial stresses .
19 of iabse report , international association of bridge and structural engineers _ , zurich , 1974 , pp . 1 - 30 . zouain n. some variational formulations of non - associated hardening plasticity .
_ mechanics of solids in brasil _ 2009 , h.s .
da costa mattos and m. alves eds .
, _ brasilian society of mechanical sciences and engineering _ , pp .
503 - 512 .
isbn 978 - 85 - 85769 - 43 - 7 . | the paper is devoted to the numerical solution of elastoplastic constitutive initial value problems .
an improved form of the implicit return - mapping scheme for nonsmooth yield surfaces is proposed that systematically builds on a subdifferential formulation of the flow rule .
the main advantage of this approach is that the treatment of singular points , such as apices or edges at which the flow direction is multivalued involves only a uniquely defined set of non - linear equations , similarly to smooth yield surfaces .
this paper ( part i ) is focused on isotropic models containing : @xmath0 yield surfaces with one or two apices ( singular points ) laying on the hydrostatic axis ; @xmath1 plastic pseudo - potentials that are independent of the lode angle ; @xmath2 nonlinear isotropic hardening ( optionally ) .
it is shown that for some models the improved integration scheme also enables to a priori decide about a type of the return and investigate existence , uniqueness and semismoothness of discretized constitutive operators in implicit form .
further , the semismooth newton method is introduced to solve incremental boundary - value problems .
the paper also contains numerical examples related to slope stability with available matlab implementation .
keywords : elastoplasticity , nonsmooth yield surface , multivalued flow direction , implicit return - mapping scheme , semismooth newton method , limit analysis |
spin transport electronics ( spintronics ) has recently attracted enormous attention mainly because of its promising applications in information storage and processing and other electronic technologies@xcite .
spin current generation , detection and manipulation are three key issues in the emerging spintronics .
large intrinsic spin hall effect ( she ) in platinum has recently been predicted@xcite and observed ( see refs . and and references therein ) . in the she ,
a transverse spin current is generated in response to an electric field in a metal with relativistic electron interaction .
the she enables us to generate and control spin current without magnetic field or magnetic materials , which would be an important step for spintronics .
furthermore , in the inverse spin hall effect , a transverse voltage drop arises due to the spin current@xcite , and this allows us to detect spin current by measuring the hall voltage .
therefore , platinum has been widely used as a spin current generator and detector in recent spin current experiments , such as spin seebeck effect@xcite , spin pumping@xcite and spin hall switching@xcite , and plays a unique role in recent developments in spintronics .
platinum is an enhanced paramagnet because its 5@xmath8-band is partially filled with a large density of states ( dos ) at the fermi level ( @xmath9 ) [ @xmath101.74 states / ev / spin ] .
consequently , it could become ferromagnetic with a significant spin magnetic moment when placed next to a ferromagnetic metal@xcite or in low - dimensional structures such as an atomic bilayer on silver ( 001 ) surface@xcite or a freestanding atomic chain@xcite .
indeed , platinum was reported to possess a magnetic moment as large as @xmath110.2 and @xmath110.5 @xmath12/atom in ni / pt and fe / pt multilayers@xcite , respectively . in a ferromagnetic metal , a transverse charge current would be generated in response to an electric field due to relativistic spin - orbit coupling ( soc ) , an effect known as the anomalous hall effect ( ahe)@xcite , discovered by hall@xcite long ago .
since the ahe is another archetypal spin - related transport phenomenon@xcite and the soc strength in pt is large , it would be interesting to study the ahe in the proximity - induced ferromagnetic platinum .
furthermore , as pointed out in ref . , the fact that the hall voltage could be generated by both the ahe and inverse she in the magnetized platinum , might complicate the detection of the pure spin current and also related phenomena using platinum . therefore , it is important to understand the transport and magnetic properties of the magentized platinum . in a ferromagnet ,
the hall voltage could also arise when a thermal gradient instead of an electric field , is applied .
this phenomenon , again due to the relativistic soc , is refered to as the anomalous nernst effect ( ane)@xcite .
interestingly , the ane could be used as a probe of the vortex phase in type ii superconductors@xcite and has been receiving considerable attention in recent years.@xcite in this context , it would be interesting to study the ane in the proximity - induced ferromagnetic platinum . on the other hand , spin seebeck effect ( another thermal phenomenon ) , which refers to the generation of a spin - motive force in a ferromagnet by a temperature gradient ,
has recently attracted considerable attention@xcite .
again , this effect is usually measured as a transverse voltage in a nonmagnetic metal such as pt in contact with the ferromagnet via the inverse she@xcite . clearly ,
if the metal is magnetized due to the magnetic proximity effect , the ane would contribute to the measured hall voltage too .
in this connection , it is imperative to understand the ane in the magnetized platinum .
palladium is isoelectronic to platinum and thus has an electronic structure similar to that of pt except a smaller soc strength ( see , e.g. , refs . and and references therein ) .
for example , like pt , pd also has a large intrinsic spin hall conductivity ( shc)@xcite and is a highly enhanced paramagnetic metal with a large @xmath102.69 states / ev / spin .
in fact , palladium possesses the largest paramagnetic susceptibility of @xmath13 emu / mole among the nonmagnetic metals@xcite and is usually considered to be nearly ferromagnetic .
it could become ferromagnetic when placed next to a ferromagnetic metal@xcite or fabricated as an atomic bilayer on silver ( 001 ) surface@xcite or a freestanding atomic chain@xcite .
recently , the ahe was observed in the pd film on an yttrium iron garnet ( yig).@xcite surprisingly , it was reported that the intrinsic anomalous hall conductivity ( ahc ) in the pd film on the yig layer has a sign opposite to that for the pt / yig bilayer.@xcite this indicates that the ahc in a magnetized nonmagnetic metal does not simply scale with the soc strength .
one would then ask what determines the ahc in the magnetized metals . in this paper , therefore , we study the ahe and ane in the proximity - induced ferromagnetic platinum and palladium within the berry phase formalism@xcite based on first - principle relativistic band structure calculations . we also perform analytic calculations to identify possible relations between the shc in an nonmagnetic metal and the ahc in the corresponding magnetized metal .
the rest of this paper is organized as follows . in the next section
, we briefly describe the berry phase formalism for calculating the ahc and anc as well as the computational details . in sec .
iii , the calculated ahc and anc will be presented .
finally , the conclusions drawn from this work will be summarized in sec .
the anomalous hall conductivity and anomalous nernst conductivity ( anc ) are calculated by using the berry - phase formalism@xcite . within this berry - phase formalism , the ahc is simply given as a brillouin zone ( bz ) integration of the berry curvature for all the occupied bands , @xmath14 } { ( \epsilon_{{\bf k}n}-\epsilon_{{\bf k}n'})^2},\end{aligned}\ ] ] where @xmath15 and @xmath16 are the fermi distribution function and the berry curvature for the @xmath17th band at @xmath18 , respectively.@xcite similarly
, the anc can be written as @xmath19,\end{aligned}\ ] ] where @xmath20 is the chemical potential and @xmath21 is the boltzmann constant.@xcite the proximity - induced ferromagnetic platinum and palladium are investigated by the constrained spin - density functional theory with the local density approximation to the exchange - correlation potential.@xcite spin - polarized self - consistent scalar - relativistic electronic structure calculations with the spin magnetic moment fixed to specified values , are performed . using the resultant self - consistent charge densities , the fully relativistic band structures
are then calculated for the ahc and anc calculations .
the highly accurate all - electron full - potential linearized augmented plane wave ( flapw ) method , as implemented in the wien2k code@xcite , is used .
the experimental lattice constants @xmath22 and 3.89 ( ) are used , respectively , for pt and pd . in both cases , the muffin - tin sphere radius ( @xmath23 ) of 2.5 a.u .
is adopted .
the wave function , charge density , and potential are expanded in terms of the spherical harmonics inside the muffin - tin spheres and the cutoff angular moment ( @xmath24 ) used is 10 , 6 and 6 , respectively .
the wave function outside the muffin - tin sphere is expanded in terms of the augmented plane waves ( apws ) and a large number of apws ( about 70 apws per atom , i. e. , the maximum size of the crystal momentum @xmath25 ) are included in the present calculations .
the tetrahedron method is used for the bz integration@xcite . to obtain accurate ground state properties , a fine 21@xmath2621@xmath2621 grid of 11616 @xmath27-points in the first bz
is used . for the ahc and anc calculations , a very find grid of 258156 @xmath27-points on the magnetic irreducible wedge ( 1/16 bz ) in the bz
this is equivalent to a large number of @xmath27-points of @xmath28 in the full bz , and corresponds to the division of the @xmath29 line into 70 intervals .
comparison with test calculations with a denser grid of 381915 @xmath27-points ( 80 divisions of the @xmath29 line ) indicates that the calculated ahc and anc converge to within a few % .
the relativistic band structure and also ahc ( @xmath0 ) as a function of the fermi energy ( @xmath9 ) for the magnetized platinum and palladium with the spin magnetic moment @xmath30 @xmath12/atom are displayed in figs . 1 and 2 , respectively .
all kramer - degenerate bands in nonmagnetic platinum ( see fig . 1 in @xcite ) and palladium ( see fig
. 1 in @xcite ) are now exchange - split due to the induced magnetization in the magnetized pt and pd .
this is clearly visible for the @xmath8-dominated bands [ i.e. , energy bands below 1.0 ev in fig .
1(a ) or 0.5 ev in fig .
2(a ) ] since the ferromagnetism is mainly caused by the exchange interaction among the @xmath8-electrons . the band spin - splittings are largest in the flat bands of almost pure @xmath8 character such as the bands around 0.5 ev ( 0.3 ev ) in the vicinity of the w - point in fig .
1(a ) [ fig .
2(a ) ] .
( color online ) ( a ) relativistic band structure and ( b ) anomalous hall conductivity ( ahc ) of the magnetized platinum with a spin magnetic moment of 0.1 @xmath12/atom .
the horizontal dotted line at the zero energy indicates the fermi level . , width=302 ] ( color online ) ( a ) relativistic band structure and ( b ) anomalous hall conductivity of the magnetized palladium with a spin magnetic moment of 0.1 @xmath12/atom . the horizontal dotted line at the zero energy indicates the fermi level.,width=302 ]
figures 3 and 4 show the calculated ahc and anc ( @xmath1 ) as well as the exchange splitting ( @xmath3 ) as a function of the induced spin magnetic moment ( @xmath7 ) in platinum and palladium , respectivelty .
@xmath3 refers to the splitting of the spin - up and spin - down bands , and we calculate @xmath3 as the spin splitting of the scalar - relativistic bands above the fermi level at the w - point [ figs .
1(a ) and 2(a ) ] .
first of all , it is clear from figs . 3 and 4 that the calculated @xmath0 and @xmath3 increase monotonically with @xmath7 .
in fact , @xmath3 is almost perfectly proportional to @xmath7 , while the amplitude of the @xmath0 increases linearly with @xmath7 for small @xmath7 values up to 0.30 and 0.25 @xmath12/atom for pt and pd , respectively .
secondly , the ahc is large . in particular ,
the magnitude of the ahc per @xmath12 ( @xmath31 ) for @xmath32 @xmath12/atom in pt and pd is , respectively , @xmath11790 and 3500 s/(@xmath33 ) , being much larger than that of @xmath11360 s/(@xmath33 ) in iron@xcite .
interestingly , the ratio @xmath31 for pt is smaller than that for pd , indicating that the ahc in a proximity - induced ferromagnetic metal is not necessarily correlated with the soc strength .
thirdly , the sign of the ahc in pt is opposite to that in pd , being in good agreement with the recent experiments on the pt / yig and pd / yig bilayers@xcite .
( color online ) ( a ) anomalous hall conductivity ( @xmath0 ) and ( b ) anomalous nernst conductivity ( @xmath1 ) as a function of the induced spin magnetic moment ( @xmath7 ) in platinum .
exchange splitting ( @xmath3 ) is displayed as a function of @xmath7 in the inset in ( a ) . in ( b ) , @xmath34 denotes temperature.,width=264 ] ( color online ) ( a ) anomalous hall conductivity ( @xmath0 ) and ( b ) anomalous nernst conductivity ( @xmath1 ) as a function of the induced spin magnetic moment ( @xmath7 ) in palladium .
exchange splitting ( @xmath3 ) is displayed as a function of @xmath7 in the inset in ( a ) . in ( b ) , @xmath34 denotes temperature.,width=264 ] in order to gain insight into the key factors that determine the ahc in a magnetized nonmagnetic metal , let us consider the two - current model to connect the conductivities for the different sorts of hall effects . within the two - current model approximation , @xmath35 and @xmath36 can be written as@xcite @xmath37 and @xmath38 , where @xmath39 and @xmath40 are the spin - up and spin - down hall conductivities , respectively . in a non - magnetic metal ,
the spin magnetic moment @xmath41 and thus , @xmath42 . in the magnetized metal , @xmath43 , where @xmath3 is the exchange splitting and is proportional to @xmath7 , as shown in the inset in fig .
3(a ) and fig .
therefore , we find @xmath44 equation ( 3 ) tells us that the ahc is proportional to the energy derivative of the spin hall conductivity [ @xmath2 ] as well as the exchange splitting ( @xmath3 ) .
interestingly , the soc strength does not appear explicitly in eq .
( 3 ) , in contrary to conventional wisdom .
we notice that platinum and palladium have similar ahc - versus - energy [ @xmath46 curves which have a prominent peak near the @xmath9 ( see fig . 1 in both @xcite and @xcite )
. however , the @xmath9 falls on the up - hill side of the peak in pt@xcite but on the down - hill side of the peak in pd@xcite , resulting in the positive @xmath6 for pt and negative @xmath6 for pd .
this , together with eq .
( 3 ) , naturally explains why both the calculated and observed ahcs in pt and pd have opposite signs .
( color online ) ( a ) spin hall conductivity ( @xmath2 ) as a function of energy in nonmagnetic pt and pd metals .
the vertical dotted line at 0 ev indicates the fermi energy ( @xmath9 ) .
@xmath47 and 1242 ( @xmath48/e)s / cm for pt and pd , respectively . @xmath49 and -4245 ( @xmath48/e)s / cm - ev for pt and pd , respectively .
( b ) spin hall conductivity as a function of the induced spin magnetic moment ( @xmath7 ) in magnetized pt and pd metals.,width=264 ] to examine quantitatively the validity of eq .
( 3 ) , here we repeat the calculations of the shc for pt@xcite and pd@xcite but using the more accurate flapw method with the same computational details as described already in sec .
the calculated shc for pt and pd as a function of energy is displayed in fig .
the @xmath2 at the @xmath9 is 2200 ( @xmath48/e)s / cm for pt and 1242 ( @xmath48/e)s / cm for pd , being in good agreement with the corresponding results calculated previously using the linear muffin - tin orbital method with the atomic sphere approximation.@xcite we then evaluate numerically the energy derivative of the shc using the @xmath50 displayed in fig . 5 .
we obtain @xmath49 and -4245 ( @xmath48/e)s / cm - ev for pt and pd , respectively .
figures 3(a ) and 4(a ) also show the @xmath0 evaluated using eq .
( 3 ) together with the calculated @xmath6 and @xmath3 .
it is clear that eq .
( 3 ) holds very well for small @xmath7 up to @xmath110.25 @xmath12/atom for pt and pd [ figs .
3(a ) and 4(a ) ] .
we have also calculated the shc in the magnetized pt and pd metals .
the calculated shc for pt and pd is shown as a function of the spin magnetic moment in fig .
in both pt and pd , the shc initially increases with @xmath7 up to @xmath110.1 @xmath12/atom and then decreases slowly as @xmath7 further increases [ fig .
nevertheless , the shc for both pt and pd remains in the same order of magnitude all the way up to @xmath7 = 0.5 @xmath12/atom .
the validity of eq .
( 3 ) may be understood at the microscopic level .
the two - current model can be derived from an approximation in which the spin - flipping part of the soc is ignored .
the spin - conserving part of the soc can still lead to nontrivial results on the transverse transport coefficients .
this non - flip approximation can be justified for crystals with inversion symmetry and in the limit of zero magnetization .
this is because that kramer s theorem implies a two - fold degeneracy of the band structure at general @xmath27-points even in the presence of the soc .
the soc term in the hamiltonian , being symmetric under spatial inversion and time reversal , must behave as a constant within the degenerate space .
therefore , it must also commute with the representation of the spin operator within the two - fold degenerate space .
in the presence of a small magnetization , the degenerate bands are split to first order in the zeeman energy according to the representation of the spin operator within each of the original degenerate space .
not being able to mix these split levels directly , the spin - flip part of the soc term in the hamiltonian can be safely discarded , because its residual effect must be of second order ( in a process going out and back to the degenerate space ) .
figures 3(b ) and 4(b ) indicate firstly that the anomalous nernst conductivity @xmath1 increases monotonically with the spin moment @xmath7 for @xmath7 up to at least 0.5 @xmath12/atom in pt and for @xmath7 up to 0.25 @xmath12/atom in pd . like @xmath0 , @xmath1 is approximately proportional to @xmath7 for @xmath51 @xmath12/atom in both pt and pd .
secondly , the calculated @xmath1 is large , especially in pt [ fig .
3(b ) ] . in fact , @xmath1 for pt at @xmath52 @xmath12/atom could be ten times larger than the intrinsic @xmath1 [ @xmath1/t = @xmath53 a/(m - k@xmath54 ) at @xmath55 k ] of iron@xcite .
the magnitude of @xmath1 for pd for @xmath56 @xmath12/atom is also several times larger than that of iron@xcite . at low temperatures , eq .
( 2 ) can be simplified as the mott relation , @xmath57 which relates the anc to the ahc .
therefore , it is not surprising that the magnetized platinum has a very large @xmath1 since the @xmath9 is located on the steep slope of @xmath58 [ fig .
1(b ) ] , resulting in a large energy derivative of @xmath58 at @xmath9 . in figs .
1(b ) and 2(b ) , the @xmath1 calculated using the mott relation [ eq .
( 4 ) ] is displayed as a function of the induced @xmath7 .
clearly , @xmath1 calculated directly [ eq . ( 2 ) ] and using the mott relation at @xmath59 k are in good agreement with each other for pt [ see fig .
3(b ) ] and also for @xmath60 @xmath12/atom for pd [ see fig .
4(b ) ] . on the other hand , @xmath1 at @xmath61 k
calculated directly differs noticeably from that from the mott relation , indicating that @xmath61 k can not be considered as a low temperature in this context .
differentiating eq .
( 3 ) and substituting the result into eq . ( 4 ) , we find @xmath62 eqs .
( 3 ) and ( 4 ) indicate that for small @xmath7 , both the ahc and anc are proportional to the exchange - splitting . as mentioned above
, the exchange - splitting is almost a perfect linear function of @xmath7 , and hence this explains why both the @xmath0 and @xmath1 are approximately proportional to @xmath7 .
furthermore , this suggests that the @xmath0 and @xmath1 are proportional to each other for small @xmath7 , as shown in fig . 3 and fig .
4 . therefore , we can rewrite eq .
( 3 ) as @xmath63m_s = \beta m_s,\ ] ] where constant @xmath64 can be determined solely by first - principle calculations for a certain spin moment @xmath65 . in the present work
, we find that @xmath66 s / cm/@xmath12 for pt and @xmath67 s / cm/@xmath12 for pd .
similarly , we can rewrite eq .
( 4 ) as @xmath68m_s = \gamma m_s.\ ] ] and using the results of the first - principle calculations , we obtain that constant @xmath69 a/(m - k@xmath70 ) for pt and @xmath71 a/(m - k@xmath70 ) for pd .
recently , the possible magnetic proximity - induced spin moment in pt films in the pt / yig bilayers was measured by magnetic x - ray circular dichroism experiments@xcite , and @xmath7 was found to be 0.054 @xmath12/atom at 300 k and 0.076 @xmath12/atom at 20 k. using @xmath72 @xmath12/atom together with eqs .
( 6 ) and ( 7 ) , we can estimate the intrinsic ahc and anc for the pt film to be @xmath0 = 40 s / cm and @xmath1 = 0.51 a/(m - k@xmath54 ) ( @xmath61 k ) .
the anomalous seebeck coefficient @xmath73 where @xmath74 is the ordinary seebeck coefficient . at @xmath61
@xmath75 @xmath76 cm and @xmath77 @xmath20v / k ( see refs . and ) .
resultantly , @xmath78v / k . using the sample sizes and the temperature gradient in the pt / yig bilayers@xcite , one
would obtain the hall voltage due to the ane in the order of @xmath110.1 @xmath20v , being comparable with the hall voltage ( @xmath110.1 @xmath20v in au / yig and @xmath111.0 @xmath20v in pt / yig ) produced by the spin seebeck effect via the inverse spin hall effect .
g.y.g . acknowledges support from the ministry of science and technology , the academia sinica thematic research program and ncts of taiwan , and thanks shang - fan lee and shiming zhou for stimulating discussions .
g.y.g . also acknowledges partial support from nbrpc ( no .
2012cb921300 and no .
2013cb921900 ) , and nsfc ( no .
91121004 ) during his visit at peking university .
q.n . was supported in part by doe - dmse ( no .
de - fg03 - 02er45958 ) and the welch foundation ( no .
f-1255 ) .
g. a. prinz , science * 282 * , 1660 ( 1998 ) ; s. a. wolf , d. d. awschalom , r. chtchelkanova , and d. m. treger , _ ibid .
_ , * 294 * , 1488 ( 2001 ) i. zutic , j. fabianm and s. d. sarma , rev .
* 76 * , 323 ( 2004 ) .
g. y. guo , s. murakami , t .- w .
chen , and n. nagaosa , phys .
lett . * 100 * , 096401 ( 2008 ) .
z. feng , j. hu , l. sun , b. you , d. wu , j. du , w. zhang , a. hu , y. yang , d. m. tang , b. s. zhang , and h. f. ding , phys .
b * 85 * , 214423 ( 2012 ) .
a. hoffman , ieee trans . mag . * 49 * , 5172 ( 2013 ) .
s. o. valenzuela , m. tinkham , nature * 442 * , 176 ( 2006 ) .
e. saitoh , m. ueda , h. miyajima , and g. tatara , appl .
* 88 * , 182509 ( 2006 ) .
k. uchida , s. takahashi , k. harii , j. ieda , w. koshibae , k. ando , s. maekawa , and e. saitoh , nature * 455 * , 778 ( 2008 ) .
y. kajiwara , k. harii , s. takahashi , j. ohe , k. uchida , m. mizuguchi , h. umezawa , h. kawai , k. ando , k. takanashi , s. maekawa , and e. saitoh , nature * 464 * , 262 ( 2010 ) . i. m. miron , k. garello , g. gaudin , p .- j .
zermatten , m. v. costache , s. auffret , s. bandiera , b. rodmacq , a. schuhl and p. gambardella , nature * 476 * , 189 ( 2011 ) .
wilhelm , phys .
lett . * 85 * , 413 ( 2000 ) .
antel , phys .
b * 60 * , 12933 ( 1999 ) .
s. blgel , phys .
b * 51 * , 2025 ( 1995 ) .
a. smogunov , a. dal corso , a. delin , r. weht , and e. tosatti , nat .
nanotechnol . * 3 * , 22 ( 2008 ) .
tung and g. y. guo , phys .
* b * 81 , 094422 ( 2010 ) .
n. nagaosa , j. sinova , s. onoda , a. h. macdonald , and n. p. ong , rev .
phys . * 82 * , 1539 ( 2010 ) e. h. hall , phil . mag . * b * 12 , 157 ( 1881 )
. s. y. huang , x. fan , d. qu , y. p. chen , w. g. wang , j. wu , t. y. chen , j. q. xiao , and c. l. chien , phys .
lett . * 109 * , 107204 ( 2012 ) .
w. nernst , ann . phys . * 267 * , 760 ( 1887 ) . z. a. xu , n. p. ong , y. wang , t. kakeshita , and s. uchida , nature * 406 * , 486 ( 2000 ) . w .-
lee , s. watauchi , v. l. miller , r. j. cava , and n. p. ong , phys . rev . lett . *
93 * , 226601 ( 2004 ) .
d. xiao , y. yao , z. fang and q. niu , phys .
* 97 * , 026603 ( 2006 ) .
t. miyasato , n. abe , t. fujii , a. asamitsu , s. onoda , y. onose , n. nagaosa , and y. tokura , phys .
lett . * 99 * , 086602 ( 2007 ) .
d. l. bergman and v. oganesyan , phys .
rev . lett . * 104 * , 066601 ( 2010 ) .
t. yokoyama and s. murakami , phys .
b * 837 * , 161407 ( 2011 ) .
s. y. huang , w. g. wang , s. f. lee , j. kwo , and c. l. chien , phys .
lett . * 107 * , 216604 ( 2011 ) . j. weischenberg , f. freimuth , s. blgel , and y. mokrousov , phys .
b * 87 * , 060406 ( 2013 ) . m. schmid , s. srichandan , d. meier , t. kuschel , j .-
schmalhorst , m. vogel , g. reiss , c. strunk , and c. h. back , phys .
* 111 * , 187201 ( 2013 ) .
k. uchida , nature mater .
* 9 * , 894 ( 2010 ) . c. m. jaworski , j. yang , s. mack , d. d. awschalom , j. p. heremans , and r. c. myers , nature mater .
* 9 * , 898 ( 2010 ) .
g. y. guo , j. appl .
* 105 * , 07c701 ( 2009 ) .
r. huguenin , g. p. pells , and d. n. baldock , j. phys .
f : metal phys . * 1 * , 281 ( 1971 ) .
z. celinski , _ et al .
lett . * 65 * , 1156 ( 1990 ) .
j. r. childress , _ et al .
_ , j. magn .
magn . mater .
* 130 * , l25 ( 1994 ) .
x. zhou , l. ma , z. shi , and s. m. zhou , arxiv:1309.4841 ( 2013 ) .
d. xiao , m .- c .
chang , and q. niu , rev .
82 , 1959 ( 2010 ) y. yao _
et al . _ ,
* 92*,037204 ( 2004 ) .
vosko , l. wilk , and m. nusair , can .
* 58 * , 1200 ( 1980 ) .
p. blaha , k. schwarz , g. madsen , d.kvasnicka , and j. luitz , wien2k , an augmented plane wave local orbitals program for calculating crystal properties ( technische university wien , austria , 2002 ) o. jepson and o. k. anderson , solid state commun .
* 9 * , 1763 ( 1971 ) j .- c .
tung , h .-
fuh , and g. y. guo , phys .
b * 86 * , 024435 ( 2012 ) . j .- c .
tung , and g. y. guo , new j. phys . * 15 * , 033014 ( 2013 ) .
note that in refs . and , @xmath79 in eq .
( 5 ) and eq .
( 2b ) , respectively , should be @xmath80 .
consequently , the signs of the spin - up and spin - down hall conductivities ( @xmath39 and @xmath40 ) as well as the spin polarization of the hall current ( @xmath81 ) listed in tables i and ii in refs . and should be reversed . y. m. lu , y. choi , c. m. ortega , x. m. cheng , j. w. cai , s. y. huang , l. sun , and c. l. chien , phys .
* 111*,147207 ( 2013 ) .
landolt - brnstein , new series , volume 15 , metals : electronic transport phenomena , subvolume a by j. bass and k. h. fischer ( springer - verlag , berlin , 1982 ) .
landolt - brnstein , new series , volume 15 , metals : electronic transport phenomena , subvolume b by j. bass , j. s. dugdale , c. l. foiles and a. myers ( springer - verlag , berlin , 1982 ) .
d. qu , s. y. huang , j. hu , r. wu , and c. l. chien , phys .
lett . * 110*,067206 ( 2013 ) .
t. kikkawa , k. uchida , y. shiomi , z. qiu , d. hou , d. tian , h. nakayama , x .- f .
jin , and e. saitoh , phys .
lett . * 110*,067207 ( 2013 ) . | we study the anomalous nernst effect ( ane ) and anomalous hall effect ( ahe ) in proximity - induced ferromagnetic palladium and platinum which is widely used in spintronics , within the berry phase formalism based on the relativistic band structure calculations .
we find that both the anomalous hall ( @xmath0 ) and nernst ( @xmath1 ) conductivities can be related to the spin hall conductivity ( @xmath2 ) and band exchange - splitting ( @xmath3 ) by relations @xmath4 and @xmath5 , respectively .
in particular , these relations would predict that the @xmath0 in the magnetized pt ( pd ) would be positive ( negative ) since the @xmath6 is positive ( negative ) .
furthermore , both @xmath0 and @xmath1 are approximately proportional to the induced spin magnetic moment ( @xmath7 ) because the @xmath3 is a linear function of @xmath7 . using the reported @xmath7 in the magnetized pt and pd
, we predict that the intrinsic anomalous nernst conductivity ( anc ) in the magnetic platinum and palladium would be gigantic , being up to ten times larger than , e.g. , iron , while the intrinsic anomalous hall conductivity ( ahc ) would also be significant . |
ever since the discovery of aggregation of cyanine dye molecules by jelley @xcite and scheibe @xcite the width of the absorption band of these linear aggregates ( the j - band ) has attracted much attention . at low temperatures , the j - band may be as narrow as a few tens of @xmath3 s ( for pseudoisocyanine ) , while at room temperature it typically is a few hundred @xmath3.@xcite the small width at low temperature is generally understood as resulting from the excitonic nature of the optical excitations,@xcite which leads to exchange narrowing of the inhomogeneous broadening of the transitions of individual molecules.@xcite as the optically dominant exciton states in inhomogeneous j - aggregates occur below the exciton band edge , their vibration - induced dephasing is strongly suppressed at low temperatures .
thus , the j - band is inhomogeneously broadened , except for a small residual homogeneous component ( @xmath4 0.1 @xmath3 ) caused by spontaneous emission of the individual exciton states underlying the spectrum . upon increasing the temperature ,
the vibration - induced dephasing of the exciton states increases , the j - band broadens , and obtains a more homogeneous character . over the past twenty years
, the temperature dependence of the j - band width and the homogeneous broadening of the exciton states , has been studied by several authors . in 1987 , de boer , vink , and wiersma @xcite performed accumulated - echo experiments to study the temperature dependence of the pure dephasing time in j - aggregates of pseudo - isocyanine bromide ( pic - br ) and found that within the temperature range 1.5 k to 100 k , this time is linearly proportional to the occupation number of a mode with an energy of 9 @xmath3 .
they assigned this to a librational mode of the aggregate .
later on , fidder , knoester , and wiersma @xcite ( see also ref . and ) showed that a similar study carried our over a wider temperature range , 1.5 k to 190 k , required the occupation numbers of three vibration modes , at 9 @xmath3 , 305 @xmath3 , and 973 @xmath3 . using the hole burning technique , hirschmann and friedrich
@xcite studied the homogeneous width of the exciton states in pseudoisocyanine iodide ( pic - i ) over the temperature range 350 mk to 80 k. they were able to fit their measurements by a superposition of two exponentials , with activation energies 27 @xmath3 and 330 @xmath3 , and they attributed the broadening to scattering of the excitons on an acoustic mode and an optical mode , respectively , of the aggregate .
finally , in 1997 renge and wild @xcite measured the temperature dependent width , @xmath5 , of the total j - band of pseudoisocyanine chloride ( pic - cl ) and fluoride ( pic - f ) over the wide temperature range 10 k to 300 k. they found that over this entire range their data closely obeyed a power - law dependence @xmath6 with the exponent @xmath7 .
they suggested the scattering of the excitons on host vibrations as a possible source of this behavior .
the above overview clearly demonstrates that the temperature dependence of the most important characteristic of j - aggregates , namely the j - band , is not understood .
it even is not clear what is the source of thermal broadening : dephasing due to vibrational modes of the aggregate itself or due to modes of the host matrix . theoretically , the study of the exciton dephasing in j - aggregates is complicated by the fact that static disorder plays an important role in these systems , as is clear from the strongly asymmetric low - temperature lineshape
. @xcite the simultaneous treatment of disorder , leading to exciton localization , and scattering on vibrational modes , is a problem that requires extensive numerical simulations .
such simulations have been used previously to model the optical response of polysilanes , @xcite the temperature dependent fluorescence @xcite and transport @xcite properties of j - aggregates , and single molecular spectroscopy of circular aggregates .
@xcite alternatively , stochastic models have been applied to describe the scattering of excitons on vibrations in disordered aggregates and the resulting spectral line shape @xcite and relaxation dynamics .
@xcite in this paper , we report on a systematic theoretical study of dephasing of weakly localized frenkel excitons in one - dimensional systems , focusing on the effect of the scattering of the excitons on the vibrational modes ( phonons ) of the host . for a chain - like configuration
it seems physically reasonable to assume that the coupling to host vibrations dominates the dynamics of the excitons . for coupling to vibrations in the chain ,
one expects self - trapped exciton states , @xcite for which in most aggregates no clear signature is found .
we describe the phonons by a debye model and consider one - phonon as well as ( elastic and inelastic ) two - phonon contributions to the dephasing .
the scattering rates between the various , numerically obtained , exciton states are derived using the fermi golden rule . due to the disorder , the dephasing rates of individual exciton states are distributed over a wide range , in particular at low temperature , making it meaningless to associate the width of individual exciton levels with the j - band width .
rather , this width is determined by direct simulation of the total absorption spectrum and is found to scale with temperature according to a power law .
the outline of this paper is as follows : in sec .
[ model ] we introduce the hamiltonian for excitons in a disordered chain and coupled to host vibrations .
general expressions for the exciton dephasing rates are derived in sec .
[ dephasing ] .
section [ homogeneouslimit ] deals with the temperature dependence of the dephasing rates for the disorder - free case , where analytical expressions can be obtained . in sec .
[ numerics ] we give the results of our numerical simulations for the dephasing rates in the presence of disorder and analyze their temperature dependence and fluctuations as well as the total j - band width .
we compare to experiment in sec .
[ experiment ] and discuss an alternative mechanism of dephasing due to the coupling of excitons to a local vibration .
in section [ concl ] we present our conclusions .
we consider an ensemble of j - aggregates embedded in a disordered host matrix .
the aggregates are assumed to be decoupled from each other , while they interact with the host .
a single aggregate is modeled as an open linear chain of @xmath8 coupled two - level monomers with parallel transition dipoles .
the interaction between a particular monomer and the surrounding host molecules in the equilibrium configuration leads to shifts in the monomer s transition energy . due to the host s structural disorder
, this shift is different for each monomer in the aggregate , giving rise to on - site ( diagonal ) disorder .
moreover , vibrations of the host couple to the aggregate excited states , because the associated displacements away from the equilibrium configuration dynamically affect the monomer transition energies .
accounting for these shifts up to second order in the molecular displacements , the resulting hamiltonian in the site representation reads : [ h ] @xmath9 with @xmath10 @xmath11 @xmath12 @xmath13 here , @xmath14 is the bare frenkel exciton hamiltonian , with @xmath15 denoting the state in which the @xmath16th monomer is excited and all the other monomers are in the ground state .
the monomer excitation energies , @xmath17 , are uncorrelated stochastic gaussian variables , with mean @xmath18 and standard deviation @xmath19 , referred to as the disorder strength .
hereafter , @xmath18 is set to zero .
the resonant interactions @xmath20 are considered to be nonrandom and are assumed to be of dipolar origin : @xmath21 @xmath22 , with @xmath23 denoting the nearest - neighbor coupling .
@xmath24 describes the vibrational modes of the host , labeled @xmath25 and with the spectrum @xmath26 ( @xmath27 ) .
the operator @xmath28 annihilates ( creates ) a vibrational quantum in mode @xmath25 . finally , the operators @xmath29 and @xmath30 describe the linear and quadratic exciton - vibration coupling , respectively , where the quantities @xmath31 and @xmath32 indicate their strengths .
we do not derive explicit expressions for these coupling strengths , as we aim to treat them on a phenomenological basis .
in particular , owing to the disordered nature of the host , we consider these strengths stochastic quantities , for which we only specify the following stochastic properties with respect to the site index @xmath16 : [ vv ] @xmath33 @xmath34 @xmath35 where the angular brackets denote averaging over realizations of @xmath31 and @xmath32 . the kronecker symbol in eqs
. ( [ < v1v1 > ] ) and ( [ < v2v2 > ] ) implies that the surroundings of different monomers in the aggregate are not correlated . in general , the hamiltonian eq . ( [ h ] )
can not be diagonalized analytically .
if the exciton - vibration coupling is not too strong , the method of choice is first to find the exciton eigenstates by numerical diagonalization and then to consider the scattering on the basis of these eigenstates .
explicitly , the exciton states follow from the eigenvalue problem [ diagonalization ] @xmath36 where @xmath37 and @xmath38 is the eigenenergy of the exciton state @xmath39 : @xmath40 in the exciton representation , the exciton - vibration interactions take the form [ v1v2 ] @xmath41 @xmath42 where @xmath43 and @xmath44 are the matrix elements of the vibration - induced scattering of an exciton from state @xmath45 to state @xmath46 , given by [ v1qv2q ]
@xmath47 @xmath48 spectroscopic data on j - aggregates clearly reveal that the exciton - vibration coupling in these systems is usually weak . for the prototypical j - aggregates of pic
, this claim is corroborated by two facts : ( i ) - the narrowness of the j - band , which only is a few tens of @xmath3 at liquid helium temperature and becomes several times broader at room temperature , and ( ii ) - the absence of a fluorescence stokes shift of the j - band ( see , e.g. , ref . ) .
the extended nature of the exciton states in j - aggregates helps to reduce the exciton - vibration coupling , as it leads to averaging of the static as well as dynamic fluctuations of the site energies , effects known as exchange @xcite and motional @xcite narrowing , respectively .
the weakness of the exciton - vibration coupling allows one to calculate the scattering and dephasing rates of the excitons through perturbation theory .
this analysis is presented in the next section .
following the arguments given at the end of sec . [ model ] , we will use fermi s golden rule to calculate the rate for scattering of excitons from one localized state , @xmath45 , to another one , @xmath46 .
the result reads @xmath49 here , the superscript @xmath50 distinguishes between one- and two - vibration - assisted exciton scattering .
furthermore , @xmath51 and @xmath52 are the energies of the vibration bath in the initial ( @xmath53 ) and final ( @xmath54 ) states , respectively , where @xmath55 denotes the set of occupation numbers of the vibrational modes .
the quantity @xmath56 is the equilibrium density matrix of the initial state of the bath .
finally , the angular brackets indicate that we average over the stochastic realizations of the surroundings of each monomer in the aggregate . in a one - phonon - assisted scattering process ,
the occupation number of one phonon mode @xmath25 increases or decreases by one , corresponding to emission and absorption of a vibrational quantum , respectively .
consequently , @xmath57 . substituting the explicit form of the operator @xmath29 from eq .
( [ v1 ] ) into eq .
( [ fgr ] ) , we obtain @xmath58\ , \delta(\omega_{\mu\nu } + \omega_q ) \nonumber\\ \nonumber\\ & + & \bar{n}(\omega_q ) \ , \delta(\omega_{\mu\nu } - \omega_q)\big ] \ , \end{aligned}\ ] ] where @xmath59^{-1}$ ] is the mean occupation number of the vibrational mode @xmath25 ( the boltzmann constant @xmath60 ) and @xmath61 . in deriving eq .
( [ w1 ] ) , we used the properties of the stochastic function @xmath31 given by eqs .
( [ < v > ] ) and ( [ < v1v1 > ] ) . defining the one - vibration spectral density as @xmath62
we can rewrite eq .
( [ w1 ] ) in the form @xmath63 as we observe , the rate @xmath64 is proportional to the overlap integral of the site occupation probabilities , @xmath65 and @xmath66 , of the exciton states involved .
first of all , this leads to a strong suppression of the scattering rate if states @xmath46 and @xmath45 overlap weakly or not at all .
second , as the low - energy exciton states in a disordered chain exhibit large fluctuations in their localization size,@xcite also the scattering rates may undergo large fluctuations ( see sec . [ numerics ] ) .
the dependence of @xmath64 on the energy mismatch @xmath67 is determined by the one - phonon spectral density @xmath68 .
characterizing this function requires knowledge of the vibrational spectrum @xmath26 as well as the @xmath25 dependence of the exciton - vibration coupling @xmath69 . for the special case of scattering on acoustic phonons of a relatively long wavelength
, we have @xmath70 .
this behavior results from the @xmath71-dependence of the density of states of acoustic phonons , combined with the fact that in the long - wavelength limit @xmath72.@xcite one may consider this a debye - like model , in which one replaces the summation over the mode index @xmath25 by an integration over the frequency @xmath26 according to the well - known rule : @xmath73 here , @xmath74 is an irrelevant constant , which we will incorporate in an overall free parameter ( see below ) , and @xmath75 is a cutoff frequency .
it is important to note that @xmath75 is not necessarily related to the debye frequency : the generally very complex density of vibrational states in a disordered solid may on average exhibit an @xmath71 scaling up to a given frequency @xmath75 . inspired by the above
, we consider a slightly wider class of one - phonon spectral densities , given by @xmath76 here , @xmath77 is a free parameter in the model , which characterizes the overall strength of the one - vibration - assisted scattering rates .
it absorbs a number of other parameters characteristic for the host lattice ( such as the velocity of sound ) , the constant @xmath74 from eq .
( [ sum - to - int ] ) , as well as the strength of the transfer interaction @xmath78 ( for details , see ref . ) .
@xmath79 is the heaviside step function .
when performing numerical simulations , we will mostly use @xmath80 , for which the spectral density of acoustic phonons in the long - wave limit is recovered . in some instances ,
however , we will discuss how results depend on the exponent @xmath81 .
several observations support considering a debye - like vibration spectral density , even for a disordered host .
thus , for strongly disordered yb@xmath82 doped phosphate glasses , a parabolic behavior of the one - phonon spectral density was found over a rather wide range of measurement ( 0 to 100 @xmath3 ) .
@xcite furthermore , closely related spectral densities of the form @xmath83 ( or linear combinations of such functions ) have been used successfully to fit the optical dynamics in photosynthetic antenna complexes ( see , e.g. , refs . ) .
one - phonon - assisted scattering results in the transition of an exciton from a given state @xmath45 to state @xmath46 , where necessarily @xmath84 .
in other words , this type of scattering changes the occupation probabilities of the exciton states and thus causes population ( or energy ) relaxation .
the population relaxation in turn contributes to the dephasing of state @xmath45 .
the corresponding dephasing rate is given by ( see , e.g. , ref . ):
@xmath85 thus , @xmath86 represents the one - phonon - assisted contribution to the homogeneous broadening of the excitonic level @xmath87 .
we note that the @xmath86 indirectly depend on temperature through the @xmath88 [ cf .
( [ w1a ] ) ] .
the temperature dependence of the sum over scattering rates in eq .
( [ gamma1 ] ) and the corresponding width of the total exciton absorption spectrum will be analyzed in secs .
[ homogeneouslimit ] and [ numerics ] . when excitons scatter on the second - order displacements of the host molecules , described by the operator @xmath30 , the occupation numbers of two phonon modes @xmath25 and @xmath89 with frequencies @xmath26 and @xmath90 change by @xmath91 .
thus , @xmath92 , where any combination of plus and minus is allowed .
the corresponding scattering rates @xmath93 are obtained from eq .
( [ fgr ] ) , taking into account the stochastic properties of @xmath32 given by eqs .
( [ < v > ] ) and ( [ < v2v2 > ] ) : @xmath94 \big[\bar{n}(\omega_{q^{\prime } } ) + 1\big ] \ , \delta(\omega_{\mu\nu } + \omega_q + \omega_{q^{\prime } } ) \nonumber\\ \nonumber\\ & + & 2 \bar{n}(\omega_q ) \big[\bar{n}(\omega_{q^{\prime } } ) + 1\big ] \ , \delta(\omega_{\mu\nu } - \omega_q + \omega_{q^{\prime } } ) \nonumber\\ \nonumber\\ & + & \bar{n}(\omega_q ) \bar{n}(\omega_{q^{\prime } } ) \ , \delta(\omega_{\mu\nu } - \omega_q - \omega_{q^{\prime}})\big ]
\ .\end{aligned}\ ] ] if in analogy to the one - vibration - assisted scattering , we define the two - vibration spectral density @xmath95 as @xmath96 the scattering rate @xmath93 takes the form @xmath97 \big[\bar{n}(\omega^{\prime } ) + 1\big ] \ ,
\delta(\omega_{\mu\nu } + \omega + \omega^{\prime } ) \nonumber\\ \nonumber\\ & + & 2 \bar{n}(\omega ) \big[\bar{n}(\omega^{\prime } ) + 1\big ] \ , \delta(\omega_{\mu\nu } - \omega + \omega^{\prime } ) \nonumber\\ \nonumber\\ & + & \bar{n}(\omega ) \bar{n}(\omega^{\prime } ) \ , \delta(\omega_{\mu\nu } - \omega - \omega^{\prime})\big ]
\ .\end{aligned}\ ] ] similar to @xmath68 [ eq .
( [ f1 ] ) ] , we will use a parametrization @xmath98 where @xmath99 is a free parameter that characterizes the overall strength of the two - vibration - assisted scattering rates [ cf . @xmath77 and the discussion following eq .
( [ f1model ] ) ] .
two main types of two - phonon - assisted processes may be distinguished .
similarly to the one - phonon case , an inelastic channel exists , where scattering occurs between different exciton states , thus giving rise to population relaxation .
however , also an elastic channel is present , in which an exciton is scattered by emitting and absorbing a phonon of the same energy , and the final exciton state is identical to the initial one .
this process results in pure dephasing of the exciton state , with a rate given by @xmath100 \
.\end{aligned}\ ] ] the quantity @xmath101 is recognized as the inverse participation ratio , @xcite which is inversely proportional to the localization size of the exciton state @xmath102 .
thus , we see that pure dephasing is suppressed for more extended states , an effect known as the motional narrowing .
@xcite like in the one - phonon assisted process , the scattering rate between different states @xmath102 and @xmath103 is proportional to the overlap of their site occupations .
the final expressions for the rates @xmath93 resulting from eqs .
( [ 1w2 ] ) and ( [ f2model ] ) are derived in appendix a ; they depend on the sign of @xmath104 as well as on the relation between @xmath105 and @xmath75 .
distinction is made between three inelastic channels : downward ( @xmath106 ) , in which two phonons are emitted , cross ( @xmath107 ) , in which one phonon is absorbed and another is emitted , and upward ( @xmath108 ) , in which two phonons are absorbed .
the fourth type of scattering is the elastic ( pure dephasing ) channel , discussed above already .
when calculating the two - phonon assisted dephasing rate @xmath109 of state @xmath102 , we will account for elastic as well as inelastic contributions : @xmath110 \ .\ ] ] this rate depends on temperature as a result of the mean occupation numbers @xmath111 and @xmath112 of the vibrational modes .
in order to gain insight in the temperature dependence of the dephasing rates and the absorption bandwidth , it is useful to start by considering a homogeneous aggregate , i.e. , @xmath113 .
analytical results can then be obtained if we restrict the resonant interactions @xmath20 to nearest - neighbor ones . in this approximation ( which we will relax in our numerical analysis ) , we have [ homogeneous ] @xmath114 @xmath115 the corresponding overlap integrals occurring in eqs .
( [ w1 ] ) and ( [ w2 ] ) now read @xmath116 \ .\ ] ] in a homogeneous linear chain , the lowest exciton state , @xmath117 , contains almost all oscillator strength , thus dominating the absorption spectrum.@xcite therefore , the dephasing rate of this state , @xmath118 , is of primary interest . as follows from eq .
( [ gamma1 ] ) , it is determined by the sum over scattering rates to the other exciton states ( @xmath119 ) , all of which are higher in energy . in order to evaluate @xmath118 , we replace the summation in eq .
( [ gamma1 ] ) by an integration , @xmath120\int dk \big)$ ] , which is allowed if @xmath121 and @xmath122 .
we will also assume that @xmath123 , which implies that the relevant exciton levels are those near the lower exciton band edge , where @xmath124 with @xmath125 .
changing the integration variable to @xmath126 , using @xmath127 and replacing the lower integration limit @xmath128 by zero , we obtain @xmath129 for temperatures @xmath130 , the upper integration limit may be extended to infinity , and we arrive at @xmath131 where @xmath132 and @xmath133 are the gamma - function and the rieman zeta - function , respectively .
thus , for @xmath130 the one - phonon - assisted dephasing rate shows a power - law temperature dependence .
note that for our model of acoustic phonons ( @xmath134 ) , @xmath118 increases quite steeply , namely as @xmath135 . from numerical evaluation of @xmath118 for a homogeneous chain with all dipole - dipole interactions , we have found that the exponent 7/2 is increased to 3.85 , mainly as a consequence of logarithmic corrections in the exciton dispersion near the lower band edge.@xcite if we go beyond the parabolic range of the energy spectrum , the growth becomes even steeper ; the exponent then tends to 4 , because the density of states becomes a constant towards the center of the band . in the opposite limit @xmath136 , the exponential in the denominator of eq .
( [ gamma1homo ] ) can be expanded in a taylor series .
up to second order , one obtains @xmath137 which simply reflects the linear high - temperature dependence of the mean occupation number @xmath111 .
obviously , this scaling also holds in the presence of disorder .
we now turn to the temperature dependence of the pure dephasing rate @xmath138 . as for a homogeneous aggregate
this rate does not depend on the state index @xmath87 , we will simply denote it as @xmath139 . using the explicit form of @xmath140 given by eq .
( [ wnunu ] ) , we arrive at @xmath141 from eq .
( [ gp ] ) it follows that for @xmath130 ( @xmath142 ) , [ gp_limits ] @xmath143 while for @xmath136 ( @xmath144 ) , @xmath145 for the case of scattering on acoustic phonons ( @xmath80 ) and @xmath130 , we thus arrive at @xmath146 .
this temperature dependence resembles that for the pure dephasing of an isolated state of a point center , derived by mccumber and sturge .
@xcite the only difference is that the exciton dephasing rate undergoes suppression by a factor of @xmath147 due to the motional narrowing effect .
we note that this narrowing is not observed for one - phonon - assisted dephasing [ see eqs .
( [ gamma1 t < ] ) and ( [ gamma1 t > ] ) ] .
the @xmath148-scaling of @xmath139 in the high - temperature limit ( @xmath149 ) results from the square of the mean phonon occupation number involved in eq .
( [ wnunu ] ) . to conclude this section
, we stress that the @xmath150 and @xmath148 scaling relations of the pure dephasing rate with temperature obtained here , also hold for disordered aggregates , because this result is determined only by the two - vibration spectral density @xmath140 .
the suppression factor , however , will then be determined by the exciton localization size , rather than the chain length . finally , we analyze the dephasing rate of the superradiant state ( @xmath151 ) resulting from the two - phonon inelastic scattering of excitons .
this rate , which will be denoted as @xmath152 , is determined by the sum of scattering rates to all higher states , @xmath153 . using eqs .
( [ 1w2 ] ) and ( [ f2model ] ) , and making the same assumptions as in the case of one - phonon - assisted dephasing ( sec .
[ 1 ] ) , we obtain @xmath154 \ .\end{aligned}\ ] ] in the low - temperature limit , @xmath130 ( @xmath155 ) , eq .
( [ gamma2_inel ] ) yields @xmath156 where the numerical factor @xmath157 is given by the double integral in eq .
( [ gamma2_inel ] ) , with both upper limits replaced by infinity .
comparing eq .
( [ gamma2_inelt < ] ) with eq .
( [ gpt < ] ) , we see that @xmath152 is characterized by a steeper temperature dependence than @xmath139 . in particular , for @xmath80 the rate @xmath158 .
this means that at higher temperatures the inelastic two - phonon channel of dephasing can compete with the elastic one ( sec .
[ 2phonon ] ) . in the high - temperature limit , @xmath136 ( @xmath159 ) ,
the rate @xmath152 is proportional to @xmath148 , which is the same scaling relation as in the case of pure dephasing [ eq . ( [ gpt > ] ) ] .
in this section , we will analyze the temperature dependence of the various dephasing contributions in the presence of disorder and their effect on the width of the j - band .
as we will see ( and already anticipated in sec .
[ linear ] ) , the dephasing rates are spread over a wide region , in particular at low temperature . as a consequence ,
it is not clear _ a priori _ what value ( mean , typical , or other ) of these rates should be related to the homogeneous width of the absorption spectrum .
therefore , the width of the j - band was obtained by direct simulation of the absorption spectrum , using the calculated dephasing rates to broaden each of the exciton transitions in the band .
explicitly , we have @xmath160 where @xmath161 is the dimensionless oscillator strength of the @xmath87th exciton state and @xmath162 is the total homogeneous width of this state . here , @xmath163 and @xmath164
are given by eqs .
( [ gamma1 ] ) and ( [ gamma2 ] ) , respectively , and @xmath165 is the radiative decay rate of state @xmath87 ( @xmath166 denotes the radiative constant of a monomer ) . in all the simulations , we will assume the limit @xmath167 , which implies that @xmath168 in eqs .
( [ f1model ] ) and ( [ f2model ] ) . as before ,
the angular brackets denote the average over the random realizations of the site energies @xmath169 .
the resulting j - bandwidth @xmath170 was determined as the full width at half maximum of the thus calculated absorption spectrum .
a similar approach has been used @xcite to simulate the absorption spectrum of thiats aggregates @xcite at room temperature . in order to study fluctuations in the dephasing rates
, we analyzed their statistics , focusing on the lowest exciton state for each randomly generated disorder realization .
this choice was motivated by the fact that the low - lying states dominate the absorption spectrum .
the dephasing rate of the lowest state is denoted @xmath171 ; its distribution , collected by considering @xmath172 disorder realizations for chains of @xmath173 molecules and a disorder strength @xmath174 is presented in figs .
[ g1](a ) and [ g1](b ) . in generating these figures , we used a one - vibration spectral density of the form eq .
( [ f1model ] ) , with @xmath175 .
furthermore , the rates were calculated for @xmath176 [ fig .
[ g1](a ) ] and @xmath177 [ fig . [ g1](b ) ] ; for the protoptypical aggregates of pseudoisocyanine ( @xmath178 @xmath3 ) , this agrees with temperatures of 50 k and 200 k , respectively .
we note that the horizontal axis of the distributions is scaled by the average value @xmath179 , so that the value of @xmath77 does not affect the figures .
of course , this scaling renders the axis temperature dependent , because @xmath179 strongly depends on @xmath180 , as we will see below ( fig .
[ g1_fwhm ] ) .
these figures clearly demonstrate that the relative spread in @xmath171 may be considerable , in particular at low temperatures .
this may be understood from the local band - edge level structure of the disordered tight - binding hamiltonian .
@xcite for temperatures smaller than the j - bandwidth , the exciton scatters between discrete levels in the vicinity of the band edge that are localized in the same region of the chain .
both the energy spacing between these states and their localization size undergo large fluctuations : @xmath181 and @xmath182 .
@xcite as a consequence , @xmath183 .
when the temperature is increased , the exciton in the lowest state may scatter to many higher - lying states , which often are delocalized over an appreciable part of the chain .
this smears the fluctuations that occur in the scattering rates between individual states , leading to a decrease in the relative spread of the dephasing rate of the lowest state .
we next turn to the absorption band calculated according to eq .
( [ a ] ) , neglecting the role of two - phonon scattering [ @xmath184 . in fig .
[ p3nocutoff](a ) this band is plotted for three temperatures at a fixed disorder strength of @xmath185 . for the one - phonon spectral density we used the form eq .
( [ f1model ] ) with @xmath186 and @xmath80 , and took @xmath187 , which is typical for j - aggregates of polymethine dyes .
chains of @xmath188 molecules were considered .
the simulated spectra clearly demonstrate the thermal broadening , caused by growing homogeneous widths of the individual exciton transitions . at low temperature ,
the homogeneous broadening is negligible , the j - band is inhomogeneous , with a width that is determined by the disorder strength . with growing temperature ,
the j - band becomes more homogeneous , as is apparent from the fact that it gets more symmetric . in fig .
[ p3nocutoff](b ) we plotted by symbols the simulated j - band width @xmath5 as a function of temperature for three values of the disorder strength : @xmath189 , @xmath190 , and @xmath191 [ all other parameters were taken as in fig . [ p3nocutoff](a ) ] .
as is seen , the @xmath5 shows a plateau at the value of the inhomogeneous width , @xmath192 , @xmath193 , and @xmath194 , respectively . beyond these plateaus
, @xmath5 goes up quite steeply , reflecting the fact that the homogeneous ( dynamic ) broadening becomes dominant .
to accurately extract at low temperatures the small homogeneous contribution to the total width , we generated up to @xmath195 disorder realizations . at higher temperatures ,
this number could be restricted to 4000 , owing to the reduction of the relative fluctuations ( cf .
[ g1 ] ) . inspired by the analytically obtained power - laws for the one - phonon - assisted dephasing rate as a function of temperature for homogeneous aggregates ( sec .
[ 1 ] ) , we considered a parametrization of the form @xmath196 for the total band width in disordered aggregates .
it turned out that the calculated line widths as a function of temperature could be fitted very well by the relation ( [ fit ] ) [ curves in fig .
[ p3nocutoff](b ) ] .
the corresponding fit parameters are @xmath197 and @xmath198 for @xmath189 , @xmath199 and @xmath200 for @xmath201 , and @xmath202 and @xmath203 for @xmath204 .
the scaling relation eq .
( [ fit ] ) turns out to hold over an even wider range of @xmath19 and @xmath205 values.@xcite this implies that , although the j - band is built up from a distribution of exciton states with different dephasing rates , the total width @xmath5 may effectively be separated in an inhomogeneous width , @xmath206 , and a dynamic contribution .
we note that the fitting exponent @xmath1 is larger than the value 3.85 found in the absence of disorder ( sec .
this increase results from downward scattering processes between optically dominant exiton states , which are possible in the presence of disorder , but not for the superradiant state in the homogeneous chain .
this claim may be substantiated by considering the dephasing rates of the lowest exciton state of each disorder realization .
we numerically generated the average of this quantity , @xmath179 , from @xmath207 disorder realizations for chains of @xmath173 molecules with @xmath208 , @xmath209 , and @xmath175 .
this average is shown as a function of temperature in fig .
[ g1_fwhm ] ( diamonds ) , together with the dynamic contribution to the total j - bandwidth , @xmath210 ( solid line ) , and the dephasing rate @xmath211 of the superradiant state for a homogeneous chain of the same length ( squares ) .
as is seen , the dynamic part @xmath210 has a larger exponent @xmath212 than @xmath179 @xmath213 .
it is remarkable , however , that @xmath179 and @xmath211 display almost identical behavior , at least in the relevant region @xmath214 , where the homogeneous contribution to the j - bandwidth is noticeable . for @xmath215 , where @xmath171 undergoes significant fluctuations ( see fig .
[ g1 ] ) , @xmath179 turns out to be much smaller than @xmath211 .
in the high - temperature regime , these fluctuations are washed out and the two quantities become almost identical .
so far , we have only presented numerical results for a one - phonon spectral density eq .
( [ f1model ] ) with the power @xmath175 , which corresponds to a debye model for the host vibrations . to end this subsection
, we will address the effect of changing the spectral density .
first , we consider the effect of the value for @xmath81 .
the diamonds in fig .
[ width_sin ] present our results for the temperature dependence of the width @xmath5 obtained for @xmath216 and @xmath217 .
as before , we found that these data may be fitted by a simple power - law of the form eq .
( [ fit ] ) ( dashed line ) . in this case
we find @xmath218 , which , again , is slightly larger than the value 3/2 found from eq .
( [ gamma1 t < ] ) , due to the correction of the dispersion relation arising from the long - range dipole - dipole interactions and downward scattering processes that contribute to the total j - bandwidth .
clearly , these data demonstrate the sensitivity of the temperature dependence of the total linewidth to the power @xmath81 .
interestingly , it turns out that while the temperature dependence of the j - bandwidth is sensitive to the overall frequency scaling of the spectral density , it is not sensitive to fluctuations of this scaling around an average power - law . to demonstrate this ,
we have considered a spectral density of the form @xmath219 $ ] , which only on average exhibits an @xmath220-dependence .
the results for the width @xmath5 obtained for @xmath221 and @xmath222 are presented as squares in fig .
[ width_sin ] .
remarkably , these results are indistinguishable from those obtained without fluctuations ( i.e. , @xmath223 : triangles ) .
the reason is that at elevated temperatures the function @xmath224 varies slowly on the scale of @xmath225 , so that the modulating function @xmath226 may be replaced by its average value , which equals unity . as we have seen in sec.[quadratic ] , the two - phonon - assisted dephasing rate @xmath227 consists of four contributions , one of which is elastic ( indicated as pure " , as it is responsible for pure dephasing ) , while the other three are inelastic and are indicated as downward ( @xmath106 ) , cross ( @xmath107 ) , and upward ( @xmath108 ) . in figs .
[ g1](c)- [ g1](f ) , [ g2p ] , and [ gamma_tmp ] we present results for the statistics of these various contributions for disordered aggregates . in all cases , we used chains of 500 molecules , a disorder strength of @xmath174 , and a two - vibration spectral density of the form eq .
( [ f2model ] ) , with @xmath175 .
the statistics are presented for the lowest exciton state in each one of @xmath228 randomly generated disorder realizations . for this state
the downward contribution vanishes and the two - phonon - assisted dephasing rate reads @xmath229 .
these three remaining contributions were calculated using the expressions derived in the appendix . from figs . [ g1](c ) and [ g1](d ) we observe that at a given temperature , the relative spread in @xmath230 is much smaller than that in @xmath231 .
the reason is that in a two - phonon - assisted upward process the exciton in the lowest state scatters to more higher - energy states than in a cross process . as a result ,
fluctuations in @xmath230 are suppressed more than those in @xmath231 . upon heating , the spread of both @xmath230 and @xmath231 reduces , which has the same explanation as given for this effect in the case of one - phonon - assisted dephasing ( sec .
[ 1phonon ] ) .
the distribution of @xmath139 does not depend on temperature at all , because , according to eq .
( [ wnunu ] ) , the rate @xmath139 fluctuates exclusively due to fluctuations in the inverse participation ratio @xmath232 .
as the latter quantity is subject to large fluctuations , @xcite this also explains the large relative spread in @xmath139 . in fig .
[ gamma_tmp ] we plotted ( on a log - log scale ) the temperature dependence of the mean values @xmath233 , @xmath234 , and @xmath235 , obtained from averaging over these rates for the lowest exciton states in the simulations discussed above .
this figure nicely shows the relative importance of the different dephasing channels .
we clearly see that @xmath236 , i.e. , the inelastic channel of dephasing due to double phonon absorption is inefficient , at all temperatures .
more importantly , we observe that the cross channel of inelastic two - phonon - assited dephasing successfully competes with the pure dephasing contribution . at low temperatures ,
pure dephasing dominates , while at higher temperatures the inelastic cross process is more important .
this correlates well with our findings for disorder - free aggregates [ see discussion below eq .
( [ gamma2_inelt < ] ) ] . for the disorder strength considered here ,
the cross - over occurs at @xmath237 ( @xmath238 k for @xmath239 @xmath3 ) .
we note that when considering two - phonon scattering , one often restricts to modelling the elastic process ( see , e.g. , ref . ) . from the above
we see that this is not justified at elevated temperatures . despite the fact that not all three curves in fig .
[ gamma_tmp ] are exactly straight lines ( deviations occur in the unimportant low - temperature part ) , they all can be fitted very well by a power - law , @xmath240 . in doing so , we obtained @xmath241 @xmath242 @xmath243 we recall that the exponent @xmath244 in the last formula is an exact result for @xmath245 and @xmath167 , as was argued at the end of sec .
[ 2 ] already . as in the case of one - phonon scattering
, we see that the inelastic two - phonon dephasing rates exhibit a steeper temperature dependence than for the homogeneous aggregate with nearest - neighbor interactions [ eq . ( [ gamma2_inelt < ] ) ] .
it is of interest to see to what extent the model we presented here is able to explain the temperature dependence of the homogeneous broadening in molecular aggregates .
as mentioned in the introduction , renge and wild @xcite found that the total j - bandwidth @xmath5 of pic - cl and pic - f over a wide temperature range ( from 10 k to 300 k ) follows a power - law scaling as in eq .
( [ fit ] ) .
although the power reported by these authors ( @xmath246 ) is smaller than the ones we derived in sec .
[ 1phonon ] , from direct comparison to the experimental data we have found that a model of one - phonon scattering with a spectral density given by eq.([f1model ] ) with @xmath80 and @xmath247 , yields an excellent quantitative explanation of the experiments over the entire temperature range , both for the shape and the width of the j - band .
the same turns out to be true for the j - bandwidth of pic - br , measured between 1.5 k and 180 k. @xcite in all these fits , @xmath19 and @xmath205 were the only two free parameters that could be adjusted to optimize comparison to experiment .
details will be published elsewhere,@xcite together with a fit of the much debated@xcite temperature dependence of the fluorescence lifetime of these aggregates . here ,
we present an explicit comparison to the hole - burning data reported by hirschmann and friedrich.@xcite using this technique , they measured the homogeneous width of the exciton states in the center of the j - band for pic - i over the temperature range 350 mk to 80 k. their data for the holewidth @xmath248 are reproduced as triangles in fig .
[ holeburning ] .
the solid line shows our fit to these data , obtained by simulating disordered chains of @xmath249 molecules with a one - phonon spectral density of the form eq .
( [ f1model ] ) with @xmath175 and @xmath250 .
the resonant interaction strength and the monomer radiative rate were chosen at the accepted values of @xmath178 @xmath3 and @xmath251 s@xmath252 , respectively .
thus , the only free parameters were the disorder strength @xmath19 and scattering strength @xmath77 .
first , we fixed the value of @xmath19 by fitting the low - temperature ( 4 k ) absorption spectrum , where the homogeneous broadening may be neglected .
this yielded @xmath253 .
next , @xmath77 was adjusted such that the measured growth of the hole width was reproduced in an optimal way .
thus , we found @xmath254 .
for each given temperature the simulated hole width was obtained as the average of the dephasing rate of the excitons found in an interval of 0.06 @xmath3 around the center of the simulated j - band at that temperature .
taking into account the fair amount of scatter in the experimental data , we conclude from fig .
[ holeburning ] that our model yields a good fit to the measurements .
as mentioned above , the same holds for the j - bandwidth in aggregates of pic - cl , pic - f , and pic - br .
this yields valuable information about the dominant mechanism of dephasing in these materials .
we conclude that this mechanism is one - phonon - assisted scattering of excitons on vibrations in the host characterized by a spectral density which ( on average ) scales as @xmath220 ; this scaling yields a natural explanation of the power - law thermal broadening of the j - band found in various experiments .
we stress that it is impossible to fit the experimental data with a spectral density that is constant or scales linearly with @xmath255 , as that yields considerably different power laws for the width [ cf .
[ width_sin ] ] .
the @xmath220 scaling of the spectral density needed to fit the experiments strongly suggests that acoustic phonons dominate the scattering process .
thus , the spectral width measured over a broad temperature range is an excellent probe for the scattering mechanism .
it is appropriate to comment on the value of @xmath77 obtained from our fit , which seems to be very large .
it should be kept in mind that @xmath77 is a phenomenological scattering strength , which combines several microscopic material properties [ see discussion below eq .
( [ f1model ] ) ] .
most importantly , the value found here is consistent with a perturbative treatment of the scattering process : the scattering rates between the optically dominant states obtained from it turned out to be much smaller than their energy separation .
we finally address an alternative mechanism of dephasing , namely scattering on local vibrations belonging to the aggregate .
this has been suggested by several authors based on activation - law fits of the measured homogeneous contribution to the j - bandwidth .
@xcite to get an estimate whether this is a reasonable mechanism , let us neglect the disorder and make the nearest - neighbor approximation for the resonance interactions @xmath20 .
then , @xmath256 and @xmath257 are given by eq .
( [ homogeneous ] ) .
close to the lower exciton band edge , the region of our interest , @xmath258 .
furthermore , let us parameterize the spectral function of a local vibration of frequency @xmath259 as @xmath260 , where @xmath261 is the coupling constant to the excitons .
we are interested in the dephasing rate @xmath262 of the superradiant state @xmath117 . using the above simpifications and replacing in eq .
( [ gamma1 ] ) the summation over exciton states by an integration , one easily arrives at @xmath263 comparing this result to the activation law @xmath264 , used in refs . and to fit the experimental data , one obtains @xmath265 .
substituting the values @xmath266 @xmath3 and @xmath267 @xmath3 from ref . , and @xmath239 @xmath3 , we obtain as estimate for the exciton - phonon coupling @xmath268 @xmath3 .
this is an enormously large number .
in particular , the stokes losses @xmath269 @xmath3 turn out to be much larger than the resonant interaction @xmath239 @xmath3 . under these conditions , a strong exciton self - trapping is to be expected , resulting in a reduction of the exciton bandwidth by a factor of @xmath270 .
@xcite coherent motion of the exciton , even over a few molecules , is then hardly possible : any disorder will destroy it .
this observation is not consistent with the widely accepted excitonic nature of the aggregate excited states , as corroborated by many optical and transport measurements.@xcite
in this paper we presented a theoretical study of the temperature dependence of the exciton dephasing rate in linear j - aggregates and the resulting width of the total absorption band ( the j - band ) .
as dephasing mechanism we considered scattering of the excitons on vibrations of the host matrix , taking into account both one- and two - vibration scattering . the excitons were obtained from numerical diagonalization of a frenkel exciton hamiltonian with energy disorder and their dephasing rates were subsequently calculated using a perturbative treatment of the exciton - vibration interaction ( fermi golden rule ) . in the absence of disorder , the lowest ( superradiant )
exciton state dominates the absorption spectrum . as a result
, the dephasing rate of this state directly gives the homogeneous width of the j - band .
we analytically calculated the temperature dependence of this homogeneous width for both one- and two - vibration scattering , assuming a debye - like model for the host vibration density of states .
it turned out that in all cases the homogeneous width obeys a power - law as a function of temperature , with the value of the exponent depending on the shape of the low - energy part of the vibronic spectrum ( sec .
[ homogeneouslimit ] ) . in the presence of disorder
the optically dominant exciton states still reside close to the bottom of the band , but their energies are now spread and their wave functions become localized on finite segments of the chain .
we have found that the various one- and two - phonon - induced contributions to the dephasing rate undergo significant fluctuations , because the exciton energies and overlap integrals vary considerably from one disorder realization to the other ( sec .
[ numerics ] ) . as a consequence
, one can not use the dephasing rates of individual states to characterize the homogeneous broadening of the j - band .
instead , we simulated the total j - band and showed that its width effectively separates in an inhomogeneous ( zero - temperature ) contribution and a dynamic ( homogeneous ) part .
the latter scales with temperature according to a power law [ eq . ( [ fit ] ) ] , with an exponent that is somewhat larger than the one found for the homogeneous broadening in the absence of disorder .
we also showed that amongst the two - vibration scattering processes , inelastic channels will at elevated temperatures dominate the usually considered pure - dephasing contribution . finally , from comparison to absorption and hole - burning experiments ( sec .
[ experiment ] ) , we found that the dominant mechanism of dephasing for j - aggregates lies in one - phonon scattering of excitons on vibrations of the host matrix , characterized by a spectral density which ( on average ) scales like the third power of the phonon frequency .
this suggests that acoustic phonons of the host play an important role in the scattering process .
all temperature dependent data available to date , are consistent with this picture . by contrast , we argued that the previously suggested mechanism of scattering on local vibrations of the aggregate , leading to an activated thermal behavior , is not consistent with the overwhelming amount of evidence that the optical excitations in j - aggregates have an excitonic character .
in this appendix we present expressions for the two - phonon - assisted scattering rate @xmath93 , starting from eq .
( [ 1w2 ] ) . after substituting the two - vibration spectral density given in eq .
( [ f2model ] ) and performing several algebraic manipulations , we arrive at @xmath271 \ , \nonumber\\\end{aligned}\ ] ] where the temperature dependent functions @xmath272 , @xmath273 , and @xmath274 distinguish between the scattering processes in which two phonons are emitted ( @xmath106 ) , one is absorbed and one is emitted ( @xmath107 ) , and two phonons are absorbed ( @xmath108 ) .
they are given by we now further analyze the three contributions to the scattering rate .
as @xmath282 describes the emission of two vibrational quanta , we have @xmath283 .
thus , performing the @xmath284 integration , we obtain | we theoretically study the temperature dependence of the j - band width in disordered linear molecular aggregates , caused by dephasing of the exciton states due to scattering on vibrations of the host matrix .
in particular , we consider inelastic one- and two - phonon scattering between different exciton states ( energy - relaxation - induced dephasing ) , as well as elastic two - phonon scattering of the excitons ( pure dephasing ) .
the exciton states follow from numerical diagonalization of a frenkel hamiltonian with diagonal disorder ; the scattering rates between them are obtained using the fermi golden rule . a debye - like model for the one- and two - phonon spectral densities
is used in the calculations .
we find that , owing to the disorder , the dephasing rates of the individual exciton states are distributed over a wide range of values .
we also demonstrate that the dominant channel of two - phonon scattering is not the elastic one , as is often tacitly assumed , but rather comes from a similar two - phonon inelastic scattering process . in order to study the temperature dependence of the j - band width
, we simulate the absorption spectrum , accounting for the dephasing induced broadening of the exciton states .
we find a power - law ( @xmath0 ) temperature scaling of the effective homogeneous width , with an exponent @xmath1 that depends on the shape of the spectral density of host vibrations .
in particular , for a debye model of vibrations , we find @xmath2 , which is in good agreement with experimental data on j - aggregates of pseudoisocyanine [ j. phys . chem .
a * 101 * , 7977 ( 1997 ) ] . |
ultracold neutral atoms are a wonderful tool to study many - body physics and strong correlation effects . since the achievement of bose - einstein condensation in alkali atoms @xcite , we have been witnesses of two major breakthroughs .
one is the cooling of fermionic atoms and the study of cooper pairing and the bcs to bec transition @xcite .
the other one is the implementation of lattice hamiltonians using bosonic @xcite and fermionic atoms in off - resonance optical lattices .
contemporary and supported by these experimental achievements , a plethora of theoretical papers has consolidated ultracold atoms as an ideal system for quantum simulations .
the goal is two - fold : amo is now capable of implementing known hamiltonians which could describe real systems in condensed matter physics , such as hubbard models @xcite and spin hamiltonians @xcite ; but it is also possible to study new physical effects , such as the quantum hall effect with bosons @xcite and lattice gauge theories @xcite . in this work
we deepen and expand the ideas presented in @xcite , where we introduced a novel mechanism for pairing based on transport inducing collisions . as illustrated in fig .
[ fig : collisions ] , when atoms collide they can mutate their internal state .
if the atoms are placed in a state - dependent optical lattice , whenever such a collision happens the pair of atoms must tunnel to a different site associated to their new state .
for deep enough lattices , as in the mott insulator experiments @xcite , this coordinated jump of pairs of particles can be the dominant process and the ensemble may become a superfluid of pairs .
the focus of this work is to develop these ideas for all possible interaction asymmetries that can happen in experiments with bosonic atoms in two internal states , where the scattering lengths among different states can be different , @xmath0 @xcite .
we want to understand the different types of correlated hopping that appear when we move beyond the limited type of interactions considered in ref .
. we will better study the resulting phases and phase transitions , using different analytical and numerical tools that describe the many body states , and evolving from few particles to more realistic simulations .
the main results will be to show that when one considers more general asymmetries , a new type of correlated hopping appears , but that both the previous @xcite and the newly found two body terms cooperate in creating a superfluid state with pair correlations .
indeed , this new state will also be shown to be more than just a condensate of pairs , based on analytical and numerical studies .
correlated hopping is not a new idea .
it appears naturally in fermionic tight - binding models , where it has been used to describe mixed valence solids @xcite and , given that they are able to mimic the attractive interactions between electrons , also high-@xmath1 superconductors @xcite . in most of these works ,
the correlated hopping appears in the form @xmath2 indicating that the environment can influence the motion of a particle .
this would seem substantially different from correlated motion of pairs of fermions .
nevertheless , even this more elaborate form of correlated hopping has been shown to lead to the formation of bound electron pairs @xcite and it has been put forward as a possible explanation for high @xmath3 superconductivity @xcite . this work is organized as follows . in sec .
[ sec : themodel ] we introduce our model of correlated hopping ( [ eq : model ] ) , qualitatively discussing how it originates and what are the quantum phases that we expect it to develop .
we present a possible implementation of this model which is based on optical superlattices and atoms with asymmetric interactions .
[ sec : exact ] includes exact diagonalizations for a small number of atoms and sites .
these calculations reveal the existence of insulating and coherent regimes , as well as pairing , and will be the basis for later analysis . in sec .
[ sec : analytics ] we study the many - body physics of larger lattices with correlated hopping , using a variety of techniques : starting with insulating regime and following with the implementation of perturbation theory and the quantum rotor model . these methods suggest a number of possible phases , including a mott insulator , a pair superfluid , a normal superfluid and a charge density wave state , and we estimate the parameters for which these phases appear . in sec . [
sec : numerics ] we develop two numerical methods to study our system , a gutzwiller ansatz and an infinite matrix product state method . with these simulations
we confirm the above mentioned phases and locate the quantum phase transitions , which are found to be of second order . finally , in sec .
[ sec : detection ] we suggest some currently available experimental methods to detect and characterize these phases .
and @xmath4 are trapped in an optical lattice and coupled by a raman laser with rabi frequency @xmath5 .
( b ) that setup is equivalent to two displaced superlattices for the dressed states @xmath6 .
( c ) when asymmetric contact interactions are considered and the hopping between superlattice cells neglected , the whole system behaves effectively as a @xmath7d array of alternating @xmath8 and @xmath9 sites , with transitions of tunneling amplitude @xmath10 between them . ]
we suggested in ref .
@xcite that the combination of atomic collisions with optical superlattices can be used to induce correlated hopping . the basic idea is shown in fig .
[ fig : superlattice]b , where atoms are trapped in two orthogonal states called @xmath11 and @xmath12 the interaction terms change the state of the atoms , forcing them to hop to a different superlattice every time they collide . in this sense , interactions are responsible for transport . in this section
we introduce the most general model of correlated hopping that can be produced by such means with two states bosons .
this model is presented in the following subsection , where we explain qualitatively the role of each hamiltonian term .
later on , in sec .
[ sec : atomic - derivation ] , we establish the connection between the parameters of this model and the underlying atomic model .
this is the foundation for the subsequent analytical and numerical studies . in this work
we study the ground state properties of a very general hamiltonian that contains different kinds of correlated hopping .
more precisely , the model will be @xmath13\bigg\}. \label{eq : model}\end{aligned}\ ] ] here , @xmath14 and @xmath15 are bosonic operators that create and anihilate atoms according to the site numbering from fig . [
fig : superlattice]b - c , and the colons @xmath16 denote normal ordering of operators @xmath17 and @xmath18 let us qualitatively explain the roles of the different terms in eq .
( [ eq : model ] ) .
the first and second terms , @xmath19 and @xmath20 are related to on - site and next - neighbor interactions .
when these terms are dominant , we expect the atoms in the lattice form an insulator . such a phase is characterized by atoms being completely localized to lattice sites , having well - defined occupation numbers , the absence of macroscopic coherence and a gapped energy spectrum . whether this insulating state is itself dominated by strong on site interactions @xmath19 or by nearest
neighbor repulsion / attraction @xmath21 will decide whether it presents an uniform density , a mott insulator ( mi ) , or a periodic density pattern , a charge density wave ( cdw ) , respectively .
the third term is the key feature of our model .
it describes the tunneling of pairs between neighboring lattice sites , with amplitude @xmath22 given @xmath23 we expect the atoms to travel along the lattice in pairs forming what we call a pair superfluid ( psf ) .
these pairs will be completely delocalized , establishing long range coherence along the lattice .
the observable @xmath24 would be the figure of merit describing this kind of delocalization , while a vanishing @xmath25 indicates the absence of the single - particle correlations appearing in a normal superfluid .
furthermore , we expect this phase to have a critical velocity , similar to that of an atomic condensate , and the energy spectrum should be gapless . unlike in ref .
@xcite , when one considers the most general kind of atomic interaction , a second kind of correlated hopping appears , described by the last term in eq .
( [ eq : model ] ) . here
, individual atoms will hop only if there is already a particle in the site they go to ( @xmath26 ) or leave at least a particle behind ( @xmath27 ) .
one might be induced to think that this term is equivalent to single
particle hopping with a strength that depends on the average density , thus giving rise to a single - particle superfluid ( sf ) phase .
however , this does not seem to be the case .
we will show that the correlated hopping @xmath28 generates a mixed phase which contains features of both the ordinary bec and the psf created by @xmath22 we now establish the relation between the model in eq .
( [ eq : model ] ) and the dynamics of atoms in an optical superlattice . the actual setup we have in mind
is shown in fig .
( [ fig : superlattice])a - b and described in more detail in [ sec : trapping ] .
it consists on a three dimensional lattice that is strongly confining along the @xmath29 and @xmath30 directions , creating isolated tubes . on top of this , we create an optical superlattice acting along the @xmath31 direction @xcite .
this superlattice traps atoms in the dressed states @xmath8 and @xmath32 while the atomic interaction is diagonal in the basis of bare states @xmath33 and @xmath34 the interaction will be described by a contact potential and parameterized by some real constants @xmath35 @xmath36 these interaction constants are functions of the s wave one dimensional scattering lengths between different species @xmath37 in general , the interaction strengths among different atomic components are different from each other , a situation that can be enhanced with feshbach resonances . we will use a parameterization @xmath38 that makes the symmetries more explicit @xmath39 the total hamiltonian combines the previous interaction with the kinetic energy and the trapping potential for one particle @xmath40\psi_s(x ) dx,\ ] ] which is written in a different basis @xmath41 since the superlattice potential @xmath42 is the dominant term , we may approximate the bosonic fields as linear combinations of the wannier modes in this superlattice and in the dressed state basis , a process detailed in [ sec : derivation ] . note that out of all the terms in the interaction hamiltonian ( [ eq : interaction2 ] ) , only the first one is insensitive to the state of the atoms .
this is important because the asymmetries @xmath43 and @xmath44 when expressed in the dressed basis , produce terms that change the state of the atoms during a collision .
once we introduce the effective interaction constants in the lattice @xmath45 where @xmath46 is the single site wannier wavefunction , we arrive to the effective hamiltonian in eq .
( [ eq : model ] ) with parameters @xmath47 and @xmath28 that relate to the microscopic model as follows @xmath48 unlike in the specific case of ref .
@xcite , the most general situation contains not only two - body correlated hopping @xmath49 but also the terms proportional to @xmath50
in this section we study the eigenstates of hamiltonian ( [ eq : model ] ) for systems that we can diagonalize exactly .
the goals are to characterize the effect of the different interaction and hopping terms , as well as to understand the structure of the ground state wavefunction .
although we are limited to a small number of particles , the following examples provide enough evidence of the roles of correlated hopping , nearest neighbor repulsion and the utility of different correlators to characterize the states .
let us take the simplest interesting case : four particles in two sites .
we write the hamiltonian in the basis @xmath51 where the notation @xmath52 stands for @xmath53 particles in the first site and @xmath54 in the second and we restrict to @xmath55 @xmath56 notice that in this particular case , @xmath19 gives rise to a global energy shift and does not affect the different eigenstates .
this is consistent with later studies where we will see that on - site interactions just add a global , density dependent contribution to the energy . to better understand the role of the remaining terms
, we will consider separately three limiting cases , two of superfluid nature and an insulating one .
[ [ limit - jneq-0-tv0-single - particle - delocalization . ] ] limit @xmath57 @xmath58 single - particle delocalization .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + in this case we take for simplicity @xmath59 and diagonalize eq .
[ eq : hfor4in2 ] , finding the normalized ground state @xmath60 note that this state is exactly a bec of @xmath61 particles spread over two sites @xmath62 this suggests that , at least in this small example , the correlated hopping proportional to @xmath28 is equivalent to the single - particle hopping in the ordinary bose - hubbard model , giving rise to the delocalization of individual particles .
however , as it will become evident later on , for larger systems and more particles this interpretation is wrong . [
[ limit - j0-t - gg - v - pair - delocalization . ] ] limit @xmath63 @xmath64 pair delocalization .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + in the presence of two particle hopping , the lowest energy state has the form @xmath65 with coefficients @xmath66 in particular , for dominant pair hopping @xmath67 this is a state of delocalized pairs @xmath68 observe that this wavefunction is not equivalent to what one would navely understand as a `` pair condensate '' from analogy with the single - particle case @xmath69 instead the previous wavefunction is isomorphic to the bec of two bosons @xmath70 under the replacement of each boson with two atoms .
it is also interesting to remark that @xmath71 has larger pair - correlations than the state in eq .
( [ eq : likebec ] ) . [ [ limit - vtoinfty - an - insulator . ] ]
limit @xmath72 an insulator .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + reusing the previous wavefunction ( [ eq : lowen - t ] ) and taking the limit of dominant nearest - neighbor interaction @xmath20 we obtain two possible states . for strong repulsion
@xmath73 states @xmath74 and @xmath75 are favored , forming a charge density wave ( cdw ) with partial filling @xmath76 on the other hand , for strong nearest - neighbor attractions @xmath77 the particles are evenly distributed forming a mott insulator @xmath78 we have seen that four particles in a two - sites lattice recreate the exact wavefunction of an ordinary bec under the replacement of single bosons with pairs .
we can test this idea for slightly bigger lattices , diagonalizing numerically the hamiltonian which only contains the pair hopping term ( @xmath79 ) .
the resulting wavefunctions are compared side by side with the bec - like ansatz we mentioned . in the case of two particles
we get indeed the expected result @xmath80 whereas for @xmath61 particles in @xmath81 sites @xmath82 we find a disagreement between the ideal case of a bec - like state with coefficients @xmath83 and the exact diagonalization with @xmath84 @xmath85 and @xmath86 we observe that when compared to the ideal bec , our paired state breaks the translational symmetry , revealing an effective attraction between different pairs , that favors their clustering . in fig .
[ fig : comppsf ] we plot the projection between these states , namely the solution of eq .
[ eq : model ] with only @xmath87 and the _ ideal _ superfluid of pairs . in the nearby plot
we also analyze two relevant correlators that will be also used later on in the manuscript , namely , a single - particle coherence @xmath88 and the pair correlator @xmath89 as it is evident from the wavefunction and from the plots , there is no single - particle coherence or delocalization because particles move in pairs .
hence , @xmath90 the other correlator , @xmath91 which we identify with the delocalization of pairs is rather large and it only decreases with increasing the lattice size because the total pair density becomes smaller .
we now study the many - body physics of our model for a much larger number of particles using exact analytical methods . we begin with the regime in which the interaction terms @xmath19 and @xmath21 dominate , obtaining the different insulator phases on the @xmath92 plane . then , using perturbation theory , we compute the phase boundaries of these insulating regions for growing @xmath28 and @xmath22 finally , we study the properties of the ground state and its excitations in the superfluid phase , with @xmath93 and dominating @xmath49 proving indeed that this region describes a superfluid of pairs . to analyze the phase diagram it is convenient to work in the grand - canonical picture , in which the occupation is determined by the chemical potential @xmath94 in this picture the ground state is determined by minimizing the free energy @xmath95 where @xmath96 is the _ total _ number of particles , including both states @xmath8 and @xmath97 the free energy has a very simple form in the absence of tunneling @xmath98 \label{eq : interaction } \\ & = & \sum_k\left [ \frac{u}{4 } ( n_k + n_{k+1})^2 + v n_k n_{k+1 } - \left(\frac{2\mu + u}{4 } \right ) ( n_k + n_{k+1 } ) \right ]
. \nonumber\end{aligned}\ ] ] this function is defined over positive occupation numbers @xmath99 a discrete minimization will determine the different insulating phases and the regions where the system is stable against collapse . for a translational invariant system with periodic boundary conditions , all solutions can be characterized as a function of two integers @xmath100 representing the occupations of even @xmath101 and odd sites @xmath102 the optimization begins by noticing that the bond energy of two sites has a quadratic form @xmath103 where physical solutions are in the sector with @xmath104 for these occupation numbers to remain bounded , the bond energy @xmath105 has to increase as @xmath106 or both grow .
this gives us two conditions that need to be fulfilled to prevent collapse .
if these conditions are not met , the ground state will be an accumulation of all atoms in the same site . in that case
, the large interaction energies and the many body losses induced by the large densities will cause the breakdown of our model and quite possibly of the experimental setup .
the first stability condition is found by studying @xmath105 along the boundaries of our domain ( @xmath107 ) .
take for instance @xmath108 this gives a total energy @xmath109n.$ ] for this function to have a local minimum at finite @xmath110 we must impose @xmath111 the second condition comes from analyzing the interior of the domain . for this
, we take the only eigenvector of @xmath112 which lies in the region of positive occupation numbers , @xmath113 this line has an energy @xmath114x^2 - [ ( u+2\mu)/4]x$ ] and , to have again a finite local minimum , we require @xmath115 [ fig : stability ] : different regions of stability and insulating phases for @xmath116,title="fig : " ] given that eq .
( [ eq : restriction1 ] ) and eq .
( [ eq : restriction2 ] ) are satisfied , the system is stable and we have two possibilities to attain the minimum energy : either at the boundaries , @xmath117 or @xmath118 or right on the eigenvector of @xmath119 inspecting @xmath120 and @xmath121 we conclude that a positive value of @xmath21 will lead to the formation of charge density waves ( cdw ) of filled sites alternating with empty sites @xmath122 if @xmath123 our energy functional will be convex and the minimum energy state will be a mott insulator with @xmath124 when @xmath125 is even , or a charge density wave with @xmath126 when @xmath125 is odd .
the actual choice between these two insulating phases is obtained by computing the energy of both states @xmath127 having @xmath128 defines the value of @xmath129 at which the state with @xmath130 particles every two sites , a mott with @xmath131 particles , stops being the ground state and becomes more favorable to acquire an extra particle to form a cdw .
the boundaries of these insulating phases for @xmath132 are given by @xmath133 thus summing up , for @xmath134 the optimal occupation is @xmath131 particles per site , forming a mott , while for @xmath135 the occupation number is @xmath136 particles spread over every two sites , having a cdw . the results of this section are summarized in fig .
[ fig : stability ] . the previous calculation can be improved using perturbation theory for @xmath137 around the insulating phases , obtaining the phase boundaries around the insulators as @xmath10 and @xmath28 are increased .
this is done applying standard perturbation theory up to second order on both variables @xcite , using as unperturbed hamiltonian the operator ( [ eq : interaction ] ) and as perturbation the kinetic energy term @xmath138 + \mathrm{h.c . }
\right\}.\ ] ] we start calculating analytically the ground state energies of the first four insulating phases according to ( [ eq : interaction ] ) , considering the perturbation @xmath139 up to second order in @xmath140 for the cdw with @xmath141 and @xmath142 this energy is obviously zero @xmath143 for the mi with one particle per site we have virtual processes of the correlated hopping @xmath144 as environment - assisted hopping starts being allowed in an uniformly filled lattice @xmath145 for the cdw with @xmath146 and @xmath147 we find some doubly occupied sites and contributions from the pair hopping @xmath10 @xmath148 finally , for the mi with two particles per site , a calculation detailed in @xcite , @xmath149 here @xmath150 is the total number of sites and all results presented in this section are for the case @xmath151
@xmath152 and @xmath153 ( left ) varying @xmath10 for @xmath154 ( right ) varying @xmath28 for @xmath155 in both cases the lowest region is a cdw with alternating @xmath156 and @xmath7 particle occupation , followed upwards by a mott of one particle per site , a cdw with @xmath7 and @xmath157 particles and the highest area a mott of two particles.,title="fig : " ] @xmath152 and @xmath153 ( left ) varying @xmath10 for @xmath154 ( right ) varying @xmath28 for @xmath155 in both cases the lowest region is a cdw with alternating @xmath156 and @xmath7 particle occupation , followed upwards by a mott of one particle per site , a cdw with @xmath7 and @xmath157 particles and the highest area a mott of two particles.,title="fig : " ] at each value of @xmath158 the boundary of an insulating phase with average density @xmath159 is given by the degeneracy condition with a compressible state @xmath160 those points correspond to the chemical potential at which a hole , @xmath161 or a particle , @xmath162 can be introduced we show here the lower and upper limits of the first four insulating regions , corresponding to the cdw with @xmath163 @xmath164 the mott with one particle per site @xmath165 the cdw with @xmath166 @xmath167 and the mi with two particles per site @xmath168 the corresponding boundaries are plotted in fig .
[ fig : ptlimits ] . for small hopping amplitude
, they match the values that are found later on with the numerical methods .
but even for larger values , this approximation anticipates that the lobes are significantly larger for pair hopping @xmath10 than for the correlated hopping @xmath50 so far we have studied the many - body physics around the limit of strong interactions .
however , the main goal of this work is to understand the effect of correlated hopping and the creation of a pair superfluid . in absence of a mean field theory , but still in the limit of dominant two - body hopping @xmath169 we can use the number - phase representation , introduced in ref .
@xcite for an ordinary bec .
note , however , that the model in ref .
@xcite can not be directly applied here . following that reference
, one would assume a large number of particles per site , @xmath170 and introduce the basis of phase states @xmath171 @xmath172 using these states , one would then develop approximate representations for the operators @xmath173 @xmath174 and @xmath175 and diagonalize the resulting hamiltonian in the limit of weak interactions .
but after a few considerations one finds that the resulting phase model does not preserve an important symmetry of our system : if @xmath93 particles can only move in pairs and the parity of each site , @xmath176 is a conserved quantity . to describe correlated hopping
we must use a basis of states with fixed parity @xmath177 @xmath178 which is @xmath179 for the ground state we are interested in . as mentioned before
, we now have to find expressions for the different operators , @xmath173 @xmath174 and @xmath180 we use the fact that our states will have a density close to the average value @xmath159 and approximate the action of the operators over an arbitrary state as @xmath181 introducing the constant @xmath182 our hamiltonian becomes similar to the quantum rotor model @xcite @xmath183\nonumber\\ & & -(u + v)l\bar{n}^2.\end{aligned}\ ] ] for small @xmath19 and @xmath20 the ground state of this model is concentrated around @xmath184 expanding the hamiltonian up to second order in the phase fluctuations around this equilibrium point , we obtain a model of coupled harmonic oscillators .
this new problem can be diagonalized using normal modes that are characterized by a quasi - momentum @xmath185 @xmath186.$ ] the result is @xmath187 with normal frequencies @xmath188 and a global energy shift @xmath189 it is evident from eq .
( [ pm_freq ] ) that our derivation is only self consistent for negative values of @xmath190 otherwise , when @xmath191 some of the frequencies become imaginary , signaling the existence of an unbounded spectrum of modes with @xmath192 and that our ansatz becomes a bad approximation of the ground state .
this strictly means that our choice @xmath193 only applies in the case of attractive nearest neighbor interactions , @xmath194 as we know that this interaction can not destabilize a translational invariant solution such as the uniform mott insulator .
however , it does not mean by itself that the whole system becomes unstable for @xmath191 indeed , we will show numerically that it remains essentially in a similar phase for all values of @xmath20 but in the case of @xmath191 the insulating phases are stable until values of the hopping slightly higher as in the @xmath195 case .
if we focus on the regime of validity , we will find that the spectrum is very similar to that of a condensate . at small momenta
the dispersion relation becomes linear , @xmath196 with sound velocity @xmath197 while at larger energies the spectrum becomes quadratic , corresponding to `` free '' excitations with some mass .
this a consequence of the similarity between our approximate model for the pairs ( [ secondpm_h ] ) and the phase model for a one - dimensional condensate .
however , we can go a step further and conclude that the similarity extends also to the wavefunctions themselves , so that the state of a pair superfluid can be obtained from that of an ordinary bec by the transformation @xmath198 this is indeed consistent with what we obtained for the diagonalization of a two - particle state in the limit @xmath199 [ see eq .
( [ condensate2 ] ) ] .
the previous sections draw a rather complete picture of the possible ground states in our model . in the limit of strong interactions we find both uniform insulators and a breakdown of translational invariance forming a cdw , while for dominant hopping we expect both single - particle superfluidity and a new phase , a pair superfluid .
we now confirm these predictions using two different many - body variational methods .
we focus on ( left ) @xmath93 and ( right ) @xmath200 separately .
upper plots show average site occupation @xmath201 ( grayscale ) , and number variance @xmath202 ( contour ) .
lower plots show single - particle coherence @xmath25 ( grayscale ) together with @xmath203 ( contour).,title="fig : " ] we focus on ( left ) @xmath93 and ( right ) @xmath200 separately .
upper plots show average site occupation @xmath201 ( grayscale ) , and number variance @xmath202 ( contour ) .
lower plots show single - particle coherence @xmath25 ( grayscale ) together with @xmath203 ( contour).,title="fig : " ] we focus on ( left ) @xmath93 and ( right ) @xmath200 separately .
upper plots show average site occupation @xmath201 ( grayscale ) , and number variance @xmath202 ( contour ) .
lower plots show single - particle coherence @xmath25 ( grayscale ) together with @xmath203 ( contour).,title="fig : " ] we focus on ( left ) @xmath93 and ( right ) @xmath200 separately .
upper plots show average site occupation @xmath201 ( grayscale ) , and number variance @xmath202 ( contour ) .
lower plots show single - particle coherence @xmath25 ( grayscale ) together with @xmath203 ( contour).,title="fig : " ] the first method that we use is a variational estimate of ground state properties based on a product state @xcite @xmath204 minimizing the expectation value of the free energy @xmath205 with respect to the variables @xmath206 under the constrain of fixed norm @xmath207 we will obtain the phase diagram in the phase space of interactions and chemical potential @xmath208 in our study we have made several simplifications .
first of all , we assumed period - two translational invariance in the wavefunction , using only two different sets of variational parameters , @xmath209 and @xmath210 in our experience , this is enough to reproduce effects such as the cdw .
next , since @xmath211 is required for the stability of the system , we have taken @xmath212 as unit of energy .
the limit @xmath213 is approximated by the limits @xmath214 in our plots . finally , in order to determine the roles of @xmath28 and @xmath49 we have studied the cases @xmath93 and @xmath215 separately .
we focus on ( left ) @xmath93 and ( right ) @xmath200 separately .
upper plots show average site occupation @xmath201 ( grayscale ) and number variance @xmath202 ( contour ) .
lower plots show single - particle coherence @xmath25 ( grayscale ) together with @xmath203 ( contour).,title="fig : " ] we focus on ( left ) @xmath93 and ( right ) @xmath200 separately .
upper plots show average site occupation @xmath201 ( grayscale ) and number variance @xmath202 ( contour ) .
lower plots show single - particle coherence @xmath25 ( grayscale ) together with @xmath203 ( contour).,title="fig : " ] we focus on ( left ) @xmath93 and ( right ) @xmath200 separately .
upper plots show average site occupation @xmath201 ( grayscale ) and number variance @xmath202 ( contour ) .
lower plots show single - particle coherence @xmath25 ( grayscale ) together with @xmath203 ( contour).,title="fig : " ] we focus on ( left ) @xmath93 and ( right ) @xmath200 separately .
upper plots show average site occupation @xmath201 ( grayscale ) and number variance @xmath202 ( contour ) .
lower plots show single - particle coherence @xmath25 ( grayscale ) together with @xmath203 ( contour).,title="fig : " ] the results are shown in fig .
[ fig : gwminus ] and fig .
[ fig : gwplus ] for @xmath195 and @xmath216 respectively .
the first interesting feature is that , as predicted by perturbation theory , we have large lobes both with integer @xmath217 and with fractional @xmath218 occupation numbers , forming uniform mott insulators and cdw , respectively .
the insulators are characterized by having a well defined number of particles per site , and thus no number fluctuations @xmath219 while the size of the lobes does not depend dramatically on the sign of @xmath20 these are significantly larger for the pair hopping @xmath10 than for the correlated hopping @xmath144 as already seen with perturbation theory . the boundary of the insulating areas marks a second order phase transition to a superfluid regime , where we find number fluctuations @xmath220 in order to characterize these gapless phases we have computed the order parameter of a single - particle condensate @xmath221 and two quantities that we use to detect pairing .
the first one is a two - particle correlation that generalizes the order parameter of a bec to the case of a pair - bec @xmath222 the second quantity @xmath223 is used to correct the previous value eliminating the contribution that may come from a single - particle condensate coexisting with the pair - bec . when @xmath93 we always find that @xmath224 even outside the insulating lobes .
this marks the absence of a single - particle bec , which is expected since we do not have single - particle hopping . on the other hand
, we now find long range coherence of the pairs and thus @xmath225 all over the non - insulating area , which we identify with the pair - superfluid regime .
the situation is slightly different for @xmath226 the single - particle order parameter @xmath25 no longer vanishes in the superfluid area , denoting the existence of single - particle coherence , but at the same time we find that the two - particle correlations exceed the contribution from the single - particle superfluid as @xmath227 which we attribute to a coexistence of both a single - particle and a pair - superfluid , or a state with both features .
this picture does not change substantially when @xmath21 is positive or negative .
the only differences are in the insulating regions , where the cdw is either due to the incommensurability of the particle number @xmath228 or really gives rise to the separation of particles alternating holes and filled sites @xmath229 however , in the superfluid regime we find no significant changes and in particular we see no breaking of the translational invariance or modulation of the coherent phase .
the previous numerical simulations are very simple and can not fully capture the single particle and two - particle correlators . to complete and verify the full picture we have searched the ground states of the full hamiltonian using the so called itebd algorithm , which uses an infinite matrix product state ansatz together with imaginary time evolution @xcite .
and on the right for @xmath230 as a function of two - particle tunneling @xmath22 the upper row shows the density ( dashed ) and particle number fluctuations ( solid ) , while the lower row shows @xmath231 ( solid ) and two - particle coherence ( dashed ) , which overlap indicating @xmath232 for each set of operators we show three lines , for @xmath233 and @xmath234 corresponding to plain line , circle and star , respectively.,title="fig : " ] and on the right for @xmath230 as a function of two - particle tunneling @xmath22 the upper row shows the density ( dashed ) and particle number fluctuations ( solid ) , while the lower row shows @xmath231 ( solid ) and two - particle coherence ( dashed ) , which overlap indicating @xmath232 for each set of operators we show three lines , for @xmath233 and @xmath234 corresponding to plain line , circle and star , respectively.,title="fig : " ] and on the right for @xmath230 as a function of two - particle tunneling @xmath22 the upper row shows the density ( dashed ) and particle number fluctuations ( solid ) , while the lower row shows @xmath231 ( solid ) and two - particle coherence ( dashed ) , which overlap indicating @xmath232 for each set of operators we show three lines , for @xmath233 and @xmath234 corresponding to plain line , circle and star , respectively.,title="fig : " ] and on the right for @xmath230 as a function of two - particle tunneling @xmath22 the upper row shows the density ( dashed ) and particle number fluctuations ( solid ) , while the lower row shows @xmath231 ( solid ) and two - particle coherence ( dashed ) , which overlap indicating @xmath232 for each set of operators we show three lines , for @xmath233 and @xmath234 corresponding to plain line , circle and star , respectively.,title="fig : " ] roughly , this ansatz is based on an infinite contraction of tensors that approximates the wavefunction of a translational invariant system in the limit of infinite size .
adapting the ansatz to our problem we write it as @xmath235 here the @xmath236 and @xmath237 are matrices that depend on the state of the odd and even sites they represent , a dependence which is signaled by the @xmath238 and @xmath239 in the previous equation .
these matrices are contracted with one - dimensional vectors of positive weights @xmath240 which are related to the coefficients of the schmidt decomposition .
this variational ansatz is known to work well for states with fast decaying correlations , but it also gives a good qualitative description of the critical phases . in order to optimize the itebd wavefunction we performed an approximate imaginary time evolution using a trotter decomposition and local updates of the associated tensors , as described in ref .
@xcite . using the canonical forms for these tensors
it is also straightforward to compute expectation values for different operators acting either on neighboring or separated sites . in fig .
[ fig : mps ] we plot the most relevant results for three cuts across the phase diagram , @xmath241 and @xmath242 so that each line crosses both an insulating plateau and the superfluid region .
we have used small tensor sizes from @xmath243 up to @xmath244 a value limited by the need of using large cutoffs for the site populations ( @xmath245 ) . as shown in the figures , when @xmath93 the single - particle correlator is zero for distinct sites , and we are left only with two - particle correlations . in the mi case
the pair correlations between neighboring sites decrease very quickly , while in the superfluid regime we see a critical behavior @xmath246 with an exponent that varies between @xmath247 and @xmath248 depending on the simulation parameters . vs. site separation , for @xmath249 @xmath250 @xmath251 @xmath252 @xmath253 @xmath254 , from top to bottom .
the line with @xmath255 ( solid , circles ) corresponds to a mi state and correlations decay to zero on the third site .
the remaining lines have been rescaled and plotted in log - log scale in the inset , which shows that the two - particle correlator decays approximately as @xmath256 with a power @xmath257
there are many ways of differentiating the phases we have found , each one having its own degree of difficulty .
the simplest and best established detection methods are related to the insulating regimes .
these phases , which involve both mi and cdw , are characterized by having a well defined number of particles at each lattice site , the lack of coherence and an energy gap that separates the insulator from other excitations .
the energy gap in these insulators may be probed either by static or spectroscopic means as it has been done in experiments @xcite , determining that indeed the system is insulating .
second , the lack of coherence will translate into featureless time - of - flight images , having no interference fringes @xcite at all .
even though there will be no fringes , the measured density will be affected by quantum noise .
the analysis of the noise correlation will show peaks at certain momenta @xcite that depend on the periodicity of the state , so that the number of peaks in the cdw phase will be twice those of the mi . in case of having access to the lattice sites , as in the experiments with electron microscopy @xcite , or in future experiments with large aperture microscopic objectives that can collect the fluorescence of individual lattice sites @xcite
, the discrimination between the mi and the cdw should be even easier , since in one case we have a uniform density and in the other a periodic distribution of atoms .
when the system enters a superfluid phase , it becomes a perfect `` conductor '' with a gapless excitation spectrum .
the lack of an energy gap , should be evident in the spectroscopic experiments suggested before .
however , we are not only interested in the superfluid nature , but rather in the fact that this quantum phase is strongly paired .
more precisely , we have found that for @xmath93 the single - particle coherence is small or zero , and that the two - particle correlator decays slowly @xmath258 the first equation implies that the time of flight images will reveal no interference fringes and will exhibit noise correlations which will be similar to the mi . in order to probe @xmath259 and confirm the pairing of the particles , we suggest to use raman photo - association to build molecules out of pairs of atoms @xcite .
for an efficient conversion , it would be best to perform an adiabatic passage from the free atoms to the bound regime . as described in @xcite
, we expect a mapping that goes from @xmath260 where @xmath130 is the number of bosons and @xmath131 the number of molecules .
more precisely , we expect the @xmath261 operator to be mapped into @xmath262 so that the pair coherence of the original atoms translates into the equivalent of @xmath263 for the molecules .
this order should reveal as an interference pattern in time - of - flight images of the molecules .
finally , in cases with @xmath57 we have found the coexistence of single - particle and two - particle coherences .
this translates also into the coexistence of interference fringes with nonzero pair correlators .
summing up , we have suggested a family of experiments with cold atoms that would produce correlated hopping of bosons .
the mechanism for the correlated hopping is an asymmetry in the contact interaction between atoms .
this asymmetry is exploited by trapping the atoms in dressed states , a configuration that gives rise to transport induced by collisions .
the main result of this paper is that there is a huge variety of interaction asymmetries that will give rise to long range pair correlations via interaction induced transport .
formally , in the resulting effective models we recognize two dynamical behaviors .
if we have a nonzero asymmetry in the interspecies interactions , @xmath264 the hamiltonian will exhibit pair hopping , while an asymmetry in the intra - species scattering lengths @xmath265 gives rise to correlated hopping . however , we have given enough evidence that * both * hamiltonian terms give rise to a novel quantum phase which we call pair superfluid .
this phase is characterized by a gapless spectrum with a finite sound speed , zero single - particle correlations and long range pair coherence .
all quantum phases are connected by second order quantum phase transitions .
these phases can be produced and identified using variations of current experiments @xcite .
the nonperturbative nature of the effect should help in that respect .
our ideas are not restricted to one dimension .
it is possible to engineer also a two
body hopping using two dimensional lattice potentials .
again , the basic ingredients would be atoms with an asymmetric interaction and an optical lattice that traps two states , @xmath8 and @xmath9 with a relative displacement .
both in the one and two dimensional cases it is a valid question to ask whether the coupling between different trapped states , @xmath266 can excite also transitions to higher bands , processes that have not been considered in the paper .
our answer here is no .
there are only two sources of coupling to higher energy bands .
one is the interaction , but we are already assuming that the interaction energies are much smaller than the band separation . following the notation from ref .
@xcite , we have the constraint that the interaction energy should be smaller than the energy separation to the first excited state in a well of the periodic potential , @xmath267 the same requisite as for ordinary bose hubbard models @xcite
. the other source of coupling to higher bands would be single particle hopping .
however , unlike @xcite , here we are assuming that these terms are strongly suppressed compared with the interaction . in other words , realizing the models that we suggest in this paper , for realistic densities
, @xmath268 and simple potentials , imposes no further constraint in current experiments .
finally , let us remark that transport inducing collisons may be implemented using other kinds of spin - dependent interactions .
for instance , correlated hopping appears naturally in state - dependent lattices loaded with spinor atoms , because their interactions can change the hyperfine state of the atoms while preserving total angular momentum @xcite .
we would like to thank miguel angel martn - delgado for useful discussions .
m.e . acknowledges support from the conquest project .
j.j.g.r acknowledges financial support from the ramon y cajal program of the spanish m.e.c .
, from u.s .
nsf grant no .
phy05 - 51164 and from the spanish projects fis2006 - 04885 and cam - ucm/910758 .
as discussed before , the main idea behind atomic correlated hopping is to trap atoms whose interaction allows them to change their state . in this section
we provide one possible implementation of this idea using state dependent superlattices that trap dressed states .
our starting point is the setup in fig .
[ fig : superlattice]a , which is itself taken from ref .
it consists on an optical lattice trapping atoms in states @xmath33 and @xmath269 together with a raman coupling between these states .
mathematically , this configuration is described by the single - particle hamiltonian @xmath270 by moving to the basis of dressed states @xmath271 we find that the trapping is effectively equivalent to two superlattices with a relative displacement as in fig .
[ fig : superlattice]b @xmath272 under appropriate circumstances , discussed in @xcite , we will find that each superlattice site has a unique ground state , energetically well differentiated from the next excited state , and which consists of a symmetric wavefunction spanning both lattice wells .
if this is the case and if all energy scales , such as the interaction and the hopping , are small compared to the separation between bloch bands , we can expand the bosonic field operators describing the atoms in terms of these localized wavefunctions @xmath273 where @xmath274 is the superlattice period , @xmath275 are bosonic operators that , for @xmath28 even ( odd ) , annihilate an atom in state @xmath8 ( @xmath9 ) in the @xmath28-th superlattice cell @xmath276 and the localized wavefunctions @xmath277 are a superposition of the wannier functions @xmath278 of the underlying lattice @xmath279.\ ] ] we will now express the interaction ( [ eq : interaction2 ] ) in the basis of dressed states .
we proceed using the change of variables in eq .
[ eq : dressed ] to find the expression of the densities @xmath280 the first obvious conclusion is that the total density is independent of the basis on which it is written , @xmath281 hence , the term of @xmath282 is insensitive to the state of the atoms . on the other hand ,
the asymmetric terms are not so simple .
the @xmath43 interaction , which is proportional to the product of densities @xmath283 gives rise to a scattering that changes the state of interacting atoms from @xmath9 to @xmath8 and viceversa , as in fig .
[ fig : collisions]a .
the term of @xmath284 has a lightly different effect , @xmath285:\ ] ] it gives rise to processes where one atom changes its state influenced by the surrounding environment . in the following subsections
we will see what happens to the interaction terms ( [ eq : total - density ] ) , ( [ eq : asym ] ) and ( [ eq : imbalance ] ) , when the atoms are confined in a lattice .
in this section we will put the previous results of this appendix together .
we will take the tight - binding expansion of the field operators ( [ eq : tight - binding ] ) and use it together with eqs .
( [ eq : total - density ] ) , ( [ eq : asym ] ) and ( [ eq : imbalance ] ) to expand the interaction hamiltonian ( [ eq : interaction2 ] ) . for convenience
, we will rename the bosonic operators as @xmath286 according to the position at which their wannier functions are centered ( see fig . [
fig : superlattice]c ) . along the derivation ,
one obtains many integrals of ground state wavefunctions @xmath287 we will only keep those integrals with a separation smaller than a superlattice period .
taking eq .
( [ eq : w ] ) , the expression for the superlattice localized states , one obtains @xmath288 where @xmath46 are the wannier wavefunctions of the underlying sublattice . using these tools ,
the symmetric interaction term becomes @xmath289 for the asymmetric parts , we first use eq .
( [ eq : asym ] ) obtaining @xmath290\end{aligned}\ ] ] and then finally the more complicated eq .
( [ eq : imbalance ] ) @xmath291 introducing constants that parameterize the on - site interactions and the strength of the underlying lattice ( [ int - constants ] ) , our final hamiltonian looks as follows @xmath292.\end{aligned}\ ] ] completing terms and replacing the sum over @xmath293 with a sum over nearest neighbors , we arrive at the desired model ( [ eq : model ] ) with the parametrization given already in eq .
( [ eq : paramet ] ) . | in this work we analyze a particular setup with ultracold atoms trapped in state - dependent lattices . we show that any asymmetry in the contact interaction translates into one of two classes of correlated hopping . after deriving the effective lattice hamiltonian for the atoms , we obtain analytically and numerically the different phases and quantum phase transitions .
we find for weak correlated hopping both mott insulators and charge density waves , while for stronger correlated hopping the system transitions into a pair superfluid .
we demonstrate that this phase exists for a wide range of interaction asymmetries and has interesting correlation properties that differentiate it from an ordinary atomic bose - einstein condensate . |
identifying codes were introduced by karpovsky et al . @xcite in 1998 to model fault diagnosis in multiprocessor systems . if we model a multiprocessor as an undirected simple graph @xmath0 , then an _
@xmath11-id code _ is a subset of the vertices of @xmath0 having the property that every collection of at most @xmath12 vertices has a non - empty and distinct set of code vertices that are distance at most @xmath13 from it . to be precise ,
let @xmath14 $ ] be the set of vertices that are within distance @xmath13 of @xmath15 ( called the `` closed @xmath13-neighborhood '' ) .
an @xmath11-id code is a subset @xmath16 of @xmath17 such that @xmath14\cap \code$ ] and @xmath18\cap \code$ ] are distinct and non - empty for all distinct subsets @xmath19 and @xmath20 with @xmath21 . in this paper
, we consider the case @xmath22 , which we denote simply as _
@xmath12-id codes _ ; we also write @xmath5 $ ] for @xmath23 $ ] .
not every graph has an @xmath12-id code .
for example , if @xmath0 contains two vertices @xmath24 and @xmath6 such that @xmath25 = n[v]$ ] , then @xmath0 can not have even a 1-id code , since for any subset of vertices @xmath16 we have @xmath25\cap \code = n[v]\cap \code$ ] .
however , if , for every pair of subsets @xmath26 with @xmath27 , we have @xmath28\not = n[y]$ ] , then @xmath0 has an @xmath12-id code , since @xmath29 is such a code .
hence we are usually interested in finding an @xmath12-id code of minimum cardinality .
the most studied case is when @xmath30 . in this case
, @xmath16 is an @xmath31-id code if and only if for all distinct vertices @xmath24 and @xmath6 in g , the intersections @xmath25\cap \code$ ] and @xmath5\cap \code$ ] are distinct and nonempty
. for a fixed subset of vertices @xmath16 , we say that vertices @xmath24 and @xmath6 are _ distinguishable _ if @xmath25\cap\code \ne n[v]\cap\code$ ] .
much work has focused on finding @xmath31-id codes for infinite grids , see @xcite . to measure how small an i d code can be , we talk about the `` density '' of an infinite grid , which , roughly speaking , is the fraction of the vertices in the graph that are in the code ( we give a formal definition of density after we prove proposition 1 ) . in 1998 ,
karpovsky , chakrabarty , and levitin @xcite considered the 6-regular , 4-regular , and 3-regular infinite grids that come from the tilings of the plane by equilateral triangles , squares , and regular hexagons .
they asked the question `` what is the minimum density of an identifying code for each grid ? ''
a short proof shows that the answer for the 6-regular grid is density @xmath32 . for the 4-regular grid , cohen et .
al @xcite constructed codes with density @xmath33 and ben - haim and litsyn @xcite proved that @xmath33 is best possible . for the 3-regular grid
, the minimum density remains unknown .
the best upper bound is @xmath8 , which comes from two codes constructed by cohen et .
al @xcite ; these same authors also proved a lower bound of @xmath34 . in this paper , we improve the lower bound to @xmath35 . before we prove our main result , which is theorem [ mainthm ] , we first prove a weaker lower bound , given in proposition [ prop1 ] .
the proof of proposition [ prop1 ] is instructive , because the proof is easy , yet the proofs of theorem [ mainthm ] and proposition [ prop1 ] use the same core idea .
we call the components of @xmath36 $ ] _ clusters _ and a cluster with @xmath37 vertices is a _ @xmath37-cluster _ (
a @xmath38-cluster has @xmath37 or more vertices ) .
[ prop1 ] the density of every vertex identifying code for the infinite hexagonal grid is at least 2/5 .
our proof is by the discharging method .
thus , @xmath16 must contain at least @xmath39 of the vertices .
we assign to each vertex @xmath6 in the identifying code @xmath16 a charge of 1 .
we will redistribute the charges , without introducing any new charge , so that every vertex ( whether in @xmath16 or not ) has charge at least 2/5 .
we redistribute the charge according to the following discharging rule : * if @xmath40 is adjacent to @xmath41 and @xmath42 has @xmath43 neighbors in @xmath16 , then @xmath6 gives charge @xmath44 to @xmath42 .
now we simply verify that after applying the discharging rule , each vertex has charge at least @xmath39 .
if @xmath45 , then @xmath6 receives charge @xmath46 . if @xmath6 is a 1-cluster , then note that each neighbor @xmath42 of @xmath6 has at least one other neighbor in @xmath16 ( otherwise @xmath6 and @xmath42 are indistinguishable ) .
so @xmath6 gives each neighbor charge at most @xmath47 .
thus , @xmath6 retains charge at least @xmath48 .
it is easy to see that @xmath16 can not contain 2-clusters .
let @xmath49 be a @xmath50-cluster , and let @xmath6 be a vertex in @xmath49 . if @xmath51 , then @xmath6 has two neighbors @xmath52 and @xmath53 not in @xmath16 .
since @xmath52 and @xmath53 are distinguishable , at least one of them has another neighbor in @xmath16 .
so the total charge @xmath6 gives to @xmath52 and @xmath53 is at most @xmath54 ; thus @xmath6 retains charge at least @xmath55 .
if @xmath56 , then @xmath6 gives away charge at most @xmath39 , so @xmath6 retains charge at least @xmath57 .
finally , if @xmath58 , then @xmath6 gives away no charge , so @xmath6 retains charge 1 . since the charge at each vertex ( whether in @xmath16 or not ) is at least @xmath39 , the density of @xmath16 is at least @xmath39 .
it is instructive to note that our proof does not rely on the structure of the grid , but only that it is 3-regular .
in fact , proposition [ prop1 ] is a special case of a more general lower bound , which follows from a similar proof : any 1-id code for any @xmath43-regular graph has density at least @xmath59 . when we study the proof of proposition [ prop1 ] , it is natural to look for `` slack '' , i.e. , vertices that have charge greater than @xmath39 after the discharging phase .
of course every vertex @xmath6 in a @xmath50-cluster @xmath49 with at least two neighbors in @xmath49 has slack .
it is clear that every @xmath50-cluster @xmath49 contains at least one such slack vertex .
our plan is to distribute this excess charge at the slack vertices among the vertices close to @xmath6 .
we must also verify that each 1-cluster and each vertex not in @xmath16 are close to some @xmath50-cluster .
this approach forms the outline of the proof of theorem [ mainthm ] .
we think our proof of theorem [ mainthm ] could be refined to give a better lower bound .
however , we think that bound would be only slightly better than theorem [ mainthm ] , and that the proof would be much more complicated , so we have not attempted it .
now we formally define density .
we fix an arbitrary vertex @xmath60 and let @xmath61 denote all the vertices that are distance at most @xmath62 from @xmath6 .
the _ density _ of a code @xmath16 is defined as @xmath63 we note that since each vertex in @xmath61 finishes with charge at least @xmath35 , the sum of the charges at vertices in @xmath61 is at least @xmath64 .
we should remark that some vertices in @xmath61 may receive charge from vertice in @xmath65 .
however , the sum of the charges sent from vertices in @xmath65 to vertices in @xmath61 is linear in @xmath62 , whereas @xmath66 is quadratic in @xmath62 .
thus , we see that @xmath35 is a lower bound on the density of @xmath16 .
the organization of the rest of the paper is as follows . in section
2 we prove the main result .
this proof consist of stating six discharging rules , and verifying that after the discharging phase , each vertex has charge at least @xmath35 .
showing that each vertex finishes with sufficient charge is a lengthy task . to simplify the analysis , we state and
prove five structural lemmas and three claims about the discharging process .
the difference between our claims and our structural lemmas is that the claims are statements about our discharging rules , whereas the lemmas are statements about any 1-id code in the infinite hexagonal grid .
so to simplify the proof of the main result , we defer the proofs of the five structural lemmas until section 3 .
let @xmath6 be a 1-cluster .
if @xmath6 has a neighbor @xmath67 , such that all three neighbors of @xmath24 are in @xmath16 , then we say that @xmath6 is _ crowded _ ; otherwise @xmath6 is _
uncrowded_. let @xmath49 be a 3-cluster and let @xmath42 be the non - leaf vertex of @xmath49 ; we call @xmath42 the _ center _ of @xmath49 .
if the neighbor of @xmath42 that is not in @xmath49 has no other neighbor in @xmath16 , then we call @xmath49 an _ open _ 3-cluster and
we call @xmath42 an _ open _ center
. otherwise we call @xmath49 a _ closed _ 3-cluster . for each leaf @xmath6 of @xmath49
, there must exist a @xmath68 at distance two from @xmath6 and not in @xmath49 ( otherwise the two neighbors of @xmath6 not in @xmath49 would be indistinguishable ) .
however , a leaf may have two or more such vertices at distance two .
if any vertex @xmath6 in @xmath49 , leaf or center , has at least two vertices @xmath69 , @xmath70 at distance two ( and neither @xmath71 nor @xmath72 is in @xmath49 ) , then we call @xmath49 _ crowded _ ; otherwise @xmath49 is _ uncrowded_. the significance of a closed or crowded 3-cluster @xmath49
is that @xmath49 will have extra help when sending charge to its adjacent vertices .
we say that an uncrowded 1-cluster @xmath73 is _ nearby _ a @xmath50-cluster @xmath74 if @xmath73 is within distance three of either a @xmath75-cluster @xmath74 , a closed 3-cluster @xmath74 , or an open center of a 3-cluster @xmath74 .
similarly , we say that an uncrowded open 3-cluster @xmath73 is _ nearby _ a @xmath50-cluster @xmath74 if @xmath73 is within distance three of either a @xmath75-cluster @xmath74 or a closed 3-cluster @xmath74 , or if both its leaves are within distance three of an open 3-cluster @xmath74 .
we will show in lemma [ lem1 ] that each uncrowded 1-cluster @xmath6 is nearby a @xmath50-cluster .
if an open 3-cluster @xmath49 is uncrowded , has no open 3-clusters within distance two , and has no closed 3-clusters or @xmath75-clusters within distance three , then we say that @xmath49 is _
threatened_. if an uncrowded 1-cluster has no @xmath75-cluster within distance three and has no nearby unthreatened 3-cluster , then we say that @xmath6 is _
threatened_. if a threatened 3-cluster @xmath49 has at least four nearby threatened 1-clusters and threatened 3-clusters , then we say that @xmath49 is _
needy_. whereas clusters that are closed or crowded already have extra help sending charge , clusters that are uncrowded , threatened , or needy will likely need to receive extra charge from elsewhere .
[ mainthm ] the density of every vertex identifying code for the infinite hexagonal grid is at least 12/29 .
our proof is by discharging .
we assign to each vertex @xmath6 in the identifying code @xmath16 a charge of 1 .
we will redistribute the charges , without introducing any new charge , so that every vertex ( whether in @xmath16 or not ) has charge at least 12/29 . the outline of the proof is as follows . consider a vertex @xmath6 not in @xmath16 .
let @xmath43 be the number of neighbors of @xmath6 that are in @xmath16 .
vertex @xmath6 will receive charge @xmath76 from each of its @xmath43 neighbors ( this is rule 1 , in the discharging rules below ) .
thus every vertex not in @xmath16 receives charge @xmath35 . clearly
every neighbor of a 1-cluster @xmath6 must have at least two neighbors in @xmath16 ; thus , each neighbor of @xmath6 will receive charge at most @xmath77 from @xmath6 .
if @xmath6 is uncrowded , then @xmath6 sends charge @xmath77 to each of its three neighbors , so @xmath6 will be left with charge @xmath78 .
hence @xmath6 needs more charge .
our plan is to send charge 1/29 to each uncrowded 1-cluster @xmath6 from a nearby @xmath50-cluster @xmath49 ; we will do this via rules 24 below .
we will also need to prove that such a @xmath50-cluster @xmath49 does not send charge to too many uncrowded 1-clusters . in verifying that each vertex finishes with charge at least @xmath35 it is convenient to count the charges of vertices in a single cluster together ; i.e. , for each cluster with @xmath79 vertices , we simply verify that the sum of the final charges of the vertices in that cluster is at least @xmath80 .
note that the charge that a closed 3-cluster @xmath49 gives away by rule 1 is at most @xmath81 .
since @xmath49 begins with charge @xmath82 and needs to keep charge @xmath83 , @xmath49 can afford to give away another @xmath84 to nearby clusters .
in contrast , the charge that an open 3-cluster @xmath85 gives away by rule 1 may perhaps be as much as @xmath86 .
thus , @xmath85 can only afford to give away another @xmath87 to nearby clusters .
below , we define rules 24 so that each uncrowded 1-cluster receives charge @xmath88 from some nearby @xmath50-clusters .
although rules 24 ensure sufficient charge for each uncrowded 1-cluster , in some cases they unfortunately require open 3-clusters to give away a total charge of @xmath89 to nearby 1-clusters .
giving away this additional charge may result in a needy 3-cluster @xmath85 with remaining charge only @xmath90 ( rather than the necessary @xmath83 ) .
thus , we add rule 5 , which supplies these needy 3-clusters with the necessary charge ( from nearby open @xmath91-clusters ) . * discharging rules * 1 .
each vertex @xmath92 that has @xmath43 neighbors in @xmath16 receives charge @xmath76 from each neighbor in @xmath16 .
each uncrowded 1-cluster that is nearby a @xmath75-cluster @xmath49 receives charge @xmath88 from @xmath49 .
each uncrowded 1-cluster @xmath6 that is nearby a closed @xmath91-cluster @xmath49 , and has not received charge by rule 2 , receives charge @xmath88 from @xmath49 .
each uncrowded 1-cluster @xmath6 that is nearby an open center in an open @xmath91-cluster @xmath49 , and has not received charge by rule 2 or 3 , receives charge @xmath88 from @xmath49 . however
, if @xmath6 is nearby a crowded 3-cluster @xmath49 , then @xmath6 receives charge 1/29 from @xmath49 and receives no charge from any other cluster .
similarly , if @xmath6 is not nearby a crowded 3-cluster , but lies on a 6-cycle with an open center in cluster @xmath49 , then @xmath6 receives charge 1/29 from @xmath49 and receives no charge from any other cluster .
if @xmath73 is a needy , open 3-cluster then it has both its leaves within distance three of another open 3-cluster @xmath74 ; @xmath73 receives charge 1/29 from @xmath74 .
however , if @xmath73 and @xmath74 are both uncrowded and they each have both leaves within distance three of the other cluster , then neither cluster sends charge to the other .
we say that such @xmath73 and @xmath74 are _ paired with each other_. note that in each rule , if cluster @xmath73 receives charge from cluster @xmath74 , then @xmath73 is nearby @xmath74 .
this is a necessary , though not sufficient , condition for receiving charge .
before we verify that each vertex has final charge at least @xmath35 , we state five structural lemmas about the relationships between uncrowded , threatened , and needy clusters and their nearby @xmath50-clusters .
we defer the proofs of these lemmas until after we complete the discharging argument .
we separate these lemmas from the rest of the present proof because they make no mention of discharging rules .
thus , they may be useful in proving a stronger lower bound .
[ lem1 ] every uncrowded 1-cluster is nearby a @xmath50-cluster .
[ lem2 ] every closed 3-cluster @xmath49 is nearby at most ten 1-clusters and open 3-clusters .
if @xmath49 is nearby exactly ten such clusters , then @xmath49 is crowded .
[ lem5 ] every needy 3-cluster has both leaves within distance three of a @xmath50-cluster .
let @xmath73 and @xmath74 be uncrowded , open 3-clusters that are paired with each other .
clusters @xmath73 and @xmath74 have at most 7 nearby threatened 1-clusters and threatened 3-clusters . if they have exactly 7 such nearby clusters , then they also have a nearby closed 3-cluster or @xmath75-cluster .
[ lem6 ] [ lem3 ] every cluster @xmath49 with @xmath79 vertices has at most @xmath93 nearby clusters .
now we verify that after the discharging phase , every vertex ( whether in @xmath16 or not ) has charge at least @xmath35 ; this proves that @xmath16 has density at least @xmath35 .
we write @xmath94 or @xmath95 to denote the charge at @xmath6 or @xmath49 after the discharging phase .
suppose @xmath45 . by rule 1 , @xmath96 , where @xmath43 is the number of neighbors of @xmath6 in @xmath16 .
now let @xmath6 be a crowded 1-cluster .
one of the neighbors @xmath24 of @xmath6 has three neighbors in @xmath16 , so @xmath24 receives only charge @xmath89 from @xmath6 .
hence @xmath97 .
finally , let @xmath6 be an uncrowded 1-cluster .
by lemma [ lem1 ] and rules 24 , @xmath6 receives charge @xmath88 from some nearby @xmath50-cluster .
thus , @xmath98 .
let @xmath49 be a closed 3-cluster ; let @xmath24 and @xmath6 be the leaves of @xmath49 , and let @xmath42 be the center . since we can distinguish between the neighbors of @xmath24 ( not in @xmath49 ) , at least one of them has a neighbor in @xmath16 other than @xmath24 ; similarly for the neighbors of @xmath6 .
since @xmath49 is closed , the neighbor of @xmath42 not in @xmath49 has another neighbor in @xmath16 .
thus , the charge given from @xmath49 to adjacent vertices is at most @xmath99 .
if @xmath49 is uncrowded , then , by lemma [ lem2 ] , @xmath49 gives charge @xmath88 to at most nine nearby clusters ; thus @xmath100 .
if @xmath49 is crowded , then the charge @xmath49 gives to adjacent vertices is at most @xmath101 .
by lemma [ lem2 ] , @xmath49 gives charge to at most ten nearby clusters , so @xmath102 .
now we consider open 3-clusters . in the remainder of this paper , we often seek to show that an open 3-cluster is not needy , thus we now study how much charge is given away by an open 3-cluster .
every open 3-cluster gives away charge at most 52/29 .
[ cla1 ] if an open 3-cluster @xmath49 has a vertex @xmath6 at distance two and @xmath6 does not receive charge from @xmath49 , then @xmath49 gives away charge at most 51/29 . similarly ,
if an open 3-cluster @xmath49 has a nearby closed 3-cluster or @xmath75-cluster , then @xmath49 gives away charge at most @xmath103 and @xmath49 is not needy .
[ cla2 ] since claims [ cla1 ] and [ cla2 ] are very similar , we only provide a proof for claim [ cla1 ] . however , a short analysis of this proof yields a proof of claim [ cla2 ] .
( 0,2.25 ) rectangle ( 11.0,10.1 ) ; in 1 , 4 , ... , 6 in 1.73205 , 3.4641 , ... , 8.7 ( , ) + ( 30:1 cm ) + ( 90:1 cm ) + ( 150:1 cm ) + ( 210:1 cm ) + ( 270:1 cm ) + ( 330:1 cm ) cycle ; in 1.73205 , 3.4641 , ... , 7 ( , ) + ( 30:1 cm ) + ( 90:1 cm )
+ ( 150:1 cm ) + ( 210:1 cm )
+ ( 270:1 cm )
+ ( 330:1 cm ) cycle ; in 1.73205 , 3.4641 , ... , 8.7 ( , 0 ) + ( 30:1 cm )
+ ( 90:1 cm ) + ( 150:1 cm ) + ( 210:1 cm )
+ ( 270:1 cm )
+ ( 330:1 cm ) cycle ; ; ( 3,2.5 ) ( 4,2 ) ( 5,2.5 ) ( 0,2 ) ( 0,1 ) ( 1,.5 ) ( 1,2.5 ) ( 2,1 ) ( 2,4 ) ( 4,4 ) ( 6,4 ) ( 7,2.5 ) ( 6,1 ) ( 7,.5 ) ( 8,1 )
( 8,2 ) ( 3,6.5 ) node[above ] 1 ( 5,6.5 ) node[above ] 2 ( 2,5 ) node[above ] 3 ( 3,5.5 ) node[below ] 4 ( 4,5 ) node[above ] 5 ( 5,5.5 ) node[below ] 6 ( 6,5 ) node[above ] 7 ( 2,4 ) node[below ] 8 ( 3,3.5 ) node[above ] 9 ( 4,4 ) node[below ] 10 ( 5,3.5 ) node[above ] 11 ( 6,4 ) node[below ] 12 ( 0,2 ) node[above ] 13 ( 1,2.5 ) node[below ] 14 ( 2,2 ) node[above ] 15 ( 3,2.5 ) node[below ] 16 ( 4,2 ) node[above ] 17 ( 5,2.5 ) node[below ] 18 ( 6,2 ) node[above ] 19 ( 7,2.5 ) node[below ] 20 ( 8,2 ) node[above ] 21 ( 0,1 ) node[below ] 22 ( 1,0.5 ) node[above ] 23 ( 2,1 ) node[below ] 24 ( 3,0.5 ) node[above ] 25 ( 4,1 ) node[below ] 26 ( 5,0.5 ) node[above ] 27 ( 6,1 ) node[below ] 28 ( 7,0.5 ) node[above ] 29 ( 8,1 ) node[below ] 30 ( 3,-.5 ) node[below ] 31 ( 4,-1 ) node[above ] 32 ( 5,-.5 ) node[below ] 33 ; ( 3,-.5 )
( 5,-.5 ) ; ( 4.4,1.7 ) node[]@xmath49 ; ( 4,-1.9 ) node[]fig .
1 : proof of claim [ cla1 ] . ;
let @xmath49 be the open 3-cluster 16 - 17 - 18 shown in fig . 1 .
let @xmath104 ; these are the vertices that may receive charge from @xmath49 and also are nearest to 16 .
similarly , let @xmath105 and let @xmath106 .
let @xmath10713 - 22 - 23@xmath108 , @xmath10921 - 30 - 29@xmath108 , and @xmath11031 - 32 - 33@xmath108 .
we now show that the charge @xmath49 sends to vertices in @xmath111 is at most @xmath112 if @xmath49 sends charge to the 3-cluster 13 - 22 - 23 in @xmath111 and at most @xmath113 otherwise ; and the same is true for @xmath114 and the 3-cluster 21 - 30 - 29 .
we also show that the charge @xmath49 sends to vertices in @xmath115 is @xmath116 if @xmath49 sends charge to both 31 and 33 , and is at most @xmath117 otherwise .
2u since we can distinguish between 9 and 15 , we know that @xmath118 . if @xmath119 , then the charge @xmath49 sends to @xmath111 is @xmath120 if @xmath49 sends charge to the 3-cluster , and @xmath113 otherwise . if @xmath121 then either each of 9 and 15 have two neighbors in @xmath16 or one of them has three neighbors and the other has only 16 .
suppose that 9 has three neighbors in @xmath16 ; note that since 9 receives charge @xmath89 from each of its neighbors , none of these neighbors can be a threatened 1-cluster .
thus the charge @xmath49 sends to @xmath111 is at most @xmath122 .
if @xmath123 , then the charge @xmath49 sends to @xmath111 is at most @xmath124 .
if @xmath125 , then the charge @xmath49 sends to @xmath111 is @xmath126 .
the same analysis holds for the charge @xmath49 sends to @xmath114 .
if @xmath127 , then the charge @xmath49 sends to @xmath115 is at most @xmath128 .
if either of 31 and 33 is not a threatened 1-cluster , then the charge @xmath49 sends is at most @xmath117 .
now we examine the total charge given away by @xmath49 .
first we consider the case when both 31 and 33 are threatened 1-clusters . now
@xmath129 and @xmath130 .
thus , @xmath49 does not send charge to the 3-cluster in either @xmath111 or @xmath114 .
hence , if @xmath49 is unpaired , then @xmath49 sends total charge at most @xmath131 .
now we consider the case when at least one of 31 and 33 is not a threatened 1-cluster .
now @xmath49 sends total charge at most @xmath132 .
however , equality holds only if @xmath133 and @xmath134 and @xmath135 ; suppose equality holds .
we consider @xmath136 $ ] .
since @xmath137 , @xmath133 , and @xmath138 , we must have @xmath139 , and specifically , 5 must be in a @xmath50-cluster @xmath73 .
if @xmath73 is a @xmath75-cluster , then @xmath73 gives charge to both 8 and 12 , so @xmath49 does nt have to .
similary , if @xmath73 is 1 - 4 - 5 or 5 - 6 - 2 , then @xmath73 is a closed 3-cluster ; again @xmath73 gives charge to 8 and 12 , so @xmath49 does nt have to .
finally , suppose that @xmath73 is 4 - 5 - 6 .
now @xmath73 is also an open 3-cluster . by the final sentence of rule 4 ,
vertices 8 and 12 each receive charge 1/29 from @xmath73 and each receive no charge from @xmath49 .
thus , again @xmath49 gives away total charge at most 52/29 .
we just proved that every open 3-cluster @xmath49 gives away charge at most 52/29 .
now lemma [ lem5 ] states that every needy 3-cluster has both leaves within distance three of a @xmath50-cluster .
so by rule 5 , every unpaired needy 3-cluster receives charge @xmath88 .
thus , for every unpaired needy 3-cluster @xmath49 , we have @xmath140 . if an unpaired open 3-cluster is not needy , then @xmath141 .
hence , we now turn our attention to paired open 3-clusters .
lemma [ lem6 ] reads : `` let @xmath73 and @xmath74 be uncrowded , open 3-clusters that are paired with each other .
either @xmath73 and @xmath74 have at most 6 nearby threatened 1-clusters and threatened 3-clusters , or they have exactly 7 such nearby clusters , but they also have a nearby closed 3-cluster or @xmath75-cluster . ''
so let @xmath73 and @xmath74 be open , uncrowded 3-clusters that are paired with each other .
we consider the two cases listed in lemma [ lem6 ] .
note that the charge that each gives to adjacent vertices not in @xmath16 is @xmath142 .
thus , if @xmath73 and @xmath74 have at most 6 nearby threatened 1-clusters and threatened 3-clusters , then @xmath143 . similarly , if @xmath73 and @xmath74 have exactly 7 nearby threatened 1-clusters and threatened 3-clusters , then @xmath144 .
finally , we consider @xmath75-clusters .
let @xmath49 be a @xmath75-cluster with @xmath79 vertices .
we will show that @xmath145 .
note that for each @xmath146 if @xmath147 , then @xmath6 gives charge at most @xmath148 to adjacent vertices not in @xmath49 .
similarly , if @xmath149 , then @xmath6 gives charge at most @xmath35 to its adjacent vertex not in @xmath49 ; if @xmath58 , then @xmath6 has no adjacent vertices not in @xmath49 .
let @xmath150 denote the number of vertices @xmath6 in @xmath49 with @xmath151 for @xmath152 .
lemma [ lem3 ] implies that @xmath49 gives charge at most @xmath153 to nearby clusters .
thus , @xmath154 . we want @xmath155 ; thus , we want to show that @xmath156 .
note that if @xmath157 , then @xmath158 , and if @xmath159 , then @xmath160 . so the desired inequality holds except when @xmath159 , @xmath161 , and @xmath162 .
now we consider the two 4-clusters with @xmath159 , @xmath161 , and @xmath162 .
[ scale = .60 ] ( 9,0.8 ) rectangle ( 12.5,10 ) ; in 1 , 4 , ... , 6 in 1.73205 , 3.4641 , ... , 7 ( , ) + ( 30:1 cm )
+ ( 90:1 cm ) + ( 150:1 cm )
+ ( 210:1 cm ) + ( 270:1 cm )
+ ( 330:1 cm ) cycle ; in 1.73205 , 3.4641 , ... , 6.5 ( , ) + ( 30:1 cm ) + ( 90:1 cm ) + ( 150:1 cm )
+ ( 210:1 cm ) + ( 270:1 cm )
+ ( 330:1 cm ) cycle ; in 1.73205 , 3.4641 , ... , 7 ( , 0 ) + ( 30:1 cm )
+ ( 90:1 cm ) + ( 150:1 cm )
+ ( 210:1 cm ) + ( 270:1 cm )
+ ( 330:1 cm ) cycle ; ; ( 5,6.5 ) node[above]1 ( 2,5 ) node[above]2 ( 3,5.5 ) node[below]3 ( 4,5 ) node[above]4 ( 5,5.5 ) node[below]5 ( 0,4 ) node[below]6 ( 1,3.5 ) node[above]7 ( 2,4 ) node[below]8 ( 3,3.5 ) node[above]9 ( 4,4 ) node[below]10 ( 5,3.5 ) node[above]11 ( 6,4 ) node[below]12 ( 7,3.5 ) node[above]13 ( -1,2.5 ) node[below]14 ( 0,2 ) node[above]15 ( 1,2.5 ) node[below]16 ( 2,2 ) node[above]17 ( 3,2.5 ) node[below]18 ( 4,2 ) node[above]19 ( 5,2.5 ) node[below]20 ( 6,2 ) node[above]21 ( 7,2.5 ) node[below]22 ( 0,1 ) node[below]23 ( 1,0.5 ) node[above]24 ( 2,1 ) node[below]25 ( 3,0.5 ) node[above]26 ( 4,1 ) node[below]27 ( 5,0.5 ) node[above]28 ( 6,1 ) node[below]29 ( 7,0.5 ) node[above]30 ( 1,-.5 ) node[below]31 ( 2,-1 ) node[above]32 ( 3,-.5 ) node[below]33 ( 4,-1 ) node[above]34 ( 5,-.5 ) node[below]35 ( 6,-1 ) node[above]36 ; ( 2,2) ( 3,2.5 ) ( 4,2 ) ( 4,1 ) ( 0,1 ) ( 1,.5 ) ( 1,-.5 ) ( 2,-1 ) ( 3,-.5 ) ( 4,-1 ) ( 0,2 ) ( -1 , 2.5 ) ( 0,4 ) ( 1,3.5 ) ( 2,4 ) ( 2,5 ) ( 4,4 ) ( 4,5 ) ( 5,3.5 ) ( 6,4 ) ( 7,2.5 )
( 6,2 ) ( 7,0.5 )
( 6,1 ) ( 6,-1 )
( 5,-.5 ) ; ( 3.8,-2.5 ) node[text width=6cm]fig .
2a : first 4-cluster with @xmath159 , @xmath161 , and @xmath162 .
; in 1 , 4 , ... , 6 in 1.73205 , 3.4641 , ... , 8.7 ( , ) + ( 30:1 cm )
+ ( 90:1 cm ) + ( 150:1 cm ) + ( 210:1 cm ) + ( 270:1 cm )
+ ( 330:1 cm ) cycle ; in 1.73205 , 3.4641 , ... , 7 ( , ) + ( 30:1 cm )
+ ( 90:1 cm ) + ( 150:1 cm )
+ ( 210:1 cm )
+ ( 270:1 cm )
+ ( 330:1 cm ) cycle ; in 1.73205 , 3.4641 , ... , 8.7 ( , 0 ) + ( 30:1 cm )
+ ( 90:1 cm ) + ( 150:1 cm ) + ( 210:1 cm )
+ ( 270:1 cm )
+ ( 330:1 cm ) cycle ; ; ( 2,5 ) node[above]1 ( 3,5.5 ) node[below]2 ( 4,5 ) node[above]3 ( 5,5.5 ) node[below]4 ( 6,5 ) node[above]5 ( 0,4 ) node[below]6 ( 1,3.5 ) node[above]7 ( 2,4 ) node[below]8 ( 3,3.5 ) node[above]9 ( 4,4 ) node[below]10 ( 5,3.5 ) node[above]11 ( 6,4 ) node[below]12 ( 7,3.5 ) node[above]13 ( -1,2.5 ) node[below]14 ( 0,2 ) node[above]15 ( 1,2.5 ) node[below]16 ( 2,2 ) node[above]17 ( 3,2.5 ) node[below]18 ( 4,2 ) node[above]19 ( 5.05,2.5 ) node[below]20 ( 6,2 ) node[above]21 ( 7,2.5 ) node[below]22 ( 8,2 ) node[above]23 ( 0,1 ) node[below]24 ( 1,0.5 ) node[above]25 ( 2,1 ) node[below]26 ( 3,0.5 ) node[above]27 ( 4,1 ) node[below]28 ( 5,0.5 ) node[above]29 ( 6,1 ) node[below]30 ( 7,0.5 ) node[above]31 ( 1,-.5 ) node[below]32 ( 2,-1 ) node[above]33 ( 3,-.5 ) node[below]34 ( 4,-1 ) node[above]35 ( 5,-.5 ) node[below]36 ; ( 2,2 ) ( 3,2.5 ) ( 4,2 ) ( 5,2.5 ) ( 2,5 ) ( 2,4 ) ( 4,5 ) ( 4,4 ) ( 6,5 ) ( 6,4 ) ( 7,3.5 ) ( 0,4 ) ( 1,3.5 ) ( -1,2.5 ) ( 0,2 ) ( 0,1 ) ( 1,.5 ) ( 1,-.5 ) ( 3,.5 ) ( 3 , -.5 ) ( 5,.5 ) ( 5 , -.5 ) ( 6,1 ) ( 7,.5 ) ( 7,2.5 ) ( 8,2 ) ; ( 4.35,1.7 ) node[]@xmath49 ; ( 4.4,-2.5 ) node[text width=6cm]fig .
2b : second 4-cluster with @xmath159 , @xmath161 , and @xmath162 . ; we begin with the 4-cluster shown in fig .
the charge that @xmath49 gives to its six adjacent vertices not in @xmath16 is at most @xmath163 . since @xmath49 begins with charge 4 and
must retain charge at least @xmath164 , the charge that @xmath49 can afford to give to nearby needy clusters is @xmath165 . note that @xmath49 has at most 11 nearby needy clusters , since all the vertices at distance 2 and 3 are covered by the 11 sets @xmath166 , @xmath167 , @xmath168 , @xmath169 , @xmath170 , @xmath171 , @xmath172 , @xmath173 , @xmath174 , @xmath175 , and @xmath176 .
if @xmath134 , @xmath177 , @xmath178 , or @xmath179 , then the charge @xmath49 gives to adjacent vertices is at most @xmath180 , and @xmath49 can give charge @xmath88 to each of the at most 11 nearby needy clusters .
hence , we assume that @xmath181 , @xmath133 , @xmath138 , and @xmath182 . under this assumption
, we will show that @xmath49 has at most 8 nearby needy clusters .
note , as follows , that the sets @xmath166 and @xmath167 intersect at most 1 needy cluster .
since @xmath137 , @xmath133 , and @xmath138 , we must have 4 in a @xmath50-cluster . since @xmath181
, we only need consider @xmath183 .
now either 2 and 4 are in the same cluster , or 4 is in a @xmath75-cluster or closed 3-cluster ( since @xmath135 ) . by symmetry ,
the sets @xmath169 and @xmath171 intersect at most 1 needy cluster .
thus , we see that @xmath49 has at most 9 nearby needy clusters .
we now show that , in fact , @xmath49 has at most 8 nearby needy clusters .
suppose to the contrary that each of the seven sets @xmath168 , @xmath170 , @xmath172 , @xmath173 , @xmath174 , @xmath175 , and @xmath176 intersects a needy cluster .
recall from claim 2 that if a @xmath50-cluster @xmath74 has a vertex @xmath6 at distance two and @xmath6 is within distance three of a @xmath75-cluster , then @xmath74 is not needy .
hence , we conclude that the cluster that intersects @xmath174 is a 1-cluster ; by symmetry , we assume that this cluster is 31 . for the same reason , we must now have the clusters 23 , 14 , and 7 .
since 7 is a 1-cluster , we must also have @xmath183 .
now if 2 and 4 lie in the same cluster , then that cluster is not needy , so the lemma is true .
similarly , the lemma is true if 4 and 12 lie in the same cluster .
hence , 4 must lie in a cluster with 5 , but not with 12 .
however , as before , the cluster containing 4 and 5 can not be needy , since @xmath135 .
thus , the lemma is true for the first 4-cluster with @xmath159 , @xmath161 , and @xmath162 .
we now consider the 4-cluster @xmath49 shown in fig .
note that @xmath49 has at most 12 nearby clusters , since each of the 12 sets @xmath184 , @xmath185 , @xmath186 , @xmath168 , @xmath187 , @xmath170 , @xmath188 , @xmath189 , @xmath190 , @xmath191 , @xmath192 and @xmath193 intersects at most one cluster , and these 12 sets cover all the vertices at distance two or three from @xmath49 . as for the previous case ,
if any of @xmath134 , @xmath177 , @xmath194 , or @xmath195 hold , then @xmath49 gives away charge at most @xmath180 to adjacent vertices , so @xmath49 can afford to give away charge @xmath88 to each of the at most 12 nearby clusters .
thus , we assume that @xmath181 , @xmath133 , @xmath196 , and @xmath197 .
we will show that the four sets @xmath184 , @xmath185 , @xmath186 , and @xmath187 intersect at most two needy clusters .
since @xmath136\cap\code\ne\emptyset$ ] , we know that @xmath91 is in a @xmath50-cluster @xmath73 ; we consider two ( non - exclusive ) cases : @xmath198 and @xmath199 .
first suppose that @xmath198 .
now the two sets @xmath184 and @xmath185 intersect at most one cluster .
further , if @xmath139 , then we can show that @xmath73 is not needy , since 5 is nearby @xmath73 but the cluster contain 5 receives no charge from @xmath73 , since it is also nearby @xmath49 .
however , if @xmath200 , then the four sets @xmath184 , @xmath185 , @xmath186 , and @xmath187 intersect at most two clusters .
suppose instead that @xmath199 .
if @xmath201 , then we can show that @xmath73 is not needy .
however , if @xmath202 , then the four sets @xmath184 , @xmath185 , @xmath186 , and @xmath187 intersect at most two clusters .
thus , the four sets @xmath184 , @xmath185 , @xmath186 , and @xmath187 intersect at most two needy clusters . by symmetry , the four sets @xmath189 , @xmath190 , @xmath191 , and @xmath192 intersect at most two needy clusters .
thus , @xmath49 has at most 8 nearby , needy clusters , so @xmath203 .
this concludes the proof of theorem [ mainthm ] , subject to proving lemmas [ lem1][lem3 ] , which we do in the next section .
[ scale = .60 ] ( 9,2.2 ) rectangle ( 10,10 ) ; in 1 , 4 , ... , 6 in 1.73205 , 3.4641 , ... , 7.0 ( , ) + ( 30:1 cm ) + ( 90:1 cm ) + ( 150:1 cm ) + ( 210:1 cm ) + ( 270:1 cm ) + ( 330:1 cm ) cycle ; in 1.73205 , 3.4641 , ... , 5.2 ( , ) + ( 30:1 cm )
+ ( 90:1 cm ) + ( 150:1 cm ) + ( 210:1 cm ) + ( 270:1 cm ) + ( 330:1 cm ) cycle ; in 1.73205 , 3.4641 , ... , 7.0 ( , 0 ) + ( 30:1 cm )
+ ( 90:1 cm )
+ ( 150:1 cm ) + ( 210:1 cm ) + ( 270:1 cm )
+ ( 330:1 cm ) cycle ; ; ( 2,5 ) node[above]1 ( 4,5 ) node[above]2 ( 2,4 ) node[below]3 ( 3,3.5 ) node[above]4 ( 4,4 ) node[below]5 ( 1,2.5 ) node[below]6 ( 2,2 ) node[above]7 ( 3,2.5 ) node[below]8 ( 4,2 ) node[above]9 ( 5,2.5 ) node[below]10 ( 2,1 ) node[below]11 ( 3,0.5 ) node[above]12 ( 4,1 ) node[below]13 ; ( 3,-1.9 ) node[]fig .
3a : proof of lemma [ lem1 ] . ; in 1 , 4 , ... , 6 in 1.73205 , 3.4641 , ... , 8.7 ( , ) + ( 30:1 cm )
+ ( 90:1 cm ) + ( 150:1 cm ) + ( 210:1 cm ) + ( 270:1 cm )
+ ( 330:1 cm ) cycle ; in 1.73205 , 3.4641 , ... , 7 ( , ) + ( 30:1 cm )
+ ( 90:1 cm ) + ( 150:1 cm )
+ ( 210:1 cm ) + ( 270:1 cm )
+ ( 330:1 cm ) cycle ; in 1.73205 , 3.4641 , ... , 8.7 ( , 0 ) + ( 30:1 cm )
+ ( 90:1 cm ) + ( 150:1 cm )
+ ( 210:1 cm ) + ( 270:1 cm )
+ ( 330:1 cm ) cycle ; ; ( 5,6.5 ) node[above]1 ( 2,5 ) node[above]2 ( 3,5.5 ) node[below]3 ( 4,5 ) node[above]4 ( 5,5.5 ) node[below]5 ( 6,5 ) node[above]6 ( 7,5.5 ) node[below]7 ( 1,3.5 ) node[above]8 ( 2,4 ) node[below]9 ( 3,3.5 ) node[above]10 ( 4,4 ) node[below]11 ( 5,3.5 ) node[above]12 ( 6,4 ) node[below]13 ( 7,3.5 ) node[above]14 ( 0,2 ) node[above]15 ( 1,2.5 ) node[below]16 ( 2,2 ) node[above]17 ( 3,2.5 ) node[below]18 ( 4,2 ) node[above]19 ( 5,2.5 ) node[below]20 ( 6,2 ) node[above]21 ( 7,2.5 ) node[below]22 ( 8,2 ) node[above]23 ( 1,0.5 ) node[above]24 ( 2,1 ) node[below]25 ( 3,0.5 ) node[above]26 ( 4,1 ) node[below]27 ( 5,0.5 ) node[above]28 ( 6,1 ) node[below]29 ( 7,0.5 ) node[above]30 ( 3,-.5 ) node[below]31 ( 5,-.5 ) node[below]32 ; ( 3,2.5 ) ( 4,2 ) ( 5,2.5 ) ( 3,.5 ) ( 3,-.5 ) ( 1,.5 ) ( 2 , 1 ) ( 0,2 ) ( 1 , 2.5 ) ( 1,3.5 ) ( 2,4 ) ( 2,5 ) ( 4,4 ) ( 4,5 ) ( 6,5 ) ( 6,4 ) ( 7,3.5 ) ( 7,2.5 ) ( 8,2 ) ( 7,0.5 ) ( 6,1 ) ( 5,0.5 ) ( 5,-.5 ) ; ( 4.35,1.7 ) node[]@xmath49 ; ( 4,-1.9 ) node[]fig .
3b : proof of lemma [ lem2 ] . ;
every uncrowded 1-cluster is nearby a @xmath50-cluster .
let @xmath204 be the uncrowded 1-cluster shown in fig . 3a .
since we can distinguish @xmath205 from its neighbors , each of these neighbors has an additional neighbor in @xmath16 . by symmetry
, we may assume that @xmath206 .
if 10 or 11 is in a @xmath50-cluster @xmath49 , then the lemma holds ; hence we may assume that 10 and 11 are 1-clusters .
since 13 is uncrowded , and @xmath207 , we know that @xmath181 . since we can distinguish 7 from 11 , we know that @xmath208 . since @xmath209\cap\code$ ] is nonempty , we know that @xmath210 . since we can distinguish 4 from 8 , we know that 4 is in a @xmath50-cluster . if 4 is in a @xmath75-cluster or closed 3-cluster @xmath49 , then the lemma holds .
hence , @xmath211 3 - 4 - 5 .
now 4 is an open center , so again the lemma holds .
every closed 3-cluster @xmath49 has at most ten nearby 1-clusters and open 3-clusters .
if @xmath49 has exactly ten such clusters , then @xmath49 is crowded .
let @xmath49 be the 3-cluster 18 - 19 - 20 shown in fig .
consider the nine bold edges and two bold isolated vertices not in @xmath49 .
these eleven sets cover all of the vertices at distance two or three from @xmath49 .
hence , @xmath49 has at most 11 distinct nearby clusters ( this is a special case of lemma [ lem3 ] ) .
note that either 4 is in a @xmath50-cluster @xmath73 or 11 is a 1-cluster .
we first consider the possibilities for @xmath73 . by symmetry , we assume that @xmath212 ; additionally , either @xmath213 , @xmath214 , @xmath215 , or @xmath216 .
suppose @xmath217 .
note that @xmath218 intersects at most one cluster ( other than @xmath73 ) .
similarly , if @xmath219 intersects two nearby clusters , then @xmath73 is a closed 3-cluster or @xmath75-cluster .
hence , @xmath49 has at most nine nearby 1-clusters and open 3-clusters .
suppose @xmath220 .
note that @xmath219 intersects at most one cluster other than @xmath73 ; similarly for @xmath218 .
hence , @xmath49 has at most nine nearby 1-clusters and open 3-clusters .
suppose @xmath221 .
assume also that @xmath222 , since this is an earlier case .
clearly , @xmath49 has at most ten nearby clusters .
furthermore , equality holds only if each bold set ( other than @xmath223 and @xmath224 ) intersects a distinct nearby cluster . under this assumption , we note the following : @xmath225 , @xmath226 , @xmath195 ( since each leaf of @xmath49 has some vertex @xmath6 at distance two , with @xmath40 ) , @xmath227 , @xmath228 ( since @xmath49 is closed ) , @xmath229 . also , @xmath134 implies that @xmath137 and @xmath230 .
but now 18 has no vertex @xmath6 at distance two , with @xmath40 .
hence @xmath49 has at most nine nearby clusters .
suppose @xmath231 .
also assume that @xmath232 , since otherwise we are in an earlier case .
note that @xmath233 , for if it is , then 14 can not be a distinct cluster and also @xmath73 is a closed 3-cluster or @xmath75-cluster . since @xmath232 and @xmath233 , we are in a similar situation to the previous case , and that argument suffices .
hence , we conclude that 11 is a 1-cluster .
first we prove that the five sets @xmath234 , @xmath235 , @xmath223 , @xmath224 , and @xmath236 intersect at most four nearby 1-clusters and open 3-clusters .
second , we assume that @xmath49 is not crowded and show that the other six bold sets intersect at most five nearby 1-clusters and open 3-clusters .
assume that both @xmath134 and @xmath225 .
since 4 is distinguishable from 11 , either @xmath139 or @xmath237 ; by symmetry , assume @xmath139 .
if @xmath232 , then @xmath224 intersects no cluster other than ( possibly ) the one that contains 14 .
however , if @xmath208 , then the cluster that contains 5 and 6 is either a closed 3-cluster or a @xmath75-cluster .
hence , @xmath49 has at most 10 nearby 1-clusters and open 3-clusters .
now assume further that @xmath49 is not crowded . since @xmath49 is closed , either @xmath228 or @xmath130 ; by symmetry , assume @xmath228 . since @xmath49 is not crowded , we know that @xmath137 , @xmath233 , @xmath230 , @xmath226 , @xmath229 , @xmath227 , and @xmath197 .
since @xmath238\cap\code$ ] is nonempty , 30 is in a @xmath50-cluster @xmath73 .
now either @xmath239 , or @xmath240 , or @xmath241 , or @xmath73 is a closed 3-cluster ( due to 32 ) or a @xmath75-cluster . in every case , the lemma is true .
every needy 3-cluster has both leaves within distance three of a @xmath50-cluster .
( 4,2.0 ) rectangle ( 6.5,8 ) ; in 1 , 4 , ... , 6 in 1.73205 , 3.4641 , ... , 8.7 ( , ) + ( 30:1 cm ) + ( 90:1 cm ) + ( 150:1 cm ) +
( 210:1 cm ) + ( 270:1 cm )
+ ( 330:1 cm ) cycle ; in 1.73205 , 3.4641 , ... , 7 ( , ) + ( 30:1 cm )
+ ( 90:1 cm ) + ( 150:1 cm ) + ( 210:1 cm ) + ( 270:1 cm )
+ ( 330:1 cm ) cycle ; in 1.73205 , 3.4641 , ... , 8.7 ( , 0 ) + ( 30:1 cm )
+ ( 90:1 cm )
+ ( 150:1 cm ) + ( 210:1 cm ) + ( 270:1 cm )
+ ( 330:1 cm ) cycle ; ; ( 5,6.5 ) node[above]0 ( 4,5 ) node[above]1 ( 5,5.5 ) node[below]2 ( 6,5 ) node[above]3 ( 7,5.5 ) node[below]4 ( 8,5 ) node[above]5 ( 4,4 ) node[below]6 ( 5,3.5 ) node[above]7 ( 6,4 ) node[below]8 ( 7,3.5 ) node[above]9 ( 8,4 ) node[below]10 ( 9,3.5 ) node[above]11 ( 6,2 ) node[above]12 ( 7,2.5 ) node[below]13 ( 8,2 ) node[above]14 ( 3,-.5 ) node[below]15 ( 5,-.5 ) node[below]16 ; ( 3,2.5 ) ( 4,2 ) ( 5,2.5 ) ; ( 3,-.5 ) ( 4,-1 ) ( 5,-.5 ) ; ( 0,2 ) ( 0,1 ) ( 1,.5 ) ; ( 8,2 ) ( 8,1 )
( 7,.5 ) ; ( 4.55,1.7 ) node[]@xmath73 ; ( 8,.95 ) node[below]@xmath74 ; ( 4,-1.9 ) node[]fig .
4 : proof of lemma [ lem5 ] . ;
let @xmath73 be the 3-cluster shown in fig . 4 . since @xmath242\cap\code$ ] is nonempty , either 1 is in a @xmath50-cluster , or 6 is a 1-cluster . in the former case , the lemma holds
, so we consider only the latter case .
suppose 6 is a 1-cluster .
we will now show that @xmath73 is not needy ; suppose , for contradiction , that @xmath73 is needy . by definition , @xmath73 is uncrowded .
this implies that neither 15 nor 16 is a 1-cluster . thus , the only candidates to be nearby threatened 1-clusters and threatened 3-clusters , other than 6 , are the three 3-clusters shown in bold . furthermore ,
since @xmath73 is needy , 6 and each of these 3-clusters must be threatened . since 6
is a 1-cluster , 1 must also have another neighbor in @xmath16 ; by symmetry , we assume @xmath183 .
now we consider whether 9 is in @xmath16 or not .
suppose that @xmath243 .
now 9 must be a 1-cluster , since @xmath74 is threatened . since 8 is distinguishable from 9 , we know that @xmath237 .
now we have @xmath183 and @xmath237 , so 2 and 3 lie together in a @xmath50-cluster @xmath244 .
note that since @xmath208 and @xmath243 , @xmath244 is either a closed 3-cluster or a @xmath75-cluster . in each case , 6 is unthreatened , so @xmath73 is not needy .
suppose instead that @xmath137 .
since @xmath245 , @xmath181 , and @xmath137 , we know that @xmath237 . similarly , @xmath181 , @xmath137 , and @xmath233 , so @xmath177 ; furthermore , 10 is in a @xmath50-cluster . since @xmath74 is threatened , 11 can not be in a @xmath50-cluster with 10 ; hence , 10 is in a cluster with 5 .
now we consider the cluster @xmath244 that contains 2 and 3 .
again , @xmath244 is either a closed 3-cluster or a @xmath75-cluster . in each case , 6 is unthreatened , so @xmath49 is not needy .
let @xmath73 and @xmath74 be uncrowded , open 3-clusters that are paired with each other .
clusters @xmath73 and @xmath74 have at most 7 nearby threatened 1-clusters and threatened 3-clusters .
if they have exactly 7 such nearby clusters , then they also have a nearby closed 3-cluster or @xmath75-cluster .
( 2,1.8 ) rectangle ( 10,10 ) ; in 1 , 4 , ... , 6 in 1.73205 , 3.4641 , ... , 10.7 ( , ) + ( 30:1 cm ) + ( 90:1 cm ) + ( 150:1 cm ) + ( 210:1 cm ) + ( 270:1 cm ) + ( 330:1 cm ) cycle ; in 1.73205 , 3.4641 , ... , 9 ( , ) + ( 30:1 cm )
+ ( 90:1 cm ) + ( 150:1 cm ) + ( 210:1 cm ) + ( 270:1 cm )
+ ( 330:1 cm ) cycle ; in 1.73205 , 3.4641 , ... , 10.7 ( , 0 ) + ( 30:1 cm )
+ ( 90:1 cm ) + ( 150:1 cm )
+ ( 210:1 cm ) + ( 270:1 cm )
+ ( 330:1 cm ) cycle ; ; ( 3,6.5 ) node[above]1 ( 4,7 ) node[below]2 ( 5,6.5 ) node[above]3 ( 6,7 ) node[below]4 ( 7,6.5 ) node[above]5 ( 8,7 ) node[below]6 ( 0,5 ) node[above]7 ( 1,5.5 ) node[below]8 ( 2,5 ) node[above]9 ( 3,5.5 ) node[below]10 ( 4,5 ) node[above]11 ( 5,5.5 ) node[below]12 ( 6,5 ) node[above]13 ( 7,5.5 ) node[below]14 ( 8,5 ) node[above]15 ( 9,5.5 ) node[below]16 ( 10,5 ) node[above]17 ( 0,4 ) node[below]18 ( 1,3.5 ) node[above]19 ( 2,4 ) node[below]20 ( 3,3.5 ) node[above]21 ( 4,4 ) node[below]22 ( 5,3.5 ) node[above]23 ( 6.1,4 ) node[below]24 ( 7,3.5 ) node[above]25 ( 8,4 ) node[below]26 ( 9,3.5 ) node[above]27 ( 10,4 ) node[below]28 ( 11,3.5 ) node[above]29 ( 2,2 ) node[above]30 ( 3,2.5 ) node[below]31 ( 4,2 ) node[above]32 ( 5,2.5 ) node[below]33 ( 6,2 ) node[above]34 ( 7,2.5 ) node[below]35 ( 8,2 ) node[above]36 ( 9,2.5 ) node[below]37 ( 10,2 ) node[above]38 ( 4,1 ) node[below]39 ( 5,0.5 ) node[above]40 ( 6.1,1 ) node[below]41 ( 7,0.5 ) node[above]42 ( 8,1 ) node[below]43 ( 9,0.5 ) node[above]44 ( 3,-.5 ) node[below]45 ( 7,-.5 ) node[below]46 ; ( 3,3.5 ) ( 4,4 ) ( 5,3.5 ) ; ( 4,1 ) ( 5,.5 ) ( 6,1 ) ; ( 5.0,-1.9 ) node[]fig .
5 : proof of lemma [ lem6 ] . ;
( 5.5,3.1 ) node[]@xmath73 ; ( 5.5,0.1 ) node[]@xmath74 ; let @xmath246 21 - 22 - 23 and @xmath247 39 - 40 - 41 be the paired , uncrowded , open 3-clusters shown in fig . 5 .
we note that the number of 1-clusters and open 3-clusters nearby @xmath73 and @xmath74 is at most 8 , as follows .
we show that @xmath73 has at most four such nearby clusters .
first suppose that at most one of 1 and 3 is a 1-cluster . now at most one nearby cluster intersects each of the following four vertex sets : @xmath248 , @xmath249 , @xmath250 , and @xmath251 .
if both 1 and 3 are 1-clusters , then the same analysis holds , except now we know that @xmath243 , so @xmath252 . since we can distinguish 24 from 33 , we know that either @xmath204 or @xmath253 .
similarly , since we can distinguish 34 from 42 , we know that @xmath254 , @xmath255 , or @xmath256 . note that if @xmath253 and @xmath254 , then 25 and 35 are in the same @xmath50-cluster and it must be a closed 3-cluster or @xmath75-cluster . furthermore , @xmath73 and @xmath74 have at most 7 other nearby clusters ; so the lemma holds .
thus , we have five possibilities : @xmath257 , @xmath258 , @xmath259 , @xmath260 , and @xmath261 .
we consider each possibility in turn , but we defer the first possibility until the end .
suppose @xmath204 and @xmath255 . since @xmath74 is uncrowded , @xmath262
; similarly @xmath229 . thus @xmath228 and @xmath263 .
now 36 and 43 lie together in a @xmath50-cluster @xmath244 . if @xmath264 36 - 43 - 44 , then @xmath244 is a closed 3-cluster and again @xmath73 and @xmath74 have at most 7 other nearby clusters .
the situation is similar if @xmath244 is a @xmath75-cluster .
hence , we assume that @xmath264 37 - 36 - 43 .
since we can distinguish between 25 and 26 , we know that 26 is in a @xmath50-cluster @xmath265 . furthermore , @xmath265 is either a closed 3-cluster or a @xmath75-cluster ; thus , both @xmath244 and the cluster containing 13 are unthreatened .
suppose @xmath204 and @xmath256 .
the analysis is similar to the previous case . since @xmath73 is not crowded , @xmath229 .
similarly , @xmath262 .
thus @xmath228 and @xmath263 and each are in @xmath50-clusters . since @xmath74 is not crowded , @xmath266 ; thus , 36 lies in a @xmath50-cluster @xmath244 with 37 .
now @xmath244 is either a closed 3-cluster or a @xmath75-cluster ; thus , the cluster containing 46 is not threatened .
similarly , the cluster containing 26 is either a @xmath75-cluster or a closed 3-cluster , so the cluster containing 13 is not threatened .
suppose @xmath253 ( and either @xmath255 or @xmath256 ) . since @xmath73 is open and uncrowded , @xmath267 and @xmath233 .
since @xmath268\cap\code\ne\emptyset$ ] , we know @xmath237 .
also 14 is in a @xmath50-cluster @xmath244 .
if @xmath244 is closed or a @xmath75-cluster , then the clusters containing 3 and 25 are unthreatened , so the lemma holds .
thus , we assume that @xmath269 5 - 14 - 15 .
since 3 is distinguishable from 12 , we know that 3 is in a @xmath50-cluster @xmath265 .
since @xmath139 , @xmath265 is unthreatened ; so we may assume that all other nearby clusters are threatened .
thus , 25 is a 1-cluster .
so , @xmath263 . if @xmath255 , then the cluster containing 36 and 43 is nearby @xmath74 and is a closed 3-cluster or @xmath75-cluster .
hence , we assume that @xmath256 .
since 25 is threatened , 5 - 14 - 15 is uncrowded .
thus , @xmath196 . so to distinguish between 36 and 37 , we must have @xmath270 .
suppose @xmath271 .
now 28 is in a @xmath50-cluster @xmath272 . recall that @xmath273 , since 5 - 14 - 15 is uncrowded .
thus , the cluster containing 28 must be either a closed 3-cluster or a @xmath75-cluster ; in each case , 25 is unthreatened , so the lemma is true .
thus @xmath274 ; we may assume 36 - 37 - 38 is an open 3-cluster , since otherwise 25 is unthreatened .
now since @xmath273 and @xmath227 , we have @xmath275 in a @xmath50-cluster . however , now 36 - 37 - 38 has a @xmath50-cluster at distance two , so 25 is unthreatened .
again the lemma holds .
( 2,1.7 ) rectangle ( 10,13 ) ; in 1 , 4 , ... , 8 in 1.73205 , 3.4641 , ... , 10.7 ( , ) + ( 30:1 cm ) + ( 90:1 cm )
+ ( 150:1 cm ) + ( 210:1 cm ) + ( 270:1 cm )
+ ( 330:1 cm ) cycle ; in 1.73205 , 3.4641 , ... , 9 ( , ) + ( 30:1 cm )
+ ( 90:1 cm ) + ( 150:1 cm )
+ ( 210:1 cm ) + ( 270:1 cm )
+ ( 330:1 cm ) cycle ; in 1.73205 , 3.4641 , ... , 10.7 ( , 0 ) + ( 30:1 cm )
+ ( 90:1 cm ) + ( 150:1 cm ) + ( 210:1 cm )
+ ( 270:1 cm )
+ ( 330:1 cm ) cycle ; ; ( 3,3.5 ) ( 4,4 ) ( 5,3.5 ) ( 6,5 ) ( 0,4 ) ( 0,5 ) ( 1,5.5 ) ( 3,6.5 ) ( 4,7 ) ( 5,6.5 ) ( 4,1 ) ( 5,.5 ) ( 6,1 ) ( 7,2.5 ) ( 2,2 ) ( 3,-.5 ) ( 9,.5 ) ( 9,-.5 ) ( 8,-1 ) ( 4,-2 ) ( 5,-2.5 ) ( 6,-2 ) ; ( 5.3,-4.9 ) node[]fig . 6 : proof of lemma [ lem6 ] ( continued ) . ; ( 3,6.5 ) node[above]1 ( 4,7 ) node[below]2 ( 5,6.5 ) node[above]3 ( 6,7 ) node[below]4 ( 7,6.5 ) node[above]5 ( 8,7 ) node[below]6 ( 0,5 ) node[above]7 ( 1,5.5 ) node[below]8 ( 2,5 ) node[above]9 ( 3,5.5 ) node[below]10 ( 4,5 ) node[above]11 ( 5,5.5 ) node[below]12 ( 6,5 ) node[above]13 ( 7,5.5 ) node[below]14 ( 8,5 ) node[above]15 ( 9,5.5 ) node[below]16 ( 10,5 ) node[above]17 ( 0,4 ) node[below]18 ( 1,3.5 ) node[above]19 ( 2,4 ) node[below]20 ( 3,3.5 ) node[above]21 ( 4,4 ) node[below]22 ( 5,3.5 ) node[above]23 ( 6.1,4 ) node[below]24 ( 7,3.5 ) node[above]25 ( 8,4 ) node[below]26 ( 9,3.5 ) node[above]27 ( 10,4 ) node[below]28 ( 11,3.5 ) node[above]29 ( 2,2 ) node[above]30 ( 3,2.5 ) node[below]31 ( 4,2 ) node[above]32 ( 5,2.5 ) node[below]33 ( 6,2 ) node[above]34 ( 7,2.5 ) node[below]35 ( 8,2 ) node[above]36 ( 9,2.5 ) node[below]37 ( 10,2 ) node[above]38 ( 4,1 ) node[below]39 ( 5,0.5 ) node[above]40 ( 6.1,1 ) node[below]41 ( 7,0.5 ) node[above]42 ( 8,1 ) node[below]43 ( 9,0.5 ) node[above]44 ( 3,-.5 ) node[below]45 ( 7,-.5 ) node[below]46 ( 8,-1 ) node[above]47 ( 9,-.5 ) node[below]48 ( 4,-2 ) node[below]49 ( 5,-2.5 ) node[above]50 ( 6,-2 ) node[below]51 ; ( 5.47,3.1 ) node[]@xmath73 ; ( 5.47,0.1 ) node[]@xmath74 ; hence we may assume that @xmath204 and @xmath254 ; by symmetry , we may also assume that @xmath276 and @xmath277 ( see fig . 6 ) .
observe that we have eight candidates to be nearby threatened clusters ; these are 13 , 30 , 35 , 45 , 18 - 7 - 8 , 47 - 48 - 44 , 3 or 3 - 2 - 1 , and 49 or 49 - 50 - 51 .
if at most six of these candidates are threatened , then the lemma holds .
hence , by symmetry , we may assume that each of 13 , 35 , 18 - 7 - 8 , and 3 or 3 - 2 - 1 is threatened .
since 1 is distinguishable from 10 , we know that 1 is in a @xmath50-cluster @xmath244 .
if @xmath278 , then @xmath244 is a @xmath75-cluster or a closed 3-cluster ; so the cluster containing 3 is unthreatened .
thus , we assume that @xmath73 is 1 - 2 - 3 .
since 13 and 14 are distinguishable , either @xmath139 or @xmath279 .
if @xmath139 , then 1 - 2 - 3 is crowded , so neither 1 - 2 - 3 nor 13 is threatened ; hence , we assume @xmath279 . since 35
is distinguishable from 25 , we know that @xmath228
. however , now the cluster containing 15 and 26 is either a closed 3-cluster or a @xmath75-cluster ; in each case both 13 and 35 are unthreatened .
every cluster @xmath49 with @xmath79 vertices has at most @xmath93 nearby clusters .
instead of proving lemma [ lem3 ] directly , we prove lemma [ lem4 ] , which is stronger , but also easier to prove by induction . [ lem4 ] if @xmath49 is a cluster with @xmath79 vertices , then we can partition the set of vertices at distances two and three from @xmath49 into @xmath93 sets , each of which consists of two adjacent vertices or a single vertex .
furthermore , if a vertex @xmath6 is distance three from @xmath49 and in fact has two disjoint paths of length 3 to the same vertex @xmath280 , then @xmath6 is in a set of size 1 in the partition .
lemma [ lem4 ] immediately implies lemma [ lem3 ] , since each set in the partition intersects at most one cluster .
if a vertex @xmath6 is distance at most three from @xmath49 , then either @xmath6 is adjacent to @xmath49 or @xmath6 is in the partition for @xmath49 ; we call such a vertex @xmath6 _ covered_.
if a vertex is not covered , then we call it _ uncovered_. ( 9.5,2.10 ) rectangle ( 14,7 ) ; in 1.73205 , 3.4641 , ... , 7.0 ( , 1 ) + ( 30:1 cm ) + ( 90:1 cm ) + ( 150:1 cm )
+ ( 210:1 cm ) + ( 270:1 cm )
+ ( 330:1 cm ) cycle ; in 1.73205 , 3.4641 , ... , 5.2 ( , 1 ) + ( 30:1 cm ) + ( 90:1 cm ) + ( 150:1 cm ) + ( 210:1 cm ) + ( 270:1 cm )
+ ( 330:1 cm ) cycle ; in 1.73205 , 3.4641 , ... , 7.0 ( , 0 ) + ( 30:1 cm )
+ ( 90:1 cm ) + ( 150:1 cm )
+ ( 210:1 cm ) + ( 270:1 cm )
+ ( 330:1 cm ) cycle ; ; ( 1,2.5 ) node[below]1 ( 2,2 ) node[above]2 ( 3,2.5 ) node[below]3 ( 4,2 ) node[above]4 ( 5,2.5 ) node[below]5 ( 2,1 ) node[below]6 ( 3,.5 ) node[above]7 ( 4,1 ) node[below]8 ( 5,.5 ) node[above]9 ( 6,1 ) node[below]10 ( 2,-1 ) node[above]11 ( 3,-.5 ) node[below]12 ; ( 3,-1.9 ) node[]fig .
7a : the first induction step . ; ( 2,-1 )
( 3 , -.5 ) ( 3,.5 ) ; in 1.73205 , 3.4641 , ... , 5.2 ( , 1 ) + ( 30:1 cm ) + ( 90:1 cm ) + ( 150:1 cm ) + ( 210:1 cm ) + ( 270:1 cm ) + ( 330:1 cm )
cycle ; in 1.73205 , 3.4641 , ... , 3.5 ( , 1 ) + ( 30:1 cm ) + ( 90:1 cm ) + ( 150:1 cm ) + ( 210:1 cm ) + ( 270:1 cm ) + ( 330:1 cm ) cycle ; in 1.73205 , 3.4641 , ... , 5.2 ( , 0 ) + ( 30:1 cm ) + ( 90:1 cm ) + ( 150:1 cm )
+ ( 210:1 cm ) + ( 270:1 cm )
+ ( 330:1 cm ) cycle ; ; ( 1,3.5 ) node[above]1 ( 3,3.5 ) node[above]2 ( 1,2.5 ) node[below]3 ( 2,2 ) node[above]4 ( 3,2.5 ) node[below]5 ( 1,.5 ) node[above]6 ( 2,1 ) node[below]7 ( 3,.5 ) node[above]8 ; ( 1.5,-1.9 ) node[]fig .
7b : the second induction step . ; we use induction on the size of @xmath49 , growing @xmath49 by one vertex at each step .
the base case @xmath281 is easy and is shown in fig .
3b . let @xmath85 be a cluster of size @xmath282 and let @xmath283 be a spanning tree of @xmath85 with a leaf @xmath6 . by deleting @xmath6 we reach a cluster @xmath49 , of size @xmath43 , for which the induction hypothesis holds ; let @xmath284 be the desired partion for @xmath49 of size @xmath43 .
we consider different induction steps , depending on whether @xmath6 has one , two , or three neighbors in @xmath49 .
note that if @xmath6 has three neighbors in @xmath49 , then @xmath284 is still a valid partition for @xmath85 .
so we consider below the cases when @xmath6 has one or two neighbors in @xmath49 .
first we consider a cluster @xmath85 that is built from @xmath49 by adding a vertex with only one neighbor in @xmath49 .
let 7 be the new vertex ( see fig .
7a ) ; by symmetry , we may assume that @xmath285 .
we assume that 1 , 3 , 5 are uncovered ( the case where one or more of these vertices is covered is easier , so we omit the details )
. we will modify @xmath284 to form a new partition that also includes 1 , 3 , 5 . beginning with @xmath284
, we delete the sets that contain 2 and 4 , and we add the sets @xmath286 , @xmath287 , and @xmath288 . as above , we assume that 10 is uncovered ( since the other case is easier ) .
since 10 is uncovered , @xmath284 contains the set @xmath289 ; replace the set @xmath289 with the set @xmath290 .
we now have a partition @xmath291 for @xmath85 and @xmath292 .
second we consider a cluster @xmath85 that is built from @xmath49 by adding a vertex with two neighbors in @xmath49 .
let 7 be the new vertex added to @xmath49 and let 6 and 8 be vertices already in @xmath49 ( see fig .
if 1 is uncovered , then 3 is distance three from @xmath49 ; similarly , if 2 is uncovered , then 5 is distance three from @xmath49 .
we assume that 1 and 2 are uncovered by @xmath49 ( the case where one or both of 1 and 2 is covered is easier , so we omit the details ) .
so 3 and 5 are both in sets of size one ; we thus replace @xmath287 with @xmath293 and we replace @xmath294 with @xmath295 . we now have a partition @xmath291 for @xmath85 and @xmath296 .
thank you to ryan martin , who brought this problem to our attention .
thank you also to an anonymous referee who noticed inconsistencies and suggested improvements .
g. cohen , s. gravier , i. honkala , a. lobstein , m. mollard , c. payan , g. zmor , improved identifying codes for the grid , comments on r19 , _ electronic journal of combinatorics _
( 1999 ) , http://www.combinatorics.org/volume_6/html/v6i1r19.html . | given a graph @xmath0 , an identifying code @xmath1 is a vertex set such that for any two distinct vertices @xmath2 , the sets @xmath3\cap\code$ ] and @xmath4\cap\code$ ] are distinct and nonempty ( here @xmath5 $ ] denotes a vertex @xmath6 and its neighbors ) .
we study the case when @xmath0 is the infinite hexagonal grid @xmath7 .
cohen et.al . constructed two identifying codes for @xmath7 with density @xmath8 and proved that any identifying code for @xmath7 must have density at least @xmath9 .
both their upper and lower bounds were best known until now . here
we prove a lower bound of @xmath10 . |
we consider the two - parameter fractional ornstein - uhlenbeck process defined as the solution of the stochastic equation @xmath7\times [ 0,s].\ ] ] here @xmath8 denotes a fractional brownian sheet with hurst parameters @xmath9 .
we also suppose that @xmath10 for every @xmath11 . our goal is to estimate the unknown parameter @xmath1 from the continuous time observation of the solution @xmath12\times [ 0,s]}$ ] .
the development of the stochastic analysis for fractional brownian motion ( fbm ) naturally led to the study of the statistical inference for stochastic equations driven by this process .
there already exists an important literature related to these aspects .
we refer , among others to @xcite , @xcite , @xcite , @xcite , @xcite .
statistical analysis of the stochastic differential equations ( sde ) driven by the fractional brownian sheet has been less considered . we refer to the paper @xcite for the study of the maximum likelihood estimator for a sde with additive fractional brownian sheet noise ( see also @xcite or @xcite for the case when the noise is a standard brownian sheet ) .
in this paper we propose a least square estimator for the unknown parameter @xmath1 following the approach in @xcite .
this estimator is obtained by formally minimizing with respect to @xmath1 the expression @xmath13 we obtain the following estimator @xmath14 the integral with respect to @xmath15 is understood as the sum of the standard lebesgue integral @xmath16 and of the stochastic integral @xmath17 which is a divergence type integral with respect to the fractional brownian sheet @xmath8 ( it will be defined in section 2 , we also refer to @xcite , @xcite , @xcite , @xcite , @xcite for the stochastic integration with respect to @xmath18 ) . using ( [ 1 ] ) and ( [ est1 ] )
we can write @xmath19 we will study the asymptotic behavior of the least square estimator @xmath6 as @xmath20 .
our tools are the multiple stochastic integrals and the malliavin calculus .
actually , the nominator and the denominator of the right hand side of ( [ theta - theta ] ) can be expressed as multiple integrals of order 2 with respect to the fractional brownian sheet and from this we will obtain concrete estimates for their moments .
we will prove that the estimator ( [ est1 ] ) is a strongly consistent estimator in the sense that it converges almost surely to the true value of the parameter @xmath1 .
this result is similar to the one - dimensional case ( see @xcite ) , however the approach presented in @xcite is not possible to be followed for the two - parameter case , instead we use among other tools , the hypercontractivity of multiples integrals . by contrary , in the two - parameter case
, the least square estimator does not preserve the asymptotic normality as in the one - parameter case .
this will be noticed at the end of our work by using criteria for the asymptotic normality of sequences of multiple integrals in terms of malliavin calculus .
our paper is structured as follows .
section 2 contains some preliminaries on multiple integrals and fractional brownian sheet . in section 3
we discuss the relation between the solution to ( [ 1 ] ) and the bessel function of order 0 .
section 4 contains the proof of the consistency of the least square estimator while section 5 is devoted to a discussion about the asymptotic normality of the estimator .
let us introduce the elements from stochastic analysis that we will need in the paper .
consider @xmath21 a real separable hilbert space and @xmath22 an isonormal gaussian process on a probability space @xmath23 , that is , a centered gaussian family of random variables such that @xmath24 .
denote by @xmath25 the multiple stochastic integral with respect to @xmath26 ( see @xcite ) .
this @xmath25 is actually an isometry between the hilbert space @xmath27(symmetric tensor product ) equipped with the scaled norm @xmath28 and the wiener chaos of order @xmath29 which is defined as the closed linear span of the random variables @xmath30 where @xmath31 and @xmath32 is the hermite polynomial of degree @xmath33 @xmath34 the isometry of multiple integrals can be written as : for @xmath35 positive integers , @xmath36 it also holds that @xmath37 where @xmath38 denotes the symmetrization of @xmath39 defined by @xmath40 the malliavin derivative acts on multiple integrals @xmath41 in the following way : for every @xmath42 @xmath43 where @xmath44 above denotes @xmath45 variables .
we recall the following hypercontractivity property for the @xmath46 norm of a multiple stochastic integral ( see ( * ? ? ?
* theorem 4.1 ) ) @xmath47 where @xmath48 is an explicit positive constant and @xmath49 . in this work
we use malliavin calculus and multiple integrals with respect to the fractional brownian sheet ( fbs ) .
let us define this process and its associated hilbert space .
the fbs with hurst parameters @xmath50 \times [ 0,s ] ) $ ] is a zero mean gaussian process with covariance @xmath51 given for all @xmath52^{2}$ ] and @xmath53^{2}$ ] .
+ we assume that @xmath54 is defined on a complete probability space @xmath55 such that @xmath56 is generated by @xmath54 .
fix a time interval @xmath57\times [ 0,s]$ ] , denote by @xmath58 the set of real valued step functions on @xmath57\times [ 0,s]$ ] and let @xmath59 be the hilbert space defined as the closure of @xmath58 with respect to the scalar product @xmath60\times[0,s ] } , 1_{[0,u]\times[0,v ] } \rangle _ { \mathcal{h}^{\alpha , \beta } } = \mathcal{r}^{\alpha}(t , u)\mathcal{r}^{\beta}(s , v)\ ] ] where @xmath61 is the covariance function of the fbs , given in ( [ cov ] ) .
the mapping @xmath62\times[0,s ] } \longmapsto b^{\alpha,\beta}_{t , s}$ ] can be extended to a linear isometry between @xmath59 and the gaussian space @xmath63 spanned by @xmath54 which is a closed subspace of @xmath64 .
we denote this isometry by @xmath65 .
fix @xmath66 , in this case we have that for every @xmath67 the scalar product has the form @xmath68 and @xmath69 .
the equation ( [ 1 ] ) has been studied in several papers ( see @xcite , @xcite ) .
it has been showed that for @xmath70 and @xmath71 equation ( [ 1 ] ) admits an unique strong solution which can be expressed as @xmath72 where @xmath73}(t_{0 } ) 1_{[0,s]}(s_{0 } ) \sum_{n\geq 0}(-1)^{n } \theta^{n } \frac{(t - t_{0})^{n } ( s - s_{0})^{n}}{(n!)^{2}}.\ ] ] we will call the process @xmath74 solution to ( [ 1 ] ) as the fractional ornstein - uhlenbeck sheet .
it is a gaussian process since it is given by a multiple integral of order 1 ( wiener integral actually ) with respect to the gaussian process @xmath18 .
we mention that the solution to ( [ 1 ] ) behaves differently as its one - dimensional counterpart which is fractional ornstein - uhlenbeck process introduced in @xcite , this will make our analysis quite different from @xcite .
for example , we note that the solution of some stochastic differential equations driven by the brownian sheet or fractional brownian sheet , which are positive in the one - parameter case , can take negative values with strictly positive probability ( see @xcite or @xcite ) .
a key element of our analysis is the fact that the solution @xmath74 ( more precisely the kernel @xmath39 of the solution ) can be expressed is terms of the bessel function of the first kind .
let us consider the bessel function of order 0 given , for every @xmath75 , by @xmath76 this bessel function admits the integral representation , for every @xmath75 @xmath77 the kernel @xmath39 in ( [ f ] ) of the solution @xmath78 \times [ 0,s]}$ ] can be expressed as @xmath79}(u ) 1_{[0,s ] } ( v ) j_{0 } \left ( 2 \sqrt { \theta ( t - u ) ( s - v ) } \right ) \label{fjo } \\ & = & 1_{[0,t]}(u ) 1_{[0,s ] } ( v ) \frac{1}{\pi } \int_{0 } ^{\pi } \cos \left(2\sqrt { \theta ( t - u ) ( s - v ) } \sin \rho \right ) d\rho.\nonumber\end{aligned}\ ] ] let us also recall the following property of the bessel function ( see e.g. @xcite ) which will play an important role for our estimates : for @xmath80 large enough @xmath81 ( the symbol @xmath82 means that the two sides have the same limit as @xmath83 ) .
in this section we study the asymptotic behavior of the estimator @xmath84 defined in ( [ est1 ] ) .
more precisely , we will show that this estimator is strongly consistent for the parameter @xmath1 , that is , @xmath6 converges to @xmath85 almost surely as @xmath20 . to this end
we will analyze separately the nominator and the denominator appearing in the right hand side of the expression ( [ theta - theta ] ) .
let us start with the study of the nominator .
it can be written as the double stochastic integral @xmath86 where the kernel @xmath39 is given by ( [ f ] ) and the integral @xmath87 acts with respect to the variables @xmath88 .
we will estimate first the @xmath89 norm of @xmath90 .
we have the following result . [ pn1 ] for every @xmath91 and for @xmath92 , @xmath93 moreover for @xmath94 large enough we have @xmath95 where @xmath96 is a strictly positive constant not depending on @xmath94 .
* proof : * we calculate the @xmath89 norm of the random variable @xmath97 . by the isometry property of multiple integrals ( [ iso ] ) and since @xmath98
this norm can be handles as follows @xmath99^{4}}dtdt_{0 } dudu_{0 } \int_{[0,s ] ^{4}}dsds_{0 } dv dv_{0}\\ & & \times f(t , s , u , v)f(t_{0},s_{0},u_{0},v_{0})\vert u - u_{0 } \vert^{2\alpha-2 } \vert v - v_{0 } \vert^{2\beta-2 } \vert t - t_{0 } \vert^{2\alpha-2 } \vert s - s_{0 } \vert^{2\beta-2 } \\ & = & \int_{0 } ^{t } dt \int_{0}^{t } du \int_{0 } ^{t } dt_{0 } \int_{0 } ^{t_{0 } } du_{0}\int_{0 } ^{s } ds \int_{0}^{s } dv \int_{0 } ^{s } ds_{0 } \int_{0 } ^{s_{0 } } dv_{0 } \\ & & \times j_{0 } \left ( 2\sqrt { \theta ( t - u)(s - v ) } \right ) j_{0 } \left(2 \sqrt{\theta ( t_{0 } -u_{0 } ) ( s_{0 } -v_{0 } ) } \right ) \\ & & \times \vert t - t_{0 } \vert ^{2\alpha -2 } \vert u - u_{0 } \vert ^{2\alpha -2 } \vert s - s_{0 } \vert ^{2\beta -2 } \vert v - v_{0 } \vert ^{2\beta -2}.\end{aligned}\ ] ] by making the change of variables @xmath100 , @xmath101 and similarly for the other variables , we obtain @xmath102 using the asymptotic behavior of the bessel function ( [ asymbessel ] ) , we have that @xmath103 for almost every @xmath104 and for every @xmath91 .
we will next apply the dominated convergence theorem . to this end , using again relation ( [ asymbessel ] ) , it suffices to show that the integral @xmath105 is finite .
this is proved in the following lemma . in the above statement
we can replace the normalization @xmath106 by @xmath107 where @xmath108 is a deterministic function which converges to zero as @xmath109 .
this is a consequence of the proof below .
[ ln1 ] let @xmath110 be given by ( [ i ] ) .
then for @xmath111 the integral @xmath110 is finite .
* proof : * consider the integral @xmath112 if @xmath113 we have @xmath114 making the change of variables @xmath115 we get @xmath116 where @xmath117 is the _ beta _ function .
the expresion above is finite for @xmath118 .
+ now , if @xmath119 @xmath120 for the first integral in the right hand side we make the change of variables @xmath121 , and for the second one we make @xmath122 then we get @xmath123 wich is finite for @xmath111 . here
@xmath124 is the _ gamma _ function , @xmath125 is the _ hypergeometric _ function , and we have make use of the property @xmath126 and the formula ( see @xcite , formula 1.6.7 ) @xmath127 proceeding in a similar way with the other seven integrals we conclude that @xmath110 is finite .
we may note that the prescence of the @xmath128 function in the asymptotic behavior of the bessel function does not allow to obtain a renormalization for @xmath90 in terms of the powers of @xmath129 and @xmath130 like in @xcite .
the following proposition is a consequence of the proposition [ pn1 ] and of the hypercontractivity property of multiple stochastic integrals .
[ pn2 ] for every @xmath91 and for @xmath111 the sequence @xmath131 converges to zero a.s . as @xmath109 .
* proof : * as in @xcite we can replace the couple @xmath132 by a discrete sequence @xmath133 such that @xmath134 converge to infinity as @xmath135 this is possible since the nominator and denominator in ( [ theta - theta ] ) are continuous a.s . with respect to @xmath132 .
indeed , the fact that the divergence integral in the nominator is continuous follows from @xcite , page 293 since the integrand @xmath74 is regular enough , and the integral @xmath136 in the nominator is clearly continuous a.s . with respect to the couple @xmath132 . for simplicity
, we will assume that @xmath137 .
let us show that @xmath138 converges to zero a.s . as @xmath139 tend to infinity .
we will use the borel - cantelli lemma . to do this
, we will estimate @xmath140 for some @xmath141 .
for every @xmath142 we have @xmath143 and since @xmath144 is a multiple integral in the second wiener chaos , the inequality ( [ hyper ] ) and proposition [ pn1 ] implies that @xmath145 putting together the two above bounds , we get @xmath146 and this series is convergent when @xmath147 for every given @xmath91 and for @xmath148 large enough we can always chose a real number @xmath149 such that @xmath150 .
this , together with the borel - cantelli lemma allows us to finish the proof .
0.3 cm the next step is to analyze the denominator in formula ( [ theta - theta ] ) .
we have the following estimate .
[ pn3 ] for any @xmath91 and for any @xmath151 @xmath152 * proof : * by the isometry of multiple integrals ( [ iso ] ) , the equation ( [ pe ] ) and the expression of the solution to ( [ 1 ] ) @xmath153 making the change of variables @xmath154 and similar for the other variables we can write @xmath155 and thus @xmath156 we will use the same idea as in the proof of proposition [ pn1 ] .
we notice first that @xmath157 converges to infinity as @xmath20 by using ( [ asymbessel ] ) ; also for @xmath158 large enough @xmath159 is bounded by @xmath160 almost everywhere @xmath161 .
also we note that the integral @xmath162 is finite for @xmath92 by using the same computations as in the proof of lemma [ ln1 ] .
+ using fatou s lemma ( we use the following version of the fatou s lemma : if @xmath163 is a sequence of functions such that @xmath164 where @xmath165 is positive and integrable , then @xmath166 ) this implies that @xmath167 0.3 cm at this point we will need the following auxiliary lemma .
[ ln2 ] consider a sequence of random variables @xmath168 such that @xmath169 for some @xmath141 and for every @xmath170 ( which implies @xmath171 almost surely as @xmath172 . )
also consider a sequence of a.s .
strictly positive random variables @xmath173 such that @xmath174 as @xmath172 . then @xmath175 * proof : * we will use again the borel - cantelli lemma .
let @xmath176 be arbitrary
. then @xmath177 using the fact that @xmath174 as @xmath172 we obtain that @xmath178 for @xmath179 large enough .
by assumption , @xmath180 .
therefore @xmath181 and therefore the conclusion follows .
+ lemma [ ln2 ] can be extended without difficulty to two - parameter sequences .
that is , if @xmath144 is a sequence of random variables such that @xmath182 and @xmath183 is a sequence of positive random variables such that @xmath184 then @xmath185 0.3 cm let us state the main result of this section .
let @xmath186 be the estimator given by ( [ est1 ] ) .
suppose that @xmath92 .
then @xmath186 is a strongly consistent estimator for the parameter @xmath1 , that is , @xmath187 * proof : * from relation ( [ theta - theta ] ) , the difference between the estimator and the true parameter is @xmath188the result is obtained by using proposition [ pn2 ] , proposition [ pn3 ] and lemma [ ln2 ] ( and the remark that follows after this lemma ) .
we proved in the previous section that the least square estimator @xmath6 is strongly consistent .
this property has been proved in the one - dimensional case in ( @xcite ) .
nevertheless , we show in this paragraph that the limiting distribution of the estimator is not the same in the one - parameter and two - parameter cases .
this different behavior is somehow expected since the fractional ornstein - uhlenbeck sheet does not keep the properties of the fractional ornstein - uhlenbeck process ( for example , the kernel @xmath39 given by ( [ f ] ) can take any real value while in the one parameter case it is positive since it is given by an exponential function ) . in order to notice the difference between the one - parameter and the two - parameter case , we will focus only on the nominator in the right hand side of ( [ theta - theta ] ) . to check the asymptotic normality we use the following criterium :
( see theorem 4 in @xcite ) .
let @xmath189 , @xmath190 ( with @xmath191 for every @xmath192 ) be a sequence of square integrable random variables in the @xmath29 th wiener chaos such that @xmath193\rightarrow 1 $ ] as @xmath194 then the following are equivalent : we obtain @xmath212 \left [ i_{1 } \left ( f(\cdot , \cdot , t_{0},s_{0})\right ) + i_{1 } \left ( f(t_{0},s_{0},\cdot , \cdot ) \right ) \right]\\ & & \times \vert t - t_{0 } \vert ^{2\alpha -2 } \vert s - s_{0}\vert ^{2\beta -2 } \\
& = & \frac{1}{\sigma ^{2 } _ { t , s } } c(\alpha ) c(\beta ) \int_{0}^{t } dt \int_{0}^{s}ds \int_{0}^{t } dt_{0 } \int_{0 } ^{s}ds_{0 } \\ & & \times \left [ i_{2 } \left ( f(\cdot , \cdot , t , s ) \otimes f(\cdot , \cdot , t_{0 } , s_{0 } ) \right ) + i_{2 } \left ( f(\cdot , \cdot , t , s ) \otimes f(t_{0 } , s_{0 } , \cdot , \cdot ) \right)\right . \\ & & \left .
+ i_{2}\left ( f(t , s , \cdot , \cdot ) \otimes f(\cdot , \cdot , t_{0 } , s_{0 } ) \right)+i_{2}\left ( f(t , s , \cdot , \cdot ) \otimes f ( t_{0 } , s_{0 } , \cdot , \cdot ) \right)\right]\\ & & \times \vert t - t_{0 } \vert ^{2\alpha -2 } \vert s - s_{0 } \vert ^{2\beta -2 } + \mathbf{e}\vert d\frac{1}{\sigma _ { t , s}}f_{t , s } \vert _ { { \cal{h}}^{\alpha , \beta } } ^{2}\\ & : = & a^{(1)}_{t , s } + a^{(2)}_{t , s } + a^{(3)}_{t , s } + a^{(4)}_{t , s}+\mathbf{e}\vert d\frac{1}{\sigma _ { t , s}}f_{t , s } \vert _ { { \cal{h}}^{\alpha , \beta } } ^{2}\end{aligned}\ ] ] let us note that that @xmath213 is equal to @xmath205 .
this follows from the fact that for any multiple integral of order @xmath29 we have @xmath214 it remains to show that the terms containing multiple integrals of order 2 does not converges to zero in @xmath198 as @xmath109 .
the first and the fourth summand are similar , as they are the second one and the third one .
we will handle only the first summand denoted by @xmath215 ( because the other three terms can be studied analogously ) .
we can write , using the definition of the scalar product in the hilbert space @xmath216 and the expression ( [ fjo ] ) of the kernel @xmath39 in terms of the bessel function @xmath217 + @xmath218 making @xmath219 and similar for the other variables we get @xmath220 using the asymptotic behavior of the bessel function when its variable is close to infinity ( see ( [ asymbessel ] ) ) we see that @xmath221 and considering the fact that the last integral is finite for @xmath222 ( the proof of this fact is similar to the proof of lemma [ ln1 ] ) it is straightforward to see that , for @xmath94 close to infinity , the quantity @xmath223 does not converges to zero . + | we will study the least square estimator @xmath0 for the drift parameter @xmath1 of the fractional ornstein - uhlenbeck sheet which is defined as the solution of the langevin equation @xmath2\times [ 0,s].\ ] ] driven by the fractional brownian sheet @xmath3 with hurst parameters @xmath4 in @xmath5 . using the properties of multiple wiener - it integrals we prove that the estimator is strongly consistent for the parameter @xmath1 .
in contrast to the one - dimensional case , the estimator @xmath6 is not asymptotically normal .
0.3 cm * 2010 ams classification numbers : * 60g15 , g0h07 , 60g35 , 62m40
. 0.3 cm * key words : * fractional brownian sheet , parameter estimation , multiple wiener - it integrals , strong consistency . |
physical objects in the world appear differently depending on the scale of observation / measurement .
take the tree as an example , meaningful observations range from molecules at the scale of nanometers , to leaves at centimeters , to branches at meters , and to forest at kilometers .
this inherent property is ubiquitous and holds equally true for natural language . on the one hand ,
concepts are meaningful only at the right resolution , for instance , named entities usually range from unigram ( e.g. , new " ) to bigram ( e.g. , new york " ) , to multigram ( e.g. , new york times " ) , and even to a whole long sequence ( e.g. , a song name another lonely night in new york " ) . on the other hand
, our understanding of natural language depends critically on the scale at which it is examined , for example , depending on how much detailed we would like to get into a document , our knowledge could range from a collection of _ keywords _ " , to a sentence sketch named
_ title _ " , to a paragraph summary named _ abstract _ " , to a page long _ introduction _ " and finally to the _ entire content_.
the notion of scale is fundamental to the understanding of natural language , yet it was largely ignored by existing models for text representation , which include simple bag - of - word ( bow ) or unigram language model ( lm ) , n - gram or higher order lms , and other more advanced text / language models @xcite . one key problem with many of these models is their inflexibility they capture the semantic structure rather rigidly at only a single resolution ( e.g. , @xmath0-gram with a single fixed value of @xmath0 ) .
however , which scale is appropriate for a specific task is usually unknown a priori and in many cases even not homogeneous ( e.g. , a document may contain named entities of different length ) , making it impossible to capture the right meanings with a fixed single scale .
scale space theory is a well - established and promising framework for multi - resolution representation , developed primarily by the computer vision and signal processing communities with complimentary motivations from physics and bio - vision .
the key idea is to embed a signal into the _ scale space _
, i.e. , to represent it as a family of progressively smoothed signals parameterized by a continuous variable of _ scale _ , where fine - resolution detailed structures are progressively suppressed by the convolution of the original signal with a smoothing kernel ( i.e. , a low pass filter with certain properties ) @xcite . in this paper
, we adapt the scale - space model from image to text signals , proposing a novel framework that enables multi - resolution representation for documents .
the adaptation poses substantial challenges as the structure of the semantic domain is nontrivially complicated than the spatial domains in traditional image scale space .
we show how this can be made possible with a set of assumptions and simplifications . the scale - space model for text
not only provides new perspectives for how text analysis tasks can be formulated and addressed , but also enables well - established computer vision tools to be adapted and applied to text processing , e.g. , matching , segmentation , description , interests points detection , and classification . to stimulate further investigation in this promising direction
, we initiate a couple of instantiations to demonstrate how this model can be used in a variety of nlp and text analysis tasks to make things easier , better , and most importantly , scale - invariant .
the notion of scale space is applicable to signals of arbitrary dimensions .
let us consider the most common case , where it is applied to 2-dimensional signals such as images .
given an image @xmath1 , its scale - space representation @xmath2 is defined by : @xmath3 where @xmath4 denotes the convolution operator , and @xmath5 is a smoothing kernel ( i.e. , a low pass filter ) with a set of desired properties ( i.e. , the scale - space axioms @xcite ) .
the bandwidth parameter @xmath6 is referred to as scale parameter since as @xmath6 increases , the derived image will become gradually smoother ( i.e. , blurred ) and consequently more and more fine - scale structures will be suppressed .
it has been shown that the gaussian kernel is the unique option that satisfies the conditions for _ linear scale space _ : @xmath7 the resultant linear scale space representation @xmath2 can be obtained equivalently as a solution to the diffusion ( heat ) equation @xmath8 with initial condition @xmath9 , where @xmath10 denotes the laplace operator which in a 2-dimensional spatial space corresponds to @xmath11 .
if we view @xmath12 as a heat distribution , the equation essentially describes how it diffuses from initial value , @xmath13 , in a homogeneous media with uniform conductivity over time @xmath6 .
as we can imagine , the distribution will gradually approach uniform and consequently the fine - scale structure of @xmath13 will be lost .
scale - space theory provides a formal framework for handling the multi - scale nature of both the physical world and the human perception . since its introduction in 1980s
, it has become the foundation of many computer vision techniques and been widely applied to a large variety of vision / image processing tasks . in this paper
, we show how this powerful tool can be adapted and applied to natural language texts .
a straightforward step towards textual sale space would be to represent texts in the way as image signal . in this section ,
we show how this can be made possible .
other alternative signal formulations will be discussed in the followed section .
let @xmath14 be our vocabulary consisting of @xmath15 words , given a document @xmath16 comprised of a finite @xmath17-word sequence @xmath18 , without any information loss , we can characterize @xmath16 as a 2d @xmath19 binary matrix @xmath13 , with the @xmath20-th entry @xmath21 indicates whether or not the @xmath22-th vocabulary word @xmath23 is observed at the @xmath24-th position , i.e. : @xmath25 , where @xmath26 if @xmath27 and 0 otherwise .
hereafter , we will refer to the @xmath24-axis as _ spatial domain _
( i.e. , positions in the document , @xmath28 ) , and @xmath22-axis as the _ semantic axis _ ( i.e. , indices in the vocabulary , @xmath29 ) .
this representation provides an image analogy to text , i.e. , a document @xmath13 is equivalent to a black - and - white image except that here we have _ one spatial _ and _ one semantic _ domains , @xmath20 , instead of two spatial domains , @xmath30 .
interestingly , scale - space representation can also be motivated by this binary model from a slightly different perspective , as a way of robust density estimation .
we have the following definition : definition 1 . _ a 2d text model _
@xmath31 _ is a probabilistic distribution over the joint spatial - semantic space _ : @xmath32 , @xmath33 , @xmath34 .
this 2d text model defines the probability of observing a semantic word @xmath22 at a spatial position @xmath24 . the binary matrix representation ( after normalization )
can be understood as an estimation of @xmath13 with kernel density estimators : @xmath35 where @xmath36 is the @xmath37-th column vector of an identity matrix , @xmath38 denotes the @xmath24-th row vector and @xmath39 the @xmath22-th column vector .
note that here the dirac impulse kernels @xmath40 is used , i.e. , words are unrelated either spatially or semantically .
this contradicts the common knowledge since neighboring words in text are highly correlated both semantically @xcite and spatially @xcite .
for instance , observing the word new " indicates a high likelihood of seeing the other word york " at the next position . as a result
, it motivates more reliable estimate of @xmath13 by using smooth kernels such as gaussian @xcite , which , as we will see , leads exactly to the gaussian filtering used in the linear scale - space theory .
the 2d binary matrix described above is not the only option we can work with in scale space . generally speaking
, any vector , matrix or even tensor representation of a document can be used as a signal upon which scale space filtering can be applied . in particular
, we use the following in the current paper : _ word - level 2d _ signal , @xmath21 , is the binary matrix we described in [ sec : word2d ] .
it records the spatial position for each word , and is defined on the joint spatial - semantic domains .
_ bag - of - word 1d _ signal is the bow representation @xmath41 , i.e. , the 2d matrix is collapsed to a 1d vector .
since the spatial axis is wiped out , this signal is defined on the semantic domain alone .
_ sentence - level 2d _ signal is a compromise between word - level 2d and the bow signals . instead of collapsing the spatial dimension for the whole document , we do it for each sentence . as a result , this signal , @xmath21 , records the position of each sentence ; for a fixed position @xmath42 , @xmath43 records the bow of the corresponding sentence .
_ topic 1d _ signal , @xmath44 , is composed of the topic embedding of each sentence and defined on the spatial domain only .
assume we have trained a topic model ( e.g. , latent dirichlet allocation ) on a universal corpus in advance , this signal is obtained by applying topic inference to each sentence and recording the topic embedding @xmath45 , where @xmath46 is the dimensionality of the topic space .
topic embedding is beneficial since it endows us the ability to address synonyms and polysemy .
also note that the semantic correlation may have been eliminated and consequently semantic smoothing is no longer necessary .
in other words , although @xmath44 is a matrix , we would rather treat it as a vector - variate 1d signal .
all these textual signals involve either a semantic domain or both semantic and spatial domains . in the following ,
we investigate how scale - space filtering can be applied to these domains respectively .
spatial filtering has long been popularized in signal processing @xcite , and was recently explored in nlp by @xcite .
it can be achieved by convolution of the signal with a low - pass spatial filter , i.e. , @xmath47 . for texts ,
this amounts to borrowing the occurrence of words at one position from its neighboring positions , similar to what was done by a cache - based language model @xcite . in order not to introduce spurious information , the filter @xmath48 need to satisfy a set of scale - space axioms @xcite .
if we view the positions in a text as a spatial domain , the gaussian kernel @xmath49 or its discrete counterpart @xmath50 are singled out as the unique options that satisfy the set of axioms leading to the linear scale space , where @xmath51 denotes the modified bessel functions of integer order . alternatively ,
if we view the position @xmath24 as a time variable as in the markov language models , a poisson kernel @xmath52 is more appropriate as it retains temporal causality ( i.e. , inaccessibility of future data ) .
semantic filtering attempts to smooth the probabilities of seeing words that are semantically correlated .
in contrast to the spatial domain , the semantic domain has some unique properties .
the first thing we notice is that , as semantic coordinates are nothing but indices to the dictionary , we can permute them without changing the semantic meaning of the representation .
we refer to this property as _
permutation invariance_. semantic smoothing has been extensively explored in natural language processing @xcite .
classical smoothing methods , e.g. , laplacian and dirichlet smoother , usually shrink the original distributions to a predefined reference distribution .
recent advances explored local smoothing where correlated words are smoothed according to their interrelations defined by a semantic network @xcite . given a semantic graph @xmath53 , where two correlated words @xmath23 and @xmath54 are connected with weight @xmath55 , semantic smoothing can be formulated as solving a graph - based optimization problem : @xmath56 where @xmath57 defines the tradeoff , @xmath58 weights the importance of the node @xmath23 .
interestingly , the solution to eqn.([eq7 ] ) is simply the convolution of the original signal with a specific kernel ) . ] , i.e. , @xmath59 .
compared with spatial filtering , semantic filtering is , however , more challenging .
in particular , the semantic domain is heterogeneous and not shift - invariant the degree of correlation @xmath55 depends on both coordinates @xmath22 and @xmath60 rather than their difference @xmath61 . as a result , kernels that provably satisfy scale - space axioms are no longer feasible . to this end
, we simply set aside these requirements and define kernels in terms of the dissimilarity @xmath62 between a pair of words @xmath22 and @xmath60 rather than their direct difference @xmath61 , that is , @xmath63 , where we use @xmath64 to denote semantic kernel to distinguish from spatial kernels @xmath65 .
for gaussian , this means @xmath66 .
scale is vital for the understanding of natural language , yet it is nontrivial to determine which scale is appropriate for a specific task at hand in advance . as a matter of fact
, the best choice usually varies from task to task and from document to document . even within one document , it could be heterogeneous , varying from paragraph to paragraph and sentence to sentence .
for the purpose of automatic modeling , there is no way to decide _ a priori _ which scale fits the best .
more importantly , it might be impossible to capture all the right meanings at a single scale .
therefore , the only reasonable way is to simultaneously represent the document at multiple scales , which is exactly the notion of _ scale space_. scale space representation embeds a textual signal into a _
continuous _ scale - space , i.e. , by a family of progressively smoothed signals parameterized by continuous scale parameters .
in particular , for a 2d textual signal @xmath21 , we have : @xmath67 where the 2d smoothing kernel @xmath48 is separable between spatial and semantic domains , i.e. , @xmath68 note that we have two continuous scale parameters , the spatial scale @xmath69 and the semantic scale @xmath70 . the case for 1d signals are even simpler as they only involve one type of kernels ( spatial or semantic ) . for a 1d spatial signal @xmath71
, we have @xmath72 , and for a semantic signal @xmath73 , @xmath74 . and
if @xmath75 is a vector - variate signal , we just apply smoothing to each of its dimensions independently .
[ [ example . ] ] example .
+ + + + + + + + as an example , figure [ fig1 ] shows four samples , @xmath76 , from the scale - space representation @xmath77 of a synthetic short text
_ new york times offers free iphone 3 g as gifts for new customers in new york _ " , where @xmath78 , the two scales are set equal @xmath79 for ease of explanation and @xmath12 is obtained based on the word - level 2d signal .
we use a vocabulary containing 12 words ( in order ) : new " , york " , time " , free " , iphone " , gift " , customer " , apple " , egg " , city " , service " and coupon " , where the last four words are chosen because of their strong correlations with those words that appear in this text .
the semantic graph is constructed based on pairwise mutual information scores estimated on the rcv1-v2 corpus as well as a large set of web search queries .
the ( 0,0)-scale sample , or the original signal , is a @xmath80 binary matrix , recording precisely which word appears at which position .
the smoothed signals at ( 1,1 ) , ( 2,2 ) and ( 8,8)-scales , on the other hand , capture not only short - range spatial correlations such as bi - gram , tri - gram and even higher orders ( e.g. , the named entities new york " and new york times " ) , but also long - range semantic dependencies as they progressively boost the probability of latent but semantically related topics , e.g. ,
iphone " @xmath81 apple " , customer " @xmath81 service " , free " and gift " @xmath81 coupon " , new " and iphone " @xmath81 egg " ( due to the online electronics store ` newegg.com ` ) .
the scale - space representation creates a new dimension for text analysis . besides providing a multi - scale representation that allows texts to be analyzed in a scale - invariant fashion
, it also enables well - established computer vision tools to be adapted and applied to analyzing texts .
the scale space model can be used in nlp and text mining in a variety of ways . to stimulate further research in this direction
, we initiate a couple of instantiations . in this section ,
we show how to make text classification scale - invariant by exploring the notion of _ scale - invariant text kernel _ ( sitk ) . given a pair of documents , @xmath16 and @xmath82 , at any fixed scale @xmath6
, the representation @xmath12 induces a single - scale kernel @xmath83 , where @xmath84 denotes any inner product ( e.g. , frobenius product , gaussian rbf similarity , jensen - shannon divergence ) .
this kernel can be made scale - invariant via the expectation : @xmath85=\int_0^\infty k_s(d , d^\prime)q(s)ds,\label{eq10}\end{aligned}\ ] ] where @xmath86 is a probabilistic density over the scale space @xmath87 with @xmath88 and @xmath89 , which in essence characterizes the distribution of the most appropriate scale .
@xmath86 can be learned from data via a em procedure or in a bayesian framework if our belief about the scale can be encoded into a prior distribution @xmath90 . as an example
, we show below one possible formulation . given a training corpus @xmath91 , where @xmath16 is a document and @xmath22 its label
, our goal in text classification is to minimize the expected classification error . to simplify matters ,
we assume a parametric form for @xmath86 .
particularly , we use the gamma distribution @xmath92 due to its flexibility .
moreover , we propose a formulation that eliminates the dependence on the choice of the classifier , which approximately minimizes the bayes error rate @xcite , i.e. : @xmath93\\ \end{split}\ ] ] where @xmath94 is a heuristic margin ; @xmath95 , called
nearest - hit " , is the nearest neighbor of @xmath96 with the same class label , whereas @xmath97 , the nearest - miss " , is the nearest neighbor of @xmath96 with a different label , and the distance @xmath98 .
this above formulation can be solved via a em procedure .
alternatively , we can discretize the scale space ( preferably in log - scale ) , i.e. , @xmath99 , and optimize a discrete distribution @xmath100 directly from the same formulation . in particular , if we regularize the @xmath101-norm of @xmath102 , eq([eq11 ] ) will become a convex optimization with a close - form solution that is extremely efficient to obtain : @xmath103 where @xmath104^\top$ ] , the average margin vector @xmath105^\top$ ] with entry @xmath106 , and @xmath107 denotes the positive - part operator .
[ [ experiments . ] ] experiments .
+ + + + + + + + + + + + we test the scale - invariant text kernels ( sitk ) on the rcv1-v2 corpus with focus on the 161,311 documents from ten leaf - node topics : ` c11 , c24 , c42 , e211 , e512 , gjob , gpro , m12 , m131 ` and ` m142 ` .
each text is stop - worded and stemmed .
the top 20k words with the highest dfs ( document frequencies ) are selected as vocabulary ; all other words are discarded .
the semantic network is constructed based on pairwise mutual information scores estimated on the whole rcv1 corpus as well as a large scale repository of web search queries , and further sparsified with a cut - off threshold .
we implemented the sentence - level 2d , the lda 1d signals and bow 1d for this task .
for the first two , the documents are normalized to the length of the longest one in the corpus via bi - linear interpolation .
we examined the classification performance of the svm classifiers that are trained on the _ one - vs - all _ splits of the training data , where three types of kernels ( i.e. , linear ( frobenius ) , rbf gaussian and jensen - shannon kernels ) were considered .
the average test accuracy ( i.e. , micro - averaged f1 ) scores are reported in table [ tab1 ] . as a reference , the results by bow representations with tf or tfidf attributes are also included .
for all the three kernel options , the scale - space based sitk models significantly ( according to @xmath108-test at @xmath109 level ) outperform the two bow baselines , while the sentence level sitk performs substantially the best with 7.8% accuracy improvement ( i.e. , 56% error reduction ) .
.text classification test accuracy .
we compared five models : the bag - of - word vector space models with tf or tfidf attributes , and the scale - invariant text kernels with bow 1d ( sitk.bow ) , lda 1d ( sitk.lda ) and sentence - level 2d ( sitk.sentence ) textual signal .
best results are highlighted in * bold*.[tab1 ] [ cols="<,^,^,^",options="header " , ] the extrema ( i.e. , maxima and minima ) of a signal and its first a few derivatives contain important information for describing the structure of the signal , e.g. , patches of significance , boundaries , corners , ridges and blobs in an image .
scale space model provides a convenient framework to obtain the extrema of a signal at different scales .
in particular , the extrema in the @xmath110-th derivative of a signal is given by the zero - crossing in the @xmath111-the derivative , which can be obtained at any scale in the scale space conveniently via the convolution of the original signal with the derivative of the gaussian kernel , i.e. : @xmath112 since gaussian kernel is infinitely differentiable , the scale - space model makes it possible to obtain local extrema / derivatives of a signal to arbitrary orders even when the signal itself is undifferentiable .
moreover , due to the non - enhancement of local extrema " property , local extrema are created _ monotonically _ as we decrease the scale parameter @xmath6 . in this section ,
we show how this can be used to detect keywords from a document in a _ hierarchical _ fashion .
the idea is to work with the word - level 2d signal ( other options are also possible ) and track the extrema ( i.e. , patterns of significance ) of the scale - space model @xmath12 through the zero - crossing of its first derivative @xmath113 to see how extrema progressively emerge as the scale @xmath6 goes from coarse to finer levels .
this process reduces the scale - space representation to a simple ternary tree in the scale space , i.e. , the so - called _ interval tree _ " in @xcite . since @xmath13 defines a probability over the spatial - semantic space , it is straightforward to interpret the identified intervals as keywords .
this algorithm therefore yields a _ keyword tree _ that defines topics we could perceive at different levels of granularities from the document . [
[ experiments.-2 ] ] experiments . + + + + + + + + + + + + as an illustrative example , we apply the hierarchical keywording algorithm described above to the current paper .
the keywords that emerged in order are as follows : scale space " @xmath81 kernel " , signal " , text " @xmath81 smoothing " , spatial " , semantic " , domains " , gaussian " , filter " , text analysis " , natural language " , word " @xmath114 . in the previous section , we show how semantic keywords can be extracted from a text in a hierarchical way by tracking the extrema of its scale space model @xmath12 . in the same spirit , here we show how topic boundaries in a text can be identified by tracking the extrema of the first derivative @xmath115 .
text segmentation is an important topic in nlp and has been extensively investigated previously @xcite .
many existing approaches , however , are only able to identify a flat structure , i.e. , all the boundaries are identified at a flat level .
a more challenging task is to automatically identify a _ hierarchical _
table - of - content style structure for a text , that is , to organize boundaries of different text units in a tree structure according to their topic granularities , e.g. , chapter boundaries at the top - level , followed in order by boundaries of sections , subsections , paragraphs and sentences as the level of depth increases .
this can be achieved conveniently by the _ interval tree _ and _ coarse - to - fine tracking _ idea presented in @xcite . in particular ,
if we keep tracking the extrema of the 1st order derivatives ( i.e. , rate of changes ) by looking at the points satisfying : @xmath116 due to the monotonicity nature of scale space representation , such contours are closed above but open below in the scale space .
they naturally illustrate how topic boundaries appear progressively as scale @xmath6 goes finer . and
the _ exact localization _ of a boundary can be obtained by tracking back to the scale @xmath117 .
also note that this algorithm , unlike many existing ones , does not require any supervision information .
[ [ experiments.-3 ] ] experiments . + + + + + + + + + + + + as an example , we apply the hierarchical segmentation algorithm to the current paper .
we use the sentence level 2d signal .
let @xmath118 denote the vector @xmath119 , where the semantic scale @xmath120 is fixed to a constant @xmath121 , and the semantic index @xmath22 enumerates through the whole vocabulary @xmath122 .
we identify hierarchical boundaries by tracking the zero contours @xmath123@xmath124 ( where @xmath125 denotes @xmath101-norm ) to the scale @xmath117 , where the length of the projection in scale space ( i.e. , the vertical span ) reflects each contour line s topic granularity , as plotted in figure [ fig2 ] ( top ) . as a reference , the velocity magnitude curve ( bottom ) @xmath126@xmath127 , and the true boundaries of sections ( red - dashed vertical lines in top figure ) and subsections ( green - dashed )
are also plotted .
as we can see , the predictions match the ground truths with satisfactorily high accuracy .
this paper presented scale - space theory for text , adapting concepts , formulations and algorithms that were originally developed for images to address the unique properties of natural language texts .
we also show how scale - space models can be utilized to facilitate a variety of nlp tasks .
there are a lot of promising topics along this line , for example , algorithms that scale up the scale - space implementations towards massive corpus , structures of the semantic networks that enable efficient or even close - form scale - space kernel / relevance model , and effective scale - invariant descriptors ( e.g. , named entities , topics , semantic trends in text ) for texts similar to the sift feature for images @xcite . | scale - space theory has been established primarily by the computer vision and signal processing communities as a well - founded and promising framework for multi - scale processing of signals ( e.g. , images ) . by embedding an original signal into a family of gradually coarsen signals parameterized with a continuous scale parameter , it provides a formal framework to capture the structure of a signal at different scales in a consistent way . in this paper , we present a scale space theory for text by integrating semantic and spatial filters , and demonstrate how natural language documents can be understood , processed and analyzed at multiple resolutions , and how this scale - space representation can be used to facilitate a variety of nlp and text analysis tasks . |
blazars are the best laboratory to study the physics of relativistic jets .
the non - thermal continuum amplified by relativistic beaming ( e.g. , urry & padovani 1995 ) , is a unique tool to probe the physical processes acting in the jet . in the last decade
the interest for blazars has been renewed by the egret discovery of 66 blazars ( hartman et al .
1999 ) as strong @xmath1-ray emitters .
in several cases the @xmath1-ray emission dominates the apparent bolometric luminosity .
moreover the short variability timescales , together with the absence of absorption of high energy photons ( through the @xmath2-pair production ) , directly imply that the @xmath1-ray emission is also relativistically beamed ( e.g. dondi & ghisellini 1995 ) . despite
the large variety of classifications , there is growing evidence that blazars form a single population .
their different spectral properties can be unified within a spectral sequence in which the leading parameter is the total luminosity ( fossati et al .
1998 ) . at the high - luminosity extreme of the sequence we find the flat spectrum radio quasars , the most luminous blazars , characterized by the presence of bright emission lines in the optical - uv spectrum and , in some cases , strong _ uv bumps _ , signatures of the underlying accretion process ( e.g. pian et al .
1999 and references therein ) .
most of the blazars detected by egret ( hartman et al .
1999 ) belong to this sub - class but not all fsrqs with comparable seds have been detected in the @xmath1-ray domain . in the study of fsrqs , observations in the x - ray band play a fundamental role . the x - ray emission from fsrq
is dominated by the continuum originating from inverse compton scattering between relativistic electrons in the jet and the soft photons produced by the disk and/or in the broad line region ( dermer & schlickeiser 1993 ; sikora , begelman & rees 1994 ) . the possibility to measure the ic component yelds important constraints on the emission models .
moreover the study of the soft x - ray spectrum ( below 1 kev ) provides insight on the presence of intrinsic absorption , on the possible contribution of the high energy end of the synchrotron emission or on the minimum energy of emitting particles .
the latter point is particularly interesting because the determination of the minimum energy of particles is a fundamental step in the estimate of the total power transported by the jet ( e.g. celotti , padovani & ghisellini 1997 ) .
the _ beppo_sax satellite has the unique capability of covering the wide energy range @xmath3 kev . for this reason it is ideal to study fsrqs .
in particular the knowledge of the spectrum in this band , together with the @xmath1-ray spectrum measured by egret , provide the most complete spectral information on the high energy component ( see e.g. , tavecchio et al . 2000 , hereafter paper i ) .
for these reasons we started an observational program with _
beppo_sax of a sample of fsrqs , ( containing a total of 50 sources ) extracted from the 2-jy sample by padovani & urry ( 1992 ) . using the flux threshold @xmath4jy
the number of sources reduces to 19 . in paper
i we discussed the case of three fsrq detected by egret , namely 0836 + 710 , 1510 - 089 and 2230 + 114 . here
we report the analysis of @xmath0sax data of 6 other sources , 3 detected by egret and three without an egret detection . adding these 6 sources to the 3 fsrqs analyzed in paper i and other 3 sources ( 3c279 , 0528 + 134 , pks 0537 - 441 ) discussed elsewhere ( maraschi et al . 1998 ; ghisellini et al . 1999 ,
pian et al . , in prep ) a total of 12 fsrqs ( out of 19 with x - ray flux @xmath5jy ) from the 2 jy sample were observed by @xmath0sax .
the sources discussed here are listed in table 1 .
partial and preliminary results were reported in maraschi & tavecchio ( 2001 ) the paper is organized as follows : in section 2 we report the analysis of _
beppo_sax observations , in section 3 we discuss the emission models for the seds and finally in sect .
4 we discuss our results .
the scientific payload of the @xmath0sax satellite ( see boella et al .
1997 ) contains four coaligned narrow field instruments ( nfis ) and two wide field cameras .
two of the nfis use concentrators to focalize x - rays : the low energy concentrator spectrometer ( lecs ) has a detector sensitive to soft - medium x - ray photons ( 0.1 - 10 kev ) , while the medium concentrator spectrometer ( mecs ) can detect photons in the medium energy range 1.3 - 10 kev .
the phoswich detector system ( pds ) , sensitive from 12 up to 200 kev , consisting of four identical units , uses rocking collimators so as to monitor source and background simultaneously with interchangeable units .
we will not be concerned here with the fourth nfi , a high pressure gas scintillation proportional counter ( hpgspc ) .
the _ beppo_sax journal of observations is reported in table 2 , with exposure times and observed count rates .
none of the sources showed significant flux variations during the observations .
we therefore obtained a cumulative spectrum for each source .
we analyzed the _
beppo_sax spectral data using the standard software packages xselect ( v 1.4b ) and xspec ( v 11.0 ) and the september 97 version of the calibration files released by the @xmath0sax scientific data center ( sdc ) . from the event files we extracted the lecs and mecs spectra in circular regions centered around the source with radii of 8@xmath6 and 4@xmath6 respectively
( see the sax analysis cookbook ) .
the pds spectra extracted with the standard pipeline with the rise - time correction were directly provided by the @xmath0sax sdc .
we used pds data rebinned with s / n@xmath7 .
for the spectral analysis we considered the lecs data in the restricted energy range 0.1 - 4 kev , because of known unsolved problems with the response matrix at higher energies .
background spectra extracted from blank field observations at the same position as the source were used .
we fitted rebinned lecs , mecs and pds spectra jointly , allowing for two variable different normalization factors to take into account uncertainties in the intercalibration of different instruments ( see sax cookbook ) . the first model chosen for the fitting procedure was in all cases a single power law plus galactic absorption . in the following we discuss in detail the results for each source
the results of the spectral fits are summarized in table 3 and 4 .
the residuals of a simple power - law model to the lecs+mecs data show the presence of a prominent absorption feature , located at @xmath8 kev , ( see fig.1 ) and an excess at about 5 kev .
inspection of the individual mecs2 and mecs3 spectra reveals that the emission feature at 5 kev is present only in the data obtained with the mecs3 indicating an instrumental origin .
only the mecs2 data are used below .
the low energy feature can be modeled either with an absorption edge ( @xmath9 ) or with an absorption trough ( @xmath10 ) with comparable probability ( @xmath11 ( @xmath12 ) and @xmath13 ( @xmath14 ) , respectively ) .
however these features find no obvious interpretation at the redshift of the source .
we then tried a warm absorber model ( absori of xspec ) , at the redshift of the quasar , obtaining a satisfactory fit ( @xmath15 ) .
the fit requires a column density of the absorber equal to @xmath16 @xmath17 and an ionization parameter @xmath18 , providing a reasonable physical interpretation .
the results of the fits are reported in table 4 .
previous rosat ( sambruna 1997 ) and asca ( kubo et al .
1998 ) spectra show a power - law x - ray spectrum with photon index @xmath19 , similar to that found with .
notably these observations ( with comparable flux ) showed a smooth continuum , without any evidence for spectral features .
the pds / mecs normalization factor required by the data ( @xmath20 ) is well above the 90% upper limit ( 0.93 ) given by the sax cookbook .
0208 - 512 is located in a rich field : although the lecs and mecs images do not show other sources , a search using the nasa extragalactic database reveals the presence of 13 sources catalogued as qsos or x - ray sources , within a circular region with radius 1 deg centered on our target .
it is then possible that the high - energy spectrum is contaminated by one or more sources present in the fov ( @xmath21 deg ) of the pds . in the fit presented here the pds data are not included .
the lecs+mecs spectrum is well fitted by an absorbed power - law ( @xmath22 [ 1.73 - 1.85 ] ) with column density consistent with the galactic value .
previous x - ray observations with exosat ( @xmath23 , pian et al .
1996 ) and rosat ( @xmath24 , sambruna 1997 ) are consistent with our results .
a power law model ( @xmath25 ) with galactic absorption fits the mecs+pds data for @xmath26 kev up to @xmath27 kev .
the inspection of the lecs image reveals the presence of several sources ( not present in the mecs image ) : in particular a low - luminosity seyfert galaxy , ngc 6212 , located 5@xmath6 away from the quasar hence within the standard extraction region of the lecs data .
we conclude that the lecs spectrum could be contaminated by the presence of other sources close to 1641 + 399 .
in fact the residuals of a fit with an absorbed power - law model of the lecs data alone ( @xmath28 , @xmath29 ) show large scatter around the model ( fig.3 ) , with the presence of possible features ( with no straightforward interpretation ) at energies @xmath30 kev and below @xmath311.0 kev .
the features can not simply be modeled as due to absorption of the continuum by a warm absorber at the redshift of the source .
a clear soft excess , interpreted as the hard tail of the uv bump , was present in the rosat data discussed by sambruna ( 1997 ) .
an absorbed power - law with galactic absorption provides a good fit to the data .
our spectrum ( @xmath32 ) is steeper than that measured on one occasion by ginga ( @xmath33 , lawson et al .
moreover we do not confirm the rosat observations that seem to indicate an absorption column higher than the galactic value at the 99% confidence level ( sambruna 1997 ) .
a power - law with intermediate spectral index ( @xmath34 ) fits well the data .
the rosat observation reported by siebert et al .
( 1998 ) showed a very steep spectrum , with @xmath35 and a flux about 2.5 times larger than during our pointing .
given the large flux variation , the steep spectrum measured by rosat could possibly represent intrinsic spectral variability possibly due to the high energy end of the synchrotron component , extending up to the soft x - ray band . as shown in table 3 a fit with a power - law model and galactic absorption
gives an unacceptable fit to the data .
using a free value for @xmath36 the fit improves significantly , requiring a column density higher than the galactic value at the 99% confidence level ( see the contour plot in fig.4 ) .
the value of the required absorption column density in the quasar rest - frame is @xmath37 @xmath17 .
an acceptable fit ( @xmath38 ) of the low - energy portion of the spectrum can also be obtained with a broken power - law model ( and galactic column density ) , with a very flat ( @xmath39 ) soft spectrum .
this , as in the case of other sources ( e.g. 0836 - 710 , paper i ) , could be due to an intrinsic break in the continuum . from a statistical point of view
both models have almost the same probability .
using @xmath0sax and historical data we assembled the seds of the sources , shown in fig.5 .
egret data are the average over the available observations as given by hartman et al .
( 1999 ) , while upper limits are from thompson et al .
( 1995 ) . for 0208 - 512
we report also the comptel detection reported in blom et al .
( 1995 ) . in the ir - optical bands , for 2223 - 052 are available quasi - simultaneous iso ( reported in haas et al . 1998 ) and optical data.for other 3 sources ( 1641 + 399 , 2243 - 123 , 2251 + 158 ) we report average optical fluxes with the observed variability range . for 1641 + 399
xie et al .
( 1999 ) report a variability amplitude in the @xmath40-band @xmath41 mag ( which is shown in fig .
5 ) , while for the other 2 sources the variability observed is rather small ( less than 1 mag . ) . for 0208 - 512 and 0521 - 365
we report only the data we could find in literature .
both sources have iue data .
references to the data are reported in the figure caption .
the double - humped shape of the spectral energy distribution ( seds ) is widely interpreted as due to synchrotron - inverse compton emission . the low energy component is synchrotron radiation from a population of relativistic electrons , while the high energy component is believed to be produced through inverse compton scattering of soft photons by the same electrons ( although other scenarios have been considered , e.g. hadronic models , see mannheim & biermann 1992 ) .
the nature of soft photons involved in the ic process is still debated . in blazars where thermal features ( emission lines , the _ blue bump _ ) are weak or even absent ( as in bl lac objects ) the energy density of target photons
is probably dominated by those produced by synchrotron ( ssc model , e.g. maraschi , ghisellini & celotti 1992 ) , while in powerful quasar ( fsrq ) photons coming into the jet from the external environment form the dominant population ( ec model , dermer & schlickeiser 1993 ; sikora , begelman & rees 1994 ) .
we have reproduced the overall seds using a synchrotron - inverse compton ( ssc+ec ) model .
the source is modeled as a sphere with radius @xmath42 , tangled magnetic field with intensity @xmath40 , in motion with bulk lorentz factor @xmath43 at an angle @xmath44 with respect to the line of sight .
relativistic effects are then regulated by the relativistic doppler factor , given by : @xmath45^{-1}$ ] . usually for blazars @xmath46 , implying @xmath47 .
relativistic electrons emit through synchrotron and ic mechanisms .
ic emissivity is calculated taking into account the full klein - nishina cross - section ( following the jones 1968 treatment ) .
the electron energy distribution is modeled with the form : @xmath48 where @xmath49 is a normalization factor , @xmath50 is the break lorentz factor , @xmath51 and @xmath52 are the spectral indices below and above the break , respectively .
this particular form for the distribution function has been assumed on a purely phenomenological basis , in order to describe the curved shape of the sed .
the external radiation field is modelled as a black - body peaking at @xmath53 hz and energy density @xmath54 .
the last quantity is calculated assuming that the radiation , with total luminosity @xmath55 ( @xmath56 , usually assumed to be @xmath57 is the fraction of the central emission reprocessed by the broad line region ) is diluted into a sphere of radius @xmath58 . for 2251 + 158 which shows a clear uv bump we take @xmath59 , while for the other cases we inferred @xmath60 from the luminosity of broad emission lines , following the approach used by celotti et al .
luminosities are reported in table 5 .
the value of @xmath58 has been choosen to fit the observed data , but we checked that it is close to the value expected from the relation between the luminosity of the blr and the radius derived ( for radio - quiet quasars ) by kaspi et al .
the external radiation energy density is amplified in the blob s rest frame by a factor @xmath61 ( e.g. ghisellini et al .
moreover , as pointed out by dermer ( 1995 ) ( and confirmed by georganopoulos , kirk & mastichiadis 2001 ) , because of relativistic effects the beaming pattern of the ec radiation is narrower than that of the ssc emission .
when @xmath62 ( as assumed here ) , the amplification is the same for both processes . with these assumptions the model
is well constrained by the observation and we can obtain fairly robust estimates of the principal physical quantities in the jet ( see paper i ) .
the values of the parameters used to reproduce the seds are reported in table 5 . in all cases
the bulk of the ic component is reproduced with the ec emission , while in some cases the ssc spectrum ( peaking at lower energies ) partly accounts for the low energy portion of the x - ray spectrum .
the cut - off in the synchrotron spectrum around @xmath63 hz is due to self - absorption .
clearly the adopted one - zone model can not account for the radio emission , widely considered to be due to the superposition of the emission produced in the outer ( @xmath64 pc ) regions of the jet ( e.g. blandford & konigl 1979 ) .
we have presented the results of the analysis of 6 fsrqs observed with @xmath0sax . the smooth power - law ( @xmath65 ) x - ray continuum is produced through inverse compton scattering on photon external to the jet , except for the `` atypical '' source 0521 - 365 . in the case of two sources , 0208 - 512 and 2251 + 158 , there is evidence of more complex x - ray spectra .
in particular 0208 - 512 shows an absorption feature at @xmath8 kev , that we suggest could be explained allowing the presence of a warm absorption intercepting the emission produced in the jet .
the soft x - ray spectrum of 2251 + 158 shows a clear deficit of photons , consistent either with an intrinsic absorption ( with rest - frame column density @xmath66 @xmath17 ) or an intrinsic break in the continuum occurring below @xmath671 kev . in the latter case
the position of the break could provide important insights into the physics of the jet , in particular regarding the value of the minimum energy of the radiating particles . on the other hand
there is growing evidence that the presence of intrinsic absorption is common in high - redshift radio - loud quasars ( e.g. , reeves & turner 2000 ) .
indeed @xmath0sax observations of two distant ( @xmath68 ) blazars ( fabian et al . 2001a ,
b ) show the presence of huge absorption , with column density of the order of few @xmath69 @xmath17 . for three sources ( 0208 - 512 , 1641 + 399 and 2251 + 158 ) we detect hard x - ray emission up to @xmath70 kev . for 1641 + 399 and 2251 + 158
the hard x - ray component appears to be the smooth extrapolation of the soft - medium x - ray continuum , while in 0208 - 512 the pds spectrum is likely contaminated by another source present in the large fov of the instrument .
we note however that 0208 - 512 is one of the few blazars detected by comptel ( blom et al .
1995 ) and because of its exceptionally intense flux in the mev band is considered as a prototype of the so - called `` mev blazars '' .
therefore we can not completely exclude the possibility that the spectral component responsible for the mev flux contributes also in the pds band , explaining the observed excess .
the modelling of the seds shows that fsrqs ( except the case 0521 - 365 , see below ) are characterized by jets with doppler beaming factors larger than 10 and magnetic field of the order of @xmath71 g , roughly in equipartition with the emitting electrons .
the size of the emitting region is of the order of @xmath72 cm , consistent with the absence of short - timescale variability ( @xmath73 day ) in this class of blazars .
these results confirm , with a larger sample , the conclusions of paper i an interesting issue for blazars is the value of the minimum lorentz factor of the emitting electrons , @xmath74 . in paper i
we showed how , from the absence of a spectral break in the soft x - ray continuum , it is possible to infer that the electron energy distribution extends down to @xmath75 .
in fact a break in the ec continuum is expected at the frequency @xmath76 ( e.g. sikora et al . 1997 ) , corresponding to an energy @xmath21 kev ( in the rest frame of the source ) .
we recall that the lack of soft x - ray photons observed in 2251 + 158 ( as for the case of 0836 + 710 reported in paper i ) could be interpreted as due to such intrinsic break in the continuum .
observations with larger signal to noise ( possible with @xmath77 or @xmath78 ) will allow to disentangle the two possible interpretations , intrinsic curved continuum or absorption . in our fits
@xmath51 , the index of the low energy portion of the electron energy distribution , is smaller than 2 .
clearly such a flat spectrum can not be produced by the cooling of high energy electrons ( in that case ine would expect @xmath79 ) .
on the other hand this spectrum is even flatter than the standard prediction @xmath79 for a population of non - thermal electrons produced through fermi acceleration by a non - relativistic shock ( e.g. blandford & ostriker 1978 ) .
however recent calculations ( see the review by kirk & dendy 2001 ) show that relativistic shocks , such as those likely present in the jet of blazars , could produce extremely flat ( @xmath80 ) spectra .
the case of 0521 - 365 is atypical .
previous observational evidence suggested that the jet forms a relatively large angle with the line of sight .
pian et al .
( 1996 ) , using independent arguments ( in particular the presence of an optical jet , e.g .
scarpa et al .
1999 ) , concluded that @xmath81 .
assuming that the bulk lorentz factor of the emitting plasma is similar to that of the other blazars , @xmath82 , this implies that the emission from 0521 - 365 is weakly boosted ( @xmath83 ) .
the sed reported in fig.5 has been calculated assuming @xmath84 and @xmath85 , implying @xmath86 .
in this situation the x - ray/@xmath87-ray continuum , contrary to the case of the other sources , is dominated by the ssc emission , while the ec spectrum gives only a small contribution in the @xmath1-ray band .
part of the motivation for this investigation was to study the difference between the egret and non - egret sources .
in the whole fsrq sample observed with there are 9 egret sources and 3 sources not detected by egret .
the average x - ray spectral indices ( @xmath88 and @xmath89 for egret and no - egret sources , respectively ) indicate that , within the statistical uncertainty , the x - ray spectral characteristics are similar for both groups . the modelling of the seds confirms the strong similarity between the two groups .
the lack of the egret detection could be attributed to the variability behaviour typical of these sources in the @xmath1-ray band ( e.g. mukherjee et al .
1997 ) rather than to an intrinsic difference in the properties of the jet .
new @xmath1-ray missions ( agile , glast ) will help to further investigate these problems .
lcccccccc + date & start & end & lecs & net cts / s@xmath90 & mecs & net cts / s@xmath91 & pds & net cts / s + & & & exp.(s ) & & exp.(s ) & & exp.(s ) & + & & & & & & & & + + 14 - 15/01/01 & 13:37:10 & 13:00:09 & 15650 & @xmath92 & 34256 & @xmath93 & 15260 & @xmath94 + + 2 - 3/10/98 & 10:16:24 & 09:46:23 & 17752 & @xmath95 & 41082 & @xmath96 & 19052 & @xmath97 + + 19/02/99 & 03:55:43 & 20:13:57 & 10934 & @xmath98 & 25866 & @xmath99 & 13052 & @xmath100 + + 10 - 11/11/97 & 13:22:27 & 00:22:40 & 9200 & @xmath101 & 16180 & @xmath102 & 6860 & @xmath103 0.08 + + 18 - 19/11/98 & 18:41:31 & 11:49:58 & 10060 & @xmath104 & 27500 & @xmath105 & 13222 & @xmath106 + + 5 - 6/6/2000 & 15:36:05 & 21:44:38 & 17425 & @xmath107 & 48515 & @xmath108 & 22439 & @xmath109
+ + + lcccc + source & @xmath110 & @xmath36 & @xmath111}$ ] & @xmath112d.o.f . + & & 10@xmath113 @xmath17 & @xmath114 erg @xmath17 s@xmath115 & + * 0208 - 512 * ( le+me2+pds)@xmath91 & @xmath116 & @xmath117 & 4.7 & 51.16/28 + ( le+me ) & @xmath118 & @xmath119 & 8.72 & 69.35/89 + ( le+me+pds ) & @xmath120 & 1.12(fix ) & 5.2 & 65.26/45 + ( le+me ) & @xmath121 & @xmath122 & 1.25 & 14./16 + -&@xmath123 & 4.87 ( fix ) & - & 14.72/17 + ( le+me ) & @xmath124 & @xmath125 & 1.83 & 18.37/23 + ( le+me+pds ) & @xmath126 & 6.5(fix ) & 10.6 & 94.2/80 + -&@xmath127 & @xmath128 & - & 77.1/79 + + lcccc + model & @xmath129 & @xmath130 & @xmath131 & @xmath112d.o.f
. + pl+edge & @xmath132 & @xmath133 & @xmath134 & 13.25/17 + pl+notch & @xmath135 & @xmath136 & @xmath137&12.85/17 + pl+absori & @xmath138 & @xmath139 & @xmath140 & 15.13/17 + + + + ccccccccc + @xmath42 & b & @xmath141 & @xmath142 & n@xmath143 & @xmath52 & @xmath49 & @xmath144 & @xmath58 + @xmath72 cm & g & & & & & @xmath145 & @xmath146 erg s@xmath115 & @xmath147 cm + & & & & & & & & + + 1.5 & 1.5 & 18 & 100 & 1.4 & 3.8 & @xmath148 & 2.1 & 0.6 + + 2.0 & 0.3 & 3@xmath149 & @xmath150 & 1.25 & 4 & @xmath151 & @xmath152 & 0.8 + + 4.0 & 2.9 & 9.75 & 200 & 1.5 & 4.2 & @xmath153 & 3.7 & 0.6 + + 4.25 & 5.6 & 17 & 135 & 1.6 & 4.3 & @xmath154 & 8.5 & 1.1 + + 3.5 & 2.5 & 15 & 250 & 1.6 & 4.3 & @xmath154 & 5.6&0.9 + + 4.0 & 1.5 & 12 & 60 & 1.8 & 3.4 & @xmath155 & 4 . & 1.0 + + | we report the @xmath0sax observations of 6 flat spectrum radio quasars .
three of them have a clear detection up to 100 kev with the pds instrument . for 4 objects the x - ray spectrum
is satisfactorily fitted by a power - law continuum with galactic absorption .
2251 + 158 show the presence of absorption higher than the galactic value , while the spectrum of the source 0208 - 512 shows a complex structure , with evidence of absorption at low energy .
we construct the spectral energy distributions adding historical data to the broad band x - ray spectra obtained with @xmath0sax and reproduce them with a one - zone synchrotron - inverse compton model ( including both ssc and external compton ) .
the implications are briefly discussed . |
the am canum venaticorum ( am cvn ) class of ultracompact binaries provides an opportunity for probing the accretion - induced heating of a white dwarf ( wd ) over six orders of magnitude in the time - averaged accretion rate , @xmath1 . in these very tight ( @xmath8 min ) binaries ( see warner 1995 for an overview ) , the donor star is a very low mass ( @xmath9 ) helium wd and the accretor is either a he or a c / o wd .
though there are a few distinct am cvn formation scenarios ( e.g. nelemans et al . 2001 and references therein ) , they all evolve as mass - transferring binaries driven by the loss of angular momentum from gravitational radiation ( e.g. deloye , bildsten & nelemans 2005 ) .
this paper is an outgrowth of townsley and bildsten s ( 2004 , hereafter tb ) work on the reheating of wds in h - accreting cataclysmic variables .
however , the case of wd reheating in am cvn binaries is different because the accreted material is pure helium , making flashes an isolated rather than constant occurence , and the binary evolution proceeds only under angular momentum loss by gravitational radiation , leading to a plummeting @xmath1 with increasing @xmath2 .
we have found that the wd is heated in the earliest phases of binary evolution and cools at later times .
there is a critical decoupling accretion rate of @xmath10 below which the reheated wd will cool even while accreting .
we find the amount of time that has passed since the binary had this @xmath1 by using the evolutionary calculations of deloye et al .
simple cooling after this point predicts the wd luminosity , as well as the expected wd @xmath0 and absolute visual magnitude , @xmath6 ( assuming , of course , a wd mass , @xmath11 , and radius , @xmath12 ) .
we show that the accreting wd can become the dominant broadband optical light source in those binaries with orbital periods in excess of @xmath13 minutes . in 2
, we discuss the physics of the wd heating due to accretion and discuss the uncertainty in the total helium layer mass expected on the wd . since the wd luminosity , @xmath14 , is nearly always less than the accretion luminosity , @xmath15 , we explain in 3 how the binary s behavior allows the wd to be directly detected at different orbital periods . at early times , during direct impact accretion ( e.g. marsh and steeghs 2002 ) , the unpulsed uv / optical light is from the hot wd , whereas at longer orbital periods , the disk is often quiescent , allowing for the underlying wd to dominate the optical continuum of a number of these binaries .
we also speculate on what could be learned if one of these wds was caught in the db pulsational instability strip .
we make direct comparisons to the observations in 4 , and conclude in 5 .
the ability for accretion to heat the deep interior of a wd has been discussed extensively ( sion 1991 ; tb ) .
indeed , accretion of pure helium at rapid rates @xmath16 is known to flash unstably and was studied as a type ia progenitor scenario ( e.g. nomoto & sugimoto 1977 ; nomoto 1982 ; woosley & weaver 1994 ) . though eventually ruled out as a major mechanism for type ia supernovae ( solheim & yungelson 2005 ) , am cvn binaries accrete pure helium for a long enough time to reheat the accreting wd .
an additional feature of these binaries is the large dynamic range in the time averaged accretion rate , @xmath1 , onto the wd , allowing us to test the theories of re - heated wds , as discussed in 3 & 4 .
here we calculate the accretor s thermal evolution along with that of the binary for several am cvn systems that represent the expected ranges for this population .
motivated by the expected mass distribution of he wd donors that evolve into am cvn systems ( nelemans et al .
2001 ) , we chose an initial donor mass of @xmath17 .
the thermal state of the donor is very important for setting the relationship between @xmath1 and @xmath2 ( bildsten 2002 ; deloye & bildsten 2003 ; deloye et al .
2005 ) . for this study we chose two thermal states found by setting the initial entropy of the donor so that the system s @xmath2 at contact is greater than the @xmath2 at contact of 30% and 90% of the systems in the deloye et al .
( 2005 ) am cvn population evolution calculation .
this provides @xmath1 as a function of time , and we then use the quasi - static envelope methods of tb to heat the accreting wd . for simplicity , this evolution is performed at constant @xmath11 and with a constant helium layer .
we chose two masses @xmath18 and @xmath19 , each case with a @xmath20 accreted helium shell ( justified below ) .
initial @xmath11 values for the binary evolution were chosen such that @xmath21 matched that of the thermal calculations at late times ( long @xmath2 ) to ensure the correct @xmath2-time relation .
using these @xmath1 trajectories ( see figure 1 ) , we identified four relevant regimes for the state of the accreting wd .
chronologically since initiation of mass transfer ( and in order of decreasing @xmath1 ) , these are : ( 1 ) rapid accretion and thermally unstable he ignition ( @xmath22 ) , ( 2 ) core heating during a phase in which @xmath14 is determined from `` compressional '' heating in the wd envelope ( @xmath23 ) , ( 3 ) a transition period where the core temperature is near an equilibrium value set by @xmath1 ( @xmath24 ) , and ( 4 ) the wd cooling regime ( @xmath25 ) where @xmath14 is determined by the time , @xmath26 , since the heating stopped at a previous epoch where @xmath27 , the decoupling mass - transfer rate .
we now describe each regime . at the very earliest times , when @xmath28 , the accreted helium is compressed rapidly and ignites in a thermally unstable manner , leading to recurrent weak explosions ( fujimoto & sugimoto 1982 ) .
( where @xmath29amu is the energy released from helium fusion ) only appear at the brief ( @xmath30 yr ) and poorly understood onset of accretion . ]
it is uncertain how much of the accreted material is ejected ( see kato & hachisu 1999 ) .
as @xmath1 drops , the compression happens more slowly so that the envelope temperature is lower , and the helium shell can become quite large . thermal instabilities at this stage likely lead to detonations ( fujimoto & sugimoto 1982 ) , such that the last thermonuclear instability might be the most violent .
however , once @xmath31 , the required unstable he ignition mass for @xmath32 wds ( nomoto 1982 ; limongi & tornambe 1991 ; goriely et al .
2002 ) exceeds the donor mass .
thus once a system reaches an orbital period of 10 minutes , it is unlikely to undergo further unstable helium ignition events .
that is , @xmath1 drops quickly enough that once @xmath1 is low enough for a strong flash , the accreted layer never becomes thick enough for it to occur . over the remaining course of the binary s evolution , about @xmath33
is accreted by the wd , motivating us to use a fixed accreted layer mass of @xmath34 to calculate the thermal evolution of the accretor .
because @xmath14 depends only logarithmically on the accreted layer mass ( tb ) , using this fixed value has only a small impact , especially when compared to the large variations in @xmath14 from the orders of magnitude variation in @xmath1 . the period of time during which @xmath35 yr@xmath36 is so short ( @xmath37 yr ) that the accretor s core is thermally unperturbed from its initial state .
however , once the helium layer begins to accumulate , thermal diffusion both out of the surface ( setting the exiting luminosity , @xmath14 ) and into the core ( acting to increase the temperature there , @xmath38 ) occurs .
we thus begin evolving the wd s thermal state @xmath39 yr after contact .
although the compressional energy release depends weakly on the accumulated he layer mass ( i.e. logarithmically , tb ) , the @xmath14-@xmath38 relation which determines the cooling is essentially independent of it . the evolution is shown in figure 1 and exhibits all of the epochs ( 2)-(4 ) including reheating , decoupling and cooling . the heat capacity and
crystallized fraction of the core was calculated using the equation of state of potekhin & chabrier ( 2000 ) , and the late - time @xmath14-@xmath38 relation ( in the limit of minimal accretion ) was adapted from hansen ( 1999 ) .
we have included a latent heat of crystallization of @xmath40 per particle , but no fractionation . at early times in the evolution ,
the wd core temperature , @xmath38 , is much lower than the maximum temperature reached at the base of the outer radiative envelope , @xmath41 . in this limit , tb showed that @xmath14 becomes independent of @xmath38 and is only a function of @xmath1 and @xmath11 .
they found that the envelope state is determined self consistently by balancing the heat released in the outer ( non - degenerate ) envelope due to fluid motion down the entropy gradient , @xmath42 , with that carried through the envelope by radiative transfer @xmath43 , where @xmath44 is the mean molecular weight in the he envelope , @xmath45 is boltzmann s constant , and @xmath46 is the baryon mass .
this same state applies here when @xmath47 and we have calculated this balance numerically using the quasi - static envelope methods outlined in tb for he accretion and found , @xmath48 this relation holds for @xmath49 and @xmath50 , although for our time histories @xmath51 by the time @xmath52 so that this form only applies to figure 1 for @xmath53 . during this early phase , the wd core is being heated as @xmath1 is dropping .
as the starting wd mass , when @xmath54 and the accreted layer is thin , their values are as much as a factor of 50 lower for @xmath55 yr@xmath36 .
we believe that @xmath14 , which is largely due to compressional heat release in the outer layers and was only of peripheral interest in their study , is not well determined by their modeling for small accreted layer masses at @xmath56 . ] as @xmath1 drops and @xmath38 increases , the accreting wd approaches the equilibrium state where the outgoing luminosity is matched by the internal energy release ( nomoto 1982 ; tb ) .
there is a brief transition from core heating to cooling which appears in figure 1 as a plateau in @xmath38 . in the equilibrium state ,
a fairly good estimate of the compressional heating luminosity is found by integrating throughout the @xmath20 accreted degenerate he layer ( identical to that performed in appendix b of tb , and nomoto 1982 ) , giving @xmath57 , where @xmath58 is the ion mean molecular weight in the helium envelope . as @xmath1 continues to drop
, this will eventually become low enough that it is exceeded by the core cooling luminosity @xmath59 . at this point the wd evolution
_ decouples _ from @xmath1 and cools much as an isolated object would . by setting @xmath60 and taking @xmath61 - 2@xmath62 k
, we find @xmath63-@xmath64 yr@xmath36 . these values are evident in figure 1 .
the precise value of @xmath65 at the plateau depends on the initial @xmath38 and when the he flashes cease , neither of which is well known .
this leads to some uncertainty in @xmath14 at the time of decoupling .
however , analogous to isolated cooling wds , the `` initial '' @xmath65 is forgotten , and @xmath14 eventually only depends on the time after decoupling , @xmath26 .
this is simply the wd cooling age from the time that heating ceased .
by @xmath66 yr , @xmath67 , already consistent with a cooling db wd of a similar age ( hansen 1999 ) . using the values tabulated by limongi & tornambe ( 1991 ,
hereafter lt91 ) for accretion of helium onto a @xmath68 wd at a fixed rate , we find that our determination of the heating time from the quasi - static model agrees well with their full numerical model .
table 2 in lt91 gives a detail of @xmath69 ( the maximum with respect to radius ) and @xmath70 for a mass history starting from @xmath68 up to about @xmath71 .
we compare our @xmath72 for a 0.9@xmath73 wd with a @xmath20 helium layer accreting at a constant rate to @xmath74 ( representative of the heat content ) from table 2 of lt91 , and find that in both cases , for this range of mass , @xmath75 is approxmately constant .
we find @xmath76 k @xmath77 and @xmath78 k @xmath77 for @xmath79 and @xmath80 respectively , compared to the values of @xmath81 k @xmath77 and @xmath82 k @xmath77 from lt91 . also , the @xmath65 reached in the plateau ( @xmath83 k ) reflects the equilibrium @xmath38 for the @xmath1 at this epoch @xmath84 .
this is consistent with the typical values of interior temperatures , @xmath85 , found by lt91 for a @xmath68 co wd accreting helium steadily at @xmath86 .
before core crystallization begins , the wd s heat capacity is well approximated by @xmath87 . with the simple relation @xmath88 , @xmath89 gives @xmath90 where @xmath91 is time in gyrs .
this agrees well between decoupling and the onset of crystallization , which occurs at @xmath92 k for @xmath93 , corresponding to a time of @xmath94 yr and an orbital period of 60(45 ) minutes for the hot donor evolution .
the effect of the latent heat release is visible in the @xmath38 history and the heavier wd cools more quickly at late times due to its earlier crystallization and the resulting decrease in total heat capacity .
the modifications due to latent heat release change the @xmath95 relation to something closer to @xmath96 and @xmath97 for @xmath18 and @xmath19 respectively in the 1 - 8 gyr region .
by contrast , the accretion luminosity @xmath15 is a smooth power law after about @xmath98 yr , given by @xmath99 for @xmath93 .
[ fig : compare ] the bottom panel of figure 1 shows the resulting @xmath100 , for which there is very little contrast with the donor entropies .
however , when displayed as a function of the observable , @xmath2 ( see figure 2 ) , a contrast immediately appears : accretors in systems with hotter donors have higher @xmath0 s .
the reason for this depends on the phase of evolution , but is essentially related to the difference in speed of evolution as set by the donor s @xmath11-@xmath12 relation . in the `` compressional heating ''
phase a lower entropy donor always produces a lower @xmath1 at a given @xmath2 , leading to a lower @xmath14 . by an orbital period of 30 - 40 minutes
the wd is cooling and the time since decoupling becomes larger than the uncertainty introduced by the unknown @xmath65 , so that at longer orbital periods we can safely predict @xmath14 .
low - entropy donors reach the decoupling @xmath1 at shorter orbital periods and evolve more slowly in @xmath101 , so that the wds at a given @xmath2 always look `` older '' ( i.e. colder ) if their donor was initially of low entropy . the track for the cool donor ( 30th percentile ) gives a good maximum @xmath26 , which for @xmath102 min can be fit by @xmath103 , and thus yields the relation for the coldest possible @xmath104 wd at these orbital periods , @xmath105 .
in order to assess the likelihood of observing the accreting wd , we must compare its luminosity to that from active accretion . when the wd is near equilibrium or simply emitting the compressional heating luminosity
, @xmath106 will be much less than the accretion luminosity , @xmath107 , since @xmath108 .
seeing the hot wd then depends on either geometry ( for example , an edge - on systmem like the newly discovered eclipsing binary sdss j0926 + 3624 ; anderson et al .
2005 ) or probes of different spectral regions .
four classes among observed systems are delineated in table 1 by horizontal lines .
the shortest period systems are in a phase where the accretion stream directly hits the wd ( e.g. marsh & steeghs 2002 ; israel et al . 2002 ) , providing the first opportunity . in this case
there is no disc and there are times when the direct impact location is out of sight , providing an excellent chance to see the ( very hot ) accreting wd . at
slightly longer @xmath2 , around 20 minutes , all systems are expected to have a consistently bright accretion disc in the optical / uv that would energetically dominate the wd ( e.g. nelemans et al . 2004 ) as noted above .
indeed , nelemans et al .
( 2004 ) showed that a simple accretion disc model at 15 minutes would give @xmath109 and temperatures of @xmath110 k , swamping the hot wd .
this model is shown by the thick grey line on the bottom panel of figure 2 .
l c c c c c c l rx j0806&5.35&@xmath111&0.125&@xmath112&21.1&&1 + v407 vul&9.49&@xmath113&0.068&@xmath114&19.9&&2 + es cet&10.3&@xmath115&0.062&130@xmath11610&16.9&&3,17 + am cvn&17.1&@xmath117&0.035&&14.0&4.8&4 + hp lib&18.4&@xmath118&0.032&&13.6&7.0&4 + cr boo&24.5&@xmath119&0.023&12&13.5 - 17.5&5.79.7&4,5,6 , 7 + kl dra&25.0&@xmath120&0.022&&16.8 - 20&&8 + v803 cen&26.9&@xmath119&0.020&14&13.2 - 17.4&5.19.3&9 + cp eri&28.4&@xmath121&0.019&17&16.5 - 19.7&&10,19 + sdssj0926 + 3624&28.3&@xmath121&0.019&18&18 - 19&&20 + 2003aw&33.8&@xmath122&0.015&17&16.5 - 20.3&&11,18 + sdssj1240 - 01&37.4&@xmath123&0.013&17&19.6&&12 + gp com&46.5&@xmath124&0.010&11&15.7 - 16&11.311.7&4,13,14 + ce-315&65.1&@xmath125&0.006&9.5&17.6&13.2&15,16 the third group of am cvn systems , with again slightly longer orbital periods , show large brightness variations due to a thermal instability in the disc ( tsugawa & osaki 1997 ) .
although at this @xmath1 the disc absolute magnitude is expected to be brighter than the wd , the low state is a time when the accreting wd can be detected , just as in the hydrogen accreting cvs ( sion 1991 ; townsley & bildsten 2003 ) .
the last group , with the longest orbital periods , again show hardly any brightness variations . in the context of the thermal instability interpretation , these systems could be on the lower , stable , branch of the instability curve and thus would be cool discs , consisting mainly of neutral helium . the steadily accreting luminosity is only @xmath126 times larger than the cooling wd ( see 2 ) , enhancing the prospects of seeing the accreting wd relative to shorter orbital periods , where the contrast is greater .
in addition , the disc temperature will generally be cooler than the wd as shown in the bottom panel of figure 2 .
in addition to directly seeing the re - heated wd in the am cvn binaries , there are also opportunities for seismological tests as the wd passes through the non - radial pulsation instability strip ( @xmath127 for isolated dbs ; beauchamp et al .
1999 ) , which we plot in figure 2 .
the pulsation periods are 200 - 1000 seconds ( see handler 2003 for a recent summary ) and are typically @xmath128 g - modes of high radial order ( e.g. kotak et al .
2003 ) .
we do not know , _ a priori _ , that the instability strip will be the same for a he accreting wd .
the clear structural difference in the accreting wd is the mass of the helium envelope , which will likely exceed @xmath20 and be seismically measurable .
it seems unlikely that the thicker he envelope would alter the instability strip , since the relevant `` pump '' for the mode excitation occurs at low pressures ( see gautschy & althaus 2002 for a recent study ) .
however , active accretion onto the outer atmosphere might make a difference , especially if the convection zone ( responsible for the mode excitation ; e.g. brickhill 1991 ) is altered .
it also seems possible that the mode pump could become active during the cooling of the outer wd layers ( piro , arras & bildsten 2005 ) following an accretion disk outburst . indeed , wood et al .
( 1987 ) reported excess power at periods of 100 - 200 seconds when cr boo had an optical magnitude of @xmath129 , which might well be the passing of the accreting wd through the instability strip following an accretion outburst .
the most dramatic alteration of the g - mode frequencies would be from wd rotation at periods comparable to the g - mode periods , as the wd has accumulated a significant amount of mass and angular momentum since the last mass ejecting he novae event .
this angular momentum either resides solely in the he layer or has been shared with the underlying c / o wd in the @xmath130 years spent at the @xmath131 minutes where @xmath0 is in the isolated db instability strip ( see figure 2 ) .
motivated by the likelihood that very weak internal magnetic fields can easily mediate angular momentum transport on this timescale ( e.g. spruit 2002 ) , we presume that all accreted angular momentum after the last he novae event is shared with the whole wd , yielding a lower limit to the rotation frequency of the he shell , where much of the g - mode energy resides .
we assume that helium accretes onto the wd with the specific angular momentum of the inner edge of the accretion disk , @xmath132 , and that the amount of accreted matter , @xmath133 .
the final wd spin frequency is then @xmath134 , where @xmath135 is the wd moment of inertia , yielding @xmath136 , @xmath137 which reaches about 100 seconds when @xmath138 .
hence , it appears very likely that the wd surface layer ( and maybe even core ) is rotating rapidly enough ( i.e. @xmath139 ) that the seismology will be dramatically altered from the isolated db case .
the discovery and study of a db pulsator in an am cvn binary would provide an unprecendented opportunity to measure the accumulated he layer mass on the c / o wd and the angular momentum transport mechanisms within it .
the most probable orbital periods for this range from 20 to 35 minutes .
the rapid surface rotation will broaden any spectral features originating there , as the resulting minimum equatorial rotational velocity is @xmath140 , nearly independent of @xmath11 .
the narrow , central line emission ( i.e. spikes ) seen in he lines of gp com , ce 315 and sdss1240 ( marsh 1999 , ruiz et al , 2001 , roelofs et al , 2005a ) are generally believed to originate on the accreting wds . however , their widths are much less than this estimate , suggesting either a different origin or an effective loss of angular momentum from the rapidly spinning accretor .
another difference with single db wds is that @xmath141n is present in the accreting material at a mass fraction of 1 - 2% ( due to cno burning in the progenitor star of the he donor ) .
this @xmath141n would rapidly sink out of the wd photosphere were it not for the constant mixing provided by the underlying convection zone of mass @xmath142 .
it is thus the sedimentation time , @xmath143 , at the base of the convection zone that sets the relevant timescale .
paquette et al . ( 1986 ) calculated @xmath143 for @xmath141n in pure he atmospheres of isolated wds , and defined an @xmath144 , below which the @xmath141n in the convective envelope will be less abundant than in the accreting material . for @xmath145 wds , @xmath146 .
though @xmath142 decreases rapidly with increasing @xmath11 ( see fontaine & van horn 1976 ) , so does the sedimentation time , making the actual value of @xmath147 not too dependent on @xmath11 for the range of @xmath0 relevant here ( see figure 14 in paquette et al . 1986 ) .
more recent calculations with updated opacities ( macdonald , hernanz , & jose 1998 ) found larger @xmath142 values by factors of 4 - 6 from paquette et al .
hence , for @xmath148 minutes , the atmospheric n abundance will be comparable to that in the accreted material . at longer orbital periods , depending on the state of the donor , it seems plausible that @xmath1 will be less than @xmath149 , and the @xmath141n will tend to sediment .
detection and study of atmospheric n ( i.e. not from the accretion disk ) in gp com , ce-315 and sdss j1240 could thus constrain the sedimentation physics .
the abundant presence of @xmath141n may simplify the spectral modeling of these wds relative to what has been done for pure he atmospheres ( see bergeron , saumon & wesemael 1995a ) .
this is because the low first ionization potential of @xmath141n will provide free electrons more readily than the he .
this may well allow the @xmath150 free - free opacity to more readily dominate for the relevant @xmath0 s , but only calculations will tell .
this could also alter the location of the pulsational instability strip .
we tabulate relevant observed and inferred properties of the am cvn binaries in table 1 .
the minimum donor mass , @xmath151 , for each system is the mass of the deloye et al .
( 2005 ) fully degenerate donor model that fills the roche lobe at the system s @xmath2 .
the minimum @xmath152 at fixed @xmath2 occurs when @xmath153 ( deloye et al 2005 ) .
we thus take @xmath154 and calculate the tabulated values of @xmath152 as follows . for systems in which the mass ratio has been inferred ( all the am cvn binaries _ except _ rx j0806 , v407 vul , es cet , and sdss j0926 + 3624 ) , we use it and @xmath155 to find @xmath11 and thus @xmath152 . when there is no mass ratio constraint , we assume that @xmath152 is given by the minimum possible value , which occurs at a mass ratio of @xmath156 ( this determined the entries for rx j0806 , v407 vul , and es cet ) , except for sdss j0926 + 3624 , for which we use the mass ratio of cp eri . our discussion in the previous sections outlined the opportunities for detecting the hot wd and we now discuss in more detail the implications of our calculations in regards to _ directly _ observing the wd in these binaries .
the evolutionary state of the two shortest period systems , rx j0806.3 + 1527 ( @xmath157 min ) and rx j1914.4 + 2456 ( @xmath158v407 vul , @xmath159 min ) is uncertain and different models have been proposed ( see e.g. wu et al . 2002 ; ramsay et al . 2002 ; marsh & steeghs , 2002 ; marsh & nelemans , 2005 ; dallosso et al . 2005 ; willems & kalogera 2005 ) .
marsh & steeghs ( 2002 ) suggested that v407 vul is a direct - impact accretor where the accretion stream directly strikes the accreting wd . noting the earlier work of townsley & bildsten ( 2002 )
, they suggested that much of the optical light could originate from the hot , accreting wd .
our high @xmath1 calculations allow us to address the direct impact scenario .
even though the accreting material arrives over a limited part of the wd , we expect complete spreading by the time it has reached depths relevant for compressional heating ( piro & bildsten 2004 ) , thus allowing our calculation to apply . at the minimum @xmath160 expected for v407 vul ( see table 1 ) , we predict that @xmath161 . the predicted range of @xmath162 and absolute magnitude , @xmath163 .
the analysis of v407 vul is complicated by the fact that the optical light is dominated by a g9v star ( steeghs et al .
, in prep )
. however , the variable part of the light clearly shows a blue color , with implied effective temperature @xmath164 k ( see table 1 ) , consistent with our calculations .
the estimated dereddened visual magnitude is 16.9 ( ramsay et al . , 2002 ) which includes a substantial g - star contribution .
combined with the estimated @xmath163 , this suggests a distance greater than 500 pc if v407 vul is a direct impact accretor . for rx j0806.3 + 1527 ( @xmath157 min ) ,
the situation is quite different .
the shorter orbital period implies a luminosity of @xmath165 , @xmath166k and @xmath167 for the accretor .
the estimated @xmath0 agrees with the observed very blue spectrum , for which israel et al .
( 2002 ) find a lower limit of 40,000 k. the apparent magnitude of 21.1 would put the system at 20 kpc , 8 kpc above the galactic plane .
a similar distance is needed to interpret the rather low x - ray flux ( israel et al .
2002 ) as being generated by @xmath1 .
it was recently suggested that the low x - ray flux could be reconciled with much smaller distances due to a temporarily lower @xmath168 or due to most of the accretion luminosity being emitted in the uv ( marsh & nelemans 2005 ; willems & kalogera , 2005 ) .
however , our calculations show that not only the x - ray luminosity , but also the optical flux tracks @xmath1 , so that the latter suggestion would still produce a much too bright wd . the compressional heating luminosity would track a drop in @xmath169 on the thermal time of the accreted layer , @xmath170 where @xmath41 is the maximum temperature , and @xmath14 is given by equation ( 1 ) .
this timescale is much longer than the timescale for which a temporarily lower @xmath169 can be sustained ( @xmath171 yrs , marsh & nelemans 2005 ) , leading to a bright wd even at low @xmath169 .
therefore , for rx j0806.3 + 1527 to be a direct impact accretor , the only solution seems to be a very large distance .
the third system in the table , es cet , is clearly a helium accreting wd .
espaillat et al .
( 2005 ) suggest it is a direct impact accretor and find @xmath172 k and radius @xmath173 cm for a distance of 300 pc .
they interpret this as the size of the impact spot , but our calculations suggest that the whole wd should be very hot .
this observed temperature is in the expected range , and the radius either suggests a massive ( @xmath174 ) wd or a less massive wd at a larger distance than they assumed . for the outbursting systems
the spectra are dominated by the accretion disc when they are in their high state .
the absolute magnitudes of the disc are at least 6 8 , much brighter than the expected absolute magnitudes of the accreting wds of about 11 ( figure 2 ) . however , in the low state , the wd can contribute significantly to the spectra , potentially producing the observed continuum . between 20 and 30 min , the expected disc absolute magnitude drops , while the wd @xmath6 stays roughly constant .
interestingly , the magnitude range observed in these outbursting systems decreases with increasing orbital period .
while cr boo has a range of 5 magnitudes , v803 cen is less , at 4.2 , and cp eri only changes by 3.2 magnitudes .
to check if the low - state spectra could be dominated by the wd , we estimate the temperature from the shape of the 40006000 continuum from the few spectra that are available .
we assume a black - body spectrum , which reasonably produces the broad band colors of theoretical db white dwarf spectra ( bergeron , wesemael & beauchamp 1995b ) .
our temperature determinations in table 1 are accurate at about the 10% level k the black body colors overestimate @xmath0 by 5001000 k , while above that temperature they underestimate the temperature by a similar amount .
in addition , in the region considered here , the change in the slope of the black body spectrum decreases with temperature above @xmath17515000 k , adding to the inaccuracy at higher temperatures ] . for cr
boo and v803 cen , the absolute magnitudes are also known ( groot et al . in prep , see table 1 ) . for cr boo and v803 cen , the temperatures are below the expected wd temperatures and indeed the absolute magnitudes in the low state are about 1 magnitude above the predicted wd brightness ( see figure 2 ) , suggesting a significant contribution from the disc ( which is cooler than the wd , but of larger area ) , even in the low - state . using the slope of spectrum in the low state
, we found a temperature of @xmath176 for cp eri , whereas a recent uv spectrum suggests @xmath177 ( sion , priv .
we have to be careful with these low - state estimates , though , as a second low - state spectrum of cr boo ( patterson et al . 1997 ) shows a very different continuum shape , suggesting a cool or hybrid cool plus hot spectrum .
the recently discovered eclipsing am cvn binary , sdss j0926 + 3624 , at @xmath178 ( anderson et al . 2005 ) provides an opportunity for probing the distinct contributions of the hot wd and the disk .
anderson et al .
( 2005 ) report that the @xmath179 minute duration of the eclipse is consistent with the @xmath180 secondary transiting a hot spot or the inner disk . it is also consistent with an eclipse of the hot wd .
fast photometry and phase resolved spectroscopy could confirm that the wd contributes and would tie the eclipse minimum to the wd rather than the disk , yielding a more robust measurement of the orbital period change expected from gravitational wave driven mass transfer ( anderson et al . 2005 ) . for gp com and ce 315 we derived the temperatures in the same way as described above and
interpret the temperature as that of the accreting wd .
for gp com , using the known distance ( d = 67 pc ; thorstensen 2003 ) , we can derive the emitting area .
thorstensen ( 2003 ) gives a typical observed v magnitude of 16.1 , but warner ( 1995 ) lists the range 15.716.0 .
we therefore use v=15.716.1 as the observed range . together with @xmath181 ( thorstensen 2003 )
this gives @xmath182 . from bergeron et al .
( 1995b ) we use the bolometric correction for a 11,000 k db model , being -0.57 .
this gives a bolometric magnitude @xmath183 , i.e. a luminosity of @xmath184 with an error of @xmath185 .
together with the effective temperature , this gives the size of the emitting region of @xmath66 cm , i.e. a typical wd radius ! using the same argument for ce 315 , with a distance of 77 pc ( thorstensen , priv .
comm . ) , using a effective temperature of 9500 k and a bolometric correction of -0.34 , we find an emitting region of @xmath186 cm , implying a more massive wd of @xmath187 . indeed
, its absolute magnitude falls close to the track for a 1.0 @xmath73 accretor ( fig .
interestingly , two systems have been recently discovered that , in their low state , clearly show the broad absorption lines of a db white dwarf , with estimated temperature of 17,000 k ( sdss j1240 - 01 and sn2003aw , roelofs et al .
2005a ; warner & woudt , 2003 , roelofs et al , 2005b ) .
based on an absolute magnitude of 11 ( fig .
2 ) , these systems should have distances of 735 pc ( sn2003aw with v = 20.3 ) and 525 pc ( sdss j1240 - 01 , with v = 19.6 ) .
we have shown ( both from theory and observation ) that thermal emission from the accreting wd within an am cvn binary is an important contributor to the observed optical emission .
for the direct impact systems with the shortest orbital periods , the hot wd is visible when the hot spot is out of the line of sight . at longer orbital periods ,
the wd can either be seen when the disk is in a low state , or at even longer orbital periods ( @xmath7 min ) , the wd always dominates the accretion luminosity in the optical bandpass .
we compared our theoretical results to the temperature and brightness of observed systems .
the accreting wd is the dominant contributor to the optical light for the direct impact accretors and constrains their distances . in particular ,
the expected brightness of the wd is so large for rx j0806 that it can only be in the direct impact phase if it is located at a distance of 20 kpc , far outside the galactic disc . for @xmath188 ,
our predicted absolute magnitude of the wd is @xmath189 .
this allows for estimates of the distance to the two newly discovered am cvn systems ( 2003aw and sdssj1240 - 01 ) in which the db white dwarf is visible .
the two outbursting systems ( cr boo and v803 cen ) with parallaxes have @xmath190 in quiescence , 1 magnitude brighter than our estimate for the wd .
this suggests that even in the low state there is a disc contribution . for gp com and ce-315
, a combination of estimated effective temperature and known distance suggests that the optical continuum emission is dominated by emission from the wd , in agreement with our calculations .
for these two systems we find consistency with a hot ( high - entropy ) donor .
whether this donor was born hot and has evolved at fairly constant entropy or is heated by ongoing processes is unknown but could be addressed by future theoretical work . our successful comparisons to observations
gives us confidence to apply our work to predict just how faint these binaries can become .
for example , either mass wd reaches @xmath191 at 5 gyrs , and at 10 gyrs , the absolute magnitudes are @xmath192 .
this will allow for meaningful constraints on the population density of such binaries from deep optical searches , both in the field ( e.g. the 3 additional candidates found by sdss , anderson et .
al . 2005 ) and in globular clusters .
much work remains to be done on the theoretical side , the most important of which is to resolve better the nature and occurrence rate of he flashes at early stages in the evolution of these binaries .
in addition to better constraining the remaining he mass on the accreting wd at late times , it would also allow us to predict the he novae rate and potentially constrain the population from the measurements of such events .
perhaps the most dramatic development would be the secure detection of non - radial oscillations in these wds , allowing for measurements of both the wd rotation rate and accumulated he layer mass .
we thank tony piro for comments on the manuscript , and ed sion for helpful input as a referee .
this work was supported by the national science foundation ( nsf ) under grants phy99 - 07949 , ast02 - 05956 and ast02 - 00876 .
is supported by the nsf physics frontier centers joint institute for nuclear astrophysics under grant phy02 - 16783 and doe under grant de - fg 02 - 91er 40606 .
gn is supported by nwo veni grant 639.041.405 . | we calculate the heating and cooling of the accreting white dwarf ( wd ) in the ultracompact am canum venaticorum ( am cvn ) binaries and show that the wd can contribute significantly to their optical and ultraviolet emission .
we estimate the wd s effective temperature , @xmath0 , using the optical continuum for a number of observed binaries , and show that it agrees well with our theoretical calculations .
driven by gravitational radiation losses , the time averaged accretion rate , @xmath1 , decreases monotonically with increasing @xmath2 , covering six orders of magnitude .
if the short period ( @xmath3 min ) systems accrete at a rate consistent with gravitational radiation via direct impact , we predict their unpulsed optical / uv light to be that of the @xmath4 k accreting wd . at longer @xmath2
we calculate the @xmath5 and absolute visual magnitude , @xmath6 , that the accreting wd will have during low accretion states , and find that the wd naturally crosses the pulsational instability strip .
discovery and study of pulsations could allow for the measurement of the accumulated helium mass on the accreting wd , as well as its rotation rate .
accretion heats the wd core , but for @xmath7 minutes , the wd s @xmath5 is set by its cooling as @xmath1 plummets .
for the two long period am cvn binaries with measured parallaxes , gp com and ce 315 , we show that the optical broadband colors and intensity are that expected from a pure helium atmosphere wd .
this confirms that the wd brightness sets the minimum light in wide am cvn binaries , allowing for meaningful constraints on their population density from deep optical searches , both in the field and in globular clusters . |
topological insulators ( ti)@xcite are a type of materials which show a gapped bulk spectrum but gapless surface states .
the topological nature of the surface states protects them against perturbations and backscattering .
@xcite in addition , the surface of a d - dimensional ti is such that the effective hamiltonian defined on its surface can not be represented by the hamiltonian of a d-1 dimensional material with the same symmetries , so the physics of a ( d-1)-surface of a topological insulator may show completely different behavior from that of a conventional ( d-1)-dimensional material .
surface states@xcite can be understood in terms of solitonic states which interpolate between two topologically different vacuums , the topological vacuum of the ti and the trivial vacuum of a conventional insulator or empty space .
the ground state of a system can be classified by a certain topological number@xcite depending on its dimensionality and symmetries present which define its topological classification . @xcite
two - dimensional single - particle hamiltonians with time reversal ( tr ) invariance are classified in a @xmath4 ( @xmath5 ) topological class.@xcite two - dimensional tr systems with nontrivial topological index ( @xmath6 ) show the so - called quantum spin hall effect ( qshe ) which is characterized by a non - vanishing spin chern number @xcite and a helical edge current .
@xcite this state has been theoretically predicted and experimentally confirmed in hgte quantum wells,@xcite as well as predicted in several materials such as two - dimensional si and ge@xcite and transition metal oxide ( tmo ) heterostructures.@xcite all these systems have in common that they present a honeycomb lattice structure with two atoms ( a , b ) as atomic basis .
in that situation , it is tempting to think that the effective hamiltonian in certain k - points could have the form of a dirac hamiltonian .
the components of the spinor would be some combination of localized orbitals in the a or b atoms , whereas the coupling would take place via non - diagonal elements due to the bipartite geometry of the lattice .
the simplest and best known example is graphene , where the hamiltonian is a dirac equation in two nonequivalent points k and k. at half filling , graphene with tr and inversion symmetry ( is ) has @xmath6,@xcite so a term which does not break those symmetries and opens a gap in the whole brillouin zone would give rise to the qshe state . the way in which an
is term can arise in a graphene hamiltonian is due to spin - orbit coupling ( soc ) , however it is known that the gap opened this way is too small.@xcite in contrast , a sublattice asymmetry , which breaks is , will open a trivial gap , as in bn.@xcite how a honeycomb structure can be constructed from a perovskite unit cell can be seen in fig .
1a and 1b , a perovskite bilayer grown along the ( 111 ) direction made of an open - shell oxide is sandwiched by an isostructural band insulating oxide .
the metal atoms of the bilayer form a buckled honeycomb lattice .
it has been shown that perovskite ( and also pyrochlore ) ( 111 ) multilayers can develop topological phases , @xcite as well as spin - liquid phases and non - trivial superconducting states.@xcite topological insulating phases have been predicted for various fillings of the d shell,@xcite here we will focus on the large soc limit ( 5d electrons ) and formal @xmath7 filling . we will study two different multilayers , @xmath2 and @xmath3 and we will focus on the realization of a nontrivial @xmath6 ground state .
@xmath8 ( @xmath9 )@xcite is a correlated metal@xcite whose lattice match with @xmath10 ( sto ) would be close enough ( @xmath11 )@xcite for them to grow epitaxially with standard growth techniques.@xcite @xmath12 has not been synthesized ( to the best of our knowledge ) but our calculations ( @xmath13 ) show a reasonable lattice match with @xmath14 would be possible ( @xmath15 ).@xcite the first multilayer is an iridate very similar to the well known @xmath16.@xcite this system presents a layered honeycomb lattice of ir atoms at @xmath0 filling , whereas the present bilayers show a buckled honeycomb lattice , and is predicted to develop the qshe , however electron correlation would lead to an antiferromagnetic order in the edges .
we will study the dependence of the topological ground state on the applied uniaxial strain and the electron - electron interaction and we will determine a transition between two topological phases in both materials .
the work is organized as follows . in section
ii we introduce a simple tight binding ( tb ) model as in ref .
focusing on the @xmath0 case . in section
iii we use density functional theory ( dft ) calculations to study the evolution of both multilayers with uniaxial strain and on - site coulomb repulsion and we determine the ground state of each material . in section iv we study the stability of the topological phase against tr and is breaking using both tb and dft calculations . finally in section
v we summarize the results obtained . .
( b ) construction of the bilayer , the tm atoms are arranged in a triangular a and b lattice in the ( 111 ) direction in such a way the two atoms will form a honeycomb lattice . the z atom corresponds to the insulating layer ( in our case @xmath10 or @xmath14 ) and does not participate in the honeycomb .
( c ) scheme of the multilayer considered in the dft calculations . ]
the qualitative behavior of this system can be understood using a simple tb model for the 5d electrons in the tm atoms , as shown in ref . . in section iia
we will give the qualitative behavior of the effective hamiltonian .
in section iib we will show numerical calculations of the full model . the octahedral environment of oxygen atoms surrounding the transition metal atoms decouples the d levels in a @xmath17 sextuplet and an @xmath18 quadruplet .
given that the crystal field gap is higher than the other parameters considered we will retain only the @xmath17 orbitals .
the hamiltonian considered for the @xmath17 levels takes the form @xmath20 is the soc term , which gives rise to an effective angular momentum @xmath21 which decouples the @xmath17 levels into a filled j=3/2 quadruplet and a half filled @xmath22 doublet .
@xmath23 is the hopping between neighboring atoms via oxygen that couples the local orbitals .
@xmath24 is a local trigonal term@xcite which is responsible for opening a gap ( as we will see below ) without breaking tr and is .
@xmath25 is a term which breaks is making the two sublattices nonequivalent tending to open a trivial gap by decoupling them . in the following discussion we will suppose that this last term is zero , but we will analyze its role in section iv .
we are interested in two different regimes as a function of soc strength : strong and intermediate .
we call strong soc to the regime where the j=3/2 and j=1/2 are completely decoupled so that there is a trivial gap between them
. we will refer to an intermediate regime if the two subsets are coupled by the hopping . the key point to understand the topological character of the calculations is that a @xmath0 configuration can be adiabatically connected from the strong to the intermediate regime without closing the gap .
the argument is the following , beginning in the strong soc regime it is expected that a four - band effective model will be a good approximation . in this regime
the mathematical structure of the effective hamiltonian turns out to be equivalent to graphene .
the trigonal term is responsible for opening a gap @xmath26 via a third order process in perturbation theory where @xmath28 is the trigonal coupling , @xmath29 is the hopping parameter and @xmath30 the soc strength
. it can be checked by symmetry considerations that the restriction of the matrix representations leads to this term as the first non - vanishing contribution in perturbation theory .
( 2 ) has been checked by a logarithmic fitting of numerical calculations of the full model .
this term will open a gap in the k point conserving tr and is and thus realizing a @xmath6 ground state.@xcite as soc decreases , the gap becomes larger while the system evolves from the strong to the intermediate regime , so the intermediate regime is expected to be a topological configuration @xcite with a non - vanishing gap .
note that even though perturbation theory will only hold in the strong soc regime , the increase in the gap as the system goes to the intermediate regime suggests that the @xmath0 configuration will always remain gapped .
this argument is checked by the numerical calculations shown below .
( left ) and strong @xmath31 ( right ) soc strength and @xmath32 .
the difference between the two cases relies on the @xmath33 invariant of the last filled band .
the red lines are the band structure with @xmath34 .
( b ) band structure zoomed for the j=1/2 bands near the k point for negative ( @xmath32 ) , zero and positive ( @xmath35 ) trigonal coupling .
in the three cases the topological invariant gives a topological ground state .
( c ) evolution of the gap in the k point with @xmath28 for the intermediate soc regime .
the two topological phases found will be identified as huti and luti in the dft calculations . ] in fig .
2 we show the results of a calculation using the tb hamiltonian proposed above .
figure 2a is the bulk band structure for strong and intermediate soc strength @xmath30 .
if a non - vanishing trigonal term is included , it opens a gap in the dirac points of the band structure generating topologically non - trivial configurations .
we can see this clearly in fig .
2b , where the band structure close to the fermi level in the vicinity of the k point is shown .
the topological character of each configuration is defined by the @xmath1 topological invariant which for a band in an is hamiltonian can be calculated as @xcite @xmath36 where the product runs over the four time reversal invariant momenta ( trim ) .
the full invariant of a configuration will be the product of the last equation over all the occupied bands @xmath37 for a @xmath0 filling the first unfilled band has always @xmath38 but the difference between strong and intermediate soc is the @xmath33 invariant of the last filled band . for strong soc the j=1/2 and j=3/2 are completely decoupled so a @xmath39 configuration would be topologically trivial , being the invariant of the fifth band @xmath38 . however , when soc is not sizable the bandwidths are large enough to couple the j=1/2 and j=3/2 levels so that the @xmath39 filling is a topological configuration . in both cases
the @xmath0 filling is topologically non - trivial .
@xcite we will see below using dft calculations that the systems under study ( tmo s with 5d electrons in a perovskite bilayer structure ) are in this intermediate soc regime .
figure 2b shows the bulk band structure , focusing now on the j=1/2 bands , for negative , zero and positive trigonal terms .
the left numbers are the @xmath33 invariant of the band while the right numbers are the sum of the invariants of that band and the bands below it .
no matter what the sign of @xmath28 is , the configuration becomes non - trivial , being its role to open a gap in the k point around the fermi level . in the dft calculations below
, it will be seen that a change of sign of the trigonal term can be understood as a topological transition between a low u topological insulating phase ( luti ) and a high u topological insulator ( huti ) , across a boundary where the system behaves as a topological semimetal ( tsm ) . where @xmath83 are the usual pauli matrices acting on spin space .
the representation follows the commutation relation @xmath84=-i\epsilon_{ijk}l_k$ ] so defining @xmath85 the usual commutation relations hold .
this change of sign introduces an overall minus sign on the eigenvalues so the spectrum results in a @xmath86 quadruplet with @xmath87 and @xmath22 doublet with @xmath88 .
the hopping takes place through the overlap of the @xmath17 orbitals of the tm and the oxygen p orbitals , it can be easily checked that the main contribution gives a matrix form that can be casted in the form @xmath91 due to the geometry of the system , a possible term in the hamiltonian that does not break the spatial symmetries of the system is a trigonal term . this term will differentiate the perpendicular direction from the in - plane directions . this term will behave as an on - site term which mixes the @xmath17 states without breaking the symmetry between them so the general form in the @xmath17 basis is the way @xmath52 enters in this term is on one hand by anisotropy of charge density due to the lack of local spatial inversion and on the other hand by distortion of the cubic perovskite edges ( and thus of the octahedral environment ) by expansion / contraction of the ( 111 ) direction .
the absence of local octahedral rotational symmetry is also responsible for the dependence of @xmath28 on u. since varying the onsite interaction will modify the local electron density , provided the local trigonal symmetry is conserved but not the local inversion ( as it is broken explicitly by the multilayer ) , this modification in the electronic density will influence the electrons across the hartree and exchange - correlation terms , by a term with those symmetries . in summary , local symmetry forces that a spin - independent perturbation , that includes @xmath52 and spin - independent u - terms , can be recasted on the previous form . whether spin - mixing trigonal terms are relevant for the effective model of this system or not should be clarified in the future .
either way , the agreement of the predictions both of the tb model and the dft calculations , in addition with the explicitly checked topological phase at high u suggests that this model is a well behaved effective model to study the nm phase of this type of systems .
katja c. nowack , eric m. spanton , matthias baenninger , markus knig , john r. kirtley , beena kalisky , c. ames , philipp leubner , christoph brne , hartmut buhmann , laurens w. molenkamp , david goldhaber - gordon , kathryn a. moler , nature materials 12 , 787 ( 2013 ) .
ab initio electronic structure calculations have been performed using the all - electron full potential code wien2k @xcite the unit cell chosen is shown in fig .
it consists of 9 perovskite layers grown along the ( 111 ) direction , 2 layers of @xmath8 ( @xmath12 ) which conform the honeycomb and 7 layers of @xmath10 ( @xmath14 ) which isolate one honeycomb from the other . for the different off - plane lattice parameters along the ( 111 ) direction of the perovskite considered ,
the structure was relaxed using the full symmetry of the original cell .
the exchange - correlation term is parametrized depending on the case using the generalized gradient approximation ( gga ) in the perdew - burke - ernzerhof@xcite scheme , local density approximation+u ( lda+u ) in the so - called `` fully located limit''@xcite and the tran - blaha modified becke - jonsson ( tb - mbj ) potential.@xcite the calculations were performed with a k - mesh of 7 @xmath40 7 @xmath401 , a value of r@xmath41k@xmath42= 7.0 .
soc was introduced in a second variational manner using the scalar relativistic approximation.@xcite the @xmath43 values used were in a.u . 1.89 for ti , 1.91 for ir , 2.5 for sr and 1.67 for o in the @xmath2 multilayer and 1.93 for ta , 1.92 for pt , 2.5 for k , 1.7 for o in the @xmath3 multilayer .
we have already discussed that the systems chosen to study a @xmath7 filling in a honeycomb lattice with substantial soc are the multilayers @xmath2 and @xmath3 formed by perovskites grown along the ( 111 ) direction .
first the structure is optimized for different @xmath44 lattice parameters respecting is using gga and without soc .
this means that mainly only the inter - planar distances in the multilayers are relaxed . for the energy minimum
the band structure is calculated turning on soc .
the band structure using three exchange - correlation schemes ( gga , lda+u and tb - mbj ) develops the same structure .
the @xmath33 topological invariant of each band is calculated as in the tb model , @xcite the topological invariant being the sum of the @xmath1 invariants over all the occupied bands .
figure 3 shows the band structure calculated with tb - mbj as well as the @xmath33 invariants also obtained ab initio .
the difference in the curvature of the bands with respect to the result obtained with the tb model is due to the existence of bands near the bottom of the j=3/2 @xmath17 quadruplet which are not considered in the tb hamiltonian .
each band has double degeneracy due to the combination of tr and is . at the optimized c ,
the gap between the last filled and the first unfilled band is located at the corner of the brillouin zone ( k - point ) . at low c ( unstable energetically but attainable via uniaxial compression ) , gga predicts that the system can become a metal by closing an indirect gap between the k and m points , however tb - mbj calculations predict that a direct gap is localized at the k point .
for all the calculations the ground state has @xmath6 and thus it develops a topological phase .
the last filled band has @xmath45 so by comparison with the tb results the system corresponds to the intermediate soc limit in which the @xmath46 and @xmath47 are not completely decoupled.@xcite if a 5d electron system like this is in the intermediate soc limit , it is hard to imagine how one can build a tmo heterostructe closer to the strong soc limit ( the only simple solution would be to weaken the hopping between the tm somehow to increase the @xmath48 ratio ) . the first unfilled band has @xmath38 as expected since a @xmath49 configuration will be a trivial insulator with a gap opened by the octahedral crystal field .
( a ) and @xmath3 ( b ) .
the right panels are the band structure zoomed in near the fermi level with the topological invariants displayed .
@xmath33 is the invariant of the band considered calculated by eq .
[ parities ] whereas @xmath1 is the sum over all the bands up to the one considered , in eq .
[ par - full ] .
] the way a trigonal field is present in the dft calculations is mainly in two ways . on one hand ,
strain along the z direction varies the distance to the first neighbors in that direction , so that the electronic repulsion varies as well .
we define this deformation as @xmath50 where @xmath51 is the off - plane lattice constant with lowest energy . on the other hand , an on - site coulomb repulsion defined on the tm by using the lda+u method
has precisely the symmetry of the bilayer , i.e. trigonal symmetry , so varying in some way the on - site potential ( always preserving parity symmetry ) will have the effect of a trigonal term in the hamiltonian ( see a.3 for further details ) . according to this ,
it is expected that in a certain regime , variations in @xmath52 can be compensated by tuning u. in this regime , similar to what we discussed above , the system will develop a transition between two topological phases : a luti and a huti . at even higher u
the system will show magnetic order .
we will address this point later and by now we will focus first on the non - magnetic ( nm ) phase . in order to study the phase diagram defined by the parameters @xmath52 and u , we will perform calculations keeping one of them constant and determine how the gap closes as the other parameter varies , keeping track of the parities at both sides of the transition . here
we will discuss the behavior of the gap in the k point with @xmath52 for various u and j values .
first we analyze the behavior of both materials in parameter space showing their similarities finishing characterizing the actual position of the ground state of the system in the general phase diagram .
first we focus on the @xmath2 case .
as shown in figure 4a , the gap closes as a function of c , so uniaxial strain can drive the system between two insulating phases just as @xmath28 does in the tb model .
however , for high u ( see below the discussion on the plausible u values ) the transition point disappears ( two such cases are plot in fig .
4a ) . for low u ( see figs
. 4 c , d ) , @xmath52 can drive the system from a positive trigonal term to a negative one .
this means that uniaxial strain can change the sign of the effective trigonal term of the hamiltonian .
however , for high u ( fig .
4a ) , strain is not capable of changing the trigonal field , so the system remains in the same topological phase for every @xmath52 . for the calculations using the tb - mbj scheme
( we will see below to what effective u this situation would correspond ) , the transition point takes place almost at @xmath53 , so based on this scheme , the ir - based multilayer would be classified rather as a tsm than as a ti .
now we will focus on the @xmath3 system ( see fig .
4 b , e , f ) . for the gga calculations the transition with @xmath52 disappears , being the system always in the huti phase . if the system is calculated using lda+u with u negative ( circles in fig .
4b ) , the behavior is similar to the previous system and the transition point across a tsm reappears . for more realistic values of u and j ( stars in fig .
4b ) the transition point disappears again .
thus , the present system ( pt - based ) , though isoelectronic and isostructural , can be understood as the strong - u limit of the previous system ( ir - based ) .
the band gap is larger , which is a sought - after feature of these ti , but not large enough to make it suitable for room temperature applications .
we observe that changing j does not vary the overall picture , just displaces slightly the phase diagram in u - space .
the hamiltonian felt by the electrons depends also on the on - site coulomb interaction between them .
if the variation in the term of the hamiltonian that controls it takes place only in the tm atoms , the symmetry of the varying term will have the same local symmetry as the tm atoms , i.e. trigonal symmetry .
thus , it is expected that a variation in u will have a similar effect as @xmath28 in the tb model , so the gap can also be tuned by the on - site interaction .
the slow increase of the gap with u suggests that the gap opened is not that of a usual mott insulator and reminds rather to the slow increase obtained in the tb model .
in fact , the calculation of the @xmath1 invariant shows again a topological phase on both sides of the transition .
both systems develop a transition between the luti to the huti by increasing u. however , the transition point of @xmath2 is at positive u s , for the @xmath3 the transition appears at negative u s , so for all possible reasonable u values the system will be in the huti phase .
tb - mbj calculations have proven to give accurate results of band gaps in various systems,@xcite including s - p semiconductors , correlated insulators and d systems , however it might give an inaccurate position of semicore d orbitals@xcite and overestimate magnetic moments for ferromagnetic metals.@xcite for @xmath2 it is possible to use the transition between the luti and the huti with the tb - mbj scheme to estimate which value of u should be used in an lda+u calculation for these 5d systems to reproduce the result of the tb - mbj calculation .
the actual value of u needed for a correct prediction of the properties under study is often a matter of contention when dealing with insulating oxides containing 5d tms.@xcite from fig .
4 c , d it can be checked that the gap closes at u=1.4 ev for j=0.95 ev and at u=1.0 ev for the more realistic j=0.27 ev , so this suggests that the values which might be used in an lda+u calculation to mimic the tb - mbj result ( a zero gap at @xmath53 ) are on the order of @xmath54 ev in agreement with ref . .
other works relate to a value of u on the order of 2 ev , @xcite however due to the well known property dependence of the value of u , @xcite it is still unclear which is the correct value to study these topological phases .
therefore , our result can serve as a reference for other ab initio based phase diagrams for iridates where topological phases have been predicted as a function of u.@xcite moreover , we study this system in a broad range of u values and using different exchange - correlation schemes to provide a broad picture of the system , rather than using a fixed u value that would yield a more restricted view of the problem . for the pt - based multilayer we could also consider the hypothetical effective value for which gap would close at negative u ( @xmath55 ) for both values of @xmath56 , we obtain the values @xmath57 ev for @xmath58 ev and @xmath59 ev for @xmath60 .
so , it is clearly seen that the gap of these systems is not only dependent on the parameter @xmath61 , but also has an strong dependence on @xmath56 , both in the realistic picture of the ir - based multilayer as well as in the negative u regime of the pt - based multilayer .
recently topological phases dominated by interactions , called topological mott insulating phases , have been theoretically found,@xcite being this term employed for physically different phenomena . in the huti phase ,
the topological gap of the systems is enhanced by increasing the u parameter so that the system seems to be robust against electron - electron interactions . in the same fashion
a usual band insulator can be connected to a mott insulating phase,@xcite the previous robustness suggests that the huti phase might be adiabatically connected to a @xmath4 non - trivial interacting topological phase.@xcite whether this is an artifact of the dft method or an acceptable mean field approach of a many body problem is something that can only be clarified with experiments .
so far we have considered a system with both tr and is . however , given that the @xmath4 classification is valid only for tr invariant systems it is necessary to determine if the ground state possesses this symmetry .
is breaking could destroy the topological phase opening a trivial gap by decoupling both sublattices , as would happen if the honeycomb is sandwiched by two different materials.@xcite tr symmetry breaking will be fulfilled by a magnetic ground state whereas is breaking will be realized by a structural instability . in this section we will study the two possibilities and conclude that both systems are structurally stable and remain nm according to tb - mbj in their ground states .
( a ) and @xmath3 ( b ) as a function of u for two different j. ( c ) phase diagram in @xmath52-u space , the approximate position of the ground state of both systems according to tb - mbj is indicated . at high
u the systems develop an af order , however at realistic u s both systems remain non - magnetic . ]
increasing electronic interactions will drive the nm ground state to a magnetic trivial mott insulating phase at very high u. for a magnetic @xmath7 s=1/2 localized - electron system , from goodenough s rules@xcite an antiferromagnetic ( af ) exchange between the two sublattices is expected which would create an af ground state breaking both tr and is . to check this result ,
we have performed dft calculations within an lda+u scheme taking @xmath62 ev and @xmath63 ev and varying u. for both systems the calculations have been carried out at different u s for the nm and af configurations at the two j s . in both cases
the ground state is af at high u. also , the sublattice magnetization increases with u. figures 5a and
5b show the evolution of the sublattice magnetic moment for both compounds .
also , a ferromagnetic ( fm ) phase has been analyzed , being the least preferred one . in the ir compound a fm phase can be stabilized but has always higher energy than the nm or af . in the pt compound a fm phase could not be stabilized for any of the u values considered .
the true ground state of the system would be a ti phase in the pt - based multilayer ( or tsm in the ir - based multilayer according to tbmbj ) depending on the value of u employed , so if the correct value to be used is larger than the critical value ( of about 3 ev , which is large according to our previous discussions ) , the topological @xmath4 phase will break down and the kramers protection of the gapless edge states will disappear ; whether the edge states would become gapped or not requires further study .
thus the experimental measurement of the magnetic moment of the ground state of these bilayers would shed light into the correct value of u which should be used in these and other similar compounds . in the pt - based multilayer
the sublattice magnetization is almost only dependent on @xmath64 as can be checked by the shifting of the curves ( fig .
5b ) , however in the ir - based multilayer there is a stronger dependence on j ( fig .
5a ) . again , simplifying the evolution in terms of the effective @xmath65 is discouraged for these systems according to our results .
the non trivial effect of j@xcite has been also observed in several compounds such as multi - band materials.@xcite to summarize , we have obtained the magnetization of both compounds as a function of u , showing the system shows a nm ground state for both compounds until a certain u ( larger than the value of u that would be equivalent to tb - mbj calculations ) where the system becomes af ( see fig .
5c ) .
the topological properties of this system rely on both tr and is .
tight - binding calculations predicted that non - invariant parity terms with energy associated of the order of magnitude of the topological gap could eventually drive the topological phase to a trivial one .
first we will discuss the @xmath2 system .
the simplest is breaking could be driven by a structural instability . to study
the structural stability , we have displaced one of the tm atoms from its symmetric position and then relaxed the structure . as a result , the structure returned to the symmetric configuration .
however , due to being in the transition point between the two topological phases , any external perturbation ( such as a perpendicular electric field ) could break inversion symmetry .
this system should be classified more as a topological semimetal rather than a topological insulator due to the ( almost ) vanishing gap of the relaxed structure .
now we will proceed to @xmath3 .
the first difference between this and the previous system is that in the present case the structure is well immersed in the huti phase .
a large is breaking is expected to drive the system to a trivial phase where the sublattices a and b would be decoupled .
however , to change its topological class , the system has to cross a critical point where the gap vanishes in some point of the brillouin zone . according to that ,
the expected behavior is that the gap closes and reopens as the sublattice asymmetry grows , going from the original topological phase to a trivial insulating phase . to check this
, we will compare the results obtained from the tb model and dft calculations . in the tb case
, we introduce a new parameter which is a diagonal on - site energy whose value is @xmath66 for a atoms and @xmath67 for b atoms .
this new parameter will break is and its value will take into account the amount of breaking . when is is broken the index @xmath1 can not be calculated with the parities at the trim s
. however , we can study the topological character searching for gapless edge states . for that sake
we calculate the band structure of a zig - zag ribbon of 40 dimers width with @xmath68 and @xmath69 .
the calculations were also carried out in an armchair ribbon and the same behavior is found ( not shown ) , however the mixing of valleys makes the band structure harder to understand . the color of the bands indicates the expectation value of the position along the width of the ribbon of the eigenfunction corresponding to that eigenvalue , it is checked that the edge states are located in the two edges ( red and blue ) whereas the rest of the states are bulk states ( green ) . according to the result of the tb model shown in figs .
6a and 6b , for small
@xmath70 the system remains a topological insulator although the introduction of @xmath70 weakens the gap . if @xmath70 keeps increasing the system reaches a critical point where the gap vanishes and if @xmath70 increases even more the gap reopens but now the edge states become gapped so the system is in a trivial insulating phase . to model the symmetry breaking in the dft calculations we move one of the pt atoms in the z direction . as the distance to the original point increases
is gets more broken .
we show the four closest eigenvalues to the fermi level in the k point for the dft calculations ( fig .
6c ) and tb model ( fig .
for the symmetric structure ( @xmath71 ) , the combination of tr and is guarantees that each eigenvalue is two - fold degenerate . once the atom is moved the degeneracy is broken and the eigenvalues evolve with the is breaking parameter .
the analogy between the two calculations suggests that in the dft calculation once the gap reopens the new state is also a trivial insulator .
the dashed lines in fig . 6c correspond to the eigenvalues at the k point for the fully relaxed structure allowing is breaking . comparing with the curves obtained for the evolution of the eigenvalues
it is observed that the relaxed structure is in a slightly asymmetric configuration but it remains in the huti phase . due to the dependence of the topological state on
is , tuning this behavior would allow to make a device formed by a perovskite heterostructure which can be driven from a topological phase to a trivial phase applying a perpendicular electric field .
the device will be formed by the tm honeycomb lattice sandwiched between layers of the same insulating ( 111 ) perovskite from above and below . in this configuration
the system will be in a huti phase .
however , a perpendicular electric field will break more the sublattice symmetry inducing a much greater mass term in the hamiltonian proportional to the applied field . modifying the value of the electric field it would be possible to drive the system from the topological phase ( @xmath72 ) , to the trivial phase ( high @xmath73 ) .
this can be exploited as an application of this ti in nanoelectronic and spintronic devices .
@xcite the sublattice asymmetry needed to make the transition is of the same order of magnitude as the topological gap , as can be checked in fig .
we have studied the gap evolution in the @xmath0 perovskite multilayers @xmath2 and @xmath3 as a function of the on - site coulomb interactions and uniaxial perpendicular strain conserving time reversal and inversion symmetry .
the behavior of the system has been understood with a simple tb model where soc gave rise to an effective j=1/2 four - band hamiltonian .
uniaxial strain and on - site interactions have been identified as a trigonal term in the tb model whose strength controls the magnitude of a topological gap .
the topological invariant @xmath1 has been calculated using the parities at the trim s both in tb and dft calculations .
comparisons between the invariants of the bands determines that both of these 5d electron systems stay in the intermediate soc regime .
the small value of the gap in the k - point comes from being a contribution of third order in perturbation theory .
in contrast , sublattice asymmetry contributes as a first order term , so the topological phase can be easily destroyed by an external perturbation that gives rise to an is - breaking term in the hamiltonian .
@xmath2 has been found to be a topological semimetal at equilibrium @xmath52 . comparing tb - mbj and lda+u calculations
reasonable results can be obtained for u in the range 1 - 2 ev . in @xmath3
a huti phase at equilibrium has been found .
this last system can be driven from topological insulating state to a trivial one by switching on a perpendicular electric field which would break inversion symmetry .
also , we have verified that the properties of these systems are dependent on both @xmath74 and @xmath56 instead of only in @xmath55 .
although the smallness of the gap ( less than 10 mev according to tb - mbj ) makes the @xmath0 configuration not particularly attractive for technological applications , the simple understanding of the system turns it physically very interesting .
the present system can be thought of as an adiabatic deformation of a mathematical realization of the four band graphene with soc , with an experimentally accessible gap , the roles of @xmath75 and @xmath20 being played now by @xmath76 and @xmath24 , but with a different physical nature of the topological state . | honeycomb structures formed by the growth of perovskite 5d transition metal oxide heterostructures along the ( 111 ) direction in @xmath0 configuration can give rise to topological ground states characterized by a topological index @xmath1=1 , as found in nature commun . 2 , 596 ( 2011 ) . using a combination of a tight binding model and ab initio calculations we study the multilayers @xmath2 and @xmath3 as a function of parity asymmetry , on - site interaction and uniaxial strain and
determine the nature and evolution of the gap . according to our dft calculations
, @xmath2 is found to be a topological semimetal whereas @xmath3 is found to present a topological insulating phase that can be understood as the high u limit of the previous one , that can be driven to a trivial insulating phase by a perpendicular external electric field . |
the observations were performed with the eso 3.6 m telescope equiped with mefos during 6 nights on november 5 - 11 , 1994 and 2 nights on november 24 - 26 , 1995 .
a description of the instrument can be found in durret et al .
the grating used with the boller & chivens spectrograph had 300 grooves / mm , giving a dispersion of 224 / mm in the wavelength region 3820 - 6100 .
the detector was ccd # 32 , with 512@xmath3 pixels of @xmath4 m .
the catalogue of galaxy positions used in this survey was obtained with the mama measuring machine and is presented in a companion paper ( slezak et al .
this catalogue gives very accurate positions , but magnitudes in the photographic band are known with an accuracy of 0.1 magnitude at best .
ccd photometry of the central regions of the cluster in the v and r bands was later performed to recalibrate these magnitudes and to obtain r magnitudes for the entire photometric sample .
we observed spectroscopically a total of 15 fields , with exposure times of 2@xmath030 minutes , and we obtained 473 spectra in total ( plus the same number of sky spectra ) .
the limiting magnitude for the galaxies observed spectroscopically was = 19 , corresponding to a recalibrated magnitude [email protected] ( see section 3 ) .
[ xy ] out of the 473 spectra obtained , we measured 410 reliable redshifts ( the other ones were discarded due to insufficient signal to noise ratio ) . since 4 is at a relatively low galactic latitude , the effects of stellar contamination of our photographic plate catalogue are large .
our star - galaxy separation software was imperfect since out of the 410 objects for which redshufts were measured only 305 were galaxies .
such a high degree of contamination of the catalogue is at least partly due to the fact that we preferred to have a galaxy catalogue as complete as possible , in order not to `` miss '' galaxies .
our catalogue includes these 305 galaxy spectra , plus those previously published by proust et al .
( 1987 ) , quintana & ramrez ( 1990 ) and malumuth et al .
( 1992 ) , and some galaxies from the cfa redshift catalogue ( huchra et al .
1992 ) , reaching a total of 466 galaxy redshifts after eliminating objects observed twice .
the positions of the objects for which we gathered reliable spectra ( either from our observations or from the literature ) are shown in fig .
these positions are relative to the cluster center taken to be the position of the maximum x - ray emission ( pislar 1998 ) : @xmath6 .
this center is within 7 arcsec of the position of the cd galaxy , a distance which is smaller than the rosat pspc pixel size , and we will therefore consider hereafter that both positions coincide .
the spectra were reduced using the iraf software .
the frames were bias corrected in the usual way .
velocities were measured by cross - correlating the observed spectra with different templates : a spectrum of m31 ( kindly provided by j. perea ) at a velocity of @xmath7 km s@xmath2 , and stellar spectra of the standard stars hd 24331 and hd 48381 , which were each observed every night during our 1994 run .
the cross - correlation technique is that described by tonry & davis ( 1979 ) and implemented in the xcsao task of the rvsao package in iraf ( kurtz et al .
the errors on the velocities derived from absorption lines are given automatically by this task .
the positions of emission lines , when present , were measured by fitting each line with a gaussian .
all the redshifts were measured by the same person ( f.d . ) in a homogeneous way .
redshifts of insufficient quality were discarded ( i.e. those with a tonry & davis parameter smaller than 2.0 , except for three galaxies with respective tonry & davis parameters of 1.5 , 1.6 and 1.9 where the absorption lines seemed to be well enough defined for these redshifts to be kept in the final catalogue ) .
[ cols= " > , > , > , > , > , > , > , > , > " , ] [ compl ] we have estimated the completeness of the final spectroscopic catalogue presented here by comparing the number of galaxies with redshifts to the total number of galaxies from our photographic plate catalogue in the same area .
results are shown in table [ compl ] .
note that the 17 galaxies in our redshift catalogue which do not have magnitudes ( see section 3 ) have been excluded from these statistics , and therefore the catalogue completeness may be slightly larger than estimated in the table .
the magnitude histograms for the galaxies and stars misclassified as non - psf - like objects are displayed in figs . [ maggal ] and [ magstars ] respectively .
the percentage of contamination by stars is very high : about 20% . for a large fraction of the galaxies taken from the literature ,
r magnitudes were made available to us by c. adami ( private communication ) . in order to give magnitudes for all galaxies in the same photometric band , we estimated the r band magnitudes for the galaxies that we observed .
this was done by identifying galaxies both in our photographic plate catalogue and in our ccd catalogue , and by finding the best linear fit ( see slezak et al .
1999 for details ) .
this best fit was : @xmath8 , with the slope fixed to 1.0 , giving an _ rms _ fluctuation of 0.04 .
the galaxies in the literature for which adami did not have magnitudes were identified with objects in our photographic plate catalogue , and we applied the same relation to derive their r magnitudes from their b@xmath9 magnitudes .
for some of the cfa galaxies , only b@xmath10 magnitudes were available ; we roughly transformed them to r magnitudes , by calculating the mean ( v@xmath11r ) colour for the 239 galaxies detected in v in our ccd catalogue ( slezak et al .
1999 ) ; this mean value is equal to 0.48 , and therefore roughly corresponds to a g0 star for which ( v@xmath11r)=0.52 and ( b@xmath11v)=0.58 ( e.g. allen 1981 ) , leading to r@[email protected] ; for these cfa galaxies , we therefore took r= b@xmath12 .
33 galaxies in our redshift catalogue have no identification with our photometric catalogue : 4 from quintana & ramrez ( 1990 ) , 14 from malumuth et al .
( 1992 ) , 1 from proust et al .
( 1987 ) and 14 from huchra et al .
they are the objects in table 4 that have labels 2 to 5 in column ( 16 ) , and their corresponding identification number in the literature , but no identification number from our photographic plate catalogue .
note that out of these 33 objects , half are located outside the area covered by our photometric catalogue , and three are located less than 1 arcmin from the cluster center , so we can not separate them from the cd .
the reason for the lack of identification of the other objects is not clear , since in all the cases where we did identify galaxies from our sample to galaxies in the literature the coordinates matched within a few arcseconds .
the most likely explanation is that the coordinates given in the literature for these objects are not accurate ; the fact that they may have not been detected in our photographic plate scan is unlikely since they are all very bright ( [email protected] ) .
[ maggal ] [ magstars ] the coordinates and magnitudes of the 101 stars misclassified as galaxies are given in table 3 ( available in electronic form only ) , to avoid further observations of these objects in galaxy surveys .
the velocity data for the galaxies in the field of abcg 496 are given in table 4 ( available in electronic form only ) .
the meaning of the columns is the following : + ( 1 ) running number ; + ( 2 ) to ( 4 ) right ascension ( equinox 2000.0 ) ; + ( 5 ) to ( 7 ) declination ( equinox 2000.0 ) ; + ( 8) heliocentric velocity ( cz ) in ; + ( 9 ) error on the velocity in ; for velocities derived from absorption lines , the error is either that stated in the literature or , for our own measurements , that given by the rvsao iraf package ; for velocities derived from several emission lines , the error was estimated from the dispersion between the velocities derived from the different emission lines ; when only one emission line was present and no absorption line redshift was available , the error on the velocity was taken to be the mean internal velocity error ; + ( 10 ) r band magnitude ( see section 3 ) ; + ( 11 ) and ( 12 ) x and y positions in arcseconds relative to the center assumed to have coordinates : + @xmath14 ; + ( 13 ) distance to the cluster center in arcseconds ; + ( 14 ) tonry & davis parameter ; + ( 15 ) label indicating the means of determination of the redshift : 0=derived from absorption lines , 1=derived from emission lines ; + ( 16 ) label indicating the origin of the data : 1=our data , 2=quintana & ramrez ( 1990 ) , 3=malumuth et al .
( 1992 ) , 4=proust et al .
( 1987 ) , 5=huchra et al .
( 1992 ) ; note that the catalogue by malumuth et al .
( 1992 ) includes previous measurements by quintana et al .
( 1985 ) and zabludoff et al .
( 1990 ) ; + ( 17 ) reference to the galaxy number in the various catalogues ; for our own data ( label 1 in col .
16 ) , the number appearing in this column is that of the photographic plate catalogue by slezak et al .
( 1999 ) , table 1 ; for data taken from the literature , the first number is that of the previously published catalogue ( or the name for cfa galaxies ) , the second that of the photographic plate catalogue by slezak et al .
( 1999 ) whenever the identification was possible .
this redshift catalogue will be used , together with our photometric catalogues ( which include both the large field catalogue obtained by scanning a photographic plate and the small ccd field catalogue , slezak et al .
1999 ) , to give an interpretation of the properties of 4 ( durret et al . in preparation ) .
they will be compared to the results already obtained from rosat pspc x - ray data ( pislar 1998 ) .
allen c.w . , 1981 , `` astrophysical quantities '' , third edition , university of london , the athlone press durret f. , felenbok p. , lobo c. , slezak e. , 1998 , a&as 129 , 281 huchra j.p . , geller m.j . , clemens c.m . ,
tokarz s.p .
, michel a. , 1992 , bull .
cds 41 , 31 kurtz m.j . ,
mink d.j . ,
wyatt w.f . , et al . , 1991 , in `` astronomical data analysis software and systems i '' , asp conf .
worrall , c. biemesderfer , & j. barnes , p. 432 - 438
malumuth e.m . ,
kriss g.a .
, van dyke dixon w. , ferguson h.c .
, ritchie c. , 1992 , aj 104 , 495 maurice e. , mayor m. , andersen j. et al . , 1984 , a&as 57 , 275 pislar v. , 1998 , phd thesis , universit paris 6 proust d. , talavera a. , salvador sol e. , mazure a. , capelato h.v . , 1987 , a&as 67 , 57 quintana h. , melnick j. , infante l. , thomas b. , 1985 , aj 90 , 410 quintana h. , ramrez a. , 1990 , aj 100 , 1424 slezak e. , durret f. , guibert j. , lobo c. , 1999 , a&as submitted tonry j. , davis m. , 1979 , aj 84 , 1511 zabludoff a.i .
, huchra j.p . , geller m.j . , 1990 , apjs 74 , 1 | we present a catalogue of velocities for 466 galaxies in the direction of the cluster 4 , in a region covering about 160@xmath0160 ( [email protected] mpc for an average redshift for 4 of 0.0331 , assuming h@xmath1=50 km s@xmath2 mpc@xmath2 ) .
this catalogue includes previously published redshifts by proust et al .
( 1987 ) , quintana & ramrez ( 1990 ) and malumuth et al .
( 1992 ) , redshifts from the cfa redshift survey , together with our new measurements .
a total of 274 galaxies have velocities in the 7800 - 11800 km s@xmath2 interval , and will be considered as members of the cluster .
4 therefore becomes one of the few clusters with a high number of measured redshifts ; its physical properties are investigated in a companion paper . |
even with decades of extensive studies , our understanding is still poor as to what physical processes produce the non - maxwellian electron velocity distribution functions ( evdfs ) in the solar wind @xcite .
observations of evdfs at heliocentric distances from 0.3 to 1 au @xcite show a prominent break " or a sudden change of slope at a kinetic energy of a few tens of electron volts , suggestive of two electron populations : the core " that dominates below the break , and the halo " that dominates at higher energies .
in addition , an anisotropic tail - like feature , skewed with respect to the magnetic field direction is found in the evdf of solar wind with speed @xmath0 .
this is called the strahl " . in solar wind coming from a sector boundary with speed @xmath1 ,
the strahl is nearly invisible and the isotropic core - halo feature dominates .
kinetic studies suggest that the strahl can result from the trapping of electrons due to large - scale focusing of the interplanetary magnetic field even in the presence of coulomb collisions and turbulent scattering @xcite . however , there is no model that produces the isotropic halo , nor one that predicts the properties of the observed core - halo feature .
@xcite discovered in the solar wind at 1 au a previously unknown population of energetic electrons even during solar quiet - time , which he called superhalo " .
it has energies of a few tens kev as was confirmed recently using data from the stereo satellite @xcite .
the superhalo electrons probably arise from nanoflares in the solar corona . in a recent paper ( * ? ? ?
* paper i ) , we postulated that nanoflares can produce electron beams with a broad range of energies .
we extrapolate the superhalo density to the nanoflare - accelerated kev electrons assuming that the beam energy flux is a flat function of electron kinetic energy .
we showed using a particle - in - cell ( pic ) simulation that the beam electrons can release their kinetic energy via an electron two - stream instability that leads to production of kinetic turbulence in the solar wind .
one of the important conclusions was the formation of an isotropic electron halo .
given that the above model is largely based on solar wind observations and extrapolations of superhalo constraints , it is critically important to examine if nanoflare - accelerated electrons are sufficient to trigger the strong electron two - stream instability in the context of existing observational and theoretical knowledge of nanoflares . in this letter , we first show that our model assumptions are consistent with existing observations and theoretical estimates of the properties of nanoflares .
we further show that the electron halo forms in the inner corona and can be persevered as the solar wind escapes along open field lines .
then we derive a set of equations that describe the heating mechanism discovered in paper i , and study how the heating mechanism determines the core - halo properties of the evdf .
we show that the solar wind core - halo temperature ratio is a function of the core - halo density ratio and the quasi - saturation condition of the two - stream instability , and is insensitive to the initial condition in the solar corona .
the result can be extended to the more general core - halo - strahl feature .
our theoretical results agree well with observations .
moreover , the formation of the electron halo in the inner corona might have important implications for the models of field line launching and accelerating of solar wind @xcite
small - scale bundled magnetic flux tubes rooted in the photosphere stretch into the solar corona .
photospheric convection leads to the shuffling of foot - points of the magnetic field , and produces magnetic reconnections that eject plasma from the photosphere into the corona along open magnetic field lines .
parker called these events nanoflares " @xcite .
recent high spatial resolution x - ray observations of the sun are providing increasing evidence supporting the existence of nanoflares @xcite .
in particular , the high - resolution coronal imager carried on a nasa sounding rocket discovered a set of fast evolving magnetic loops in the inner - most region of the corona that had lifetimes , sizes and impulsive heating consistent with those expected for nanoflares @xcite .
typical energy released by nanoflares is @xmath2 ergs , @xmath3 orders of magnitude lower than that of flares @xcite .
the duration of nanoflares is typically a few seconds and the size of the loops is typically a few hundred kilometers , about one thousandth the size of flare loops @xcite .
the fundamental difference between nano- and normal flares is that nanoflares can continuously occur everywhere in the quiet sun , even in corona holes , while flares only occur in active regions .
the occurrence rate of nanoflares is nearly @xmath4 events / s over the whole sun , even during solar quiet - times .
similar to flares , nanoflares accelerate particles , and the characteristic energy of the accelerated electrons is in the kev range @xcite .
therefore , nanoflares form a quasi - continuous source of free energy .
in paper i we showed that kinetic alfvn wave ( kaw ) and whistler wave turbulence , as well as an electron halo , can be produced if the kinetic energy of electron beam is @xmath5 times the thermal energy of the core , and the electron beam density is @xmath6 of the core electron density .
we now examine if these conditions can be satisfied based on estimations using our current knowledge of nanoflares .
we also show where the core - halo forms and whether the feature can survive collisions in the corona before escaping into space .
nearly 50% of the released energy from solar flares is in the kinetic energy of accelerated particles @xcite .
the density of accelerated electrons can be comparable to the coronal density @xcite . given the similarity between normal and nanoflares , for simplicity we can assume that all the energy released by nanoflares is converted into the electrons kinetic energy of @xmath7 kev .
for the event rate of @xmath8 @xmath9s@xmath10@xcite , the number of kev electrons produced by nanoflares at the corona base should be @xmath11 @xmath9s@xmath10 .
we assume that these kev electrons are uniformly distributed in a thin layer above the surface of the sun with the thickness equals to the average height of nanoflare loops @xmath12 km @xmath13@xcite .
given that the typical duration of nanoflares is a few seconds , the density of kev electrons should be @xmath14@xmath15 . in the inner corona ,
the ambient electron density is @xmath16 @xmath15 and hence @xmath17 . the observed electron temperature in the inner corona ranges from a few times @xmath18 k to @xmath4 k @xcite .
the electron temperature of the ambient coronal gas before heating should be @xmath19 k. the kinetic energy of beam electrons is then @xmath20 times higher than the thermal energy of the coronal electrons before the onset of turbulent heating .
therefore , the conditions we adopted in papar i is likely to be satisfied . in paper
i , we showed that kev electrons can generate kinetic turbulence and produce an electron halo and become demagnetized and thermalized within @xmath21 ion gyro - periods or @xmath22 s , assuming a magnetic field of 1 g @xcite .
the kev electrons can travel @xmath21 km before thermalization , a distance much smaller than the height of typical magnetic loops .
if the electrons are produced in the inner corona and travel along either magnetic loops or open field lines , the electrons are released to the corona and become the halo within @xmath23 .
the electron coulomb collision rate in the inner corona is @xmath24
s@xmath10 and the coulomb heating time scale is @xmath25 seconds . during this period
the solar wind travels @xmath26 with a velocity of @xmath20 .
thus collisions are insufficient to scatter halo electrons into the core to form a single maxwellian distribution before the expansion of corona convects the turbulent plasma into the outer corona at around @xmath27 where the electron density drops from @xmath28@xmath15 to @xmath29@xmath15 @xcite and can be neglected .
therefore , the core - halo shape is stable to collisional scattering as the solar wind being advected outward to 1 au and beyond @xcite . to facilitate our discussion of the properties of evdf , we first summarize the pic simulation reported in paper i. the 2.5d pic simulation is initialized in the solar wind reference frame with a uniform magnetic field @xmath30 .
the initial evdf has two maxwellian components that represent the ambient thermal electrons in the corona and the nanoflare - accelerated kev electron stream , respectively . the ratio of the density of the electron beam @xmath31 and that of the ambient electrons @xmath32 is 0.1 .
the relative drift @xmath33 between the two populations is along @xmath34 .
the ratio of the mean kinetic energy of beam electrons @xmath35 and the thermal energy of ambient electrons @xmath36 is 72 ( @xmath37 is the boltzmann constant ) .
we now look at the evolution of evdf in the simulation . in fig .
[ dist ] , we show the evdfs @xmath38 and @xmath39 at four times . at the beginning ,
two stream instability is triggered and quickly damps most of the kinetic energy of the electron beam , part of the beam electrons join the ambient electrons and the remaining form a long tail along @xmath40 ( green dashed ) .
this instability nearly saturates on the time scale of the electron plasma frequency @xmath41 . a weibel - like instability
is then triggered and converts the electrostatic waves into electromagnetic waves , which in turn scatters the beam electrons in direction perpendicular to the magnetic field and broadens the narrow tail .
the breakup of the weibel - like waves producing irregular wave - wave interactions further scatters the electrons into different directions . at the final stage
the kaw and whistler wave turbulence evolve to an equilibrium state .
nearly all ( @xmath42 96% ) of the energy in electromagnetic fluctuations is returned to heat the electron tail .
the tail is scattered into a nearly isotropic halo , resulting in a core - halo " structure . in fig .
[ fhalo ] we show @xmath43 and @xmath44 , cross - sections that cut through the center of the simulated 2d evdfs parallel and perpendicular to @xmath34 , respectively , after the core - halo feature becomes stable . for theoretical simplicity and to follow the observational convention of @xcite each of these 1-d evdfs is fitted with two maxwellian functions representing the core and the halo respectively , even though the halo distribution is known to be non - maxwellian . because both the core and the halo are nearly isotropic , the differences between the two 1-d evdfs are very small except for the small drift between the halo and core in the @xmath45-direction .
we found the halo temperature @xmath46 to be @xmath47 times the core temperature @xmath48 , and the core temperature @xmath48 increases by a factor of 4 from the initial temperature @xmath49 .
the corresponding thermal velocity of the core is @xmath50 which equals the relative core - halo drift and is the condition for saturation of the electron two - stream instability .
the ratio between the halo density @xmath51 and the core density @xmath52 is @xmath53 , which is about half the initial density ratio between the electron beam and the ambient electrons . from what we have learned about the turbulence heating processes from our simulation
, we can derive the general properties of core - halo features and relations between these properties .
the evdf break " is simply the velocity at which the core and halo maxwellians are equal and can be found as @xmath54^{1/2 } v_{te , c } , \label{brk}\ ] ] where we have used the facts that the drift energy of the halo is much smaller than @xmath55 and @xmath56 . for the parameters we adopted in our simulation
, we have @xmath57 .
the efficient heating and scattering by kinetic turbulence allows us to determine with relative ease the final temperature of the core @xmath48 and @xmath58 . during development of the kinetic turbulence ,
the total momentum in the center - of - mass frame is zero .
the two - stream instability quickly achieves quasi - saturation at @xmath59 and the exponential decay ends , where @xmath60 is the parallel temperature .
if the fraction of kinetic energy converted into the heat of core electrons is @xmath61 , we have @xmath62 , where @xmath63 is the relative drift when the two - stream instability is at its quasi - saturation .
consequently we have @xmath64 the perpendicular temperature , on the other hand , is still approximately @xmath49 .
later , wave - wave interactions scatter electrons and transfer the thermal momentum from the parallel to perpendicular direction . part of the beam electrons joins the ambient electrons to form a new core with density @xmath52 .
thus the decrease of the new core s parallel thermal energy from @xmath65 to @xmath66 is equal to the increase in perpendicular thermal energy from @xmath67 to @xmath66 , which leads to @xmath68 .
the final isotropic temperatures of the core @xmath48 and halo @xmath46 satisfy the following energy equations ( ignoring the small fraction of energy stored in the turbulence ) : @xmath69 @xmath70 where @xmath71 is the initial temperature of beam electrons . since @xmath72 and @xmath73 , from eqs .
( [ tcl])([tc2 ] ) we obtain @xmath74 the final temperatures are much higher than the initial temperatures , then eq . ( [ dthc1 ] ) can be further simplified as @xmath75 note that in general @xmath76 and hence eq .
( [ dthc2 ] ) slightly overestimates @xmath77 .
( [ dthc2 ] ) shows that @xmath77 depends on the final density ratio @xmath78 and @xmath61 , but is insensitive to the initial condition due to the strong heating process . if @xmath79 then @xmath80 , meaning that if all of the electron beam kinetic energy is converted into the heat of electrons by the two - stream instability , then the temperature ratio is simply determined by the quasi - saturation condition of the two steam instability , the lower limit of @xmath77 . for a conversion fraction @xmath81
as shown in our simulation , the temperature ratio varies from @xmath82 for the reasonable range of @xmath83 $ ] .
one of the unique predictions of our model is the existence of a relic relative drift between the core and the halo along the magnetic field .
the drift equals the core thermal velocity , i.e. , @xmath84 when the two - stream instability fully saturates .
in this letter , based on the existing observations and theoretical constraints of nanoflare properties , we show that nanoflare - accelerated electron streams are likely to trigger sufficiently strong electron two - stream instabilities and produce a core - halo feature of evdf in the inner corona .
the core - halo feature can survive the corona collisions and escape to space from the open field lines .
we show in eq .
[ dthc2 ] that the temperature ratio @xmath77 is related to the quasi - saturation property of the two - stream instability and the density ratio @xmath85 when the turbulence fully saturates .
the `` break '' velocity of the core - halo feature is related to @xmath77 and @xmath85 by eq .
( [ brk ] ) .
a unique feature of our model is a relative drift between core and halo comparable to the core thermal velocity , a relic of the fully saturated electron two - stream instability .
we now compare the results with the observed evdfs of slow solar wind with speed @xmath86 300 because the slow wind tends to be observed near sector boundaries , where the radial magnetic field is near zero @xcite , and thus the magnetic focusing effect is weak and less likely to create distortions to the evdf as the solar wind travels away from the sun .
features of evdf such as @xmath77 and @xmath78 could be preserved if the core and halo evolve the same way as the solar wind travels into space .
we found that the following observations in the slow wind with speed @xmath86 300 by @xcite at 1 au agree well with our model 1 ) @xmath87 ; 2 ) the break " is at @xmath88 6km / s @xmath89 ( about 100 ev ) ; 3 ) pilipp et al . who noted that the relative drifts between the cores and halos in evdf along the solar wind magnetic field existed , but provided no explanation .
we find that the drifts agree well with the thermal velocities of the observed cores in this paper a direct consequence of the saturation of the two - stream instability .
the strahl in the solar wind evdfs is likely produced by magnetic focusing as the solar wind travels outward from the sun .
recent studies suggest that the strahl can form at @xmath90 in fast wind @xcite , well beyond the site where the halo forms .
this is consistent with observations showing that strahl originates from both the core and halo @xcite .
as the solar wind travels to larger heliocentric distances the strahl should become more and more anisotropic .
however , observations show that from 0.3 au to 1 au electrons are actually scattered from the strahl into the halo @xcite .
the cause of the scattering is unclear .
the anisotropic strahl may generate new unstable processes such as kaw or whistler wave turbulence when the evdf is stretched sufficiently @xcite . during the strahl formation and the strahl - induced electron scattering ,
the total energy is conserved .
observations of halo and strahl electron densities in the fast wind found that total density of the hot components ( halo and strahl ) is nearly constant @xcite . the temperature of the hot component electrons @xmath91 where @xmath92 represents the strahl .
then we have @xmath93 where @xmath94 .
( [ dthc2 ] ) can be extended to core - halo - strahl evdfs , i.e. @xmath95 this ratio is nearly independent to the distance from the sun and even in the fast wind @xmath96 , consistent with the observations @xcite .
further observational test for the evolution of @xmath97 with the the distance to the sun is needed .
nanoflare - accelerated kev electrons are quickly converted into halo electrons in the inner corona .
this heating process is so rapid that the kev beam electrons may not produce sufficient non - thermal xray bremsstrahlung radiation to be observed consistent with rhessi observations that found little non - thermal xray bremsstrahlung radiations at a few kev @xcite
. the electron halo may play a very important role in the models of field line launching and accelerating of the solar wind .
@xcite proposed a new solar wind model which can produce the observed anti - correlation between the electron temperature and speed of solar wind @xcite .
this model proposed that the magnetized plasma is preheated in the coronal loops and released by reconnection with open magnetic field lines .
@xcite discovered that the electron temperature in corona magnetic loops is higher than that of ambient electrons , confirming the existence of preheating process .
however , the nature of this preheating is not well understood .
the fast thermalization of electron streams along loops in our model provides a possible explanation for such preheating .
electron temperature at the corona source of the solar wind can be obtained from the measurements of charge states of solar wind ions .
it has been found that the electron halo in evdf can affect the ionic charge states and hence ionic charge states can be used to infer properties of electron halo @xcite .
hc is grateful to the helpful discussions with drs .
j. klimchunk , b. dennis , e. marsch and g. gloeckler .
the authors thank the anomalous referee whose comments helped in making this manuscript more clear and readable .
this research was supported by the nasa postdoctoral program at nasa / gsfc administered by oak ridge associated universities and nasa grant nnh11zda001n .
resources supporting this work were provided by the nasa high - end computing ( hec ) program through the nasa advanced supercomputing ( nas ) division at ames research center . ,
p. , de pontieu , b. , martnez - sykora , j. , deluca , e. , hansteen , v. , cirtain , j. , winebarger , a. , golub , l. , kobayashi , k. , korreck , k. , kuzin , s. , walsh , r. , deforest , c. , title , a. , & weber , m. 2013 , , 770 , l1 ( panel a ) and @xmath98 ( panel b ) at @xmath99 ( black solid line ) , 0.32 , two - stream instability nearly saturates and slowly decay ( green dashed ) , 0.64
, the weibel instability is near the peak ( red dash - dotted ) , and 10.56 , turbulence is in full saturation ( blue dots - dashed ) .
[ dist ] ] ( parallel to @xmath34 ) and @xmath100 ( perpendicular to @xmath34 ) at @xmath101 .
the dot - dashed lines delineate the core maxwellian distribution functions with @xmath102 and the dashed lines represent the halo distribution functions with @xmath103 .
, scaledwidth=90.0% ] | the formation of the observed core - halo feature in the solar wind electron velocity distribution function is a long - time puzzle . in this letter based on the current knowledge of nanoflares
we show that the nanoflare - accelerated electron beams are likely to trigger a strong electron two - stream instability that generates kinetic alfvn wave and whistler wave turbulence , as we demonstrated in a previous paper .
we further show that the core - halo feature produced during the origin of kinetic turbulence is likely to originate in the inner corona and can be preserved as the solar wind escapes to space along open field lines .
we formulate a set of equations to describe the heating processes observed in the simulation and show that the core - halo temperature ratio of the solar wind is insensitive to the initial conditions in the corona and is related to the core - halo density ratio of the solar wind and to the quasi - saturation property of the two - stream instability at the time when the exponential decay ends .
this relation can be extended to the more general core - halo - strahl feature in the solar wind .
the temperature ratio between the core and hot components is nearly independent of the heliospheric distance to the sun .
we show that the core - halo relative drift previously reported is a relic of the fully saturated two stream instability .
our theoretical results are consistent with the observations while new tests for this model are provided .
# 1to 0pt#1 # 1 # 1 to size # 1 to size # 1 = cmr10 scaled2 = cmbx10 scaled2 = cmti10 scaled2 ps . |
the helicity density matrix of a generic momentum - helicity wave packet under the action of the lorentz group is ill - defined .
the underlying reason is the explicit helicity dependence on the momentum eigenstates when transformed from one reference frame into another .
consequently , tracing over momentum is an act of violence against the rules of quantum mechanics because one intends to sum helicity density matrices from different hilbert spaces .
it makes no sense to talk about transformation properties of the helicity density matrices and there is no group representation the helicity matrix can transform according to .
the situation is further worsened by attempts to measure such a state to recover the encoded information .
this problem was first systematically studied in @xcite but some explicit comments were already made in @xcite and probably even before .
a partial remedy was found in @xcite too where the authors first realized the importance of a specific helicity measurement and then introduced the concept of a three - dimensional helicity density matrix to deal with the problems mentioned above . regarding the helicity measurement , it is a simple projective measurement onto the plane perpendicular to the direction of propagation of the wave packet and due to its simplicity
it will be used in this work as well .
we will refer to this kind of measurement as the peres - terno ( measurement ) scheme .
we will , however , abandon the concept of three - dimensional helicity density matrix .
the additional longitudinal degree of freedom introduced in @xcite for the sake of having at least definite transformation properties under ordinary rotations will be of no use here .
moreover , the introduction of a longitudinal degree of freedom sounds like there is something forbidden going on but it is not the case .
this expression , however , simply refers to the third degree of freedom of a specific povm measurement .
one could in principle have povms with arbitrarily many outcomes to be able to reconstruct the wave packet to an arbitrary precision .
the whole problem of the helicity density matrix has been further clarified and generalized to other ( non - relativistic ) situations in @xcite and @xcite .
problems of density matrix construction were considered in other scenarios , too @xcite . for the purpose of this article
we stick to the relativistic context .
there exists a class of wave packets where the definition of the helicity density matrices causes no troubles .
these are linear polarization wave packets in which case , loosely speaking , all the momentum three - vectors point in the same direction ( the direction of propagation ) .
this class of wave packets has very simple transformation properties and allows for introduction of the invariant helicity density matrix @xcite . that is , tracing over momentum is allowed and by invariance we mean that the coefficients of the resulting density matrix are independent of the reference frame .
states transforming covariantly under a general action of the lorentz group are a subject of interest .
the classification of two - photon states has been provided in ref .
let us emphasize that contrary to our paper only states with sharp momenta were considered so the physical picture considerably differs .
the aim of this paper is twofold . in the first part
we construct localized wave packets where the information is encoded in the helicity degrees of freedom ( sec .
[ sec : invariantpacket ] ) .
the effect of an arbitrary lorentz transformation is studied and the conditions for invariant transformation of these wave packets are found . the helicity density matrix derived from these states
is well defined and is invariant under the general action of the lorentz group . finally , the effect of the peres - terno helicity measurement on the receiver s side is investigated concluding the perfect discriminability of two initially orthogonal states in contrast to the original solution .
appendix a contains some further technical details .
the second part is solely aimed at non - experts on the problem of the photonic wave packets construction and their relativistic transformations .
it consists of appendices b and c where we first briefly review wigner s phenomenal contribution to the understanding of the role which is played by the poincar group in quantum field theory in minkowski spacetime @xcite . appendix c contains a discussion of a rather complicated issue of measurement for generic photonic wave packets and the solution of this problem found in @xcite .
we adopt the following notation throughout the article .
the components of a general momentum four - vector ( briefly four - momentum ) @xmath0 is written as @xmath1 .
we recall that indices from the greek alphabet are reserved for labeling all four components of the four - momentum .
the four - momenta @xmath1 living in the four - momentum space can be of null ( @xmath2 ) , timelike ( @xmath3 ) or spacelike ( @xmath4 ) character ( omitting the zero vector ) .
for the purpose of this article we are interested in the first class .
hence we set @xmath5 and normalize energy such that @xmath6 so we get @xmath7 ( @xmath8 ) for all null four - momenta from the future light cone . for further details about the poincar group and how it acts the reader can take a look at appendix b. unless explicitly mentioned the manifold we will consider in this paper is the four - momentum space and not the usual minkowski spacetime ( one temporal and three spatial coordinates ) .
the difference is again sketched in appendix b. for a certainly better exposition , one might also consult extensive literature on the subject of the representations of the poincar group , for example , in @xcite .
a generic single - particle pure photonic wave packet can be written in the form @xmath9 where @xmath10 is the lorentz invariant measure and @xmath11 is an envelope ( scalar ) function therefore staying invariant under the lorentz transformation @xcite .
the helicity eigenstates span a two - dimensional complex hilbert space and these degrees of freedom are perfectly suited for transmission of quantum or classical information in free space .
we keep the zeroth component of all four - momentum vectors equal to one ( or equivalently that @xmath7 holds ) so we restrict ourselves to monochromatic wave packets .
we did not make our considerations less general by restricting to monochromatic wave packets .
the measurement introduced later will simply be color - blind , that is , it will act trivially on the color subspace @xmath12 for every direction @xmath0 .
furthermore , we naturally assume that @xmath13 are identical for all momentum directions so we may drop the subscript .
the preparation of the wave packet is assumed to be under our control .
moreover , it would be a strange way of encoding quantum information into the wave packet if the quantum state differed for different momentum four - vectors already at the preparation stage .
however , the momentum directions ( the envelope function @xmath14 ) continue to be distributed arbitrarily .
we thus rewrite eq .
( [ eq : genwavepacket ] ) as @xmath15 let us see what happens if we first just rotate the wave packet @xmath16}{\mathop{\left|q,+\right>}\nolimits}+\b\int{\rm d}\mu(q)f(q)\exp{[-i\t_{p , q}]}{\mathop{\left|q,-\right>}\nolimits},\ ] ] where @xmath17 determines the spatial rotation of the coordinate system ( or of the wave packet if we adopt the active point of view ) in the direction given by @xmath18 and @xmath19 and from the scalar character of the envelope function it is understood that @xmath20 .
the transformation in eq .
( [ eq : wavpackrotated ] ) is indeed nothing else than a @xmath21 transformation and the parameter is the wigner angle from eq .
( [ eq : unitarylorentz ] ) . also , from eq .
( [ eq : unitarylorentz ] ) follows that the wigner angle is explicitly dependent on @xmath0 and so every ket under the above integral acquires a different phase . as pointed out by the authors of @xcite when we trace over the non - interesting degrees of freedom ( the momentum ) the resulting two - dimensional helicity density matrix does not transform according to a representation of the @xmath22 group . it does not actually have transformation properties at all .
we can get an insight into why tracing over the momentum is ill - defined .
take just two different directions @xmath23 in eq .
( [ eq : genwavepacket_rewritten ] ) .
the rotation of the wave packet given by @xmath17 results in an unequal phase change and thus the coefficients originally equal to each other will generally become different .
so by a suitable rotation two vectors with the same components can be made orthogonal and viceversa .
note , however , that even before the rotation the operation of tracing over the momentum is not valid despite of bringing up a reasonable density matrix . we will take a slightly different path towards a reasonable wave packet density matrix .
while we stick to the helicity measurement studied by peres and terno @xcite we change the building block from which a wave packet is constructed .
the choice of the peres - terno measurement scheme is well motivated .
this helicity measurement is probably the simplest one possible .
we recall the whole issue for readers unfamiliar with the problem in appendix c. the fixed projection measurement breaks the rotational symmetry .
this is the reason why we can not hope in a construction of a covariant density matrix .
nevertheless , we will be able to construct an invariant density matrix .
that is a matrix whose components will stay constant independently on the receivers reference frame .
how do we achieve that ?
contrary to @xcite we use helicity - momentum entangled two - mode states as the construction blocks .
we let the helicity - momentum entangled two - mode states propagate in the @xmath24 direction in a fixed coordinate system .
we first find the conditions for the entangled states under which they do not acquire any phase in a different ( spatially rotated ) reference frame .
having these conditions imposed , we investigate the effect of the helicity measurement on the entangled states and also see the effect of a boost in the @xmath24 direction . based on the results
, we shall construct wave packets which can perfectly transfer quantum or classical information even in the relativistic regime . for the purpose of the helicity wave packet construction introduced later we will now show that for the bell states ( eq .
( [ eq : bellstates ] ) ) with specifically correlated momenta the effect of a spatial rotation is harmless .
following appendix b , let us introduce the helicity eigenstates of the standard momentum @xmath25 @xmath26 the axial component can be disregard as a result of the application of the coulomb gauge which explicitly sets the component to zero .
the helicity eigenstates can be transformed into the linear polarization eigenstates corresponding to the standard momentum @xmath25 by means of the following unitary operation @xmath27 we will need this operation to transform the rotation operator from the linear polarization basis @xmath17 @xcite to the helicity basis @xmath28 .
therefore , to determine the helicity vector for an arbitrary direction parametrized by two angles @xmath29 we transform the helicity eigenstates as @xmath30 .
then @xmath31 and so throughout the article we work strictly in the helicity basis .
before proceeding , let us mention that there exist many conventions for the matrix @xmath32 .
one can , for example , often find @xmath33 the rotation matrix then looks different as well as other objects we will meet later but the physical consequences are equivalent . by working with eq .
( [ eq : pmatrix ] ) we employed the convention of @xcite . ] .
we write down the usual four bell states [ eq : bellstates ] @xmath34 where @xmath35 using the notation of appendix b ( from now on , the latin subscripts label corresponding four - vectors and not their components ) . no need to repeat that the states above are not wave packets yet . using eqs .
( [ eq : pmatrix ] ) and ( [ eq : helrotation ] ) we rewrite states ( [ eq : bellphipm ] ) and ( [ eq : bellpsipm ] ) .
the helicity triads then read @xmath36 an arbitrary rotation can be decomposed into rotations around the @xmath24 axis followed by another rotation around the @xmath37 axis . following
the prescription sketched in remark [ rem : wignerangle ] ( or looking into @xcite or eventually @xcite where the wigner angle was calculated in the representation of the cover of the lorentz group the group @xmath38 ) we find that the rotation @xmath39 around the @xmath24 axis through angle @xmath40 induces the wigner phase angle @xmath41 for all @xmath0 as long as @xmath42 . exactly the measure zero possibility @xmath43 is excluded from the definition of @xmath44 because that would imply @xmath45 and @xmath44 to be converted into @xmath46 ^ 2+[a^{\dagger}(k,-)]^2\big){\mathop{\left|vac\right>}\nolimits}$ ] due to indistinguishability of such photons .
hence the bell states transform as @xmath47 the rotation @xmath48 around the @xmath37 axis results in a more complicated formula for the wigner angle @xmath49}.\ ] ] to tackle the momentum dependence we will assume the momenta @xmath50 to be correlated such that the net effect of the rotation will be the overall phase equal to zero for all @xmath51 .
that implies for eq .
( [ eq : bellphipm ] ) to find such @xmath52 that the corresponding wigner phases are conjugated ( because the helicity is equal for each product ) . having @xmath53 this condition implies that we have to find solutions of the equation @xmath54 where @xmath55 .
this equation deserves an explanation . to find a solution means to find @xmath56 independent on @xmath57 . in other words , we want to locate all fixed points .
note that these points are not completely fixed since they vary with @xmath58 .
similarly , for the second couple eq .
( [ eq : bellpsipm ] ) we have to find @xmath50 such that the wigner phases are equal ( the helicity is opposite for each product ) @xmath59 [ fig : fixedpoints1 ] there exist four fixed points for eqs .
( [ eq : fixedpointsofphipm ] ) and ( [ eq : fixedpointsofpsipm ] ) .
( [ eq : fixedpointsofphipm ] ) can be cast into the form @xmath60=-\sin{(\vp_1+x)}\sin{\vt_1}-\sin{\vp_1}\sin{(\vt_1+y)}.\ ] ] to have solutions independent on @xmath57 and satisfying the above equation the only possibility is when both sides are equal to zero .
that implies @xmath61 it is easy to enumerate all possibilities for which , for example , the first equation is satisfied and then to check if it also holds for the second one .
we end up with four different solutions ( modulo @xmath62 ) @xmath63 having the solutions of eq .
( [ eq : fixedpointsofphipm ] ) we immediately get the solutions for eq .
( [ eq : fixedpointsofpsipm ] ) .
the later equation is obtained from the former one by a mere shift @xmath64 .
hence @xmath65 in figs .
[ fig : fixedpointsphi ] and [ fig : fixedpointspsi ] we see the position of four fixed points for some @xmath66 . if we had in fig .
[ fig : fixedpointspsi ] the same parameters @xmath66 as in fig .
[ fig : fixedpointsphi ] the picture would indeed be only shifted along the @xmath37 axis .
we can check that all the solutions satisfy the corresponding eqs .
( [ eq : fixedpointsofphipm ] ) and ( [ eq : fixedpointsofpsipm ] ) .
the first solution eq .
( [ eq : phipma ] ) is the most interesting one since @xmath67 dwells in the same hemisphere as @xmath51 .
commenting on the rest of the solutions , eq .
( [ eq : phipmb ] ) is equivalent to eq .
( [ eq : phipma ] ) and the remaining two solutions are a bit awkward because @xmath67 points in the opposite direction .
we might , however , say that @xmath68 from the physical point of view there is a noteworthy asymmetry in the solutions for eq .
( [ eq : fixedpointsofpsipm ] ) .
the first two fixed points again correspond to a wave packet with the second momentum pointing in the opposite direction and the last two solutions are just an identity operation ( @xmath69 ) .
so there are two solutions where @xmath70 holds but , for the practical purposes , the opposite momentum direction is an obvious obstacle .
the conclusion is that as long as we keep the momenta in the bell states correlated as described above it is not necessary to introduce three - dimensional helicity density matrices since there is no phase change due to the spatial rotation at all ( cf .
( [ eq : zrotatedphipm ] ) , ( [ eq : zrotatedpsipm ] ) , ( [ eq : yrotatedphipm ] ) and ( [ eq : yrotatedpsipm ] ) ) .
the way of getting rid of the wigner phase for spatial rotations is reminiscent of the method presented in @xcite . the authors constructed wave packets by entangling two spatially distinguishable wave packets carrying the same momentum and opposite helicities .
every spatial rotation induces an opposite wigner phase for each wave packet so the resulting phase is zero . as we already stressed , however , the bell states we work with so far are not wave packets so the construction presented here is qualitatively different .
the projection present in the peres - terno measurement process ( eq . ( [ eq : projpol ] ) ) acts by ` cutting off ' the @xmath24 component of eqs .
( [ eq : mom - heleigsrepresented1 ] ) and ( [ eq : mom - heleigsrepresented2 ] ) . henceforth , after the proper normalization we get @xmath71 where @xmath72 .
note that @xmath73 for all @xmath0 but @xmath74 as expected .
the ` floored ' kets indicate the action of the projection @xmath75 on a momentum - helicity state .
interesting things start to happen when we ask how the projection @xmath75 acts on the bell states .
first of all , these states are two - mode states so the projection is actually @xmath76 where the superscripts indicate which mode is being measured
. the action of the projection can be found in appendix a. in general , these states are subnormalized and non - orthogonal and thus resembling a general situation studied in @xcite .
but , intriguingly , when plugging in some of the solutions for the momentum correlations ( namely eqs .
( [ eq : phipma ] ) , ( [ eq : phipmc ] ) , ( [ eq : phipmd ] ) , ( [ eq : psipma ] ) and ( [ eq : psipmb ] ) ) found from the study of fixed points for arbitrary rotations about the @xmath37 axis we find that ( i ) the resulting floored bell states are orthogonal for all @xmath66 and ( ii ) one of the states is always ( up to a normalization ) invariant .
this has some interesting consequences particularly for the solution eq .
( [ eq : phipma ] ) . recall that this is the only non - trivial solution where the resulting state ` points ' in the same direction as the propagation direction . from now on we will stick just to this solution since it will be later relevant for our discussion of localized wave packets ( the analysis for the rest of the non - trivial solutions gives similar results but their use as localized wave packets is obviously none ) .
when the previous claim written in detail for @xmath77 we get from eqs .
( [ eq : appphiprojplus ] ) and ( [ eq : appphiprojminus ] ) @xmath78 @xmath79 where @xmath80 and @xmath81 .
we denoted @xmath82 and similarly for the rest of the states .
note that both output states are orthogonal for all @xmath83 and @xmath84 .
also , the resulting states are subnormalized even though states ( [ eq : mom - heleigscutoff1 ] ) and ( [ eq : mom - heleigscutoff2 ] ) are normalized .
this is again the effect of cutting off ( alias projection ) the ` longitudinal ' part of eqs .
( [ eq : mom - heleigsrepresented1 ] ) and ( [ eq : mom - heleigsrepresented2 ] ) . the normalization factor for eq .
( [ eq : phiplusproj ] ) and eq .
( [ eq : phiminusproj ] ) is @xmath85 and @xmath86 , respectively . recall that the measurement indeed acts as a trace - decreasing non - completely positive map @xcite .
we also want to draw attention to the similarity between the behavior of the investigated bell states under the irrep of the product @xmath87 and the projection @xmath76 . in the fist case
the singlet is preserved and the triplet states transform among themselves .
this is a starting point to considerations about reference frames and decoherence - free subspaces @xcite . in our case one of the triplets is invariant meanwhile the second triplet mingles with the singlet and the remaining triplet state .
an important comment also related to decoherence - free subspaces has to made .
as we have seen , the bell states remain orthogonal after the projection @xmath88 however , the action of the projection is not @xmath87 covariant . in other words , there is a preferred basis spanned by states from eq .
( [ eq : bellstates ] ) .
one faces a slightly analogous situation encountered in the study of decoherence - free subspaces @xcite .
however , the presented results still hold for any orthogonal basis @xmath89 forming the actual entangled states in eq .
( [ eq : bellstates ] ) . the only difference is given by the fact that our underlying hilbert space is infinite - dimensional .
the found result has several serious consequences .
first , if we prepare two orthogonal pure qubits having the basis formed by the bell pair @xmath90 , they will be perfectly distinguishable by the measurement .
this was not the case for the original peres - terno scheme .
but more important , if we boost along the @xmath24 axis and then perform the same projective measurement we see that state ( [ eq : phiplusproj ] ) remains invariant in the sense that the boost only modifies @xmath91 and keeps @xmath92 intact .
of course , by the change of @xmath93 the four - momenta @xmath50 change too as eq .
( [ eq : particle ] ) dictates .
they symmetrically close in or open up depending on the observer s velocity direction . in a real experiment this would be registered as the color change ( this is the doppler effect which we formally suppress by the renormalization of the new four - vectors ) .
hence the effect of a boost in the @xmath24 direction @xmath94 followed by the projective measurement @xmath76 is @xmath95 let us conclude this section with two comments .
( i ) as in the previous case all output states are orthogonal for all boosts and every @xmath51 .
we have to stress , however , that the above transformations are not unitary .
we recall that the normalization of eqs .
( [ eq : mom - heleigscutoff1])-([eq : phiminusproj ] ) is explicitly @xmath96dependent .
it would be a clear sign of inconsistency if we had a finite - dimensional unitary representation for boost transformations the boost generators are the noncompact part of the lorentz algebra .
similar inconveniences were encountered in @xcite but in our case it does not cause serious troubles .
( ii ) for a state propagating in the @xmath24 direction the most general lorentz transformation can be expressed as @xmath97 .
hence , we have shown that the bell states propagating in the @xmath24 direction and with the appropriately correlated momenta are unaffected by an arbitrary rotation ( and therefore a general boost ) if measured according to the peres - terno scheme
. the real touchstone for quality will be the behavior of wave packets composed from the previously studied entangled states under the lorentz transformation .
as indicated , we will demonstrate the construction and properties of invariant wave packets with @xmath90 serving as the logical basis for construction of a ( pure ) wave packet qubit .
we will not have anything specific to say about the actual feasibility of producing of such states in a laboratory .
also , the dispersion effects of other than relativistic origins are not considered here .
let us follow the prescription and notation for the single - mode case eq .
( [ eq : genwavepacket ] ) @xmath98 similarly to the situation of ordinary wave packets we assume @xmath99 and @xmath100 to be momentum independent ( see the discussion below eq .
( [ eq : genwavepacket_rewritten ] ) ) .
taking into account eqs .
( [ eq : zrotatedphipm ] ) and ( [ eq : yrotatedphipm ] ) and how the boost acts we easily find @xmath101 where @xmath102 . as discussed earlier in eq .
( [ eq : wavpackrotated ] ) the measure and the envelope function are lorentz invariant ( @xmath103 ) .
hence , we call the transformed wave packet invariant because the state amplitudes remain unaltered by an arbitrary lorentz transformation .
in other words , otherwise disastrous decoherence between helicity and momentum degrees of freedom is kept under control by a careful choice of the logical basis . as a final step ,
the measurement described by @xmath76 is realized , due to eqs .
( [ eq : phiplusproj ] ) and ( [ eq : phiminusproj ] ) , by the projections @xmath104 to be more precise , we have seen that the map related to the measurement operator @xmath76 is trace - decreasing .
consequently , this measurement is a conditional ( or post - selected ) measurement .
the overall probability of measurement is lower than one and depends on the details of the wave packet , namely on the choice of its envelope function .
when the measurement event occurs the observer can always perfectly distinguish both orthogonal basis states since the two subspaces from eq .
( [ eq : wavpackprojs ] ) are orthogonal and independent on the observer s reference frame .
note that we tacitly but naturally assume the existence of a global coordinate system .
there exists a class of photonic states we call the realistic wave packets .
it is a weighted and normalized superposition of common eigenvectors of the momentum and helicity operators .
the reason we call them realistic is twofold . from the mathematical point of view , if the corresponding envelope function is square integrable ( but otherwise almost arbitrary ) function , such an object lives in a hilbert space . from the physical point of view
, this object can in principle be prepared , transmitted , manipulated in a localized manner and finally detected in a laboratory .
the problem appears when we consider wave packets in a relativistic regime .
the general lorentz transformation related to the change of a reference frame acts individually on every momentum and induces momentum - helicity entanglement .
the situation is even worse when we try to recover the information by an act of measurement .
the simplest ( and probably the only feasible ) way is the peres - terno scheme of the helicity projection onto the plane perpendicular to the direction of propagation . surprisingly , in this paper we have shown that under these unfavorable conditions we are still able to prepare localized wave packets for which the information can be , at least in principle , transmitted and recovered in every reference frame with perfect fidelity .
there are two key ingredients .
first , the logical qubit basis are two - mode maximally entangled helicity states .
second , the momentum degrees of freedom of these states are correlated in a precise manner .
this results in canceling the explicit momentum dependence for an arbitrary rotation and boost .
intriguingly , the same condition implies that during the measurement according to the above scheme the orthogonal logical basis states forming a qubit are projected into two orthogonal subspaces enabling us to perfectly distinguish them .
it has been known from the previous analysis that the map governing the evolution and measurement of generic wave packets in the relativistic setting is trace - decreasing and non - completely positive .
it is a consequence of the non - covariant helicity measurement scheme and this leads to all aforementioned effects and troubles .
here we have encountered precisely the same behavior but contrary to the previous work these effects are now harmless .
consequently , using the investigated encoding the problem of non - covariant transformation of the von neumann entropy of helicity density matrices disappears .
finally , let us stress that the choice of the peres - terno measurement ( arguably spoiling all nice transformation properties of the momentum - helicity eigenstates ) is directly motivated by its relative feasibility in a laboratory
. should more sophisticated helicity measurements be available we might obtain more elegant results .
it is a pleasure to thank roco juregui and prakash panangaden for comments on the draft .
the work was supported by a grant from the office of naval research ( n000140811249 ) .
99 a. peres and d. r. terno , journal of modern optics 50 , 1165 ( 2003 ) .
arvind and n. mukunda , pramana 45 , 347 ( 1996 ) .
a. aiello and j. p. woerdman , physical review a 70 , 023808 ( 2004 ) .
a. aiello and j. p. woerdman , arxiv : quant - ph/0503124v1 .
n. t. lindner and d. terno , journal of modern optics 52 , 1177 ( 2005 ) .
t. setl , a. shevchenko , m. kaivola , and a. t. friberg , physical review e 66 , 016615 ( 2002 ) .
p. m. alsing and g. j. stephenson , arxiv:0902.1399 .
m. czachor and m. wilczewski , physical review a 68 , 010302(r ) ( 2003 ) .
p. caban and j. rembieliski , physical review a 68 , 042107 ( 2003 ) .
p. caban and j. rembieliski , physical review a 76 , 052102 ( 2007 ) .
e. p. wigner , annals of mathematics 40 , 149 ( 1939 ) .
s. weinberg , the quantum theory of fields ( cambridge university press , cambridge , 1995 ) .
tung , group theory in physics , ( world scientific , philadelphia , pa , 1985 ) .
g. w. mackey , bulletin of the american mathematical society 69 , 628 ( 1963 ) .
r. u. sexl and h. k. urbantke , relativity , groups , particles , ( springer , 2001 ) .
r. m. gingrich , a. j. bergou and c. adami , physical review a 68 , 042102 ( 2003 ) .
f. r. halpern , special relativity and quantum mechanics ( prentice - hall , englewood cliffs , nj , 1968 ) .
l. c. biedenharn and j. d. louck , angular momentum in quantum physics , vol .
8 of encyclopedia of mathematics and its applications ( addison - wesley , reading , ma , 1981 ) .
s. d. bartlett and d. r. terno , physical review a 71 , 012302 ( 2005 ) .
s. d. bartlett , t. rudolph , and r. w. spekkens , reviews of modern physics 79 , 555 ( 2007 ) .
s. d. bartlett , t. rudolph , and r. w. spekkens , physical review letters 91 , 27901 ( 2003 ) .
a. shaji and e. c. g. sudarshan , physics letters a 341 , 48 ( 2005 ) .
the projection corresponding to the peres - terno measurement scheme acts as @xmath105 where the ` floored ' quantities indicate the projection action . denoting @xmath106
we get @xmath107 @xmath108 @xmath109 and @xmath110
the purpose of the following two appendices is to recall the problem of the polarization encoding for realistic photonic wave packets .
readers not familiar with this problem full of intricate details might find this short review useful .
the poincar algebra has two casimir operators @xmath111 and @xmath112 where @xmath113 is the pauli - lubanski vector , @xmath114 are the total angular momentum generators and @xmath115 are the boost generators .
we recall that @xmath116 is the levi - civita tensor and @xmath117 .
the corresponding poincar group induces translations @xmath118 ( in the coordinate representation ) where @xmath119 is a lorentz transformation @xmath120 and @xmath121 .
it is well known that the poincar group is a semidirect product of the proper orthochronous lorentz group and a group of translations @xmath122 .
the group acts on set @xmath123 which is a linear vector space equipped with the minkowski metric with the signature @xmath124 .
hence the set @xmath123 is a manifold rich in structure .
note that in the case of minkowski spacetime the action of the poincar group is uninteresting since it acts transitively .
it is not so in the four - momentum space .
the lorentz group acts in a similar manner for both spaces but the action of the translation operator differs .
one can see it easily ( in an admittedly handwaving way ) realizing that the fourier transform of a function @xmath125 for some constant @xmath126 is a simple phase transformation in the three - momentum space .
the analysis of the four - momentum space @xmath123 shows that there are six orbits .
the whole classification from the physical point of view can be found , for instance , in @xcite .
we are interested in the orbit which physically corresponds to null vectors with a positive zero component of the momentum four - vector .
they correspond ( after quantization ) to the scarce but important class of free massless particles containing photons and possibly gravitons . by the very definition of orbit
no lorentz transformation can change the character of a null vector .
in other words , photons always travel at the speed of light no matter the reference frame a potential observer resides in .
we are free to generate the ` null orbit ' from any null momentum four - vector but it is customary and advantageous to choose the simplest possible one the standard four - momentum @xmath127 . note that other choices are completely equivalent as discussed in detail in @xcite . from the physical perspective , the standard direction corresponds to ( eventually quantized ) plane electromagnetic waves ( photons ) traveling at the normalized speed of light along the @xmath24 axis .
nevertheless , the word ` physical ' should be used carefully .
it is not , in principle , possible to generate such a field ( classical or quantum ) .
the reason is that the sharp momentum value implies that the classical wave is ` infinitely ' spread in position and one would need an infinite amount of energy to prepare such a field . in the quantum case the corresponding state does not even occupy a hilbert space of square integrable functions and is completely delocalized in minkowski spacetime .
classical or quantum fields created in a laboratory are in reality wave packets with a finite spread in momentum and position . using another property of the algebra @xmath128=0 $ ] and @xmath129 which holds in the massless case
only we denote a ` single - particle ' state @xmath130 as an eigenstate of @xmath131 where @xmath1 are eigenvalues of @xmath131 .
other possible degrees of freedom related to @xmath132 rather than to @xmath133 ( note that @xmath134=0 $ ] ) are gathered in @xmath135 @xmath136 to be able to use quantum mechanics in the relativistic context as we habitually do in ordinary quantum mechanics we need to introduce a hilbert space and a unitary representation in it which respects the composition law for the poincar group @xmath137 the appearance of the phase is due to the projective nature of quantum states ( rays instead of vectors ) . for simply - connected groups it is possible to get rid of the phase factor ( @xmath138 ) .
this is not the case of the lorentz ( or poincar ) group and it will have some impact at a later stage . the poincar group is non - compact and therefore there is no faithful finite - dimensional unitary representation .
this might have been a huge obstacle for studying the evolution of relativistic quantum states but wigner realized how to circumvent this problem .
he found that the unitary action of the lorentz group is governed by the irreps of its stabilizer subgroups @xcite . loosely speaking ,
the little group ` induces ' its action on the lorentz group .
the method is called the method of induced representations and was later studied and generalized by @xcite and others .
the wigner method prescribes that the unitary action of a lorentz transformation @xmath139 reads @xmath140 where @xmath141 is the irrep of the stabilizer group @xmath142 keeping the standard null direction @xmath143 intact .
the stabilizer group can be written @xmath144 where @xmath145 takes the standard four - vector to any other null vector ( i.e. , to a vector on the same orbit ) .
it turns out that the stabilizer group for the standard null orbit is the euclidean group @xmath146 ( also known as @xmath147 ) transforming a two - dimensional ( euclidean ) plane into itself .
it is a three parametric group generated by translations in two directions and a rotation around the axis perpendicular to the plane .
the components of the pauli - lubanski vector are the generators of the corresponding lie algebra .
the group is noncompact so it might seem that we did not improve our situation but we actually did . to make the long story short
@xcite it appears that the noncompact part of the stabilizer group is not physically relevant to the evolution of the photon under the lorentz group so the only @xmath148 subgroup of @xmath146 remains .
the corresponding algebra generator is @xmath149 and we find @xmath150 we can see from eq .
( [ eq : heli ] ) that @xmath151 so we define helicity as the projection of the angular momentum along the direction of motion and @xmath135 is clearly a relativistic invariant .
the @xmath148 group element is as usual recovered by exponentiation @xmath152 so for the purpose of eq .
( [ eq : wigneraction ] ) we get @xmath153}\delta_{\s\s'}\ ] ] and so for an arbitrary lorentz group @xmath154 @xmath155}\ketq.\ ] ] [ rem : wignerangle ] we will call @xmath156 the wigner angle and it can be explicitly calculated for a given lorentz transformation @xmath154 from eq . ( [ eq : stabilizer ] ) by setting @xmath157 .
notice that the phase factor @xmath158}\in u(1)$ ] is indeed a unitary representation of @xmath159 .
the two groups are known to be isomorphic .
the parameter @xmath135 can take just integer or half - integer values as the result of the inability to set the phase in eq .
( [ eq : unitaryrepre ] ) to zero @xcite .
moreover , the momentum projection in two opposite directions @xmath160 and @xmath161 are related by the parity operator so for photons as massless vector bosons we get @xmath162 .
we see that it is necessary to enlarge the poincar group by including the parity transformation @xcite . clearly , there is no helicity equal to zero since there is no rest frame for a photon .
a state @xmath130 coincides with the action of creation operators @xmath163 to the vacuum state from the usual quantization procedure of a free massless vector field yielding a single particle state of a momentum @xmath160 and helicity @xmath162 ( alias circular polarization ) .
it is straightforward to generate multi - particle states by a repeated application of the creation operators for different modes satisfying @xmath164=\d ( p'- p)\d_{\s'\s}$ ] and thus to create the familiar fock space as a direct sum of the completely symmetric hilbert space of @xmath165 photons @xmath166 .
also notice that one of the consequences of the wigner procedure is a clear indication that there are just two ` spin ' degrees of freedom and they are always perpendicular to the direction of motion and to each other .
we know this fact without mentioning the coulomb gauge whatsoever .
the coulomb gauge can achieve the same goal but one has to pay the price of not having a manifestly covariant gauge condition .
that is , the gauge condition must be imposed for every reference frame separately . on the other hand ,
the coulomb gauge is important and useful in quantum field theory from a broader point of view .
the authors of @xcite proposed a very simple measurement which corresponds to what actually might happen in a laboratory when measuring the helicity ( polarization ) degrees of freedom .
assume that the direction of propagation of the wave packet is the standard direction @xmath25 .
it is customary to place a polarization analyzer perpendicular to the direction of motion .
our helicity eigenstates @xmath167 can be represented as two orthogonal complex four - vectors @xmath168 . for a free field
the coulomb gauge @xmath169 implies the axial component of the helicity vector to be zero @xmath170 ( holds for all @xmath0 because just rotations are considered now ) .
quantum electrodynamics indicates what happens next .
we summon up the well - known relation for the helicity eigenvectors @xcite @xmath171 where the overbar indicates complex conjugation . following the notation of eqs .
( [ eq : mom - heleigsrepresented1 ] ) and ( [ eq : mom - heleigsrepresented2 ] ) and appendix b the helicity measurement for @xmath43 corresponds to the trivial projection @xmath172 where special symbols have been assigned for the helicity eigenstates because we work with them in the main body .
the problem is that the projection is of rank two .
when a single photon arrives in a direction @xmath173 it has a longitudinal component and hence it gets cut during the course of measurement .
then , to have a consistent definition of the helicity density matrix in a rotated reference frame , one has to reconstruct the longitudinal part because that is the place where some parts of the wave function ` got lost ' .
this basically means to measure the helicity in the planes perpendicular to the @xmath174 and @xmath37 axes in our fixed coordinate system to be able to build a three - dimensional helicity density matrix . finally , we get a two - dimensional effective density matrix @xmath175 by cutting a @xmath176 block from the whole three - dimensional matrix .
the resulting state is positive - semidefinite and subnormalized .
it is the consequence of the lemma following the definition .
[ lem : submatrix ] let @xmath178 be a density matrix written in terms of ( not necessarily orthogonal ) pure states @xmath179 such that @xmath180 .
then every leading principal submatrix of @xmath178 is positive semidefinite .
taking the @xmath181dimensional leading principal submatrix of @xmath179 means to replace the last @xmath182 entries of @xmath179 by zeros .
we will label the resulting vectors as @xmath183 .
the corresponding matrices @xmath184 are still positive semidefinite and the positivity stays preserved by taking their convex combination @xmath185 .
the resulting density matrix is subnormalized .
the projection @xmath186 does exactly the job of cutting out the principal two - dimensional submatrix from the @xmath187 helicity matrix ( recall that the axial part is set to zero for all @xmath160 by the gauge condition ) .
therefore @xmath188\pi _ k,\ ] ] where @xmath189 is a normalization constant .
( [ eq : rhoeff ] ) says nothing other than if we measure the helicity perpendicular to the propagation direction ( the @xmath24 axis in this case and so @xmath75 is of the simple form eq .
( [ eq : projpol ] ) ) a statistical mixture of all @xmath56 helicity components for all @xmath160 is generated .
this is exactly the result from @xcite .
of course , we may decide to measure along a general axis @xmath190 . in this case eq .
( [ eq : rhoeff ] ) is still valid if we take @xmath191 and render the helicity vectors in the new basis .
one can adopt the approach of @xcite where the above peres - terno measurement scheme was reformulated in terms of measurement of the stokes parameters of the incoming wave packet .
we will not go into detail but we just point out the lie algebraic aspect of this approach . in a similar manner as above
the authors basically constructed three different effective helicity matrices which correspond to the above mentioned measurement of the helicity in three orthogonal spatial directions . from these they constructed a three - dimensional helicity density matrix which in general can be written in terms of the @xmath192
lie algebra generators @xmath193 @xcite @xmath194 where @xmath195 are the actual measured stokes parameters .
we now realize that the @xmath192 lie algebra is composed of three mutually dependent @xmath196 lie algebras and they precisely form the three effective helicity density matrices calculated in @xcite ( and one of them is eq .
( [ eq : rhoeff ] ) ) .
we should stress that in the above process one indeed gets a longitudinal part of the wave function but there is no need to call it a non - physical situation @xcite .
we simply get a helicity component in the plane not parallel to the plane where the helicity is measured . in principle
, this situation can not be avoided unless one works with single - particle states or very narrow photonic wave packets . also , the three - dimensional helicity density matrix concept has only little to do with the bosonic nature of the photon .
it is just a coincidence that we measure along three perpendicular directions corresponding to our usual spatial dimensions .
if we had had a more sophisticated measurement device we could have measured the helicity in each and every plane perpendicular to every @xmath160 direction and map it to some other ( non - photonic ) multi - level quantum system .
we would avoid projecting it down to our usual three - dimensional space where helicity measurement devices usually operate . pushing it to the limit , this way we would have been able to recover the whole density matrix because @xmath197 holds in the infinite - dimensional helicity - momentum hilbert space .
so far we talked just how a wave packet transforms under an ordinary rotation .
the remedy was to introduce a three - dimensional helicity density matrix or to eventually create a narrow wave packet where the ` longitudinal ' component is negligible @xcite .
the effect of a boost on a wave packet is more disastrous .
assume again that the wave packet eq .
( [ eq : genwavepacket ] ) is propagating along the @xmath24 axis with the standard momentum @xmath143 and an observer is also moving along the @xmath24 axis with a constant velocity @xmath198 .
having @xmath199 the boost @xmath200 where @xmath201 does not induce any phase change @xcite so the boost transforms a general single - particle state as @xmath202 .
the effect of the boost on a four - momentum vector is twofold : ( i ) the magnitude of the three - vector and the zeroth component are no longer equal to one .
this is the doppler effect we suppress by renormalization as discussed below eq .
( [ eq : genwavepacket ] ) .
( ii ) the gauge condition is violated since it is not a lorentz covariant condition and must be imposed for every @xmath203 as is usually done in quantum electrodynamics @xcite .
thus @xmath204 is followed by @xmath205 where @xmath206 ( @xmath207 following from ( i ) ) . the gauge condition eq .
( [ eq : gaugecond ] ) assures that the new helicity four - vectors stay orthogonal in a plane orthogonal to @xmath203 .
this is a final blow to our effort to have nice transformation properties of the helicity density matrix between two different frames of reference .
the things get more technically complicated if we allow for a boost in a general direction
. then , a wigner phase appears what can be easily seen if we realize that such a boost can be decomposed in a boost in the @xmath24 direction followed by a rotation in the required direction . | we present a photonic wave packet construction which is immune against the decoherence effects induced by the action of the lorentz group .
the amplitudes of a pure quantum state representing the wave packet remain invariant irrespective of the reference frame into which the wave packet has been transformed .
transmitted information is encoded in the helicity degrees of freedom of two correlated momentum modes .
the helicity encoding is considered to be particularly suitable for free - space communication .
the integral part of the story is information retrieval on the receiver s side .
we employed probably the simplest possible helicity ( polarization ) projection measurement originally studied by peres and terno .
remarkably , the same conditions ensuring the invariance of the wave packet also guarantee perfect distinguishability in the process of measuring the helicity . |
binary systems are among the best known sources of gravitational waves . specially interesting
are the double neutron star systems ( nsns systems ) , the binaries formed by a neutron star and a black hole ( bhns systems ) and the binaries of black holes ( bhbh systems ) , because their emissions have a high probability of being detected in the near future .
besides , it is well known that binaries lose energy and momentum via emission of gravitational waves , which cause reductions in their orbital distances , consequently increasing the orbital frequencies . moreover , for systems in circular orbits , the frequency of the emitted waves is twice the orbital frequency .
the process continues until the systems reach the coalescing phase , where the systems leave the periodic regime and start the merging phase . on the other hand ,
concerning the study of the gravitational radiation itself , the stochastic backgrounds are of special interest .
backgrounds can be generated when , for example , there is a superposition of signals of several sources , resulting in smooth - shaped spectra spanning a wide range of frequencies . in particular ,
we are concerned in this paper with the backgrounds generated by a population of coalescing compact binaries formed from redshifts ranging from @xmath0 up to @xmath1 , i.e. , cosmological binaries .
we can find in the literature some very interesting works on this issue , such as @xcite , where the authors calculated the backgrounds generated by coalescing nsns systems . generally speaking , they used monte carlo techniques to simulate the extragalactic population of compact binaries . besides , these authors considered the time evolution of the orbital frequency by using the delay time " ( that is , the interval of time between the formation and the coalescence of the systems ) in their calculations . in our method , we consider the time evolution in an explicit form , where the equation describing the evolution of the orbital frequency is taken at each instant of life of a given binary . in zhu
et al.@xcite , the authors consider the spectra generated by coalescing bhbh systems . in this paper , they assume average quantities for the energy emissions of single sources ; on the other hand , in our calculations we consider all the values for the orbital parameters a system can have , in the form of distribution functions . still concerning the spectra generated by coalescing bhbh systems , marassi et al .
@xcite adopted an updated version of the seba population synthesis code , in which the masses of the black holes range from @xmath2 and @xmath3 . roughly speaking , the various papers cited above show results that are characterized by spectra with frequencies ranging from @xmath4hz and @xmath5hz and with maximum amplitudes located in the interval ranging from @xmath6hz to @xmath7hz . as shown below
, the backgrounds we generated have similar forms to the ones we mentioned , though our results show , in general , higher amplitudes .
this difference will be discussed timely .
further , as it will be shown , our method has the very useful characteristic of being numerically simple , in the sense that it does not demand a heavy computational work . in this paper
, the spectra generated by the coalescing binaries will be calculated by means of@xcite @xmath8 where @xmath9 represents the dimensionless amplitude of the spectrum , @xmath10 is the observed frequency , @xmath11 is the amplitude of the signals generated by each source and @xmath12 is the differential rate of generation of gravitational waves .
it is worth pointing out that ( [ bg ] ) was deduced from an energy - flux relation .
in fact , in a paper by de araujo et al.@xcite , the authors gave a detailed derivation of this equation , showing its robustness .
actually , one can use ( [ bg ] ) in the calculation of different types of stochastic backgrounds , provided that one knows @xmath11 and the corresponding @xmath12 to the case one is dealing with . here , @xmath11 has the form@xcite @xmath13 where @xmath14 is the emitted frequency that , in this case , is the frequency emitted by a coalescing system ( note that @xmath14 and @xmath10 are related to each other by [ redshift ] ) , @xmath15 is the reduced mass and @xmath16 is the total mass .
the differential rate @xmath12 in writing in following form @xmath17 where @xmath18 is the comoving volume element , @xmath19 is the redshift and @xmath20 and @xmath21 are the mass distribution functions of the components of the systems . in fact , for neutron stars such distributions are given by dirac s delta functions , since we are considering that all these objects have the same mass of @xmath22 ; for black holes we use the function@xcite @xmath23\mbox{,}\ ] ] where @xmath24 is given in solar mass units .
the term @xmath25 is known from cosmology , and the problem of determining @xmath12 comes down to the calculation of @xmath26 . at this point , one could ask in what the present study differs from the previous ones .
we will see that the main difference has to do with the application of a new method to calculate @xmath12 developed in our previous papers@xcite .
the paper is organized as follows : in section [ sec2 ] , we show the main steps to obtain @xmath12 ; then , with ( [ bg ] ) , @xmath12 and ( [ source ] ) at hand , we calculate @xmath9 for the three families of compact binaries , which are explained in section [ sec3 ] ; in section [ sec4 ] we present the results and discuss , in particular , the detectability of the backgrounds studied here by the interferometric detectors laser interferometer space antenna ( lisa ) ( now evolved lisa ( elisa ) ) , big bang observer ( bbo ) , deci - hertz interferometer gravitational wave observatory ( decigo ) , advanced laser interferometer gravitational wave observatory ( aligo ) , einstein telescope ( et ) , and the cross - correlation of pairs of aligos and et ; finally , in section [ sec5 ] we present our conclusions .
we present here the main steps for the calculation of @xmath26 ( we refer the reader to refs .
@xcite for details ) .
we start by writing @xmath26 in the form @xmath27 where the expression in the right - hand side is the formation rate of systems per comoving volume that reach the frequency @xmath14 at the instant @xmath28 ( see the derivation in appendix [ apena ] and also in ref.@xcite ) .
also , @xmath29 refers to the instant of birth of the systems . in ( [ dr2 ] ) , @xmath30 is the frequency distribution function of the binaries , which has the form ( we refer the reader to appendix [ apenb ] and ref.@xcite for the derivation of this equation ) : @xmath31^{1/3}\nu^{-11/3}\nu_{0}^{2}\mbox{exp}\left[\frac{-(r-\bar{r})^{2}}{2\sigma^{2}}\right ] , \label{dist2}\end{aligned}\ ] ] where @xmath32 is the initial frequency ; @xmath33 , @xmath34 and @xmath35 are constants given in table [ tab2 ] ; and @xmath36 is the orbital distance , related to @xmath14 by means of kepler s third law . on the other hand , @xmath37 is the binary formation rate density that , for nsns systems is given bys @xmath38 here , @xmath39 refers to the redshift of birth of the systems , which is related to @xmath29 via the usual expression that can be found in any textbook of cosmology ; and @xmath40 is the star formation rate density ( sfrd ) .
there are , in the literature , many different proposals to the sfrd , although they do not differ from each other very significantly . here
we adopt , as a fiducial one , that given by springel and hernquist@xcite , namely @xmath41 where @xmath42 , @xmath43 , @xmath44 and with @xmath45 fixing the normalization .
besides , in ( [ bfrd ] ) , @xmath46 ( see @xcite ) is the mass fraction of stars that is converted into neutron stars , where @xmath47 is the fraction of binaries that survive to the second supernova event ; @xmath48 gives the fraction of massive binaries ( that is , those systems where both components can generate supernovae ) and @xmath49 is the mass fraction of progenitors that originates neutron stars , which , in the present case , is calculated by @xmath50 where @xmath51 is the salpeter mass distribution@xcite with @xmath52 and @xmath53 .
numerically , we have @xmath54 , @xmath55 and @xmath56 . now , considering ( [ bfrd ] ) , we can write ( [ dr2 ] ) in the form @xmath57 where , following the notation adopted by zhu et al@xcite , @xmath58 is calculated by @xmath59dt_{0}.\ ] ] in this context , following wu et al@xcite , @xmath60 is given by @xmath61 where @xmath62 is the local coalescence rate and @xmath63 has the form @xmath64_{z=0}dt_{0}.\ ] ] for bhns and bhbh systems , we have similar expressions , but with different values of @xmath37 .
such values can be estimated by means of the results found in belczynski et al.@xcite , where these authors claim that the population of binaries is formed by @xmath65% of nsns , @xmath66% of bhbh and @xmath67% of bhns binaries .
therefore , @xmath68 and @xmath69 can be related to @xmath70 by means of these proportions .
it is worth mentioning that belczynski et al.@xcite studied compact binaries with merger times lower than @xmath71yr ; that is , they considered coalescing binaries .
basically , in the simulations they used , the binaries are formed through several different channels .
specifically , nsns systems are formed through @xmath72 different channels , where there is a predominance of channels containing hypercritical accretion between a low - mass helium giant and its companion neutron star . on the other hand , bhns and bhbh systems
are formed through just for and three channels , respectively , where there is a moderate predominance of mass transfer events .
in this section , we present the calculations of the spectra generated by nsns , bhbh and bhns systems .
although we are using ( [ bg ] ) for the three cases , the calculations are different for each family of binaries .
therefore , we will show the calculations separately .
neutron stars , according to theories of stellar evolution and observations ( see , e.g. , ref .
@xcite ) have characteristic masses that fall in a narrow interval around @xmath22 .
so , in this paper we are considering that all neutron stars have masses of @xmath22 , which is a realistic choice , and at the same time , a simplification .
one could ask how the results would be affected by the choice of the neutron star ( ns ) equation of state ( eos ) .
since we are considering the coalescing phase , the results depend mainly on the mass of the nss .
nsns systems are characterized by a specific coalescence frequency that , according to , e.g. , ref .
@xcite , may be considered as @xmath73hz for a pair of two @xmath22 nss .
the choice of the eos is certainly important for the subsequent phase of evolution of the system , when the merger phase takes place .
although there is just a value for the coalescence frequency , the spectrum will be spread over a wide range of frequencies .
this behavior is due to the cosmic redshift , because systems emitting at the same frequency , but at different redshifts , will generate signals with different observed frequencies , obeying @xmath74 where , in this case , we are considering @xmath75hz .
moreover , as we are considering that the redshift has the minimum and maximum values given by @xmath76 and @xmath77 , respectively , the observed frequencies will have minimum and maximum values of @xmath78hz and @xmath79hz , respectively .
since the masses of the components of the systems and their coalescing frequencies are the same , the spectra will only depend on the redshift @xmath19 .
however , from ( [ redshift ] ) one notices that there will be just one value of @xmath10 for each value of @xmath19 , such that ( [ bg ] ) must be handled in a particular way .
first , let us rewrite ( [ bg ] ) in the form @xmath80 where @xmath81 , and @xmath82 is the value of redshift corresponding to each observed frequency by means of ( [ redshift ] ) .
now , in order to calculate ( [ coalnsns1 ] ) , @xmath12 can be written in the following form @xmath83 where @xmath84 is the value of the comoving volume at @xmath82 and @xmath85 is the dirac s delta function .
therefore , substituting ( [ coalnsns2 ] ) in ( [ coalnsns1 ] ) and integrating , we have @xmath86 the study of the coalescence of bhns systems is more complicated than the case of nsns systems , because the frequency of coalescence depends on the mass of the black hole .
it is usually assumed that the coalescence occurs when the neutron star reaches the innermost stable circular orbit ( isco ) of the black hole .
so , using kepler s third law and recalling that the emitted frequency is twice the orbital frequency , we have @xmath87 where @xmath88 is the radius of the isco of the black hole , which is related to the schwarzschild radius , namely @xmath89 substituting ( [ schw ] ) in ( [ kepler ] ) we have @xmath90 since we are considering that black holes have masses in the interval
@xmath91 @xmath92 @xmath93 , the emitted frequencies will have minimum and maximum values given by @xmath94hz and @xmath95hz , respectively .
moreover , considering ( [ redshift ] ) and ( [ massbh ] ) , one notes that for each value of @xmath10 , one has continuous ranges of values for @xmath19 and @xmath96 .
however , in the case of bhns systems , ( [ bg ] ) must be integrated over @xmath19 but not over @xmath96 , since these variables are not independent .
moreover , for a given value of @xmath10 one needs to determine the maximum and minimum limits of integration , which are given by @xmath97 when @xmath98 or @xmath99 , we set @xmath76 and @xmath77 , respectively . once the interval of integration is set , the variables @xmath96 and @xmath14 in the integral can be written as functions of @xmath19 by solving ( [ massbh ] ) and ( [ redshift ] ) . in addition , note that in this case one needs to consider in the calculation of ( [ bg ] ) the distribution function given by ( [ blackhole ] ) .
as a result , we have @xmath100 in this case , the frequency of coalescence depends on the values of the two components of the system .
following marassi et al.@xcite , @xmath14 is given by @xmath101 where @xmath102 is the symmetric mass ratio and @xmath16 is the total mass .
the polynomial coefficients are @xmath103 , @xmath104 and @xmath105 . according to ( [ bhbh1 ] ) , for each value of @xmath14
, there will be a continuous set of pairs of values for @xmath106 and @xmath107 that satisfies this equation .
moreover , the masses are not independent , in fact they are related to each other by ( [ bhbh1 ] ) .
therefore , we should integrate over one of the masses ( or one parameter which describes both masses . )
however , it is not possible to solve ( [ bhbh1 ] ) analytically , i.e , to write the masses as functions of @xmath14 .
therefore , we need to use approximations . first , consider that @xmath106 and @xmath107 ( which we refer as to @xmath108 and @xmath109 ) are related to each other by @xmath110 where the variable @xmath111 is greater than or equal to one . substituting ( [ bhbh2 ] ) in ( [ bhbh1 ] ) we can write the masses @xmath108 and @xmath109 as functions of @xmath14 and @xmath111 , namely @xmath112\\\nonumber m_{2}&=&\frac{c^{3}}{g}\frac{1}{\pi \nu}\left[\frac{a_{0}k^{2}}{(1+k)^{5}}+\frac{b_{0}k}{(1+k)^{5}}+\frac{c_{0}}{1+k}\right].\end{aligned}\ ] ] a suitable approximation for ( [ bhbh1 ] ) is given by @xmath113 since @xmath114 for all values of @xmath108 and @xmath109 .
considering that @xmath115 , we use ( [ bhbh4 ] ) in order to estimate a first value for @xmath108 ; next , we take this pair of values for the masses and calculate a first value for @xmath116 ; we then correct the value for @xmath108 using @xmath117 with this new value of @xmath108 , we repeat the above process : we calculate @xmath116 again and use ( [ bhbh5 ] ) , bearing in mind that this process may be performed an arbitrary number of times in order to yield more accurate values for @xmath108 . in the cases where we have @xmath118 at the end of the process , we consider @xmath119 and perform an analogous process to find out @xmath109 . with the pair @xmath120 at hand ,
we calculate the maximum value of @xmath111 : @xmath121 thus , in order to cover all possible values of @xmath108 and @xmath109 , we consider that @xmath111 , in ( [ bhbh3 ] ) , ranges form one to @xmath122 . finally , ( [ bg ] ) can be written as follow @xmath123
as usual , in the literature , we represent the backgrounds in terms of the strain amplitude @xmath124 , which is given by @xmath125 and also in terms of the energy density parameter @xmath126 , which reads ( see , e.g. , ref .
@xcite ) @xmath127 the spectra are shown in fig .
[ fig1 ] and fig .
[ fig2 ] for @xmath124 and @xmath126 , respectively .
note that these spectra are mainly compared to the sensitivity curves of et@xcite and aligo@xcite , since the spectra are in their frequency bands .
the sensitivity curves of the proposed space - based antennas lisa , elisa@xcite , bbo@xcite , decigo@xcite are also shown , but given their frequency bands they can not detect the spectra studied here . from fig .
[ fig1 ] , one notices that the background generated by the three families of compact binaries are below the sensitivity curves of the interferometric detectors . besides , one notices that the background generated by the bhns systems have higher amplitudes when compared to the ones generated by nsns and bhbh systems . on the other hand ,
the spectrum corresponding to bhbh systems has the lowest amplitudes .
since our calculations depend on some parameters and functions , it is worth investigating how our results are affected by different choices of these quantities .
first , let us consider the masses of the components : in ( [ source ] ) , if we multiply both masses by a factor of @xmath128 , @xmath11 will be multiplied by a factor of @xmath129 .
important variations in the amplitudes would occur only if @xmath130 or @xmath131 . from ( [ bg ] ) , ( [ bfrd ] ) and ( [ strain ] ) ,
one notices that @xmath132 and @xmath133 .
therefore , if ones multiply @xmath40 or @xmath60 by , say , a factor of @xmath134 , the amplitudes shown in fig . [ fig1 ] will increase by a factor of @xmath135 .
therefore , for realistic scenarios , different choices for @xmath40 and @xmath60 would have small effects on the amplitudes of the backgrounds . comparing fig .
[ fig2 ] with similar studies found in the literature ( see , e.g. , refs .
@xcite ) , one sees a good agreement concerning the shapes of the spectra , although our results show higher amplitudes .
for example , in @xcite one sees that @xmath136 at @xmath137hz for the backgrounds generated by nsns systems , while for our corresponding spectrum , shown in fig .
[ fig1 ] , we have @xmath138 at the same frequency . comparing our results for bhbh systems ( see fig .
[ fig2 ] ) with the results found in @xcite , one can note some similarities : the amplitudes increase until a maximum value in the range @xmath139hz and then they have a sharp decrease . concerning the amplitudes , we have a maximum value of @xmath140 , while in zhu et al the value is @xmath141 .
marassi et al @xcite also study backgrounds generated by bhbh binaries .
these authors discussed different models , and the resulting spectra present maximum amplitudes ranging from @xmath142 for a frequency band around @xmath143hz .
it is worth mentioning that , generally speaking , the spectra is model dependent .
therefore , different assumptions lead to different backgrounds . in ref.@xcite , for example , the population of binaries is such that the maximum probability of coalescence is around @xmath144 . therefore , for @xmath145 there is a relatively small proportion of coalescing systems emitting ; in our calculations we do not consider such a behavior .
this difference in the proportion of systems at lower redshifts could explain our higher amplitudes as compared to the ones of ref .
@xcite .
although the spectra ( signals ) shown in fig .
[ fig1 ] are below the sensitivity curves of the detectors , it could well be possible detect them by correlating the outputs of two or more detectores . for the correlation of two interferometers , the detectability of a given signal can be quantified by means of the so called signal - to - noise ratio ( s@xmath146n ) , namely @xcite : @xmath147,\ ] ] where @xmath148 and @xmath149 are the spectral noise densities , @xmath150 is the integration time , and @xmath151 is the overlap reduction function , which depends on the relative positions , spatial orientation , and distances of the detectors ; and @xmath126 is given by ( [ omega ] ) . in table
[ tab1 ] one can see the s@xmath146n for the three families of compact binaries , in particular for pairs of aligos and et .
lll system & aligo & et + nsns & @xmath152 & @xmath153 + bhns & @xmath154 & @xmath155 + bhbh & @xmath156 & @xmath157 + from table [ tab1 ] , one notices that et could in principle detect the backgrounds where the spectrum generated by bhns systems would have higher probability of detection ; for pairs of aligos , the low values of the s@xmath146n ratio indicate a non detection .
in this paper , we calculate the stochastic background of gravitational waves generated by coalescing compact binaries , using a new method developed in our previous studies @xcite .
we show that , of the three spectra considered in this paper , the one generated by bhns systems has the highest amplitudes , while the background by bhbh systems show the lowest amplitudes . moreover , one notices slight differences in the forms of the spectra , which are due to the different methods used to calculate them by means of ( [ bg ] ) .
we found that the backgrounds calculated here would not be detected by the interferometric detectors such as ligo and et , although thanks to the cross - correlation of signals et could , in principle , detect such signals .
particularly , we found that the spectrum generated by bhns systems have the highest s@xmath146n ratio , while the one corresponding to bhbh systems presents the lowest s@xmath146n .
concerning the dependence of our results on the parameters used in the calculations , we found that the masses of the components of the binaries , as well as @xmath40 and @xmath60 , do not strongly influence the backgrounds . besides , a particular choice for the ns eos does not affect the results either .
we compared the spectra studied here with some interesting results found in the literature .
one notices similarities in their shapes , namely : maximum frequencies of @xmath5hz and maximum amplitudes in the range @xmath158hz . roughly speaking ,
these characteristics are common to the three families of binaries . on the other hand ,
our amplitudes given in terms of @xmath159 are in general higher than the ones found in the literature by roughly one order of magnitude .
we concluded that such a difference is mainly due to population characteristics assumed .
therefore , generally speaking , the spectra is model dependents .
efde would like to thank capes for support and jcna would like to thank fapesp and cnpq for partial support .
finally , we thank the referee for the careful reading of the paper , the criticisms , and the very useful suggestions which greatly improved our paper .
in this appendix , we present the main steps of the derivation of @xmath26 . for further details ,
we refer the reader to refs .
@xcite .
we derive @xmath26 by means of an analogy with a problem of statistical mechanics . in this problem ,
the aim is to calculate the number of particles that reach a given area @xmath160 in a time interval @xmath161 , i.e. , the objective is to calculate the flux @xmath162 of particles .
basically , this flux is calculated by counting the particles inside the volume @xmath163 , adjacent to the area @xmath160 , that are moving towards @xmath160 with velocity @xmath164 , where @xmath165 obeys a distribution function @xmath166 .
hence , the flux is obtained by integrating over all the positive values of @xmath165 . with some modifications ,
this method can be used to determine @xmath26 .
first , we substituted the spatial coordinate @xmath167 by the frequency @xmath14 and the velocity @xmath165 by the time variation of the frequency , which is defined by @xmath168 .
therefore , the number of systems in the interval @xmath169 adjacent to a particular frequency @xmath14 is given by considering that the distribution @xmath172 gives the number of systems which have @xmath173 in the interval @xmath174 , the number of systems in @xmath169 and with values of @xmath173 in the interval @xmath174 is given by notice that the denominator of the term between parenthesis in ( [ back2 ] ) is the total number of systems .
now , using the function given by ( [ xi ] ) and changing the differential @xmath169 by means of the chain rule , ( [ dmu ] ) assumes the form where @xmath204 , @xmath108 and @xmath109 are the masses of the components of the system and @xmath205 is the initial frequency .
so , carrying out a change of variables via @xmath206 , where @xmath205 was associated with the variable @xmath202 in @xmath207 , one has finally , it would be necessary to perform a further coordinate transformation in order to write @xmath209 as a function of the emitted frequency @xmath14 .
such a transformation , calculated by means of @xmath210 , is trivial . besides
, @xmath36 is written as a function of @xmath202 by means of kepler s third law . | gravitational waves are perturbations in the spacetime that propagate at the speed of light .
the study of such phenomenon is interesting because many cosmological processes and astrophysical objects , such as binary systems , are potential sources of gravitational radiation and can have their emissions detected in the near future by the next generation of interferometric detectors .
concerning the astrophysical objects , an interesting case is when there are several sources emitting in such a way that there is a superposition of signals , resulting in a smooth spectrum which spans a wide range of frequencies , the so - called stochastic background . in this paper
, we are concerned with the stochastic backgrounds generated by compact binaries ( i.e. binary systems formed by neutron stars and black holes ) in the coalescing phase .
in particular , we obtain such backgrounds by employing a new method developed in our previous studies .
example.eps 20 20 moveto 20 220 lineto 220 220 lineto 220 20 lineto closepath setlinewidth gsave .4 setgray fill grestore stroke grestore |
the possibility to trap and manipulate the atoms in a bose - einstein condensate using standing wave laser beams @xcite has led to a renewal of the interest in basic solid state models . in such systems ,
the atoms experience a periodic potential from an optical lattice leading to formation of band structure in the energy spectrum .
these bands have been investigated in experiments @xcite . in the spectroscopy experiments in ref .
@xcite , the atoms experienced a periodic potential in only one direction , being free to move on a much larger length scale in the other directions .
this implied that interactions between atoms could be ignored . if the atoms are confined to reside on the sites of a lattice in three dimensions , interactions become important . as a result , it was shown theoretically @xcite , and subsequently also experimentally @xcite , that a system of interacting cold atoms , residing in the lowest bloch band of the periodic potential , maps onto a bosonic hubbard model .
this model is of great theoretical interest since it exhibits a quantum phase transition @xcite between ground states where the atoms are localized ( mott - insulator ) and where they are delocalized ( superfluid ) as the strength of the hopping relative to the inter - atomic interaction is varied .
the dynamics of particles under the influence of changes in the hamiltonian ( such as lattice tilts or rapid changes in the particle interaction strength ) has also proved interesting @xcite .
another development is an interest in the idea of mixing bosonic atoms of different flavors in the lattice @xcite .
several ways of achieving multiple flavors have been suggested including using atoms of different species and exploiting different internal atomic states .
so far , experiments on strongly interacting atoms in three dimensional optical lattices have been restricted to atoms in the lowest ( zeroth ) bloch band .
recently scarola and dassarma considered the possibility of novel supersolid phases within the first excited bloch band of an optical lattice .
@xcite in this paper , the theory of atoms in the two lowest ( zeroth and first ) bloch bands of a three dimensional optical lattice is considered .
we show here , that due to the lack of available phase space for the decay products , such excited states can ( in some parameter ranges ) have life times much longer than the characteristic time scales for inter - site hopping . thus it should be possible to establish quasi - equilibrium within the manifold of these metastable states .
we find that it is possible in this way to realize novel effective multi - species bosonic hubbard hamiltonians .
depending on the choice of lattice depths the number of degenerate bands varies and we find effective models involving @xmath0 flavors of bosons , where @xmath0 can be 1 , 2 , or 3 .
these flavors correspond to the three different possible nodal planes in the excited state wave function such as the one illustrated in fig .
( [ fig : wavefuns ] ) . we will show that a characteristic of these hamiltonians is that ( to a good approximation ) each flavor can only hop in one direction ( _ i.e. _ , @xmath1 ( nodal plane ) particles can only hop in the @xmath2 direction , etc . ) . neglecting interactions
we would then have @xmath0 interpenetrating one - dimensional free bose gases , one for each column ( or row ) in the lattice . allowing intra - species interactions converts these one - dimensional gases into luttinger liquids ( or , if the interactions are strong enough , and the mean particle number per site is an integer , into mott insulators ) .
we show below that , besides intra - species interactions , the full interaction also includes on - site inter - species conversion terms that allow atoms to change flavor in pairs .
thus for example , two @xmath1 particles constrained to move along a single @xmath2 column can collide , turn into @xmath3 particles and move away along a @xmath4 column .
such processes lead to novel quantum dynamics for this coupled set of interpenetrating luttinger liquids .
( plotted in the plane @xmath5 ) formed by the zeroth band bloch functions and the wave function @xmath6 formed by the zeroth band bloch functions in the y- and z- directions and the bloch functions from the first band in the x - direction .
these are approximately harmonic oscillator states , @xmath7 $ ] , and @xmath8 $ ] where the parameter @xmath9 is determined by the curvature of the optical lattice potential near its minima . similarly @xmath10 and @xmath11 .
[ fig : wavefuns],title="fig:",width=264 ] as will be seen , the anisotropic nature of the hopping in conjunction with the pairwise conversion leads to hamiltonians with an infinite but subextensive set of @xmath12-gauge symmetries intermediate between local and global .
such infinite symmetries have been found in certain frustrated spin models @xcite and in a ` bose metal ' model @xcite and are known to cause dimensional reduction in some cases .
@xcite we will see below how this dimensional reduction appears in a simple way in this system .
a related _ global _ @xmath12 symmetry and associated ising order parameter appear in problems involving boson pairing due to attractive interactions mediated by feshbach resonances . in that case
the symmetry appears due to a conversion term that connects pairs of bosons with a distinct molecular field .
this can lead to exotic states in which pairs of bosons are condensed but single bosons are _ not _ and in which half vortices are permitted @xcite further , due to strong interatomic repulsion , the ground state in 3d ( three flavors ) breaks a kind of chiral symmetry and displays an additional accidental ground state degeneracy at the mean field level .
a similar situation occurs for special parameter values in frustrated xy - models , where parallel zero energy domain walls can be inserted @xcite .
the outline of this paper is as follows : in section [ sec : lattice_ham ] the appropriate generalization of the bosonic hubbard model is introduced along with numerical values of the parameters entering the hamiltonians obtained from band - structure calculations for various lattice depths .
then , in section [ sec : effective_hams ] , the aforementioned effective hamiltonians for atoms in the first band are derived for three particular choices of relative lattice depths in the @xmath13 directions . using simple mean - field theory we sketch the ground state phase - diagrams in section [ sec : mft ] and in section [ sec : interference ] we discuss how the superfluid phases are reflected in the interference pattern in an experimental situation . finally , in section [
sec : lifetime ] , treating the interaction perturbatively to second order , we estimate the lifetime of a population inverted state ( all atoms residing entirely in the first excited band ) .
the starting point is the hamiltonian for weakly interacting bosons of mass @xmath14 in an external potential @xcite @xmath15 where @xmath16 is the @xmath17-wave scattering length .
the external potential has two contributions @xmath18 and @xmath19 corresponding to the lattice potential and the magnetic trapping potential . denoting the wavelength of the lasers by @xmath20 , @xmath21 being the lattice spacing ,
the former can be written @xmath22 the positional dependence of the magnetic trapping potential @xmath23 is much weaker than that of the lattice , _ i.e. _ @xmath24 and will be ignored in the remainder of this paper .
one should be aware though , that this term has been shown to influence for instance the phase diagram of the single flavor bosonic hubbard model @xcite . for the cubic lattices considered here , the wannier functions corresponding to the noninteracting part of the hamiltonian in eq .
( [ eq : hamiltonian ] ) can be written @xmath25 here the bold face vectors @xmath26 and @xmath27 are integer triplets @xmath28 and @xmath29 which represent band indices and lattice sites respectively , i.e. @xmath30 these functions are to a good approximation described by localized harmonic oscillator wave functions sketched in fig .
[ fig : wavefuns ] .
the completeness of the wannier functions allows the field operators to be expanded as @xmath31 the operators @xmath32 and @xmath33 , which are the creation and annihilation operators of bosons at site @xmath34 and with band index @xmath35 , obey bose commutation relations @xmath36=\delta_{\bf
n,\bf n^\prime}\delta_{{\bf m},{\bf m}^\prime}.$ ] ignoring all hopping other than nearest neighbor hopping and all interactions other than on - site interactions , the hamiltonian in eq .
( [ eq : hamiltonian ] ) can be written @xmath37\nonumber \\ & + & \frac{1}{2}\sum\limits_{{\bf n}_{1},{\bf n}_{2},{\bf n}_{3},{\bf n}_{4}}\sum\limits_{\bf m}u({\bf n}_{1},{\bf n}_{2},{\bf n}_{3},{\bf n}_{4})\left[\hat{d}^\dagger_{{\bf n}_{1}}({\bf m } ) \hat{d}^\dagger_{{\bf n}_{2}}({\bf m})\hat{d}_{{\bf n}_{3}}({\bf m})\hat{d}_{{\bf n}_{4}}({\bf m})\right].\end{aligned}\ ] ] here , the on - site interaction energies are defined as @xmath38 while the energies @xmath39 and the hopping energies @xmath40 are given by @xmath41 @xmath42 note that the energies @xmath43 for hopping in the @xmath44-direction depend only on the lattice depth @xmath45 in the corresponding direction and the @xmath46:th component @xmath47 of the band index @xmath35 .
the notation @xmath48 in eq .
( [ eq : latticehamiltonian ] ) indicates that the sum should be carried out over nearest neighbor sites @xmath27 and @xmath49 in the @xmath44-direction .
one could for instance write , @xmath50 it is straight forward to numerically solve the noninteracting schrdinger equation and find the energies in expressions ( [ eq : unnnn])-([eq : tn ] ) above . in doing so , it is convenient to first switch to dimensionless units
. thus , we measure length in units of the inverse wave vector and potential depth in units of the recoil energy @xmath51 , i.e , @xmath52 and @xmath53 , with @xmath54 the hopping energies for the two lowest bands , obtained from band - structure calculations , are shown in fig .
[ fig : j12 ] as functions of lattice depth . to get the on - site interaction ( eq .
[ eq : unnnn ] ) in a suitable form to use later on in the paper we define dimensionless overlap integrals @xmath55 where dimensionless wannier wave functions @xmath56 have been introduced .
the dependence on @xmath57 in these functions is parametric , i.e. @xmath58 is the wannier function corresponding to the one - dimensional non - interacting problem with potential depth @xmath59 .
the variable @xmath0 denotes the band index . approximating the wannier functions with harmonic oscillator wave functions corresponding to the curvature at the bottom of each well
, one finds approximate values for the overlap integrals @xmath60 a comparison between these values for the overlap integrals and those obtained from band - structure calculations is shown in fig .
[ fig : overlaps ] . and @xmath61 in units of the recoil energy as functions of lattice depth @xmath45 in the hopping direction .
the upper line is the hopping energy @xmath62 for atoms in the first bloch band hopping between nearest neighbor wells while the lower line , @xmath61 , corresponds to atoms in the zeroth band .
[ fig : j12],title="fig:",width=302 ] defined in eq .
( [ eq : overlaps ] ) .
solid lines , from top to bottom , @xmath63 , obtained from numerical calculations .
the dashed lines correspond to the values in eq .
( [ eq : hoapprox ] ) obtained by using harmonic oscillator wave functions determined from the curvature of the potential at the well bottom .
[ fig : overlaps],title="fig:",width=272 ]
in this section we will focus on the meta - stable situation having all atoms in the first bloch band(s ) of the lattice .
it is quite easy to achieve such a situation , i.e. consider an initial moment of time when the optical lattice has been loaded with atoms in the lowest bloch band , @xmath64 .
the anharmonicity of the lattice well potential allows one to treat the vibrational degree of freedom as a two level system .
if one singles out , say , the @xmath2-direction , then , by applying an appropriate vibrational @xmath65-pulse , i.e. shaking " the lattice in this direction with a frequency on resonance with the transition @xmath66 , the state can be inverted and the atoms excited to states with band index @xmath67 .
a strong @xmath65-pulse can achieve this inversion in a time short compared to the inter - well hopping time so that the dispersion of the upper band is not an issue in the inversion process .
the simplest starting state would be the mott insulator state with one boson per site in the lowest band . in a typical experimental setup ,
the parabolic confining potential will cause the population of each well to vary and the system will be in a state with regions of mott insulators with different filling factors .
it is however easy to confirm ( by direct simulation ) that even in this case , taking interactions into account , a pulse shape can be tailored that will invert the population simultaneously for regions with different filling factors provided a deep enough lattice is used .
another way of preparing the initial state is to use the method recently demonstrated by browaeys _
_ @xcite . by loading a condensate into a moving 1d lattice and applying a subsequent acceleration
the condensate can be prepared in the lowest energy state ( quasimomentum @xmath68 ) in the first bloch band .
the situation desired in this paper can then be obtained by ramping up the lattice in the two remaining ( perpendicular ) directions adiabatically .
the natural question regarding the lifetime of the resulting meta - stable state will be considered in sec .
[ sec : lifetime ] .
the subsequent dynamics of atoms in the first band(s ) is then predominantly governed by some subset of the terms in eq .
( [ eq : latticehamiltonian ] ) . this relevant subset will be referred to as the effective hamiltonian .
we will treat three different regimes of values for the lattice potentials @xmath45 which lead to effective hamiltonians with one , two , and three flavors respectively .
the three scenarios are : 1 .
@xmath69 ( 1d , single flavor ) .
2 . @xmath70 ( 2d , two flavors ) .
3 . @xmath71 ( 3d , three flavors ) .
as indicated , the number of particle flavors as well as the dimensionalities in the effective hamiltonians vary . the reason for the different numbers of flavors becomes clear if one considers the restrictions on the final states into which two atoms may scatter due to the interatomic interaction ; the presence or absence of such states can be inferred from the presence or absence of degenerate , or nearly degenerate , levels in the energy spectrum of the noninteracting system .
take , for example the second scenario above with @xmath72 and all atoms initially in a state with index @xmath67 , then , due to the on - site inter - atomic interaction these atoms can scatter elastically into a state with index @xmath73 through a first order process the connecting different degenerate states .
further , it is easy to show that scattering resulting in states with other indices , for instance @xmath74 , is only possible through higher order processes if energy ( and also parity ) is to be conserved and can safely be ignored if the gas is dilute .
hence , the atoms can , at a formal level , be divided into two flavors : an @xmath1 flavor corresponding to atoms in @xmath67 and a @xmath3 flavor in @xmath73 . by the same argument
one can see how the one- and three - flavor situations arise .
apart from having different number of flavors the dimensionalities of the effective hamiltonians differ . to understand this consider again the second case above , @xmath70 , with particles in the excited bands @xmath67 and @xmath73 corresponding to @xmath1- and @xmath3-flavors
. for the @xmath1 flavor , hopping in the @xmath2-direction has a matrix element @xmath75 while hopping in the @xmath4- and @xmath76- direction have matrix elements @xmath77 and @xmath78 respectively . looking at fig .
[ fig : j12 ] it is then clear that , to a good approximation , the @xmath1 particles can only hop in the @xmath2-direction while hopping in the @xmath4- and @xmath76-directions is strongly ( exponentially ) suppressed .
similarly the @xmath3 particles can only hop in the @xmath4-direction and all hopping occurs only in the @xmath79 plane , hence the 2d character .
a similar argument holds for the three flavor case where in addition to the @xmath1 and @xmath3 particles , there are @xmath80- particles hopping in the @xmath76-direction
. the effective hamiltonians also contain terms arising from the on - site interaction .
apart from the terms that repel atoms from each other , the symmetry of the on - site interaction allows , say , two @xmath1 particles moving in the @xmath2-direction to collide and convert into two @xmath3 particles which thereafter move off in the @xmath4-direction .
the time reversed process can of course also occur .
thus , the number of particles of each flavor is not conserved and there is a pairwise exchange of particles of different flavors .
the anisotropy of hopping and the flavor conversion process is schematically depicted for the 2d ( two flavors ) case in fig .
[ fig:2dillustr ] .
particles of @xmath1 flavor are shown in gray while the @xmath3 flavor is drawn in black .
below , we give the effective hamiltonians for all three different cases listed above .
the first case to be considered is when @xmath82 and only states with band index @xmath67 ( and possibly some residual atoms in @xmath83 ) are occupied . in anticipation of the other effective hamiltonians
it is convenient to introduce for the @xmath1 flavor the creation and destruction operators @xmath84 and @xmath85 , _
i.e. _ : @xmath86 the effective hamiltonian is then essentially that of a quasi one - dimensional bosonic hubbard model @xmath87\right)\nonumber \\ & -&t\sum\limits_{\left<{\bf m , m^\prime}\right>_x}\left[\hat{x}^\dagger_{\bf m}\hat{x}_{{\bf m}^\prime}+{\rm h.c.}\right]\nonumber\end{aligned}\ ] ] the energies @xmath88 , @xmath89 , and @xmath90 arise from the inter - well tunneling and the inter - atomic interaction respectively .
the presence of atoms residing in the lowest band leads to an additional effective on - site energy and can be absorbed in the on - site energies @xmath91 , @xmath92 .
although this single - flavor model is equivalent to a single - flavor model in the zeroth bloch band , the additional random on - site potential resulting from residual atoms could be exploited in the study of the disordered bose - hubbard system .
the parameters entering the hamiltonian ( [ eq:1dlatticehamiltonian ] ) are conveniently expressed as @xmath93 @xmath94 to simplify the notation for the case @xmath96 , we introduce new letters for the creation / annihilation operators @xmath97 , @xmath98 .
then , the hamiltonian governing atoms in the excited bands becomes @xmath99\nonumber\\ -t\sum\limits_{\left<{\bf m , m^\prime}\right>_y}\left[\hat{y}^\dagger_{\bf m}\hat{y}_{\bf m^\prime}+{\rm h.c.}\right]-t\sum\limits_{\left<{\bf m , m^\prime}\right>_x}\left[\hat{x}^\dagger_{\bf m}\hat{x}_{\bf m^\prime}+{\rm h.c.}\right]\nonumber \\
+ u_{xy}\sum\limits_{{\bf m}}\hat{n}^{(x)}_{\bf m}\hat{n}^{(y)}_{\bf m}+\frac{u_{xy}}{2}\sum_{\bf m}[\hat{x}^\dagger_{\bf m}\hat{x}^\dagger_{\bf m}\hat{y}_{\bf m}\hat{y}_{\bf m}+{\rm h.c.}].\nonumber\\ \label{eq:2dham}\end{aligned}\ ] ] again , the energy @xmath100 arises from the interatomic interaction and depends on the lattice depth .
note that this two flavor bosonic hubbard hamiltonian differs in an important aspect from previously studied two flavor systems : the presence of the last term that mixes the two flavors .
hence , the inter - atomic interaction leads to a josephson term " that allows for the conversion of two @xmath1-atoms into two @xmath3-atoms and vice versa .
the coefficients @xmath101 are given by the same expression as in the 1d case while @xmath102 figure [ fig:2dillustr ] illustrates the dynamics in the 2d ( two flavors ) situation .
the generalization of the above hamiltonian to the case when @xmath104 is straight forward . introducing a third flavor @xmath105 , @xmath106 , one may write an effective hamiltonian as @xmath107\right)\nonumber \\ &
+ & \sum\limits_{s\neq s^\prime}\sum\limits_{{\bf m}}u_{ss^\prime}\left(\hat{n}^{(s)}_{\bf m}\hat{n}^{(s^\prime)}_{\bf m}+\frac{1}{2}[\hat{s}^\dagger_{\bf m}\hat{s}^\dagger_{\bf m}\hat{s^\prime}_{\bf m}\hat{s^\prime}_{\bf m}+{\rm h.c.}]\right)\nonumber\\ & -&t\sum\limits_{s = x , y , z}\sum\limits_{\left<{\bf m , m^\prime}\right>_s}\left[\hat{s}^\dagger_{\bf m}\hat{s}_{\bf m^\prime}+{\rm h.c.}\right].\label{eq:3dham}\end{aligned}\ ] ] here @xmath108 with @xmath90 and @xmath100 given by ( [ eq : uaa ] ) and ( [ eq : uabexp ] ) with @xmath109 . because of overall number conservation the hamiltonian has the usual global @xmath110 symmetry .
however , because the flavor conversion occurs pairwise and locally ( _ i.e. _ , on site ) , the hamiltonians described above also exhibit an infinite number of @xmath12 gauge symmetries corresponding to conservation modulo 2 of the number of @xmath1 particles in any given column of the lattice running in the @xmath2-direction ( and similarly for @xmath3 and @xmath80 particles ) .
these symmetries correspond to invariance under each of the transformations @xmath111\ ] ] @xmath112\ ] ] @xmath113,\ ] ] where the integer pair @xmath114 in the superscript of each @xmath115 determine the location of a column .
the first transformation for example takes @xmath116 for all @xmath117 in the column specified by @xmath118 and @xmath119 .
since @xmath84 and @xmath85 operators always appear pairwise , the hamiltonian is invariant under this class of @xmath12 transformations .
these @xmath12 symmetries are in a sense intermediate between local and global .
while the number of such symmetries is infinite ( in the thermodynamic limit ) it is of course sub extensive and thus not large enough to fully constrain the system ( or to make it integrable for example ) . as mentioned in the introduction , such symmetries have been found in certain frustrated spin models @xcite and in a ` bose metal ' model @xcite and are known to cause dimensional reduction in some cases .
@xcite because introducing a defect across which the sign of the @xmath12 order parameter changes along any given single column costs only finite energy , the system will , like the 1d ising model , disorder at any finite temperature thereby restoring the @xmath12 symmetry .
we will see below how this reduced dimensionality physics appears in a simple way in this system .
having derived effective hamiltonians in one , two , and three dimensions , we turn now to the investigation of their ground states .
the 1d , single flavor hamiltonian has been extensively studied ( see for instance ref .
@xcite and references therein ) and needs no further discussion here .
the other two hamiltonians in eqs .
( [ eq:2dham ] ) and ( [ eq:3dham ] ) deserve some attention though .
the 2d hamiltonian ( [ eq:2dham ] ) is a two flavor bosonic hubbard hamiltonian , a system that has recently received much attention and shown to have a rich phase diagram @xcite . in this section
we will investigate the ground state of the hamiltonian in eq .
( [ eq:2dham ] ) using simple mean field theory .
the hamiltonian ( [ eq:2dham ] ) differs from those previously studied in two aspects : the presence of pairwise inter flavor mixing and the anisotropic tunneling .
we follow here the method suggested in ref .
@xcite ( see also refs .
we consider the possibility that the global @xmath110 and columnar @xmath12-symmetries discussed in section [ sec : z2 ] are spontaneously broken by introducing complex scalar columnar order parameter fields @xmath120 and @xmath121 , _
i.e. _ one for each @xmath2-column and one for each @xmath4-column .
these fields should then satisfy the self consistency conditions @xmath122 for all @xmath117 in the @xmath2-column specified by @xmath118 , and , @xmath123 for each @xmath118 in the @xmath4-column specified by @xmath117 . for simplicity
we omit in this discussion of the 2d ( two flavors ) case the @xmath76 component @xmath119 of the position vector @xmath34 .
the possibility that fluctuations restore the symmetry will be discussed further below .
mean field theory results from decoupling sites in the same column by neglecting fluctuations in the kinetic energy .
for instance , for the @xmath2-column specified by a particular value of @xmath118 one has @xmath124 thus the sites along each column decouple . doing the same for the @xmath3:s and writing the hamiltonian in dimensionless form where all energies are scaled by @xmath90 , _
i.e. _ , @xmath125 , @xmath126 and @xmath127 , we obtain @xmath128 here , the on - site mean field hamiltonians are given by @xmath129 -2\tilde{t}\left[\psi_y(m_x)\hat{y}^\dagger_{\bf m}+\psi_y^*(m_x)\hat{y}_{\bf m}\right]+\tilde{u}_{xy}\hat{n}^{(x)}_{\bf m}\hat{n}^{(y)}_{\bf m}\nonumber \\ & + & \sum\limits_{s = x , y}\left(\frac{1}{2}\hat{n}^{(s)}_{\bf m}[\hat{n}^{(s)}_{\bf m}-1]-\tilde{\mu}\hat{n}^{(s)}_{\bf m}+2\tilde{t}|\psi_s({\bf m})|^2\right)+\frac{\tilde{u}_{xy}}{2}[\hat{x}^\dagger_{\bf m}\hat{x}^\dagger_{\bf m}\hat{y}_{\bf m}\hat{y}_{\bf m}+ \hat{y}^\dagger_{\bf m}\hat{y}^\dagger_{\bf m}\hat{x}_{\bf m}\hat{x}_{\bf m}],\nonumber\\ \label{eq:2dham2}\end{aligned}\ ] ] where @xmath130 serves as a common chemical potential .
the on site hamiltonians satisfy the eigenvalue relations @xmath131 for two arbitrary complex fields .
an eigenstate of the full mean field hamiltonian can be written as a product state of such eigenstates @xmath132 where the fields satisfy the self consistency conditions in eqs .
( [ eq : scons1])-([eq : scons2 ] ) .
the mean field ground state is obtained by globally minimizing the energy @xmath133 with respect to the fields and the set of eigenstates @xmath134 .
this is most easily done by numerical diagonalization in a truncated hilbert space where each site can hold at most a total of @xmath135 atoms . since @xmath136}&&\left(\sum\limits_{\bf m}\epsilon_{n_{\bf m}}(\psi_x(m_y),\psi_y(m_x))\right)\nonumber \\ & \ge&\sum\limits_{\bf m}\min\limits_{\left[n,\psi_x,\psi_y\right]}\epsilon_{n}(\psi_x,\psi_y)\nonumber\end{aligned}\ ] ] and @xmath137 of the order parameter fields @xmath138 and @xmath139 .
solid arrows correspond to @xmath140 and dashed to @xmath141 .
[ fig:2dphases],width=302 ] it is enough to minimize the ground state energy of a single site with respect to the fields and then find the largest manifold of states compatible with having columnar order parameters fields . carrying out this scheme reveals two different scenarios for the minimum of each on - site energy @xmath142 ; either @xmath143 and @xmath144 is integer ( incompressible ) , or , @xmath145 ( compressible ) .
the former case corresponds to a mott insulating state while the latter suggests a superfluid phase . due to the positivity of @xmath100
the last term in the mean field hamiltonian is minimal whenever @xmath146 and @xmath147 , on the same site , have a phase difference of @xmath148 . for the ( mean field )
ground state manifold we must have in this phase @xmath149 for all @xmath150 and @xmath151 column order parameters . requiring the phases of all @xmath146 in each @xmath150 and all @xmath147 in the @xmath151 columns to be the same while fixing the relative phase between @xmath146 and @xmath147 to @xmath152 results in configurations as the one shown in fig .
[ fig:2dphases ] . here
the phases of @xmath146 and @xmath147 are shown represented as arrows ( planar spins ) .
the direction of the arrows defining the angle . clearly , this phase shows a breaking of the global @xmath110 symmetry .
the meaning of the quasi - local nature of the @xmath12 symmetries discussed above becomes clear . although the phases of @xmath146 in each @xmath150 column are the same there is no energy cost associated with flipping all the spins @xmath2-spins in a single column or all the @xmath4-spins in a @xmath4-column .
the ordering between different columns is thus nematic .
one should note here , that since the only energy cost associated with flipping a single spin , say an @xmath2-spin in an @xmath150column , is given by the states of the neighboring @xmath2 spins in the column , the situation is essentially that of a 1d ising model along each column .
hence , at any finite temperature , domains of flipped spins will proliferate and the @xmath12 symmetries will be restored .
this essentially one dimensional behavior is an example of the dimensional reduction mentioned above .
the model under consideration is highly anisotropic .
mainly since @xmath1 particles can only hop in the @xmath2-direction , it seems to be impossible to develop phase coherence among @xmath1 particles in different @xmath2-columns ( and similarly for the other flavors ) .
suppose however that , as discussed above , the flavor exchange interaction term causes the relative phase of two flavors , say @xmath1 and @xmath3 , to lock together so that @xmath153 condenses @xmath154 in this case the mean field decomposition of the exchange interaction yields terms of the form @xmath155 which permit individual particles to change flavor and hence phase coherence can freely propagate in all directions throughout the lattice via a kind of ` andreev ' process ( _ i.e. _ self - energy off - diagonal in flavor index ) in which an @xmath1 particle can turn into a @xmath3 particle when it needs to travel in the @xmath4 direction .
to understand this isotropic superfluid phase , it is convenient to consider a phase only representation with compact phase variables on each site @xmath156 and @xmath157 . the flavor exchange ( ` josephson ' ) term then becomes ( for the 2d ( two flavors ) case ) @xmath158).\ ] ] defining @xmath159 we have @xmath160 assuming that the relative phase of the condensates is locked together by this josephson term is equivalent to assuming that ( for @xmath161 ) the fluctuations of @xmath162 away from the ground state value @xmath163 ( or its equivalent @xmath164 under the @xmath12 gauge symmetry ) are massive and can be ignored .
thus we obtain @xmath165 the continuum limit of the anisotropic kinetic energy @xmath166 then becomes @xmath167 and we immediately see that the anisotropy has effectively disappeared at long wavelengths and we have a superfluid . a vortex configuration in @xmath168 can be viewed as a bound state of two half vortices in the @xmath169 and @xmath170 fields .
the columnar @xmath12 symmetry allows the @xmath146 field to have a phase jump of @xmath65 across a cut parallel to the @xmath2 axis and similarly for @xmath147 .
thus half vortices are permitted .
if the two order parameter phases are locked together ( @xmath162 fluctuations are massive ) then the two half vortices are confined to each other as shown in fig .
( [ fig : half - vortex ] ) .
such a vortex has an energy which scales ( as usual ) only logarithmically with system size , despite the semi - infinite branch cut ( @xmath65 phase jump ) of @xmath146 running horizontally out to the right from the vortex center and of the similar branch cut in @xmath147 running vertically out above the vortex center .
( black solid arrows ) and @xmath147 ( red dashed arrows ) containing a half vortex . notice the branch cuts indicated by the long solid black and dashed red lines .
the @xmath12 symmetry means that these branch cuts have zero ` string tension ' and contribute only a finite core energy to the vortex .
the half winding number can be seen by going around a loop a - b - c - d - a and calculating the total phase twist @xmath171 .
this twist is calculated by summing the changes in @xmath170 when going vertically and @xmath169 when going horizontally .
[ fig : half - vortex],width=302 ] to see that such a vortex is topologically well defined despite the @xmath12 symmetry one can consider a loop around the vortex core as shown in fig .
( [ fig : half - vortex ] ) . in going around the loop
we add up the phase twist @xmath171 and map onto the complex plane . to calculate @xmath171 along the loop the changes in @xmath170 has to be added when going vertically and the changes in @xmath169 when going horizontally .
the net results is that going once around the vortex core the phase winds by @xmath65 .
if one applies a @xmath65-flip in all the @xmath169 ( @xmath170 ) phases in any row ( column ) the mapping onto the complex plane will remain invariant .
the mott insulating states , having integer number of atoms in each well , are best characterized by the @xmath172 eigenstates . these are product states @xmath173 where @xmath174 and the integer
@xmath175 is the total number of particles @xmath176 in each well .
the index @xmath46 runs from @xmath177 to @xmath175 for each @xmath175 and for the three lowest values of @xmath175 the eigenstates are @xmath178 @xmath179 @xmath180 @xmath181 @xmath182 @xmath183 in the plane of @xmath184 , the scaled chemical potential @xmath185 , and @xmath186 , the scaled hopping energy . here
calculated for the experimentally relevant ratio @xmath187 using a truncated hilbert space with at most 10 particles per site .
lobes of mott - insulating states , of successively increasing integer filling factor with increasing chemical potential @xmath184 , are surrounded by a superfluid phase .
the superfluid phase is characterized by columnar order parameter fields @xmath120 and @xmath121 , one for each @xmath2- and @xmath4- column respectively .
all @xmath146 and @xmath147 have equal , nonzero , magnitudes while their relative phases are either @xmath177 or @xmath65 .
( see fig . [ fig:2dphases ] ) .
[ fig : phasediag2d],width=302 ] in fig .
[ fig : phasediag2d ] the mean field phase diagram has been drawn for the physically relevant value @xmath188 which is characteristic for the proposed setup . the lobes marked mi correspond to incompressible mott insulating phases with integer filling factors .
the remaining part of the diagram , marked sf , corresponds to a superfluid phase with the columnar nematic ordering ( see fig .
[ fig:2dphases ] ) discussed above . considering the @xmath189 eigenstates above two things become clear . trivially , if @xmath190 the lowest lobe , and all other odd filling lobes , vanish and the model reduces to two noninteracting single flavor models as expected . secondly , at @xmath191 there is a level crossing between @xmath192 and @xmath193 .
it follows that the size of the lowest odd filling lobes increases with increasing values of @xmath194 up until @xmath191 after which it starts to decrease again . by considering fluctuation effects higher order in the tunneling amplitude
, we can demonstrate that the permutational symmetry between the @xmath1- and @xmath3- flavors can be broken in the mott insulator phase . in the absence of tunneling , the single - particle states @xmath195 and
@xmath196 are degenerate . taking tunneling into account
breaks this degeneracy and to second order in @xmath197 ( using for instance the schrieffer - wolff transformation @xcite ) an effective ( pseudo ) spin-@xmath198 hamiltonian for the interaction between neighboring sites can be found @xmath199 the up- and down- states of the pseudo spin operators @xmath200 correspond to the site @xmath34 being occupied by one @xmath1-atom or one @xmath3-atom respectively .
the effective magnetic interaction is @xmath201 there is a critical value of the inter flavor interaction @xmath202 for which @xmath203 vanishes . for @xmath204 ,
the system is ferromagnetic and spontaneously favors one flavor over the other . for @xmath205
the system is anti - ferromagnetic and favors an ordering with @xmath1- and @xmath3-atoms on alternating sites .
thus , we conclude that at integer filling factor the permutational symmetry between @xmath1 and @xmath3 flavors , ( or equivalently , the cubic symmetry of the underlying lattice ) is always broken in the mean field ground state .
further , in the anti - ferromagnetic state , sublattice ( _ i.e. _ translation ) symmetry is broken as well .
, @xmath206 and @xmath207 of the order parameter fields @xmath208 , @xmath209 and @xmath210 . as in the 2d ( two flavors ) case
the underlying symmetry of the hamiltonian allows for flipping say all the @xmath140 along any @xmath2-column by @xmath65 to obtain another ground state configuration .
in addition to the ground state degeneracy obtained from such operations , an accidental degeneracy associated with parallel planes of different chirality is present . in this figure
the middle @xmath211 plane has a different chirality than the other two @xmath211 planes.[fig:3dphases],width=264 ] , width=302 ] using the same type of mean field theory as for the 2d ( two flavors ) case , the 3d ( three flavors ) case can be treated as well .
the resulting phase diagram for @xmath188 is shown in fig .
[ fig : phasediag3d ] .
again , mott - lobes with integer filling factors are seen surrounded by a superfluid phase where all order parameters @xmath212 have equal magnitude , _
i.e. _ , @xmath213 . due to the positivity of the coefficient @xmath100 in the josephson term in eq .
( [ eq:3dham ] ) the relative phases of the three condensates are frustrated .
thus , writing @xmath214 , @xmath215 one finds @xmath216 .
an interesting effect here is that the on site frustrated phase configurations come in two different ` chiralities ' that can not be converted into each other by shifting any one of the phases by the @xmath65 shift allowed by the @xmath12 gauge symmetry . to see this
one may consider the current flowing between the condensates of different flavors on a given site .
the current flowing between the @xmath1 and @xmath3 condensates on a particular site is determined by @xmath217)$ ] . in a right handed configuration with , say , @xmath218
there is an on - site current flowing from @xmath219 the situation is different in a left handed configuration with @xmath220 , where the current is now flowing in the opposite direction , _
i.e. _ @xmath221 adding an arbitrary phase of @xmath65 ( _ i.e. _ invoking the @xmath12 symmetry ) to any of the phases does not affect these currents . starting from a ground state with the same chirality throughout the system one can choose a set of parallel planes and change the chirality of each plane individually .
such changing of chirality of a plane requires that the whole plane has the same chirality .
this additional ground state degeneracy is not associated with any symmetry of the hamiltonian but is an accidental one .
a similar situation occurs for special parameter values in frustrated xy - models , where parallel zero energy domain walls can be inserted @xcite .
one should note that such accidental degeneracies at the mean field level may be lifted by fluctuation effects associated with collective modes such as spin waves .
as in the 2d ( two flavors ) case , the smaller mott - lobes , corresponding to integer filling factors not divisible by the dimensionality of the system , are degenerate in the @xmath172 limit .
this degeneracy is lifted due to tunneling , leading to ( pseudo ) magnetic ordering like that demonstrated for the 2d ( two flavors ) case . to fully lift the degeneracy one has to employ fourth order perturbation theory .
the resulting hamiltonian will include terms acting simultaneously on three and four sites .
however , such fourth order corrections are very small and may be difficult to observe in the proposed experimental situation .
they can however lead to novel physics and can be intentionally generated @xcite . beyond @xmath222 , superfluid phases with broken permutational symmetry can be achieved .
shown here are the order parameters sorted according to magnitude , @xmath223 , @xmath224 for an interspecies interaction @xmath225 as a function of @xmath197 .
the total number of particles is shown as a dashed line .
[ fig : cubsym],width=302 ] before leaving this section , we comment on the possibility of breaking the permutational symmetry among the flavors in the superfluid phase .
as is well known , large interspecies interaction strength in the two flavor bosonic hubbard model leads to phase separation . a phenomena occurring also here if @xmath226 .
however , due to the positive constant in front of the josephson ( flavor changing ) term , another phenomenon can take place in the 3d ( three flavors ) model . as an example consider fig .
[ fig : cubsym ] . here
@xmath227 and @xmath225 .
as can be seen for small @xmath197 the system is in a mott insulating state with filling factor 2 .
as @xmath228 increases the system becomes superfluid .
this occurs in two steps .
first , mean field theory predicts a second order transition to a state with only one flavor superfluid and then a first order transition to a state with two nonzero superfluid order parameters of equal magnitude .
increasing the hopping strength further does not seem to make the third flavor superfluid .
we attribute this to the large energy cost associated with having the phases of the three order parameters in a frustrated configuration .
the traditional way of detecting superfluidity is by releasing the trap and looking at the density distribution of the expanding cloud .
provided that the cloud expands many times its initial diameter , the final position of a particle is determined by its momentum rather than its initial position .
hence this expanded real - space density distribution provides a direct picture of the momentum - space distribution of the trapped system .
more precisely , the density distribution a time @xmath228 after trap release is related to the momentum density of the trapped state @xmath229 as @xmath230 where @xmath231 . it is useful to think of this spatial distribution as resulting from interference of matter waves radiated by the different lattice sites when the trap is released .
the one - dimensional character of the @xmath12 gauge symmetry means that thermal fluctuations can destroy the long range order phase order by allowing the phase on an arbitrary site to flip by @xmath232 . if the system disorders in this way , any interference pattern in the radiated matter waves will be destroyed as well . in this case
, further information about the correlations in the system can be obtained by looking at the density fluctuations ( noise ) in the released cloud @xcite in a hanbury - brown twiss like statistical measurement .
we begin this section by looking at the zero temperature momentum distribution and then consider the density fluctuations of the expanded cloud around its mean .
although any real experiment is conducted at finite temperature , the zero temperature columnar phase ordering may prevail for a finite system at low enough temperatures .
the zero temperature momentum distribution is thus of interest and we will estimate it by using a single macroscopically occupied wave function corresponding to the superfluid states in the two- and three - flavor cases .
the details of the calculations can be found in the appendix and we here only state the main results .
we begin by considering a single 2d plane with @xmath233 sites at zero temperature in the two - flavor system and model the superfluid state with a macroscopically occupied wavefunction @xmath234 here @xmath235 is the vacuum state of the lattice , _ i.e. _ no atoms present , while @xmath236 is the creation operator @xmath237 the subscripts @xmath14 and @xmath0 denote the coordinates , rows and columns , in the lattice while @xmath9 and @xmath238 are phase factors ( @xmath239 ) determining the phase of the wavefunction on a given site . at zero temperature
the phases of @xmath1- particles are ordered along rows while the phases of @xmath3-particles are ordered along columns , _ i.e. _ , @xmath240 and @xmath241 ( cf .
[ fig:2dphases ] and fig . [ fig : fields ] ) . for a macroscopic occupation @xmath242 the observed density distribution in a single shot in the @xmath79 plane
after expansion is proportional to the momentum distribution @xmath243 as shown in the appendix we have for a single 2d plane in the two - flavor system @xmath244 where @xmath245 the functions @xmath246 and @xmath247 are the fourier transforms of the on - site wannier functions and @xmath248 and @xmath249 are @xmath250-periodic random functions with typical magnitude of order unity which depend on which of the degenerate ground states is observed ( see fig .
[ fig : random ] and the appendix for details ) . from the above equations the interference pattern from a single 2d plane in the two - flavor system
can bee seen to be a grid like structure as shown in fig .
[ fig : momdist2d ] where the interference pattern has been calculated numerically for a 40x40 lattice .
the appearance of lines , rather than points as in a single flavor 2d system , stems from the one - dimensional character of the superfluid state with phases only being aligned along rows ( columns ) but randomly distributed between rows ( columns ) .
the randomness in the distribution between the rows ( columns ) show up as the random interference pattern along grid lines .
, width=302 ] in an experiment one typically does not probe a single plane but it is the integrated density of a large number of planes that is imaged
. for imaging in the plane parallel to the 2d planes the integrated column density ( intensity in the absorption image ) is for a @xmath251 lattice with @xmath242 atoms in each 2d plane @xmath252 here the line over the quantum mechanical averaging denotes the averaging over the different ground state configurations allowed by the @xmath12 symmetry . since @xmath253 ( see appendix ) the random interferences seen in fig .
[ fig : momdist2d ] will be averaged out and a grid of smooth lines , void of interference , will be seen .
another source of smoothing out the random interferences comes from limited detector precision . for a large system ,
the random oscillations becomes increasingly rapid and only an average over nearby momenta can be probed . for a @xmath254 lattice for a specific realization of @xmath255 note that @xmath256 is symmetric under inversion .
( b ) the random function @xmath257 for a @xmath254 lattice for a specific realization of @xmath258 and @xmath259 .
note that @xmath260 is not symmetric under inversion .
[ fig : f_2],width=264 ] in the 3d ( _ i.e. _ three flavors ) case , the situation is very similar .
special care have to be taken with accidental symmetry breaking of the ground state giving rise to planes of different chirality .
if we assume that planes with uniform chirality have normals in the @xmath150direction the momentum distribution can be written @xmath261 + 2{\rm re } \left[\tilde{\psi}_y({\bf q})^*\tilde{\psi}_z({\bf q})\right]\nonumber\\ & + & 2{\rm re}\left[\tilde{\psi}_z({\bf q})^*\tilde{\psi}_x({\bf q})\right ] \label{eq : x}\end{aligned}\ ] ] where @xmath262 @xmath263 @xmath264 again , since long range order is only aligned along 1d strips , the released cloud will be a set of intersecting perpendicular planes with intersections at positions corresponding to odd momenta @xmath265 .
the planes in each direction will have a random intensity modulation specified by the random functions @xmath256 , @xmath266 and @xmath267 ( see appendix for details ) .
examples of these distribution functions @xmath256 and @xmath260 are shown in fig .
[ fig : f_2 ] .
the last three terms in eq .
[ eq : x ] randomly modulate the distribution along the intersections of the planes .
if a single shot measurement is made , the integrated column density will show a pattern of grid lines similar to that in fig .
[ fig : momdist2d ] , the grid lines showing random interference patterns . between the lines a periodic random distribution ( of lesser intensity than the lines ) will be present
. this latter distribution will be either @xmath256 or @xmath260 depending on the orientation of the planes with uniform chirality .
thus if the absorbtion image is taken in the same plane as the planes with uniform chirality this background modulation will be symmetric under space inversion ( cf .
fig [ fig : f_2](a ) ) whereas if it is taken perpendicular there will be no such symmetry in the random modulation ( cf .
fig [ fig : f_2](b ) ) .
as pointed out above , the dimensional reduction present in the system means that finite temperatures can destroy the 1d ising - like ordering of phases along columns and that the individual phases at any one site can be flipped @xmath268 , i.e. , the @xmath12 gauge symmetry is restored . in this case
there will be no visible interference pattern although atoms are delocalized , _
i.e. _ the delta peaks will be smeared and a random density distribution will be seen each shot . to illustrate the usefulness of correlation measurements we consider a single @xmath233 2d plane in the two - flavor system at unit filling ( @xmath269 ) . if the temperature is finite , not only may the @xmath12 symmetry in the superfluid state be restored but it is also possible for the unit filling mott state to be disordered .
there are then four different possible states the system can be in 1 .
superfluid with restored @xmath12 symmetry .
ferromagnetic mott insulator ( all atoms of the same flavor ) .
3 . anti - ferromagnetic mott insulator ( alternating flavors on alternating sites ) .
disorder mott insulator ( each site having one atom but with random flavor ) .
if one makes multiple single shot measurements and averages the density distribution obtained in each shot , one obtains a measure of the average momentum distribution ( see appendix ) @xmath270 as _ disorder averages_. to distinguish the four states one can instead measure the hbt - like density - density correlations of the expanding cloud @xcite , @xmath271 here @xmath272 is the density of atoms at point @xmath273 a time @xmath228 after the trap has been switched off averaged over many experimental realizations ( see appendix ) . to measure @xmath274
one calculates the product of the observed densities @xmath275 in each shot and averages over several experimental runs .
just as for the density distribution , the correlation function @xmath276 provides a measure of the momentum correlations @xmath277\ ] ] prior to trap release . to get a qualitative understanding of how the superfluid state can be detected by correlation measurements we first return to the @xmath278 result in the previous section and look at the periodic function @xmath279 .
this random modulation arose because phases of @xmath3:s were uncorrelated between rows .
since the relative phases of the @xmath3:s in rows is @xmath268 this function is even in @xmath280 and along any given grid line the quantity @xmath281 is thus strongly correlated when @xmath282 .
averaging over many realizations one sees that ( along a grid line of constant @xmath283 ) @xmath284 in the thermally disordered superfluid state the phase on any site is allowed to flip by @xmath65 ( restoring @xmath12 symmetry ) .
this destroys the delta peaks in @xmath285 and gives instead a random modulation given by the function @xmath286 .
since only @xmath65-flips are allowed this modulation is symmetric @xmath287 and we get strong correlations in @xmath288 @xmath289 where @xmath290 is a reciprocal lattice vector . on a technical level ( see appendix ) this can be seen to arise since the disorder averaged propagator for single particles is necessarily short - ranged due to the random @xmath65 phase changes at finite temperature , while the disorder averaged propagators for _ pairs _ of particles can still be long - ranged .
in appendix we have calculated @xmath291 for the four scenarios for a single plane in the two - flavor system and we here give the qualitative results .
the superfluid state is , confirming the qualitative discussion above , characterized by peaks at @xmath292 ( see eq .
( [ eq : gsf2d ] ) ) while the ferromagnetic mott state and the disordered mott state has peaks only at @xmath293 ( see eqs .
( [ eq : gfm2d ] ) and ( [ eq : gdo2d ] ) ) . the correlation function in these mott states
are distinguished by having different background intensities between peaks and different peak strengths . the antiferromagnetic mott state has peaks at half reciprocal lattice vectors @xmath294 ( see eq .
( [ eq : gafm2d ] ) ) . in the three flavor system
the situation is similar to the two flavor scenario discussed above . in the presence of thermal disordering of the superfluid state
the interference patterns of the mott state and the superfluid states become indistinguishable .
again , in the three flavor case the pair propagators will be nonzero in the disordered superfluid state and @xmath291 will have peaks at @xmath292 ( see eq .
( [ eq : gdo3d ] ) ) while there should be no problem to measure the correlation functions for a system with three flavors , the two - flavor system poses a problem of technical nature since in an experiment several uncorrelated 2d planes will be created .
suppose one has @xmath175 uncorrelated planes .
if one detects one atom at position @xmath273 and another at @xmath295 in a single experiment the atoms could have come from either the same plane or different planes .
measuring the product of densities in each shot and averaging over several experiments one will for the case with @xmath175 2d planes actually measure @xmath296 rather than @xmath297 thus the signal - to - noise ratio scales as @xmath298 requiring many experimental runs for large systems .
in the previous sections , effective hamiltonians for atoms in the first band(s ) of the optical lattice were introduced and the mean field ground state phase diagrams drawn . in doing so
, it was assumed that the interaction terms in the original hamiltonian ( [ eq : latticehamiltonian ] ) responsible for scattering particles between bands could be ignored . in this section , these interactions are taken into account perturbatively and the lifetime of atoms in the first band is estimated . the obtained ( inverse ) lifetime should be compared to other energy scales in the problem , most importantly the smallest one , the hopping energy .
if the lifetime turns out to be long compared to the time scale of hopping , the novel states described in the previous sections should be possible to realize in experiment . to simplify matters , the discussion will be restricted to the 1d - case .
the ensuing results are expected to agree well , both qualitatively as well as quantitatively , with the 2d and 3d cases to lowest order in perturbation theory .
this follows from taking parity considerations into account when determining the allowed transitions .
thus , ignoring tunneling in the @xmath299 directions and measuring distance in units of the lattice spacing ( see section [ sec : lattice_ham ] ) the 1d hamiltonian can be written @xmath300 with @xmath301 and @xmath302 here @xmath303 creates an atom at @xmath304 in the band of the 1d - system and @xmath305 . apart from the field operators @xmath306 it is convenient to define boson operators in two other bases
first , we have the basis of bloch functions @xmath307 with band index @xmath0 and lattice - momentum @xmath308 .
these functions satisfy @xmath309 and are associated with the field operators @xmath310 .
second , we have the wannier - functions @xmath311 defined in section [ sec : lattice_ham ] . in this section
we will denote the corresponding field operators by @xmath312 .
note that these definitions depart from the conventions in previous sections and that operators corresponding to bloch functions and wannier functions are distinguished in the number of subscripts .
begin by looking at the case when the second term @xmath313 in eq .
( [ eq:1d ] ) is small compared to @xmath314 and consider an initial state where all @xmath175 atoms reside in the lowest lying bloch state of the first band , @xmath315 a first order decay process is then one where two atoms in the first band collide , promoting one to the second band , the other to the zeroth band , i.e. the final state is @xmath316 the first order matrix element for this transition is occurring in eq .
( [ eq : decay ] ) for the first order decay rate calculated in the wide - band limit . in this limit
, there are no available energy states for the first - order decay provided @xmath317.(dashed line ) the dashed line shows the maximum value of the ratio @xmath318 for which the wide - band analysis is valid . [ fig : lifetime],width=302 ] @xmath319 if the filling factor ( atoms / well ) of the first band is @xmath320 , and the density of states of the @xmath0:th band is @xmath321 the transition rate per well becomes @xmath322 defining @xmath323 this can be compactly written as @xmath324 in fig .
[ fig : lifetime ] @xmath325 obtained from numerical calculation is shown . for convenience
the inverse of @xmath326 has been plotted . as can be seen
, the lifetime goes to zero for small and large @xmath57 .
this is a result of the diverging density of states at the band edges .
above @xmath327 the first order process is no longer energetically possible and higher order perturbation theory has to be applied .
the validity of the wide - band calculation relies upon the assumption that the inequality @xmath328 is satisfied .
this condition can be used to obtain an upper bound on the ratio @xmath318 by assuming the lattice depth to be the same in all directions , i.e. @xmath329 , which yields @xmath330 this quantity is shown for filling factor @xmath331 as the dashed line in fig .
[ fig : lifetime ] . from the discussion above it is clear that for deep enough potentials the validity of the wide - band analysis breaks down unless @xmath332 is extremely small .
an alternative starting point is when @xmath333 while the filling factors for the zeroth and the second band are small , i.e. , @xmath334 .
keeping terms of order @xmath335 , the relevant unperturbed hamiltonian to start from is in this case one where tunneling events in the two lowest bands are completely ignored whereas interactions between atoms is only considered for atoms interacting with particles in the first band .
hence , one finds , @xmath336 here the number operators @xmath337 have been introduced .
the initial state is a product of fock - states with definite numbers of particles in the first band of each well as depicted in fig .
[ fig : inv ] . here , each well @xmath14 initially has @xmath338 atoms in the first band , i.e. @xmath339 the final state is one where the population has changed such that , for a particular well , denoted by @xmath340 , one particle has decayed from the first band down to the zeroth while another atom , in order to conserve energy , ends up in a bloch state of the @xmath0:th ( @xmath341 ) band , i.e. @xmath342 this state is not an exact eigenstate of the unperturbed hamiltonian in eq .
( [ eq : unpert ] ) but an approximate one .
the correction to the bloch wave functions for @xmath341 , which will later occur in the overlap integrals , is however only of the order @xmath343 and can thus be ignored .
what is more important is the associated energy shift since this affects the position of the band edges of the @xmath344 band .
this , in turn , can have impact on the lifetime since it affects the final density of states .
hence , one can replace the two last terms in the unperturbed hamiltonian ( [ eq : unpert ] ) by a term diagonal in the band index @xmath0 ; @xmath345 here , the first order ( hartree ) shift @xmath346 in energy due to interactions between an atom in the @xmath344 bloch band and the atoms in the first band have been incorporated .
@xmath347 for the lifetime estimate in the narrow band limit .
all atoms are residing in the first band , localized in the wells .
in this particular case the filling factors are @xmath348 , @xmath331 and @xmath349 .
[ fig : inv],width=302 ] for first order decay one needs only the matrix element @xmath350 from which the rate follows ; @xmath351 to present a comprehensive numerical analysis of this decay rate is prohibitive due to the large number of parameters entering expression eq .
( [ eq : narrowdecay ] ) .
thus , for sake of illustration , we will here restrict the discussion to unit filling factor in the second band , i.e. @xmath331 .
further , we use @xmath352 which is a reasonable value from an experimental point of view .
the lattice depths in the transverse directions will be chosen slightly larger than in the @xmath2-direction , i.e. chose @xmath353 the results of the calculation , using wave functions obtained from band - structure calculations , are shown in fig . [
fig : decaynarrow ] which plots the ratio between the hopping rate and decay rate , @xmath354 .
the different solid lines correspond to different number of particles initially in the well .
the cases @xmath355 are shown , @xmath356 having the longest lifetime and @xmath357 having the shortest .
the dashed line shows the ratio @xmath358 which should be less than unity for the expression to be valid . as a comparison , the resulting lifetime obtained in the wide - band limit eq .
( [ eq : decay ] ) is also shown as the dash - dotted line . for a 1d system with filling factor @xmath331 and @xmath352 in the narrow - band limit according to eq .
( [ eq : narrowdecay ] ) .
the solid lines show the ratio between the lifetime and the time scale for hopping ( @xmath359 ) for @xmath360 particles in well @xmath340 . from top to bottom ; ( blue ) @xmath356 ; ( green ) @xmath361 ; ( red ) @xmath362 ; ( black ) @xmath357 .
the dashed line shows the ratio @xmath363 which should be less than unity for perturbation theory to be valid .
the dot - dashed line shows the result obtain by using the wide - band formula in eq .
( [ eq : decay ] ) using the same parameters.[fig : decaynarrow],width=264 ] the most interesting part of the result shown in fig .
( [ fig : decaynarrow ] ) is the sudden decay of the lifetime .
this is , as was the case in the wide - band limit , a result of the diverging density of states @xmath364 near the band edge . for lattice potentials deeper than @xmath365
, there is no phase space ( no available final energy levels for the excited particle ) , available for the first order decay . to find out the lifetime for larger values of @xmath366 ,
second order perturbation theory is needed .
consider again the same initial state @xmath367 as above .
adhering to energy conservation arguments , there are three different , mutually orthogonal , final states reachable through a second order process , @xmath368 @xmath369 @xmath370 the corresponding decay rates @xmath371 are obtained from the text - book relation @xmath372 numerical evaluations of the decay rates reveal that the dominant contribution to the total decay rate @xmath373 comes from @xmath374 .
the reason for this is easily understood ; the contribution from @xmath375 is small due to destructive interference of time reversed processes while the smallness of @xmath376 , is due to smallness of overlap integrals , which in turn can be understood from parity considerations . the decay rate @xmath374 is shown in fig .
[ fig : decaynarrow2 ] . as can be seen , it is possible to achieve lifetimes considerably larger than the inverse of the hopping energy , thus justifying the validity of the hamiltonians in eqs .
( [ eq:2dham ] ) and ( [ eq:3dham ] ) . for a 1d system with filling factor @xmath331 and @xmath352 in the narrow - band limit according to eq .
( [ eq : narrowdecay ] ) .
the solid lines show the ratio between the lifetime and the time scale for hopping ( @xmath359 ) for @xmath360 particles in well @xmath340 . from top to bottom ; ( green ) @xmath361 ; ( red ) @xmath362 .
[ fig : decaynarrow2],width=287 ]
by extending the usual mapping to the bosonic hubbard model of ultra cold atoms in an optical lattice to incorporate higher bloch bands , effective hamiltonians governing the dynamics of atoms in the first bloch band(s ) have been obtained .
these hamiltonians resemble previously studied bosonic hubbard hamiltonian but differ in two important respects : * atoms in the first excited band are labelled by three possible flavors @xmath377 .
the dynamics is such that @xmath1 particles can ( to a good approximation ) move only in the @xmath2 direction , etc .
* flavor changing collisions of atoms on the same site leading to conversion of the form @xmath378 , etc . occur .
by appropriate choices of the lattice depths in the different directions the number of flavors and the effective dimensionality ( equal to the number of flavors ) of the system can be changed . to obtain values of the relevant parameters , such as hopping energy and interaction energies , entering these effective hamiltonians
we have solved the time independent schrdinger equation ( mathieu equation ) .
the effective hamiltonians in two and three dimensions also show , apart from the usual global @xmath110 gauge symmetry , a set of @xmath12-gauge symmetries intermediate between local and global .
the ground state in the 3d ( three flavors ) case also displays a chiral symmetry breaking and an additional accidental ground state degeneracy associated with different planar chiral ordering .
the phase diagrams for two particular cases relevant for experiment have been sketched using mean field theory , indicating quantum phase transitions between mott - insulating and superfluid states . using time dependent perturbation theory up to second order in the interatomic interactions the lifetime of the atoms in the excited bands
have been estimated .
the results show that life times considerably longer ( orders of magnitude ) than relevant dynamical time scales can obtain .
this suggests that it may be possible to realize quasi - equilibrium in the subspaces of meta stable states spanned by the effective hamiltonians .
finally , we would like to stress that the mean field theory used to draw the phase diagram is only able to describe the most simple scenario with a transition from a mott - state to a superfluid state with order parameter @xmath379 .
it is well known [ ] that other multi - flavor bosonic hubbard models such as the 2-species bose - hubbard model shows a rich phase diagram with phases which can not be described in this simple approximation .
the present model , already rich at the mean - field level warrants further study .
in particular , we have pointed out potential connections to certain classes of models of frustrated spins @xcite and bose metals @xcite that also have an infinite but subextensive number of @xmath12 gauge symmetries and as a result exhibit dimensional reduction and exotic phases . with the microscopic hamiltonian developed here , these connections can and
should now be pursued in detail .
the authors wish to acknowledge m .-
cha , k. sengupta , n. read and s. sachdev for many useful discussions .
ai was in part supported by the swedish foundation for international cooperation in research and higher education ( stint ) and smg by the national science foundation through nsf dmr-0342157 .
here we provide a detailed derivation of the density distribution one expects to observe in the various phases and the different ways of measuring the density density correlations in the released cloud of atoms .
if the system is in a many body quantum state @xmath229 when the trap is released at time @xmath189 the density distribution of atoms at a later time @xmath228 is given by @xmath380 where @xmath381 is the time - evolution operator of the released system @xmath382 . to measure the quantity in eq .
( [ eq : densdistr ] ) one has to , in general , perform several measurements starting with the same trapped state @xmath229 each time .
an exception to this is when the ground state is to a good approximation a macroscopically occupied single particle state .
this is typically the case for a superfluid system and a single measurement gives a good approximation of @xmath383 . for a weakly interacting dilute gas of atoms
the interactions between atoms can be ignored during the expansion of the cloud and the time evolution operator in eq .
( [ eq : densdistr ] ) can be replaced by the free time - evolution operator @xmath384 .
expanding in the momentum components one finds @xmath385 for a system of linear size @xmath386 and for times @xmath387 the stationary phase approximation gives @xmath388 measuring the density of atoms after a long time of flight @xmath228 thus corresponds to a measurement of momentum distribution of the state @xmath229 prior to trap release . in a typical experiment one
takes an absorbtion image of the released cloud .
this means that the only the integrated column density is measured , _
i.e. _ , if an image of , say , the @xmath79 plane is taken , one measures @xmath389 in the next subsection we derive the momentum distribution @xmath390 for the superfluid states in the two- and three - flavor systems at zero temperature where @xmath12 symmetry is broken . for the two - flavor case
the system is comprised of 2d planes with uncorrelated ground states . a superfluid state of a single 2d plane can be described by a wave function with @xmath242 particles in a single state @xmath391 @xmath392 the subscripts @xmath14 and @xmath0 denote the coordinates , rows and columns , in the lattice while @xmath9 and @xmath238 are phase factors @xmath239 determining the phase of the wavefunction on a given site . to evaluate the momentum distribution we expand the field operators @xmath393 and @xmath394 in terms of the localized creation and destruction operators @xmath395 _ etc .
_ where the subscripts @xmath14 and @xmath0 respectively denote the row and column for the site on which the operator is acting . for a general state @xmath229 ( not necessarily the state in eq .
( [ eq : phistate ] ) ) we find the localized wannier orbitals @xmath397 and @xmath398 can be rewritten @xmath399 with the prefactors @xmath400 coming from the gauge choice in the inital way of writing the hamiltonian in equation [ eq : latticehamiltonian ] . carrying out the fourier integrals we find @xmath401\right.\nonumber\\ & \times&\left.\left[x_{m_2n_2}(-1)^{n_2}{\tilde{\phi}_0^x}({\bf q})+ y_{m_2n_2}(-1)^{m_2}{\tilde{\phi}_0^y}({\bf q})\right]\big|\phi\right>. \label{eq : app1}\end{aligned}\ ] ] here the position vectors @xmath402 and @xmath403 are shorthand for the lattice vectors @xmath404 and @xmath405 denote the fourier transform of the onsite wavefunctions . in the harmonic oscillator approximation
these are given by @xmath406 to evaluate @xmath407 we need to calculate the expectation values of the kind @xmath408 .
for the superfluid state @xmath229 in eq .
( [ eq : phistate ] ) it is easily verified that in terms of the in - plane density @xmath409 one gets @xmath410 _ etc .
_ hence the terms in eq .
( [ eq : app1 ] ) factors and one can write it conveniently as @xmath411 \label{eq : app3}\end{aligned}\ ] ] where we have defined @xmath412 for a system at absolute zero the phase factors @xmath9 and @xmath238 are aligned along rows and columns respectively but are randomly distributed between the lines and columns . to describe this situation
we introduce two sets of fields , @xmath413 and @xmath414 which can take on values @xmath415 .
the relation between these values of the fields and the phases along rows and columns is shown in fig .
[ fig : fields ] .
thus we can write @xmath416 consider now the summations needed to evaluate @xmath417 @xmath418 the summation over columns ( @xmath0-summation ) converges in the large @xmath175 limit to a sequence of delta - functions @xmath419 and a similar equation can be obtained for @xmath420 .
hence @xmath421 introducing @xmath422 the last summations can be rewritten @xmath423 where we have defined the random momentum distribution function @xmath424 . with the aid of eq .
( [ eq : doublesum ] ) we now deduce some properties of @xmath424 .
we begin with the magnitude of the function for any value of @xmath425 . for each nonzero value of @xmath426
the last summation is over an uncorrelated sequence of integers @xmath415 and can be viewed as a 1d random walk for which we have that @xmath427 the summation over @xmath426 contains @xmath428 terms which for each value of @xmath425 are random of magnitude @xmath429 .
this is again a random walk with @xmath428 steps and we conclude that the whole expression in eq .
( [ eq : doublesum ] ) is of order @xmath175 .
this can also be seen by noting that @xmath430 thus for each configuration @xmath413 we have a randomly oscillating function @xmath431 of unit magnitude .
an example of @xmath424 obtained for a specific realization of @xmath413 with @xmath432 is shown in fig .
[ fig : random ] . from eq .
( [ eq : doublesum ] ) it is also clear , since @xmath413 and @xmath433 are uncorrelated for nonzero @xmath426 that the average over allowed ground state configurations is @xmath434 an important property of @xmath424 is that it is even in @xmath425 @xmath435 this is a result of the nematic ordering between rows .
@xmath436 can be calculated the same way as @xmath437 and we get @xmath438 the interference term , the last part of eq .
( [ eq : app3 ] ) , for the momentum distribution vanishes . to see
this one can make use of equations ( [ eq : phasefields ] ) and ( [ eq : psi_x ] ) @xmath439\nonumber\\ & = & 4\pi^2\rho{\tilde{\phi}_0^x}\tilde{\phi}_0^y { \rm re}\left [ \sum_{mn}\delta\left(aq_x-{(2n+1)\pi}\right)e^{i{mq_ya}}\eta_{m}^x\right.\nonumber\\ & \times&\left .
i\sum_{mn}\delta\left(aq_y-{(2m+1)\pi}\right)e^{-i{nq_xa}}\eta_{n}^y\right]=0\nonumber\end{aligned}\ ] ] for the three flavor case at t=0 we consider again a state of the kind in eq .
( [ eq : phistate ] ) but with @xmath440 the subscript @xmath441 denotes collectively the @xmath2- @xmath4- and @xmath76-coordinates in the 3d lattice . as in the two flavor case , the observed momentum distribution can be written as @xmath411\nonumber\\ & + & 2{\rm re}\left[\tilde{\psi}_y({\bf q})^*\tilde{\psi}_z({\bf q})\right]\nonumber\\ & + & 2{\rm re}\left[\tilde{\psi}_z({\bf q})^*\tilde{\psi}_x({\bf q})\right ] \label{eq : app5}\end{aligned}\ ] ] with @xmath442 @xmath443 @xmath444 here the subscripts @xmath445 refer to the @xmath150 @xmath151 and @xmath446 coordinates in the lattice respectively . to see how to handle the phase factors in the three flavor case , we begin with a state without accidentally broken chiral symmetry @xmath447 where the random fields @xmath448 can again take on values @xmath415 . since the accidental chiral symmetry breaking occurs in parallel planes we can without loss of generality single out the @xmath2-direction as the direction in which planes have uniform chirality ( to compare with fig [ fig:3dphases ] make the rotation of axes @xmath449 in fig . [ fig:3dphases ] ) .
we thus introduce an additional field @xmath259 taking values @xmath415 for planes with different @xmath2-coordinate @xmath14 .
the corresponding phase factors for such a state will be @xmath450 we can now evaluate @xmath451 in the same way as for the two flavor case @xmath452 in eq .
( [ eq:2dx ] ) @xmath453 has been introduced @xmath454 the random distribution function @xmath256 is the two - variable analog of the function @xmath424 above .
an example of @xmath256 for a 40x40 lattice is shown in fig .
[ fig : f_2](a ) .
just as @xmath424 , @xmath256 obeys a sum rule @xmath455 is symmetric under inversion @xmath456 and has an average equal to unity when averaged over ground states @xmath457 the expressions for @xmath458 and @xmath458 are similar but the accidental ground state degeneracy modifies the random distribution functions .
explicitly we have @xmath459 } \eta_{m_1o_1}^y\eta_{m_2o_2}^ye^{-i\frac{2\pi}{3}(\sigma_{m_1}-\sigma_{m_2})}\nonumber\\ g_2^z&\equiv&\sum_{\stackrel{m1n_1}{m_2n_2}}e^{ia[(m_1-m_2)q_x+(n_1-n_2)q_y ] } \eta_{m_1n_1}^z\eta_{m_2n_2}^ze^{-i\frac{4\pi}{3}(\sigma_{m_1}-\sigma_{m_2})}.\nonumber\end{aligned}\ ] ] en example of the distribution function @xmath266 is shown in fig .
[ fig : f_2](b ) . note that due to the fields @xmath460 characterizing the different chirality of planes this distribution function is _ not _ symmetric under inversion .
finally we look at the interference terms in eq .
( [ eq : app5 ] ) . in the above equation
the summations over @xmath462 and @xmath463 constitute random walks .
for the @xmath462 summation this is a random walk on a line with @xmath175 unit steps @xmath415 giving rise to , for each value of @xmath464 a random term of order @xmath429 .
the sum over @xmath465 can also be viewed as a random walk for each value of @xmath466 but in the complex plane . each step being of unit length in any of the four directions @xmath467 and @xmath468 . summing over @xmath462 and @xmath465 thus yields , for each @xmath469 a random term of magnitude @xmath175 with a completely random phase .
thus the interference terms in eq .
( [ eq : app5 ] ) ( the other two terms can be treated similarly ) give rise to a three dimensional grid of lines in the released cloud where the density along any given line is randomly distributed . if the density is averaged over several shots , with different ground states we have no contribution from the interference terms since @xmath470 .
if @xmath472 is large enough for thermal fluctuations to restore @xmath12 symmetry but still small enough to preserve the distinction between the mott state and the superfluid state , measuring the density distribution alone does not suffice since the delta - peaks will be smeared .
instead correlations can be measured . to this end , assume we have a single physical system . at finite @xmath472
this system undergoes transitions in a manifold of @xmath473 states .
denote this manifold by the states @xmath474 . in a single shot a single one of these states will be probed . in an infinite series of experiments
each of these states will be probed an infinite number of times and one can thereby measure the quantity @xmath475 here we have ignored the boltzmann factors since the manifold we are looking at is nearly degenerate . in reality only a finite sequence of @xmath476 experiments can be carried out and the fluctuations in @xmath477 is of concern .
there are two sources fluctuations ; first , for each state @xmath478 there is quantum shot noise .
second , since not all of the @xmath479 states will be probed there will be deviations due to not sampling the entire distribution . if the manifold of states probed are superfluid states then @xmath480^{nd})$ ] with @xmath481 being the dimensionality of the system .
since superfluid states are to a good approximation macroscopically occupied single particle states , fluctuations due to shot noise are reduced .
the remaining fluctuations are classical and expected to scale as @xmath482 and @xmath477 should in principle be possible to measure . on the other hand , if the state measured is a mott state the manifold @xmath474 consists typically of only a few states in which case multiple measurements reduces the quantum shot noise since each quantum state will be probed many times .
we thus conclude , that by making repeated measurements and averaging the results , one can measure @xmath477 .
a quantity of interest to measure in this way is the correlation function @xmath483 again , if @xmath387 this is to a good approximation the same as @xmath484.\ ] ] the disorder - averages of the momentum density distributions are easy to calculate .
for instance , for the two flavor superfluid state in eq .
( [ eq : phistate ] ) we have @xmath485.\label{eq : app4}\end{aligned}\ ] ] the averages in eq .
( [ eq : app4 ] ) can be calculated using the representation in eqs .
( [ eq : psirep1 ] ) and ( [ eq : psirep2 ] ) @xmath486 ( and all @xmath487 ) .
it follows @xmath488 and @xmath489 hence , for the two flavor case we find @xmath490 whereas for three flavors we have @xmath491 we now turn to the evaluation of the two point correlator which we begin by normal ordering @xmath492 the normal ordered expectation value @xmath493 can be written in a form analogous to eq . ( [ eq : app1 ] ) @xmath494\left[x^\dagger_{k}(-1)^{n_k}{\tilde{\phi}_0^x}({\bf q'})^*+ y^\dagger_{k}(-1)^{m_k}{\tilde{\phi}_0^y}({\bf q'})^*\right]\right.\nonumber\\ & \times&\left.\left[x_{j}(-1)^{n_j}{\tilde{\phi}_0^x}({\bf q})+ y_{j}(-1)^{m_j}{\tilde{\phi}_0^y}({\bf q})\right]\left[x_{l}(-1)^{n_l}{\tilde{\phi}_0^x}({\bf q'})+ y_{l}(-1)^{m_l}{\tilde{\phi}_0^y}({\bf q'})\right]|\phi\right>.\label{eq : app7}\end{aligned}\ ] ] for the two flavor superfluid state in eq .
( [ eq : phistate ] ) ( a single plane with @xmath233 sites having a total of @xmath242 particles ) it is easy to verify that the expectation values of on - site operators are given by expressions of the type @xmath496 to calculate the average over disorder we have to average over @xmath497 and @xmath498 . the nonzero averages are easily seen to be @xmath499\label{eq : xxyy}\\ \overline{\left < x^\dagger_ix^\dagger_k y_j y_l\right>}&=&\frac{\rho^2}{4}\overline{\alpha_i^*\alpha_k^*\beta_j\beta_l } = -\frac{\rho^2}{4}\delta_{ik}\delta_{jl}\\ \overline{\left < y^\dagger_ix^\dagger_k y_j x_l\right>}&=&\frac{\rho^2}{4}\overline{\beta_i^*\alpha_k^*\beta_j\alpha_l } = \frac{\rho^2}{4}\delta_{ij}\delta_{kl}\\ \overline{\left < y^\dagger_ix^\dagger_k x_j y_l\right>}&=&\frac{\rho^2}{4}\overline{\beta_i^*\alpha_k^*\alpha_j\beta_l } = \frac{\rho^2}{4}\delta_{il}\delta_{kj}\\ \overline{\left < x^\dagger_iy^\dagger_k y_j x_l\right>}&=&\frac{\rho^2}{4}\overline{\alpha_i^*\beta_k^*\beta_j\alpha_l } = \frac{\rho^2}{4}\delta_{il}\delta_{kj}\\ \overline{\left < x^\dagger_iy^\dagger_k x_j y_l\right>}&=&\frac{\rho^2}{4}\overline{\alpha_i^*\beta_k^*\alpha_j\beta_l } = \frac{\rho^2}{4}\delta_{kl}\delta_{ij}\\ \overline{\left < y^\dagger_iy^\dagger_k x_j x_l\right>}&=&\frac{\rho^2}{4}\overline{\beta_i^*\beta_k^*\alpha_j\alpha_l } = -\frac{\rho^2}{4}\delta_{ik}\delta_{jl}\label{eq : yyxx}\\ \overline{\left < y^\dagger_iy^\dagger_k y_j y_l\right>}&=&\frac{\rho^2}{4}\overline{\beta_i^*\beta_k^*\beta_j\beta_l}\nonumber\\ & = & \frac{\rho^2}{4}\left[\delta_{ik}\delta_{jl}+\delta_{ij}\delta_{kl}+\delta_{il}\delta_{kj}\right]\label{eq : yyyy}.\end{aligned}\ ] ] note that eqs . ( [ eq : xxxx]),([eq : xxyy]),([eq : yyxx ] ) and ( [ eq : yyyy ] ) have contributions that correspond to pairs of particles propagating .
the disorder average of single particle propagation being zero due to the random orientation of phases . using eqs .
( [ eq : app7])-([eq : yyyy ] ) we can evaluate the terms in eq .
( [ eq : app7 ] ) which are nonzero .
we state each term contributing to the correlator in eq .
( [ eq : app7 ] ) separately using subscripts to denote the specific ordered combination of operators from which the term derives .
@xmath500\\ \overline{\left<\phi\right|\psi_{\bf q}^\dagger\psi_{\bf q'}^\dagger\psi_{\bf q}\psi_{\bf q'}\left|\phi\right>}_{xxyy}&=&-{\tilde{\phi}_0^x}({\bf q})^*{\tilde{\phi}_0^x}({\bf q'})^*{\tilde{\phi}_0^y}({\bf q}){\tilde{\phi}_0^y}({\bf q'})\frac{\rho^2}{4}\sum_{ij } e^{i({\bf q}+{\bf q}')\cdot({\bf r}_i-{\bf r}_j)}\\ \overline{\left<\phi\right|\psi_{\bf q}^\dagger\psi_{\bf q'}^\dagger\psi_{\bf q}\psi_{\bf q'}\left|\phi\right>}_{yxyx}&=&{\tilde{\phi}_0^y}({\bf q})^*{\tilde{\phi}_0^x}({\bf q'})^*{\tilde{\phi}_0^y}({\bf q}){\tilde{\phi}_0^x}({\bf q'})\frac{m^2}{4}\\ \overline{\left<\phi\right|\psi_{\bf q}^\dagger\psi_{\bf q'}^\dagger\psi_{\bf q}\psi_{\bf q'}\left|\phi\right>}_{yxxy}&=&{\tilde{\phi}_0^y}({\bf q})^*{\tilde{\phi}_0^x}({\bf q'})^*{\tilde{\phi}_0^x}({\bf q}){\tilde{\phi}_0^y}({\bf q'})\frac{\rho^2}{4}\sum_{ij } e^{i({\bf q}-{\bf q}')\cdot({\bf r}_i-{\bf r}_j)}\\ \overline{\left<\phi\right|\psi_{\bf q}^\dagger\psi_{\bf q'}^\dagger\psi_{\bf q}\psi_{\bf q'}\left|\phi\right>}_{xyyx}&=&{\tilde{\phi}_0^x}({\bf q})^*{\tilde{\phi}_0^y}({\bf q'})^*{\tilde{\phi}_0^y}({\bf q}){\tilde{\phi}_0^x}({\bf q'})\frac{\rho^2}{4}\sum_{ij } e^{i({\bf q}-{\bf q}')\cdot({\bf r}_i-{\bf r}_j)}\\ \overline{\left<\phi\right|\psi_{\bf q}^\dagger\psi_{\bf q'}^\dagger\psi_{\bf q}\psi_{\bf q'}\left|\phi\right>}_{xyxy}&=&{\tilde{\phi}_0^x}({\bf q})^*{\tilde{\phi}_0^y}({\bf q'})^*{\tilde{\phi}_0^x}({\bf q}){\tilde{\phi}_0^y}({\bf q'})\frac{m^2}{4}\\ \overline{\left<\phi\right|\psi_{\bf q}^\dagger\psi_{\bf q'}^\dagger\psi_{\bf q}\psi_{\bf q'}\left|\phi\right>}_{yyxx}&=&-{\tilde{\phi}_0^y}({\bf q})^*{\tilde{\phi}_0^y}({\bf q'})^*{\tilde{\phi}_0^x}({\bf q}){\tilde{\phi}_0^x}({\bf q'})\frac{\rho^2}{4}\sum_{ij } e^{i({\bf q}+{\bf q}')\cdot({\bf r}_i-{\bf r}_j)}\\ \overline{\left<\phi\right|\psi_{\bf q}^\dagger\psi_{\bf q'}^\dagger\psi_{\bf q}\psi_{\bf q'}\left|\phi\right>}_{yyyy}&=&|\phi_0^y({\bf q})|^2|\phi_0^y({\bf q}')|^2\frac{\rho^2}{4}\sum_{ij}\left [ 1+e^{i({\bf q}+{\bf q}')\cdot({\bf r}_i-{\bf r}_j)}+e^{i({\bf q}-{\bf q}')\cdot({\bf r}_i-{\bf r}_j)}\right].\label{eq : app8}\end{aligned}\ ] ] where the factor of proportionality is @xmath502 .
the fourier sums give , in the limit of an infinite lattice , sequences of delta functions @xmath503-{\bf g}_i)\nonumber\end{aligned}\ ] ] where @xmath504 are reciprocal lattice vectors .
the most interesting part of eq .
( [ eq : gsf2d ] ) is the second line which comes from the pair like propagation .
this can be used as a signature to detect the superfluid phase even if thermal disorder has restored the @xmath12 symmetry . for comparison
we also look at the 2d ( two flavors ) mott state . for simplicity
we consider unit filling .
there are three scenarios for the unit filling mott state that need to be considered ; ( a ) fm mott state , _
i.e. _ all atoms of the same flavor ( b ) afm mott state , x - flavor and y - flavor on alternating sites and ( c ) thermally disordered mott state with random occupation of x- and y- flavor on each site . in the ferromagnetic mott state at unit
filling @xmath269 we have two degenerate ground states @xmath505 and @xmath506 and the average in eq .
( [ eq : disorderaverage ] ) is trivial to evaluate @xmath507 the momentum correlator can be calculated using eq .
( [ eq : app7 ] ) . for the state
@xmath508 this equation reduces to there are two pairings of operators that contribute to the average @xmath510 the term @xmath511 results from having only one particle at each site but we will ignore this term ( and terms similar to it in what follows ) since its relative contribution is of order @xmath512 .
the disorder average contains only two states yielding @xmath513.\label{eq : app10b}\end{aligned}\ ] ] the correlation function for the ferromagnetic mott state is thus @xmath514 in the antiferromagetic mott state the disorder average is again over two states . dividing the 2d lattice into two sublattices @xmath515 and @xmath516 these states are @xmath517 and @xmath518 . for the momentum density we have @xmath519 ) and
has 6 nonzero contributions .
the disorder average over the two states will in this case make no difference since the two different states always give the same contribution and it is enough to consider one of them .
hence , we find for the 2d antiferromagnetic mott state at unit filling the correlation function @xmath521\sum_{i\in aj\in b } \cos\left[{({\bf q - q'})\cdot({\bf r}_i-{\bf r}_j)}\right]\nonumber\\ & + & \left(|{\tilde{\phi}_0^x}({\bf q})|^2|{\tilde{\phi}_0^x}({\bf q'})|^2+|{\tilde{\phi}_0^y}({\bf q})|^2|{\tilde{\phi}_0^y}({\bf q'})|^2\right)\sum_{i\in aj\in a } e^{i({\bf q - q'})\cdot({\bf r}_i-{\bf r}_j)}.\label{eq : gafm2d}\end{aligned}\ ] ] the fourier sums converges in the limit of large @xmath175 to @xmath522=\frac{\pi^2n^2}{2a^2}\sum_{mn}[(-1)^n+(-1)^m]\delta(q_x-\frac{n\pi}{a})\delta(q_y-\frac{n\pi}{a})\ ] ] @xmath523\delta(q_x-\frac{n\pi}{a})\delta(q_y-\frac{n\pi}{a})\ ] ] finally we look at the disordered mott state where each site holds one atom but whether it is an @xmath1 or a @xmath3 is random .
the manifold of states to average over thus contains @xmath524 states .
such a state can be written as @xmath525\left|0\right>\ ] ] where @xmath526 is a random field taking on values @xmath415 on each site @xmath46 .
the disorder averaged momentum distribution is again the same as before @xmath527 we now look at the momentum correlations in the thermally disordered three flavor superfluid phase . to evaluate the correlation function @xmath529 one can write down the extension of eq .
( [ eq : app7 ] ) .
there will be overall 81 terms in the expansion to evaluate . when taking the disorder average only 21 terms are nonzero .
note that when taking the disorder average in the three flavor model one has to average not only over all possible @xmath65 flips of the phases but also over the symmetry breaking field @xmath259 ( see eq .
( [ eq : chiralfield ] ) ) to account for the chiral symmetry breaking as well as over the 3 directions directions in which chiral symmetry is broken .
the nonzero averages one obtains in this way are shown below @xmath530\\ \overline{\left < x_i^\dagger x^\dagger_k y_j y_l\right>}&= & \overline{\left < x_i^\dagger x^\dagger_k z_j z_l\right>}=-\frac{1}{4}\frac{\rho^2}{9}\delta_{ik}\delta_{jl}\\ \overline{\left < y_i^\dagger y^\dagger_k x_j x_l\right>}&= & \overline{\left < y_i^\dagger y^\dagger_k z_j z_l\right>}=-\frac{1}{4}\frac{\rho^2}{9}\delta_{ik}\delta_{jl}\\ \overline{\left < z_i^\dagger z^\dagger_k y_j y_l\right>}&= & \overline{\left < z_i^\dagger z^\dagger_k x_j x_l\right>}=-\frac{1}{4}\frac{\rho^2}{9}\delta_{ik}\delta_{jl}\\ \overline{\left < x_i^\dagger y^\dagger_k x_j y_l\right>}&= & \overline{\left < x_i^\dagger z^\dagger_k x_j z_l\right>}=\frac{\rho^2}{9}\delta_{ij}\delta_{kl}\\ \overline{\left < y_i^\dagger x^\dagger_k y_j x_l\right>}&= & \overline{\left < y_i^\dagger z^\dagger_k y_j z_l\right>}=\frac{\rho^2}{9}\delta_{ij}\delta_{kl}\\ \overline{\left < z_i^\dagger x^\dagger_k z_j x_l\right>}&= & \overline{\left < z_i^\dagger y^\dagger_k z_j y_l\right>}=\frac{\rho^2}{9}\delta_{ij}\delta_{kl}\\ \overline{\left < x_i^\dagger y^\dagger_k y_j x_l\right>}&= & \overline{\left < x_i^\dagger z^\dagger_k z_j x_l\right>}=\frac{\rho^2}{9}\delta_{il}\delta_{kj}\\ \overline{\left < y_i^\dagger x^\dagger_k x_j y_l\right>}&= & \overline{\left < y_i^\dagger z^\dagger_k z_j y_l\right>}=\frac{\rho^2}{9}\delta_{il}\delta_{kj}\\ \overline{\left < z_i^\dagger x^\dagger_k x_j z_l\right>}&= & \overline{\left < z_i^\dagger
y^\dagger_k y_j z_l\right>}=\frac{\rho^2}{9}\delta_{il}\delta_{kj}\end{aligned}\ ] ] and the desired correlator is obtained f. s. cataliotti , _ et al .
_ , science * 293 * 843 ( 2001 ) .
d. jaksch , c. bruder , i. cirac , p. zoller , phys .
* 81 * , 3108 ( 1998 ) . c. orzel , a. k. tuchman , m. l. fenselau , m. yasuda and m. a. kasevich , science * 291 * 2386 ( 2001 ) .
fisher , p. b. weichman , g. grinstein and d. s. fisher , phys .
b. * 40 * , 546 ( 1989 ) .
s. sachdev , _ quantum phase transitions _
( cambridge university press , cambridge 1999 ) .
d. van oosten , p. van der straten , and h. t. c. stoof , phys .
a , * 63 * , 053601 ( 2001 ) .
d. van oosten , p. van der straten , and h. t. c. stoof , phys .
rev . a , * 67 * , 033606 ( 2003 ) .
g. h. chen and y. s. wu , phys .
a , * 67 * , 013606 ( 2003 ) .
a. b. kuklov and b. v. svistunov , phys .
, * 90 * , 100401 ( 2003 ) .
a. kuklov , n. prokofev and b. svistunov , phys .
lett . * 92 * , 050402 ( 2004 ) .
e. altman , w. hofstetter , e. demler and m. d. new j phys * 5 * , 113 ( 2003 ) .
l. m. duan , e. demler and m. d. lukin , phys .
lett . , * 91 * , 090402 ( 2003 ) .
a. isacsson , m. c. cha , k. sengupta , s. m. girvin , ( unpublished ) . v. w. scarola and s. das sarma , cond - mat/0503378 . cenke xu and j. e. moore , phys .
lett . * 93 * , 047003 ( 2004 ) .
cenke xu and j. e. moore , cond - mat/0405271 .
w. zwerger , j. opt .
b .- qunatum s. o. * 5 * , s9 ( 2003 ) .
k. ziegler , phys .
a * 68 * , 053602 ( 2003 ) .
k. sheshadri , h. r. krishnamurthy , r. pandit , adn t. v. ramakrishnan , europhys . lett .
* 22 * , 257 ( 1993 ) .
j. r. schrieffer and p. a. wolff , phys . rev . * 149 * , 491 ( 1966 ) .
e. altman , e. demler , and m. d. lukin , phys .
a. * 70 * , 013603 ( 2004 ) .
m. greiner , c. a. regal , j. t. stewart , and d. s. jin , cond - mat/0502411 .
s. flling , et al .
, nature * 434 * , 481 ( 2005 ) [ cond - mat/0503587 ] . | we propose that by exciting ultra cold atoms from the zeroth to the first bloch band in an optical lattice , novel multi - flavor bosonic hubbard hamiltonians can be realized in a new way . in these systems
, each flavor hops in a separate direction and on - site exchange terms allow pairwise conversion between different flavors . using band structure calculations , we determine the parameters entering these hamiltonians and derive the mean field ground state phase diagram for two effective hamiltonians ( 2d , two - flavors and 3d , three flavors ) .
further , we estimate the stability of atoms in the first band using second order perturbation theory and find lifetimes that can be considerably longer than the relevant time scale associated with inter - site hopping dynamics , suggesting that quasi - equilibrium can be achieved in these meta - stable states . |
numerous observations indicate that galaxies and clusters of galaxies have gravitational fields that are stronger at large radii than the standard theory of gravity predicts from the conjecture that light is a fair tracer of mass .
the standard interpretation of this phenomenon is that @xmath0 of the mass in the universe is contributed by particles that are dark because they do not interact electromagnetically .
the alternative hypothesis , that the standard theory of gravity fails at low accelerations , was advanced by @xcite .
subsequently @xcite proposed a modification of poisson s equation that put the proposal on a quantitative basis .
recently interest in modified newtonian dynamics ( mond ) has increased with a growing awareness that relativistically covariant theories are possible that reduce in the non - relativistic limit to the bekenstein - milgrom theory @xcite .
one way to rule out mond would be to detect the particles that are supposed to contribute most of the universe s mass , and several experiments are endeavouring to do this under the assumption that the particles have weak interactions .
another way to rule out mond would be to show that it is inconsistent with observations of the growth of cosmic structure ( e.g. * ? ? ? * ) or the dynamics and evolution of galaxies . in standard gravity , dynamical friction ( df )
is thought to play a major role in several astrophysical contexts ( e.g. * ? ? ?
ciotti & binney ( 2004 ; hereafter cb04 ) showed that df is a more potent phenomenon in mond than in standard gravity , and argued that it would cause the globular clusters of dwarf spheroidal galaxies to spiral to the centres of their hosts within a dynamical time .
however , the results in cb04 were obtained under rather restrictive assumptions and by a non - standard argument that involves the fluctuation - dissipation theorem ( see * ? ? ?
in particular , cb04 gave results for plane - parallel systems that are in the `` deep - mond '' regime .
given the prediction of cb04 that df is much more powerful in mond than in newtonian gravity , it is important to test their results with a different approach and in less specialized contexts . in this paper
we use n - body experiments to explore df in regimes in which departures from newtonian gravity are of varying magnitude .
the paper is organized as follows . in section 2
we outline our methodology explain our choice of problem .
section 3 gives details of the code and the initial conditions of the simulations .
section 4 presents the results and section 5 relates them to other recent work .
the conclusions and their astronomical implications are summarized in section 6 .
two fundamental features of mond preclude a direct extension of the standard chandrasekhar spitzer derivation of df @xcite : in mond ( i ) forces are not additive and ( ii ) all two - body orbits are bound , so individual encounters do not have a finite duration . on the other hand , there is no fundamental obstacle to direct simulations of df in mond : the drag experienced by a massive object as it moves through a swarm of less massive objects can be measured in an n - body simulation . in the classical approach to df one imagines the background swarm to be homogeneous and perpetrates the jeans swindle to neglect the mean gravitational field of the swarm . even in newtonian gravity
this step is of questionable validity , and it is inadmissible in mond because ( i ) the long - range nature of two - particle interactions in mond implies that distant encounters are liable to dominate ( in the newtonian case they nearly do ) and ( ii ) the non - linearity of the bekenstein
milgrom field equation implies that the impact of an encounter depends on the nature of the mean field .
therefore , any attempt to extend to mond the classical approach to df leads to the consideration of the drag that a massive body experiences as it moves through a self - gravitating system of finite size . in order to compare the action of df in mond and newtonian systems as cleanly as possible ,
we compare df in mond with df in the _ equivalent newtonian system _ ( ens ) ; that is the newtonian system in which the visible matter has exactly the same phase - space distribution as in the mond system @xcite . in the ens
the visible matter is enveloped in a dark - matter halo that contributes to the df experienced by a massive body .
there are two astrophysically important problems for which df is crucial : ( i ) the inward spiralling of a small system that has fallen into a larger one ( e.g. * ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , and ( ii ) the slowing of a massive bar by the material in which it is embedded ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* and references therein ) .
our mond - enabled n - body code ( n - mody ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) is better adapted for problem ( ii ) because it uses a polar grid and therefore has higher resolution at small radii than at large .
so we study the effect that mond has on the rate at which the rotation of a rigid bar is slowed by df .
with @xcite we assume that poisson s equation @xmath1 should be replaced by the non - relativistic field equation @xmath2 = 4\pi g \rho , \label{eqmond}\ ] ] where @xmath3 is the standard euclidean norm , and @xmath4 is the gravitational potential for mond ; the gravitational acceleration is @xmath5 just as the newtonian acceleration is @xmath6 . for a system of finite mass , @xmath7 as @xmath8 .
the function @xmath9 is constrained by the theory only to the extent that it must run smoothly from @xmath10 at @xmath11 ( the so - called deep - mond regime ) to @xmath12 at @xmath13 ( the newtonian regime ) , with the transition taking place at @xmath14 , i.e. , when @xmath15 is of order the characteristic acceleration @xmath16 . in the present work
we adopt @xmath17 @xcite . from poisson s equation and equation ( [ eqmond ] )
it follows that the mond and newtonian gravitational accelerations are related by @xmath18 , where @xmath19 , and @xmath20 is a solenoidal field dependent on the specific @xmath21 considered . since in general @xmath22 , standard poisson solvers can not be used to develop mond n - body codes ; equation ( [ eqmond ] ) must be solved at each time step ( brada & milgrom 1999 , nipoti et al .
2007a ) .
n - mody is a parallel , three - dimensional particle - mesh code that can be used to run either mond or newtonian simulations ( nipoti et al .
2007a , * ? ? ?
in the present study the spherical grid has 128 radial nodes , 64 nodes in colatitude @xmath23 and 128 nodes in azimuth @xmath4 , and the total number of particles is in the range @xmath24 .
we verified with convergence experiments that these numbers of particles and grid points are sufficient to exclude that our results are significantly affected by discreteness effects . in previous papers (
* nipoti et al .
2007ac ) we have used n - mody to demonstrate significant differences in the operation of violent relaxation in systems in which mond is important and in their newtonian equivalents .
we refer readers to these papers and to @xcite for details of the code and its tests .
the baryonic component of the initial conditions of the simulations is described by a spherical @xmath25-model @xcite with density distribution @xmath26 where @xmath27 is the total stellar mass , @xmath28 is the scale radius , @xmath25 is the inner logarithmic slope , and we consider the cases @xmath29 and @xmath30 @xcite . for a given model , the mond potential
is easily calculated , and to each mond model with potential @xmath4 corresponds an ens with @xmath31 , thus having a _ total _
density @xmath32 . in principle , such a distribution would have infinite mass , so we truncate it at @xmath33 . the particles of the stellar component are distributed in phase - space with the standard rejection technique , restricting for simplicity to the fully isotropic case .
the mond distribution function is obtained numerically with an eddington inversion ( e.g. * ? ? ?
* ) @xmath34 where the upper integration limit reflects the far - field logarithmic behaviour of the mond potential . in the equivalent newtonian models , at variance with nipoti et al .
( 2007c ) , we do not distinguish between dark and stellar particles , but we consider a single component distributed according to the numerical isotropic distribution function @xmath35 note that the upper limit of integration is due to the finite mass of the system ( a consequence of the density truncation at @xmath36 ) .
the physical scales of the problem are introduced as follows .
first we identify each mond initial condition by fixing a value for the dimensionless internal acceleration parameter @xmath37 , so that @xmath27 and @xmath28 are not independent quantities : in physical units , @xmath38 , where @xmath39 . the time and velocity units are @xmath40 , and @xmath41 ( nipoti et al .
the simulations are evolved up to @xmath42 ( where @xmath43 is the initial rotation period of the bar , see section [ secbar ] ) , with timestep @xmath44 .
we verified that both the mond system and its ens were in equilibrium by running them without the bar for several dynamical times .
the dark and luminous matter distributions of the considered enss are shown in fig .
[ figmlratio ] , plotting , as functions of radius , their dark - matter density @xmath45 ( top panel ) and their total ( dark - matter plus stars ) mass - to - light ratio @xmath46 , where @xmath47 is the stellar mass - to - light ratio ( bottom panel ) .
all the considered enss are dark - matter dominated , because the corresponding mond cases have internal accelerations @xmath48 .
focusing on the central regions of the enss , we note that , for fixed @xmath49 , models with cored stellar profile ( @xmath29 , dashed curves in fig .
[ figmlratio ] ) have higher @xmath50 and shallower inner dark - matter profile than models with cuspy stellar profile ( @xmath30 , solid curves in fig . [ figmlratio ] ) .
in particular , we have @xmath51 when @xmath52 and @xmath53 when @xmath54 , where @xmath55 is the inner logarithmic slope of the dark - matter density distribution .
the density distribution of the rigid bar is a prolate ellipsoid of density distribution @xmath56 where in the cartesian system co - rotating with the bar @xmath57 in all the cases here presented we use @xmath58 , @xmath59 and @xmath60 : with this choice , the length of the bar s semi - major axis is @xmath28 .
we adopted the distribution ( [ eqrhob ] ) because the gaussian dependence on @xmath61 produces a bar well defined spatially , but at the same time it avoids a density discontinuity that could produce spurious behaviour in the potential solver at the bar s edge . in all the simulations the bar rotates around the @xmath62 axis .
the lagrangian formulation can be used to derive the bar s equations of motion in an inertial cartesian frame @xmath63 such that @xmath64 : with @xmath65 the angle in the @xmath66 plane between the @xmath67 and @xmath68 axes , we have @xmath69 where @xmath70 and from equation ( [ eqrhob ] ) @xmath71 in order to start the numerical simulations smoothly , we take the density of the bar to be the product of equation ( [ eqrhob ] ) and the function @xmath72 where the growth time is @xmath73 with @xmath74 and @xmath75 is the initial angular velocity of the bar s frame . as a consequence , when the bar mass is very small , the underlying density distribution is only slightly affected by its gravitational field , and the integral at the r.h.s of equation ( [ eqphib ] ) almost vanishes . of course , in the mond simulations , on the r.h.s . of equation ( [ eqmond ] )
we have @xmath76 .
in all the presented cases we use @xmath77 , where @xmath78 is the circular velocity at @xmath28 ( i.e. at the edge of the bar ) .
for all the simulations , the inner logarithmic slope @xmath25 , the internal acceleration ratio @xmath49 , the total dark to luminous mass ratio @xmath79 , the initial angular velocity of the bar @xmath75 and the total number of particles @xmath80 are given in table 1 .
cb04 found that the df timescales in a ( deep ) mond system ( @xmath81 ) , and in the _ equivalent _ newtonian system ( @xmath82 ) are related by @xmath83 where @xmath84 with @xmath85 the modulus of the mond field and @xmath86 the modulus of the newtonian field _
generated by the baryonic distribution_. we recall again that the treatment of cb04 was carried out for a plane - parallel distribution of field particles . in the present case
the situation is more complicate , as @xmath87 is not constant in the system .
thus , for each pair of simulations with the same value of @xmath49 and @xmath25 ( and therefore with the same gravitational field ) we assume as fiducial value @xmath88 , i.e. we evaluate @xmath87 at the edge of the bar .
models with lower values of the parameter @xmath49 are in deeper - mond regime , so @xmath87 increases for decreasing @xmath49 . in particular , we considered the cases @xmath89 , @xmath90 and @xmath91 , so all the models presented have internal accelerations @xmath48 . for each simulation , fig .
[ figtomega ] plots against time the angular frequency of the bar @xmath92 ; models with constant - density cores are on the left and those with cuspy cores are on the right .
systems with smaller @xmath49 and therefore more mond / dark - matter dominated dynamics are lower down . in each panel
the full curve shows the mond system , and the dotted line its newtonian equivalent . in every case
the bar slows down more in the mond system than in its ens .
the instantaneous df timescale is @xmath93 , but in practice this quantity fluctuates strongly , so is is not a useful measure of the df timescale .
a more robust measure is the time @xmath94 for @xmath95 to reach 70% of its initial value , so @xmath96 .
figure [ figkratio ] is a plot against @xmath97 of the mond value of @xmath94 divided by @xmath94 of the ens .
this ratio decreases with @xmath49 as predicted by cb04 , and in the deep - mond regime the rate of decrease with increasing @xmath97 is consistent with that predicted by cb04 ( shown by the full line ) .
the vertical offset between the full line and the points could partly reflect the arbitrariness of our definition of @xmath94 and our assignment of values of @xmath87 to simulations .
however , we verified that for any sensible definition of @xmath87 , the experimental ratio @xmath98 is always closer to unity than cb04 predicted .
presumably this finding arises because we are considering spherical systems , while cb04 assumed plane - parallel symmetry .
@xcite used n - body simulations similar , in some respect , to those reported here to compare the dynamics of galactic bars in mond and in equivalent newtonian systems .
they found that df had a much _ smaller _ effect on bars in mond than in the newtonian cases .
this finding is not in conflict with our opposed finding for this reason : whereas our bars contain only five percent of the galaxy s baryonic mass , in the tiret & combes simulations the bars contain the majority of the baryonic mass .
consequently , there is significant background mass for the tiret & combes bars to interact with through df only when dark matter is present .
by contrast our bars interact with a background that contains 95 percent of the baryonic mass , so df is effective .
similar reasoning applies to the findings of nipoti et al .
( 2007c ) and @xcite that merging timescales are much longer in mond than in newtonian gravity with dark matter : in the newtonian case there is an abundance of dark matter to absorb energy and angular momentum from the stellar systems , while in mond the energy and angular momentum has to be absorbed by the outer parts of the stellar systems themselves .
consequently , in the newtonian case a lot of matter takes up a relatively small amount of energy per particle , and the concept of df is applicable , while in the case of mond a small amount of matter takes up a lot of energy per particle , and df is not an appropriate concept .
the prolonged merging timescale in mond is a consequence of the need to completely transform the orbits of the stars that are absorbing the original orbital energy .
our fully non - linear simulations of the mond dynamics of realistic stellar systems nicely confirm the analytical predictions of cb04 .
in particular , the simulations confirm the predicted scaling of the df timescale with the parameter @xmath97 that measures the extent to which the actual acceleration exceeds that generated by the baryons alone in newtonian gravity . consequently , the astrophysical consequences listed in cb04 are confirmed .
prominent among these was the implication of short df timescales for the existence of globular clusters in dwarf spheroidal galaxies .
even simple newtonian models of dwarf spheroidals and dwarf ellipticals predict that many observed globular clusters should spiral to the galaxy centres within a hubble time @xcite .
therefore the existence of objects such as the fornax dwarf spheroidal , which has five globular clusters but no stellar nucleus that could be the remains of clusters that have spiralled inwards , is a challenge to both newtonian gravity and mond . in the context of the standard cold - dark - matter theory ,
the survival of fornax s globular cluster against df is problematic because it can be explained only if the dark matter halo has an extended core , and not a central cusp as predicted by the theory @xcite . from the point of view of mond ,
dwarf spheroidals are in the deep - mond regime ( e.g. * ? ? ?
* ) , so the mond inspiralling time of their globular clusters is as short as the dynamical time .
snchez - salcedo et al .
( 2006 ) evaluated in detail the predictions of mond and newtonian gravity for the fate of the globular clusters of the fornax dwarf spheroidal galaxy , and from the results of cb04 concluded that fornax is much more problematic for mond than for dark matter models .
this conclusion gains strength from our confirmation of the central result of cb04 .
we thank scott tremaine for useful suggestions , and the anonymous referee for helpful comments .
some of the numerical simulations were performed at cineca , bologna , with cpu time assigned under the inaf - cineca agreement 2007/2008 .
arena s.e .
, bertin g. , 2007 , a&a , 463 , 921 bekenstein j. , 2004 , phys .
d , 70 , 3509 bekenstein j.d .
, maoz e. , 1992 , apj , 390 , 79 bekenstein j.d . , milgrom m. , 1984 , apj , 286 , 7 binney j. , tremaine s. , 2008 , galactic dynamics 2nd ed . , princeton university press , princeton bontekoe t.r . ,
van albada t.s . , 1987 , mnras , 224 , 349 brada r. , milgrom m. , 1999 , apj , 519 , 590 chandrasekhar s. , 1942 , principles of stellar dynamics , dover , new york ciotti l. , binney j. , 2004 , mnras , 351 , 285 ( cb04 ) ciotti l. , londrillo p. , nipoti c. , 2006 , apj , 640 , 741 ciotti l. , nipoti c. , londrillo p. , 2007
, in bertin g. , pozzoli r. , rom m. , and sreenivasan k.r .
, eds . , collective phenomena in macroscopic systems , world scientific , singapore , p. 177
debattista v.p . ,
sellwood j. , 1998 , apj , 493 , l5 dehnen , w. 1993 , mnras , 265 , 250 gerhard o.e . , spergel d.n .
, 1992 , apj , 397 , 38 goerdt t. , moore b. , read j.i . , stadel j. , zemp m. , 2006 , mnras , 368 , 1073 hernquist l. , 1990 , apj , 356 , 359 hernquist l. , weinberg m. , 1989 , mnras , 238 , 407 milgrom m. , 1983 , apj , 270 , 365 milgrom m. , 2001 , mnras , 326 , 1261 londrillo p. , nipoti c. , 2008 , mem .
soc . astron .
italiana supp .
, in press ( arxiv:0803.4456 ) lotz j.m .
, telford r. , ferguson h.c . ,
miller b.w . , stiavelli m. , mack j. , 2001 , apj , 552 , 572 nipoti c. , londrillo p. , ciotti l. , 2007a , apj , 660 , 256 nipoti c. , londrillo p. , zhao h.s . , ciotti l. , 2007b , mnras , 379 , 597 nipoti c. , londrillo p. , ciotti l. , 2007c , mnras , 381 , l104 snchez - salcedo f.j . , reyes - iturbide j. , hernandez x. , 2006 , mnras , 370 , 1829 sellwood j. , 2006 , apj , 637 , 567 spitzer l. , ( 1987 ) dynamical evolution of globular clusters , princeton university press , princeton tiret o. , combes f. , 2007a , a&a , 464 , 517 tiret o. , combes f. , 2007b , preprint ( arxiv:0712.1459v1 ) tremaine s.d . , 1976 ,
apj , 203 , 345 tremaine s.d .
, ostriker j.p . , spitzer l. , 1975 , apj , 196 , 407 tremaine , s. , richstone , d.o . , yong - ik , b. , dressler , a. , faber , s.m . , grillmair , c. , kormendy , j. , & laurer , t.r . 1994 , aj , 107 , 634 weinberg m. d. , 1985 , mnras , 213 , 451 zlosnik t.g . , ferreira p.g . , starkman g.d . , 2007 , preprint ( arxiv:0711.0520v1 ) | we have tested a previous analytical estimate of the dynamical friction timescale in modified newtonian dynamics ( mond ) with fully non - linear n - body simulations .
the simulations confirm that the dynamical friction timescale is significantly shorter in mond than in equivalent newtonian systems , i.e. systems with the same phase - space distribution of baryons and additional dark matter .
an apparent conflict between this result and the long timescales determined for bars to slow and mergers to be completed in previous n - body simulations of mond systems is explained .
the confirmation of the short dynamical - friction timescale in mond underlines the challenge that the fornax dwarf spheroidal poses to the viability of mond .
gravitation stellar dynamics galaxies : kinematics and dynamics |
luminous supersoft x - ray sources were discovered with the einstein observatory and were established as an important new class of x - ray binaries based on rosat observations of 18 sources in the milky way and the magellanic clouds ( greiner 1996 ; kahabka & van den heuvel 1997 and reference therein ) .
these sources have extremely soft spectra with equivalent blackbody temperatures of 15 - 80 ev and are highly luminous with bolometric luminosities of @xmath6 erg s@xmath7 . they are thought to be white dwarfs that are steadily or cyclically burning hydrogen accreted onto the surface .
remarkably , the accretion rates must be in a narrow range of @xmath8 , and the resultant luminosities are below @xmath9 erg s@xmath7 as observed .
white dwarfs with steady nuclear burning are promising type ia supernova progenitors because , unlike explosive nova events , they can retain the accreted matter and their mass can increase till it approaches the chandrasekhar limit .
more and more supersoft sources have been discovered in distant galaxies with the advent of the chandra and xmm - newton x - ray observatories .
these distant supersoft sources , as compared to the well - studied ones in the milky way and magellanic clouds , are generally hotter , more luminous and largely associated with spiral arms , suggesting they are young and massive systems ( di stefano & kong 2004 ) .
some have bolometric luminosities , as derived from blackbody models , far above the eddington luminosity ( @xmath10 erg s@xmath7 ) for white dwarfs , which we call ultraluminous supersoft sources ( uls ) .
one prototype uls is m101-ulx1 , which showed supersoft spectra of @xmath11 ev , 0.3 - 7 kev luminosities of @xmath12 erg s@xmath7 and bolometric luminosities of @xmath13 erg s@xmath7 during its 2004 outburst ( kong et al .
2004 ) .
while the supersoft spectra of ulss can be explained by a white dwarf burning accreted materials on its surface , the white dwarf model can not explain the extremely super - eddington luminosities .
on the other hand , an intermediate mass black hole ( imbh ) of @xmath14 , as kong et al .
suggested , can naturally explain both the high bolometric luminosities and the supersoft spectrum given the scaling relation @xmath15 between the black hole mass and the accretion disk temperature . in the case of m101-ulx1 , the imbh nature is further evidenced by the observed spectral state transitions , which are reminiscent of the high - soft to low - hard transitions for stellar black hole binaries .
such imbhs may have played an important role in the hierarchical merging scenario of galaxy formation ( madau & rees 2001 ) , and are under hot pursuit by workers in the fields of massive star clusters and of ultraluminous x - ray sources ( miller & colbert 2004 ) .
swartz et al . (
2002 ) observed the nearby spiral galaxy m81 with chandra acis , and discovered a bright supersoft source in the bulge .
its supersoft spectrum can be fitted by a blackbody model with a temperature of @xmath16 ev and a bolometric luminosity of @xmath17 erg s@xmath7 .
this source is thus a uls , which we designate as m81-uls1 in this paper .
given the high luminosity , swartz et al . suggested that it could be a white dwarf burning helium instead of hydrogen , or an intermediate mass black hole ( imbh ) of @xmath18 with @xmath19 as the inclination angle .
a sudden flux drop by a factor of @xmath20 , swartz et al . reported , occurred in less than one hour in the observation obsid 390 .
this resembles the ingress of an eclipsing x - ray binary with a short period of several to several tens of hours , comparable to the periods for some canonical supersoft sources ( greiner 1996 ) .
here we report the follow - up x - ray study of m81-uls1 with the 17 chandra acis observations accumulated since its discovery . in 2 we describe the reduction of the observations and the spectral and temporal analyses of uls1 .
presented in 3 are the searches for possible periods for uls1 , which fail to obtain any period .
the results are summarized and discussed in 4 .
the distance of m81 is taken to be 3.63 mpc ( @xmath21 mag ; freedman et al .
1994 ) in this work .
there have been 17 chandra observations of m81 nuclear region , where uls1 resides . as listed in table 1 ,
these include two observations ( obsid 390 , 2.4 ksec , 2000 - 03 - 21 ; obsid 735 , 50 ksec , 2002 - 05 - 07 ) analyzed by swartz et al .
( 2002 ) , and 15 monitoring observations of 10@xmath2212 ksec each from 2005 - 05 - 26 to 2005 - 07 - 06 with an average separation of 3 days ( pi : d. pooley ) .
all these observations were analyzed uniformly with ciao 3.4 .
point sources were detected with wavdetect on the individual and combined chandra images .
uls1 , at r.a.=09:55:42.2 , decl.=69:03:36.5 , appeared on the s3 chip in all observations , and is about @xmath23 away from the m81 nucleus .
its photon lists for timing and spectral analysis were extracted from the source eclipses enclosing 95% of the total photons as reported by wavdetect .
the light curve of uls1 from the 17 observations over six years is shown in figure 1 .
the count rates in these observations were corrected for deadtime and vignetting , which are usually less than 10% as listed in table 1 .
the 2005 monitoring observations caught the source in a relatively low intensity state ( @xmath24 0.01 count / sec ) on 7 occasions , a high state ( @xmath25 count / sec ) on 7 occasions , and a low to high transition in obsid 5944 .
these high states , intermittent with 2 low states , clustered in about a month and appeared as a period of super activity out of low states .
it is , however , not clear whether the low states or the high states represent the `` normal '' condition of the source due to the limited number of observations . inspection of the individual light curves shows that the flux was quite steady during most observations .
however , sudden flux drop and rise on time scales of @xmath4 sec , as shown in the figure 1 inserts , were present in obsid 390 and 5944 , resembling ingress / egress in eclipsing x - ray binaries .
in addition , the light curve in obs i d 5940 exhibited a few bins above 0.01 ct / sec before its gradual decrease to below 0.01 ct / sec , and could have caught the tail of an ingress .
the binary properties will be significantly constrained if an orbital period can be found , and the next section is devoted completely to searches for such a period .
uls1 has been persistently supersoft in all 17 observations spanning six years .
as shown in table 1 , the majority of the photons are in the soft band of 0.3 - 1.1 kev . for the 14 observations with @xmath2630 photons , the numbers of hard photons above 1.1 kev are only @xmath242% of the total photons in 12 observations ; the percentages are about 5% and 15% for obsid 5944 and obsid 390 , which happen to be the two observations that show `` eclipses '' in their light curves . while photons are expected to harden in the eclipses , 12 out of the 13 photons in the `` eclipse '' of obsid 390 ( i.e. , in the low state ) are below 1.1 kev , thus the photon energy distribution is no harder than for photons out of `` eclipse '' ( ie .
, in the high state ) . similarly for obsid 5944 , 14 out of the 15 photons in the `` eclipse '' are below 1.1 kev .
further comparison between the energy distributions for photons in the low - state observations and for photons in the high - state observations reveals no significant difference .
spectra with @xmath26200 photons are fitted to absorbed blackbody models within 0.3 - 2 kev , and the results are listed in table 1 .
the blackbody fits are quite acceptable , with temperatures clustered around 75 ev with a 11 ev standard deviation .
the x - ray luminosities in 0.3 - 2 kev are in the range of 1.8 - 2.8 @xmath27 erg s@xmath7 , and the bolometric luminosities are in the range of 1.9 - 7.2 @xmath28 erg s@xmath7 .
the neutral hydrogen column densities appear to cluster around two values , @xmath29 @xmath30 and @xmath31 @xmath30 , but there are no apparent correlations with temperatures or count rates .
the spectral parameters are best constrained with the longest observation obsid 735 .
this spectrum , as plotted in figure 2 , can be well fitted in the band of 0.3 - 2 kev by an absorbed black body model , with @xmath32 , @xmath33 @xmath30 , @xmath34 ev , @xmath35 erg s@xmath7 , and @xmath36 erg s@xmath7 .
the spectral fit shows residuals in the two bins above 1 kev , suggesting presence of a hard component barely detected by the acis observation .
note that our spectral fits are slightly different from swartz et al .
( 2002 ) because of different choices of energy bands and changes in the calibration files over the years .
like in canonical supersoft sources , the blackbody emission can come from the nuclear burning on the surface of a white dwarf .
for such a system , the surface temperature increases monotonically with the white dwarf mass as illustrated in iben ( 1982 ; figure 2 ) .
the best - fit blackbody temperature would imply a white dwarf mass of @xmath37
. the spectrum can be slightly better fitted by a multicolor disk model , with @xmath38 , @xmath39 @xmath30 , @xmath40 ev and @xmath35 erg s@xmath7 . as shown in figure 2 , the difference between the blackbody and multicolor disk models is hardly discernible below 0.8 kev . above 0.8 kev ,
the diskbb model begins to predict slightly larger count rates , by 10% at 1 kev . as in the blackbody fit
, an insignificant hard component is revealed only from the fitting residuals in two bins above 1 kev . for a thermal accretion disk ,
the black hole mass can be expressed as @xmath41 , with @xmath42 as the radiative efficiency , @xmath43 as the hardening factor , and @xmath44 as introduced to normalize the bolometric luminosity ( soria 2007 and references therein ) .
the best fit inner edge temperature and the model luminosity would imply a black hole mass @xmath45 with all factors set to their default values .
alternatively , the model normalization ( @xmath46 ; @xmath47 is the distance in 10 kpc ) gives the inner disk radius @xmath48 km ; assuming @xmath49 corresponds to the last stable orbit , this implies an imbh of @xmath50 , quite consistent with the estimate from the disk temperature .
the light curves of uls1 during obsids 390/5944 exhibited `` eclipses '' suggestive of an eclipsing binary . to search for possible periodicity from the 17 observations , we binned the individual light curves to have an average of @xmath51 counts per bin for low - state ( @xmath52 ct / sec ) observations , to have a total of 30 bins for high - state ( @xmath53 ct / sec ) observations , and adjusted the bins to exhibit the sudden flux drop / rise in obsids 390/5944 . if the variations are mainly caused by the the donor eclipsing the accretor , and the bins in the low states of obsids 390/5944 and in 7 low - state observations are in the eclipse , the low - state bins should cluster together to form only one eclipse when individual light curves are folded with the true period
were uls1 an eclipsing binary , the period should be longer than the longest observation obsid 735 ( 50 ksec ) , during which the flux was steadily at a high state of 0.07 ct / sec .
also , the period should be shorter than @xmath2450 days given the alternating low / high states during the 2005 monitoring observations .
two techniques were applied on the 17 binned light curves to search for possible periods .
the first was the lomb - scargle method ( lomb 1976 ; scargle 1982 ) , which was devised to work on unevenly spaced data as in this case .
this method calculates the lomb periodogram for a range of trial periods and quantifies the significance as the probability for the periods to originate from a purely random signal .
the lomb periodogram for uls1 is plotted in figure 3 , which shows a multitude of significant candidate periods with the probability from randomness smaller than @xmath54 .
these candidate periods cluster around @xmath55 hours and @xmath4 hours , which is about the length of the 2005 monitoring observations . to check how the time windows of the observations affect the results , we computed the lomb periodogram for monte carlo permutations of the observed light curves ( i.e. , re - assigned the count rate for one time bin to another bin randomly ) . as shown in figure 3 , the power for the monte carlo permutations is far less than for the observations , indicating that the resulted candidate periods come from the source variability rather than the time windows .
we also applied the phase dispersion minimization method as demonstrated in stellingwerf ( 1978 ) . for each trial period
, this method folds the light curves and computes the dispersion of observations in phase bins .
the phase dispersion was calculated for a period range of 15 - 1200 hours , and the result is plotted in figure 4 . a multitude of candidate periods are revealed as deep minima of @xmath56 , a measure of the phase dispersion . given the number of data points in this calculation , @xmath57 corresponds to the probability of @xmath58 for the candidate period and associated phase dispersion to come from random fluctuations .
similar to the experiments for the lomb - scargle method , we find that the phase dispersion for the monte carlo permutated data is much higher than for the observed data , indicating that the resulted candidate periods come from the source variability rather than the time windows .
the reported significance of the candidate periods , however , is only testing the probability for the power or phase dispersion to come from random fluctuations .
the small probability only negates their coming from random fluctuations , but is not always a guarantee that they come from truly periodic variations . to check the authenticity of these candidate periods , we utilize the criterion that , as naturally expected , there should be only one eclipse in the folded light curve .
eclipses are defined as dips of one or more low - state bins enclosed by high - state bins in the folded light curve .
technically , we assign 0/1 to low - state / high - state bins , and count an eclipse when a jump from 0 to 1 is detected in the folded light curve .
this eclipse counting method is conceptually similar to the string - length method ( dworetsky 1983 ) , but is computationally more efficient .
we have computed the numbers of eclipses for several tens of significant candidate periods and visually checked their folded light curves .
interestingly , none of them could give only one eclipse in their folded light curves .
plotted in figure 3 as an example is the folded light curve with four `` eclipses '' for the candidate period of 96.291 hours , the most significant from the phase dispersion minimization method .
apparently , such candidate periods are not qualified as the true orbital period for uls1 .
the failure to qualify these significant candidate periods led to the suspicion that there was no binary periodicity in these observations at all .
however , a true period might have escaped the detection of the two employed methods for some reasons , we thus tried to examine a continuous range of periods from 50 ksec to 50 days with the one - eclipse criterion in search for a true period .
first , periods should be excluded if they make a high - state light curve and a low - state light curve overlapping in phase .
assume a low state curve from @xmath59 to @xmath60 in time trails a high state curve from @xmath61 to @xmath62 .
the shortest period is @xmath63 to avoid overlapping in phase , and the maximum number of periods between the two curves would be @xmath64 . to avoid overlapping after subtracting @xmath65 periods ( @xmath66 ) from the trailing light curve
, periods should not satisfy + @xmath67 , + or @xmath68 , + or @xmath69 , + or @xmath70 . + these requirements lead to forbidden periods in ranges + @xmath71 , + and @xmath72 , + and @xmath73 , + and @xmath74 , + which can be reduced to @xmath75 . in practice , we exclude periods that satisfy @xmath76 for @xmath77 , with @xmath78 allowing for slight overlapping of low / high states .
the forbidden period ranges were computed for each combination of a high - state light curve and a low - state light curve .
the light curves for obsids 390/5944 were split into two separate light curves , a high - state one and a low - state one .
the allowed period ranges was obtained by excluding all the forbidden period ranges from the range of 50 ksec to 50 days .
the continuous intervals of allowed periods was discretized with the period resolution set to @xmath79 . under this period increment ,
the relative phase changes @xmath80 for the bins spanning @xmath81 ( about six years ) do not exceed @xmath82 , a tiny amount as compared to the width of possible eclipses .
such discretization makes the search for periodicity exhaustive , and guarantees not to miss the true period if it exists .
this exhaustive search , however , failed to find any period that gave only one eclipse in the folded light curve ; in fact , all discretized periods gave three or more eclipses as shown in figure 6 .
an assumption for the periodicity search was that all low - state bins were caused by eclipsing .
if some low - state bins were caused by reasons other than eclipsing , they might have appeared as separate eclipses in addition to the true eclipse , thus disqualifying some trial periods as the true period . since all discretized periods gave three or more ``
eclipses '' , at least two out of the nine low - state observations ( including obsid 390/5944 ) should be considered as out - of - eclipse to possibly obtain a folded light curve with only one `` eclipse '' .
the continuous period range of 50 ksec to 50 days was tested with different combinations of two low - state observations excluded , and two combinations were found to result in viable periods with only one `` eclipse '' in the folded light curves ( figure 7 ) . the first combination of obsids 5935/5944 led to viable periods around 96.7 hours , but the folded light curves showed an unreasonably wide eclipse ( @xmath83 ) .
the second combination of obsids 5940/5944 led to eight intervals of viable periods evenly spaced in the range of 30 - 35 days , reflective of the duration of the super activities ( with the two intervening low - state observations obsids 5940/5944 excluded ) in the 2005 monitoring observations .
six years of chandra acis observations have revealed the ultraluminous supersoft source m81-uls1 as persistently supersoft despite the dramatic flux changes . the highest quality spectrum for m81-uls1
can be well fitted by a multicolor disk around an imbh of @xmath84 .
the spectrum can also be well fitted by a blackbody model as for the nuclear burning on the surface of a white dwarf of @xmath37 , but the bolometric luminosity is 10 times higher than the eddington luminosity for white dwarfs steadily burning hydrogen on the surface , and 5 times higher than for white dwarfs steadily burning helium ( iben & tutukov 1989 ) . the super - eddington problem , however , may be solved by adopting more sophisticated white dwarf atmosphere models , experiments with which showed that they could reduce the bolometric luminosities by a factor of ten ( van teeseling et al .
1996 ) . to summarize , the x - ray spectrum is consistent with m81-uls1 being either an imbh of @xmath84 or a massive white dwarf .
x - ray observations thus lack the definitive power to distinguish between an imbh accretor or a white dwarf accretor , and observations in lower energies are needed to understand the nature of the accretor .
the accretion disk should behave quite differently in presence of different accretors for the same observed x - ray luminosities .
if the accretor is a white dwarf , the accretion disk , illuminated by the soft x - ray emission from the white dwarf surface , will flare up geometrically , intercept a large fraction of the x - ray emission and re - emit in lower energies ( popham & di stefano 1996 ) .
the emergent spectrum will take a form of @xmath85 in the optical ( fkr 2002 ) . if the accretor is an imbh and
thus the high x - ray luminosity is less than a few percents of the eddington luminosity , the accretion disc will be a standard geometrically - thin optically - thick one ( shafee et al .
the spectrum of such a thin thermal disk , conventionally described by a multi - color disc model ( mitsuda et al .
1984 ) , is a powerlaw of @xmath86 in the optical .
such a difference may be picked up by optical and infrared observations .
indeed , optical studies of m81-uls1 have identified m81-uls1 as a bright point - like object , the spectral energy distribution of which can be decomposed into an agb star as the companion and a blue component ( liu & di stefano 2007 ) .
the blue component is consistent with a @xmath87 powerlaw as from the accretion disk expected for an imbh , but not consistent with a @xmath88 powerlaw expected for a white dwarf .
this is strong observational evidence that m81-uls1 is an imbh instead of a white dwarf .
the light curves of uls1 exhibited in two observations sudden flux drop / rise that resembled the eclipse ingress / egress for eclipsing x - ray binaries , suggestive of periodicity due to eclipsing .
this tempted us to search for periodicity from the 17 observations for uls1 .
the candidate periods obtained with the lomb - scargle and phase dispersion minimization methods , however , are all false periods because they do not satisfy the criterion that a true period should have only one `` eclipse '' in their folded light curves . further examinations on possible periods in a continuous range of 50 ksec to 45 days , discretized in an exhaustive way , also fail to find any true period .
we conclude that no periodicity is revealed over six years observations for m81-uls1 despite the presence of apparent eclipse ingress / egress .
indeed , the optical identification of the secondary with an agb star implies an orbital period of 30 years or longer if the agb star overfills / underfills its roche lobe given its supergiant size ( @xmath89 ; liu & di stefano 2007 ) .
such a long orbital period is consistent with the absence of periodicity over six years observations . on the other hand ,
it is possible to obtain folded light curves with only one `` eclipse '' for some periods scattered in the ranges of 30 - 35 days , if we dismiss the two low - state obsids 5940/5944 .
however , there are no compelling reasons to dismiss these observations , and the resulted periods are inconsistent with the long orbit period expected given the agb secondary .
what are the origins for the dramatic flux drop / rise within an hour observed in obsids 390/5944 ?
the failure to detect any periodicity suggests that they are not eclipses from the companion star blocking the primary ; indeed , ingress / egress for an orbital period of 30 years or longer is expected to last @xmath5 days ( or @xmath90 ) , three orders of magnitudes longer than observed in obsids 390/5944 .
if they originate from eclipsing by an intervening thick cloud of gas , this gas must have a huge size orders of magnitudes larger than the x - ray emitting region ( @xmath91 km ) , unbounded by the primary , move really fast at @xmath92 km / sec , yet block the primary repetitively , and cause changes in @xmath93 ; we can not think of a reasonable scenario to produce such a cloud , and no correlated changes in @xmath93 were observed .
this variability may be reminiscent of the stochastical oscillation between high state and low state of some stellar mass black hole x - ray binaries ( e.g. , grs 1915 + 105 , cyg x-1 ) , but uls1 has never shown the accompanied hard / soft spectral state transition as observed for these sources .
were uls1 a white dwarf , such dramatic flux changes may be comparable to those observed in other supersoft sources .
for example , nova v382 vel showed a flux drop by a factor of 2 within 1.5 hours ( orio et al .
2002 ) , nova vl494 aql exhibited a supersoft flare with flux increase by a factor of 6 that lasted about 15 minutes ( drake et al .
2003 ) , and nova v4743 sagittarii showed the flux dropped to nearly zero within a few ks ( ness et al .
2003 ) . like in the case for m81-uls1
, the physical explanations are lacking for the the dramatic variability in these sources .
dramatic changes in the x - ray flux have also been observed , sometimes in anti - correlation with the optical fluxes , for canonical supersoft sources such as rx j0513.9 - 6951 ( reinsch et al .
1996 ) and cal 83 ( greiner & di stefano 2002 ) .
they are usually understood as shifts of the spectrum peak due to the white dwarf photosphere expansion and contraction as a result of accretion rate changes , accompanied by the induced changes in the irradiation and heating of the accretion disk .
the time scales for such changes , however , are a few days or longer rather than hours as observed in m81-uls1 . the short timescale variability , however , can in principle be explained by the accretion disk changes if the accretor is an intermediate mass black hole . in response to the variations in the accretion rate , the accretion disk structure and temperature can change in as short as @xmath4 seconds even if the response time is as long as @xmath94 the dynamical timescale , which is short ( @xmath95 second ) for black hole accretors . because the accretion disk temperature is rather low ( @xmath96 ev ) , the x - ray observations are only catching the exponential tail of the emergent spectrum from the accretion disk , and a small change in temperature can lead to dramatic changes in the x - ray flux .
we note that this mechanism has been used to explain the dramatic x - ray variability in narrow line seyfert i nuclei and in some canonical supersoft sources as mentioned above .
we would like to thank rosanne di stefano , jeff mcclintock , ramesh narayan and his friday group meeting for helpful discussions .
jfl acknowledges the support for this work provided by nasa through the chandra fellowship program , grant pf6 - 70043 .
di stefano , r. and kong , a. , 2004 , apj , 609 , 710 drake , j. , wagner , r. , starrfield , s. , et al .
2003 , apj , 584 , 448 dworetsky , m. , 1983 , apj , 203 , 917 frank , j. , king , a. , and raine , d , 2002 , _ accretion power in astrophysics _ , 5.10 [ fkr 2002 ] freedman , w.,hughes , s. , et al
. 1994 , apj , 427 , 628 greiner , j. 1996 , p299 in _ supersoft x - ray sources _ , ed . by greiner , j. greiner , j. & di stefano , r. , 2002 , a&a , 387 , 944 iben , i. 1982 , apj , 259 , 244 iben , i. and tutukov , a. 1989 , apj , 342 , 430 kahabka , p. and van den heuvel , e. , 1997 , ara&a , 35 , 69 kong , a. , di stefano , r. & yuan , f. , 2004 , apjl , 128 , 2783 liu , j. and di stefano
, r. , 2007 , submitted to apj letters liu , j. , di stefano , r. , mcclintock , j. , et al .
2006 , apj , 653 , 602 lomb , n.r . , 1976 , astrophysics and space sciences , 39 , 447 madau , p. and rees , m. , 2001 , apj , 551 , 27l miller , m. coleman , and colbert , e. , 2004 , ijmpd , 13 , 1 mitsuda , k. , inoue , h. , koyama , k. , et al . 1984 ,
pasj , 36 , 74 ness , j. , starrfield , s. , burwitz , v. et al .
2003 , apjl , 594 , l127 orio , m , parmar , a. greiner , j. et al .
2002 , apjl , 333 , l11 popham , r. and di stefano , r. , 1996 , p65 in _ supersoft x - ray sources _ , ed . by greiner , j. reinsch , k. , van teeseling , a. , et al .
1996 , a&a , 309 , l11 scargle , j.d . , 1982 ,
apj , 263 , 835 shafee , r. , narayan , r. & mcclintock , j. 2007 , astroph/0705.2244 soria , r. 2007 , aps&s , 311 , 213 stellingwerf , r. , 1978 , apj , 224 , 953 swartz , d. , ghosh , k. , et al .
2002 , apj , 574 , 382 van teeseling , a. , heise , j. , & kahabka , p. , 1996
, iaus , 165 , 445 acis390 & 2000 - 03 - 21 & 2.5 & 0.91 & 2.4 & 127 & 107 & 58 @xmath97 5.2 & 5.32e-13 & & & + acis735 & 2000 - 05 - 07 & 3.2 & 0.97 & 49.9 & 3182 & 3155 & 66 @xmath97 1.2 & 5.28e-13 & 73 @xmath97 1.5 & 8.6 @xmath97 0.9 & 1.196/38 + acis5935 & 2005 - 05 - 26 & 1.0 & 0.92 & 11.0 & 10 & 7 & 1.0 @xmath97 0.3 & 9.23e-15 & & & + acis5936 & 2005 - 05 - 28 & 1.0 & 0.90 & 11.4 & 10 & 9 & 1.0 @xmath97 0.3 & 9.39e-15 & & & + acis5937 & 2005 - 06 - 01 & 0.9 & 0.99 & 12.0 & 20 & 18 & 1.7 @xmath97 0.4 & 1.53e-14 & & & + acis5938 & 2005 - 06 - 03 & 0.9 & 0.99 & 11.8 & 456 & 448 & 39 @xmath97 1.8 & 3.77e-13 & 72 @xmath97 5.8 & 20.6 @xmath97 100 & 1.421/16 + acis5939 & 2005 - 06 - 06 & 0.8 & 0.99 & 11.8 & 329 & 326 & 28 @xmath97 1.6 & 2.82e-13 & 73 @xmath97 7.4 & 9.5 @xmath97 6.3 & 1.519/14 + acis5940 & 2005 - 06 - 09 & 0.8 & 0.99 & 12.0 & 65 & 63 & 5.5 @xmath97 0.7 & 5.55e-14 & & & + acis5941 & 2005 - 06 - 11 & 0.8 & 0.99 & 11.8 & 387 & 381 & 33 @xmath97 1.7 & 2.98e-13 & 78 @xmath97 5.9 & 17.6 @xmath97 6.0 & 0.602/15 + acis5942 & 2005 - 06 - 15 & 0.8 & 0.99 & 12.0 & 371 & 364 & 31 @xmath97 1.6 & 3.19e-13 & 65 @xmath97 4.4 & 17.7 @xmath97 5.8 & 0.575/15 + acis5943 & 2005 - 06 - 18 & 0.8 & 0.99 & 12.0 & 511 & 498 & 43 @xmath97 1.9 & 3.77e-13 & 87 @xmath97 5.5 & 11.2 @xmath97 4.3 & 1.129/22 + acis5944 & 2005 - 06 - 21 & 0.8 & 0.99 & 11.8 & 334 & 312 & 29 @xmath97 1.6 & 2.78e-13 & 99 @xmath97 8.9 & 19.3 @xmath97 8.2 & 0.921/15 + acis5945 & 2005 - 06 - 24 & 0.7 & 0.99 & 11.6 & 380 & 376 & 33 @xmath97 1.7 & 2.90e-13 & 65 @xmath97 4.8 & 17.8 @xmath97 6.3 & 0.810/14 + acis5946 & 2005 - 06 - 26 & 0.7 & 0.99 & 12.0 & 253 & 252 & 21 @xmath97 1.3 & 2.06e-13 & 68 @xmath97 5.8 & 9.2 @xmath97 3.9 & 0.683/10 + acis5947 & 2005 - 06 - 29 & 0.6 & 0.99 & 10.7 & 31 & 30 & 2.9 @xmath97 0.5 & 2.50e-14 & & & + acis5948 & 2005 - 07 - 03 & 0.6 & 0.99 & 12.0 & 60 & 59 & 5.1 @xmath97 0.7 & 4.80e-14 & & & + acis5949 & 2005 - 07 - 06 & 0.6 & 0.99 & 12.0 & 41 & 40 & 3.5 @xmath97 0.5 & 2.99e-14 & & & + | luminous supersoft x - ray sources found in the milky way and magellanic clouds are likely white dwarfs that steadily or cyclically burn accreted matter on their surface , which are promising type ia supernova progenitors .
observations of distant galaxies with chandra and xmm - newton have revealed supersoft sources that are generally hotter and more luminous , including some ultraluminous supersoft sources ( ulss ) that are possibly intermediate mass black holes of a few thousand solar masses . in this paper
we report our x - ray spectral and timing analysis for m81-uls1 , an ultraluminous supersoft source in the nearby spiral galaxy m81 .
m81-uls1 has been persistently supersoft in 17 chandra acis observations spanning six years , and its spectrum can be described by either a @xmath0 ev blackbody for a @xmath1 white dwarf , or a @xmath2 ev multicolor accretion disk for a @xmath3 intermediate mass black hole . in two observations , the light curves exhibited dramatic flux drop / rise on time scales of @xmath4 seconds , reminiscent of eclipse ingress / egress in eclipsing x - ray binaries .
however , the exhaustive search for periodicity in the reasonable range of 50 ksec to 50 days failed to reveal an orbital period .
the failure to reveal any periodicity is consistent with the long period ( @xmath5 yrs ) predicted for this system given the optical identification of the secondary with an asymptotic giant star . also , the eclipse - like dramatic flux changes in hours are hard to explain under the white dwarf model , but can in principle be explained by disk temperature changes induced by accretion rate variations under the intermediate mass black hole model . |
for a number of years , the @xmath0 radio galaxy 3c 356 has been considered a good test case for understanding the peculiar properties of distant powerful radio galaxies , and for using such objects to find and study distant clusters .
the peculiar properties of these objects are typified by the so - called ` alignment effect ' , i.e. the tendency for their optical and radio axes to be aligned ( mccarthy et al .
1987 ; chambers , miley & van breugel 1987 ) .
studies of 3c 356 provided one of the first examples of an alignment effect seen at near - infrared wavelengths ( eales & rawlings 1990 ) and the large ( 4.8arcsec ) separation of the two primary aligned components ( denoted ` a ' and ` b ' by eales & rawlings ) made it ideal for ground - based studies . however , the picture is complicated by the fact that both galaxies display radio emission ( fernini et al . 1993 ) and consequently 3c 356 remains one of the very few 3c objects where the site of the powerful jet - producing central engine has not yet been unambiguously identified with a specific host galaxy .
eales & rawlings ( 1990 ) proposed that the jet - producing engine was housed in component ` b ' ( brighter in both the radio and optical ) , and favoured a model in which component ` a ' was a starburst triggered by the advancing radio source .
lacy & rawlings ( 1994 ) extended this hypothesis by suggesting that components ` a ' and ` d ' ( a further aligned component located on the other side of ` b ' ) had undergone two episodes of star formation driven by interactions between gas clouds and precessing antiparallel jets .
these ideas were strongly challenged by the detection of broad emission lines in the polarized - light spectrum of 3c 356a ( cimatti et al .
1997 ) , indicating the presence of a _ bona fide _ agn in this component .
the idea that powerful radio galaxies can be used to locate distant clusters of galaxies has its basis in both theory and observations .
the luminosities of these objects imply the existence of extremely massive black holes at their centres , which in turn implies the host galaxies themselves are massive ( e.g. haenhelt & kauffmann 2000 ; laor 2001 ) , a result supported by observations of radio galaxies at all redshifts . in addition , the red colours of radio galaxies and the small scatter in the @xmath3-band hubble diagram of their hosts ( e.g. eales & rawlings 1996 ; simpson , rawlings & lacy 1999 ) imply that their stellar populations are old and consequently formed at an early epoch and , in models of structure formation , galaxies form first in the most overdense regions .
3c 356 was the first distant radio galaxy to be detected in x - rays ( crawford & fabian 1993 ) and when follow - up observations showed this emission to be diffuse , it was assumed that the radio source is embedded in a rich cluster ( crawford & fabian 1996 ) . since then , other distant 3c objects have been detected ( e.g. worrall et al .
1994 ; crawford & fabian 1995 ) , including the @xmath4 radio galaxy 3c 294 ( fabian et al .
2001 ) , currently the most distant intracluster medium detected .
statistical studies reveal an excess of faint galaxies around the @xmath5 3c radio galaxies , consistent with a richness of abell class 0 ( best 2000 ) , and therefore similar to the average found for slightly lower - redshift objects ( e.g. hill & lilly 1991 ; wold et al .
yet despite the moderate number of x - ray detections and the statistical evidence , only 3c 324 has had its location in a rich cluster confirmed ( dickinson 1997 ) , permitting a more detailed study of its environment ( e.g. kajisawa et al .
2000a , b ) .
nevertheless , it has been suggested ( bremer , fabian & crawford 1997 ) that the cluster environment is important in triggering and fuelling the radio source beyond merely providing a site for a massive elliptical galaxy capable of hosting a powerful radio source .
the interplay between powerful radio sources and their environments therefore warrants further investigation , and 3c 356 is once again a suitable test case for such a study . in this paper
, we present ( section 2 ) new deep , wide - field near - infrared images of 3c 356 and new near - infrared spectra of both the principal components .
we combine our imaging data with archival _ hst _
wfpc2 images in section 3 to investigate the clustering environment of the radio galaxy . in section 4 , we reconsider the nature of galaxies ` a ' and ` b ' . in section 5
we discuss the x - ray properties of 3c 356 and propose a model for the triggering of powerful radio galaxies , within the framework of which we attempt to explain the properties of other well - studied powerful radio galaxies . finally in section 6
we provide a brief summary of our conclusions . throughout this paper ,
we adopt @xmath6kms@xmath7mpc@xmath7 , @xmath8 , and @xmath9 . for this cosmology ,
1arcsec represents a projected distance of 8.1kpc at the redshift of 3c 356 .
deep near - infrared images at @xmath10 and @xmath11 of 3c 356 were obtained with the cooled infrared spectrograph and camera for ohs ( cisco ; motohara et al.1998 ) , mounted at the nasmyth focus of subaru telescope , on the night of ut 2000 june 17 .
cisco has a field of view of @xmath12arcmin square .
eight - point jittered mosaics were taken at each of four locations around the radio galaxy , to cover a square region slightly over 3arcmin on a side .
the data were dark - subtracted and a flatfield created for each eight - point mosaic from median - filtered , object - masked images . since the bias level of cisco does not stabilize immediately after the telescope has nodded to a new position , only the last few images at each position were used to construct the flatfield .
an attempt was made to correct for the variable bias level in the following manner .
the sky level was estimated and an appropriately scaled version of the flatfield subtracted from the image .
an estimate of the residual signal in each column was then determined and subtracted from that column , and the final image flatfielded .
the success of this method was determined by fitting a parabola to a log - histogram of the sigma - clipped pixels to estimate the noise in object - free areas of the frame , and the procedure repeated iteratively to minimize this noise .
this procedure was extremely successful for the @xmath10-band data , but failed to produce an acceptable result for the first @xmath11-band image at each position , and one - eighth of the @xmath11-band images had to be discarded .
all the images taken in each filter ( which had already been sky - subtracted by the above procedure ) were then mosaicked with bad pixel rejection , using offsets computed from the centroids of the brightest unsaturated objects .
the typical exposure times per point were 2560s at @xmath10 and 1120s at @xmath11 , while the central region received the total exposure time which was four times greater .
the fwhm of the final images were 0.42arcsec and 0.55arcsec in @xmath10 and @xmath11 , respectively .
the @xmath10-band image was aligned with and then smoothed to the poorer resolution of the @xmath11-band image .
the near - infrared images were flux calibrated using the ukirt faint standard fs 27 ( hawarden et al .
2001 ) . while fs 27 is much bluer than a @xmath1 galaxy , colour terms for cisco are negligible compared to the ukirt photometric system ( f. iwamuro , private communication ) .
the systematic uncertainties in the photometric calibration are @xmath13mag at @xmath10 and @xmath14mag at @xmath11 . with our photometric calibration ,
the @xmath15 limits in a 2-arcsec aperture are @xmath16 and @xmath17 . to ensure a similar detection threshold across the entire image
, we consider only that part of the images which received at least the exposure times listed above .
the @xmath11-band image of this field , measuring 8.436arcmin@xmath18 and centered approximately 9arcsec north of component ` a ' , is shown in fig .
[ fig : kimage ] .
we also retrieved the _ hubble space telescope _
wfpc2 f622w and f814w images presented in best , longair & rttgering ( 1997 ) from the _ hst _ data archive ( pid 1070 ; p.i .
longair ) .
the images were combined and sky - subtracted using the iraf / stsdas task _
crrej_. the _ wmosaic _ task was used to correct for geometric distortions so that a simple transformation between the coordinate frames of the _ hst _ and subaru data could be made . during the longer f622w exposure , an object moved across the field of view , leaving a bright track about 2arcsec wide .
fortuitously , this track passed through the photometric aperture of only one source , which was not affected by cosmic rays in the shorter exposure .
we were therefore able to obtain a photometric measurement for this source , although the uncertainty is much larger due to the shorter exposure time . for consistency with our near - infrared images
, we also calibrate the _ hst _ images on the vega magnitude scale , as opposed to the ab scale which is more frequently used .
the offsets for the f622w and f814w filters are 0.08 and 0.36mag , respectively , in the sense that @xmath19 . we do not correct for the small galactic extinction ( @xmath20 ; simpson et al .
1999 ) .
we obtained near - infrared spectra of components ` a ' and ` b ' on the night of ut 1999 may 13 using the cgs4 spectrograph on ukirt .
a 0.6-arcsec wide slit was used in conjunction with the 40linesmm@xmath7 grating in second order to provide a resolution of @xmath21kms@xmath7 at the redshifted wavelength of [ o iii ] @xmath225007 . after peaking up on the bright star to the south of component
` b ' , the autoguider crosshead was offset 11.3 arcsec south and 0.3 arcsec west and this star was reacquired and used for guiding during the observation .
a position angle of @xmath23 ensured that both galaxies ` a ' and ` b ' were included within the slit .
a total of 112minutes was spent on source , with spectra being taken at two positions along the slit with the usual ` abba ' technique ( e.g.eales & rawlings 1993 ) .
bad pixel masking and interleaving of the @xmath24 oversampled spectra was performed within cgs4dr and the 2-d spectrum was exported to iraf for further reduction .
the flux scale was tied to the ukirt bright standard hd 162208 .
a spatial cut of the standard star spectra revealed a fwhm of 2 pixels ( 1.2 arcsec ) , although the true seeing would be less than this due to the coarse pixel scale .
spectra were extracted along a 3-pixel ( 1.8-arcsec ) aperture .
the spectra of components ` a ' and ` b ' thus extracted are shown in fig .
[ fig : spectra ] . both components of the [ o iii ] @xmath254959,5007 doublet
are clearly detected in component ` a ' , but no line emission is seen in ` b ' . the flux of the @xmath225007 line in ` a ' is @xmath26wm@xmath27 , and it is resolved spectrally with @xmath28kms@xmath7 . assuming a similar width for the narrow line emission in 3c 356b , we can place a @xmath15 upper limit of @xmath29wm@xmath27 on the strength of this line in our extraction aperture .
we note that the [ o iii ] @xmath225007 line flux we measure for component ` a ' alone is very similar to that determined by jackson & rawlings ( 1997 ) for ` a ' and ` b ' combined .
however , jackson & rawlings spectrum was taken with 3-arcsec pixels , and therefore they had extreme difficulty in separating the line emission from the two components .
our 2-d spectrum clearly shows that all the [ o iii ] emission arises in component ` a ' .
to investigate the environment of 3c 356 , we used sextractor v2.0.21 ( bertin & arnouts 1996 ) .
we ran the object detection algorithm on the @xmath11-band image , since this filter samples a rest wavelength of @xmath30 m , and use ` two - image mode ' to obtain photometry in the other filters .
photometry was performed in 2-arcsec apertures centered on the @xmath11 barycentres .
we adopted an object detection threshold of 10 connected pixels which exceeded the local sky level by @xmath31 .
we investigated incompleteness in the standard manner of taking isolated galaxies , reducing their brightnesses , and then adding them back in to the image at random positions .
we estimated the completeness as the fraction of these sources which are recovered by the detection algorithm .
our catalogue is 100 per cent complete down to @xmath32 and drops to zero completeness by @xmath33 .
the appropriate @xmath34-correction for a present - day elliptical galaxy is @xmath35 ; we are therefore complete to @xmath36mag fainter than @xmath37 ( gardner et al . 1997 ) at @xmath38 .
alternatively , a present - day 13-gyr - old @xmath39 galaxy would have @xmath40 at @xmath0 , accounting for passive evolution according to the pegase models of fioc & rocca - volmerange ( 1997 ) , and we should therefore be able to detect galaxies as faint as @xmath41 at the redshift of 3c 356 .
since the fractional completeness is not uniform across our image at @xmath42 due to differences in exposure time , we restrict future discussion in this paper to those objects which lie above our completeness limit . of the 135 objects with @xmath43 , 74 lie within the wfpc2 field of view , and all are readily identifiable in one or both of the _ hst _ images . of these
, we classify 30 as stars based on their having a fwhm of @xmath36 pixels or being saturated and displaying diffraction spikes ; the other 44 are all resolved . to classify the remaining 61 objects not within the wfpc2 field of view , we computed the inverse second moment of the 132 unsaturated objects , @xmath44 ( kron 1980 ) , from the unsmoothed @xmath10-band image and plotted this against the @xmath45 colour in fig .
[ fig : stargal ] . known stars and galaxies are clearly separated in this figure , and we classified the other objects according to their locations .
our complete sample thus consists of 88 galaxies and 47 stars .
we note first the small number of galaxies in our field : the number counts of gardner , cowie & wainscoat ( 1993 ) predict 130 galaxies with @xmath46 ( approximately the same depth as our complete sample after accounting for aperture losses ) .
this indicates the need to image a moderately wide field to obtain an accurate field correction when attempting to detect an excess of galaxies around a high - redshift object .
however , the fraction of red galaxies is large ; elston , eisenhardt & stanford ( in preparation ) find that only 44 per cent of galaxies with @xmath47 have @xmath48 ( s. a. stanford , private communication ) , whereas we find the fraction is 56 per cent ( assuming @xmath49 ) , inconsistent at the 99 per cent level .
in addition , a plot of the radial distribution of galaxies ( fig .
[ fig : radial]a ) clearly shows a highly significant excess of objects around the radio galaxy .
we choose 3c 356b as the origin for investigating a radial excess as it has the brightest total @xmath3-band magnitude , and is therefore presumably the most massive component .
the excess of galaxies near 3c 356 is even more significant when only the reddest galaxies are considered ( fig . [ fig : radial]b ) .
we classify as ` red ' those galaxies with @xmath50 ; this is the colour of a 4.3-gyr old s0 galaxy at @xmath0 , according to the pegase models , and 4.3gyr is the elapsed time since a plausible formation redshift of @xmath51 . while the probability of 9 randomly selected galaxies all having @xmath50 is less than 1 per cent , all 9 galaxies within 20arcsec of 3c 356b are this red .
we label the objects above our completeness limit in fig .
[ fig : kimage ] with symbols according to their nature .
the dashed line is the locus of main sequence stars from the spectral library of fioc & rocca - volmerange ( 1997 ) .
the solid lines are the loci of 4.3-gyr old e and s0 galaxies for @xmath52 , also from fioc & rocca - volmerange .
the e locus is the upper of the two , and crosses are placed at redshift intervals of 0.2 . ]
we interpret the excess of galaxies around 3c 356 , which is visible in both panels of fig .
[ fig : radial ] , as part of a group or cluster around the radio galaxy .
we confirm this in fig .
[ fig : colcol ] , where it is apparent that the galaxies which are red in @xmath45 are also red in @xmath53 , and so can only be early - type galaxies at @xmath54 .
( it is also clear that our star
galaxy separation has worked well . )
we estimate the richness of the cluster using hill & lilly s ( 1991 ) @xmath55 statistic .
this is the excess of galaxies within 0.5mpc ( 61arcsec ) of the cluster centre and brighter than @xmath56 , where @xmath57 is the magnitude of the brightest cluster member .
hill & lilly discuss the advantages of this method over measurements which rely on the magnitude of the third brightest cluster member , @xmath58 ( e.g.abell 1958 ; bahcall 1981 ) . using our aperture magnitudes , component ` a '
is the brighter of the two main components , and hence @xmath59 .
although @xmath60 is therefore slightly fainter than our completeness limit , we do not apply a correction as there are only 3 galaxies with @xmath61 and we estimate the completeness in this range to be @xmath62 per cent .
although fig .
[ fig : radial ] clearly shows an excess of ( red ) objects within @xmath63kpc of the radio galaxy , this excess is diluted when considering a 0.5-mpc radius circle .
based on the mean number density at projected distances greater than 0.5mpc , we calculate that @xmath64 ( @xmath15 ) .
the shapes of the histograms in fig .
[ fig : radial ] suggest that there are unlikely to be many cluster members at radial distances greater than 0.5mpc , which would contaminate our background counts and hence cause us to underestimate the cluster richness .
we therefore conclude ( from table 4 of hill & lilly 1991 ) that 3c 356 lies in an environment poorer than abell richness class 0 , although fig .
[ fig : radial ] provides compelling evidence that it is not isolated .
this environment is perhaps somewhat poorer than the average of moderate - redshift radio - loud agn ( e.g. hill & lilly 1991 ; roche et al . 1999 ; best 2000 ; wold et al .
2000 , 2001 ) , but certainly not unusual given the large spread . however , the environment of 3c 356 represents a much poorer cluster than all 14 of the distant ( @xmath65 ) x - ray selected clusters of mcnamara et al .
( 2001 ) , which have similar x - ray luminosities , @xmath66w , to that detected by crawford & fabian ( 1993 ) .
there seems therefore to be a discrepancy between the x - ray and optical / infrared properties of the environment of 3c 356 , which we will address later .
as mentioned in section 1 there has been considerable debate concerning whether galaxy ` a ' or galaxy ` b ' hosts the jet - producing agn responsible for feeding the large - scale radio emission .
we reconsider the location of the radio source s central engine in this section , and to aid the discussion we show close - ups of the region around the two galaxies in all four filters in fig .
[ fig : closeup ] .
we first consider the morphologies of both galaxies .
radial surface brightness profiles were constructed in the @xmath11 band and fit using the two - dimensional routine of simpson , ward & wall ( 2000 ) .
the results are shown in fig .
[ fig : radprof ] , and we obtained half - light radii of 036 ( 2.9kpc ) and 161 ( 13.1kpc ) for ` a ' and ` b ' , respectively . in neither case was a significant nuclear excess required to obtain a good fit .
both galaxies lie on the kormendy relation for massive ellipticals after accounting for passive evolution in the manner of best , longair & rttgering ( 1998 ) , and , assuming no further merger activity , will evolve into @xmath67 galaxies by the present epoch . our radius for `
a ' is much smaller than the 09 determined by best et al .
( 1998 ) ; we attribute their result to their much poorer seeing and the use of a one - dimensional fitting algorithm which is extremely sensitive to errors in the sky level and companion objects , and improperly weights the data ( see simpson et al . 2000 and taylor et al .
1996 for more details ) .
the profile of ` b ' has an excess at radii @xmath68 , similar to that of a cd galaxy ; however , a visual inspection of our image clearly reveals that this excess is not azimuthally - symmetric and is caused by filamentary - like structures emanating from galaxy ` b ' . while lacy & rawlings ( 1994 ) favoured ` b ' as the source of the radio jets and suggested that ` a ' was the site of a jet
cloud interaction , later spectropolarimetric observations have revealed scattered broad line emission from ` a ' , and energetic arguments indicate that these lines originate locally ( cimatti et al .
galaxy ` a ' is therefore a true agn , whose optical polarization has its electric vector position angle accurately perpendicular to the axis of the galaxy s ` dumbbell ' structure ( best et al .
the dumbbell axis is in turn roughly aligned with the radio axis and , of course , the vector joining ` a ' and ` b ' .
together with the high optical polarization of ` a ' , these facts suggest that much of the blue emission from ` a ' is reflected from a luminous buried quasar ; this also explains the much bluer colour of object ` a ' ( @xmath69 , compared to @xmath70 for ` b ' ) .
the discovery of weak optical polarization with a broadly similar electric vector position angle in ` b ' could mean that some of this light is being scattered by material associated with ` b ' ( cimatti et al . 1997 ) .
the low excitation [ o ii ] line emission in ` b ' could also plausibly be excited by the quasar in ` a ' .
it is the nature of galaxy ` b ' which is therefore unclear , and we investigate further here .
we first briefly revisit the three arguments of lacy & rawlings ( 1994 ) to favour ` b ' as the radio galaxy .
one was based on the @xmath3-band magnitudes of the two objects . according to eales & rawlings ( 1990 ) photometry ,
` b ' lay on the @xmath3@xmath71 relation for 3c radio galaxies , while ` a ' was too faint .
more recent data indicate that the magnitudes of components ` a ' and ` b ' in standard 63.9-kpc apertures are @xmath72 and @xmath73 , respectively ( from the ukirt data of simpson et al .
1999 ; the increased accuracy of our subaru data is outweighed by the uncertainty in the @xmath74 colours ) , with a small ( @xmath75% ) amount of emission - line contamination to the broad - band magnitude of ` a ' .
both are therefore consistent with the @xmath3@xmath71 relation for powerful radio galaxies ( e.g. simpson et al .
lacy & rawlings ( 1994 ) also noted the 15 displacement of ` a ' from the line joining the radio hotspots , but this is not unusual among powerful radio galaxies ( e.g. , 3c 55 and 3c 265 from fernini et al .
1993 ) . the remaining argument of lacy & rawlings , which we study in more detail here , is the flat radio spectrum of ` b ' compared to the steep spectrum of ` a ' ( fernini et al.1993 ) .
powerful jet - producing agn have traditionally been believed to have flat - spectrum radio cores so the spectrum of ` b ' seems typical whilst that of ` a ' between 5 and 8.4ghz seems unusually steep ( @xmath76 , @xmath77 ; fernini et al .
however , athreya et al.(1997 ) studied 12 powerful high - redshift ( @xmath78 ) radio galaxies and found that three have cores at least as steep as that of galaxy ` a ' , even allowing for the slightly higher rest - frame frequencies probed by their study , this no longer seems to be a strong argument against having ` a ' as the source of the radio jets . particularly as its spectrum flattens dramatically below 5ghz ( @xmath79 ; fernini et al .
1993 ) .
having negated the arguments of lacy & rawlings ( 1994 ) against ` a ' being the core of the radio source , we note that the observed arm - length asymmetry in 3c 356 favours it .
if ` a ' is the radio galaxy , the fractional separation distance ( bahati 1980 ) is @xmath80 , while if ` b ' is the correct i d , @xmath81 . in the sample of 95 frii 3crr radio sources ( 72 radio galaxies and 23 quasars ) studied by best et al .
( 1995 ) , only 4 radio galaxies and 2 quasars have @xmath82 , while there are 14 radio galaxies and 9 quasars with @xmath83 .
we find similar numbers from the slightly larger sample of arshakian & longair ( 2000 ) .
we will discuss the arm - length asymmetry further in section [ sec : timescales ] .
we also note that our new near - infrared spectroscopy clearly places ` a ' on the [ o iii]radio luminosity correlation of jackson & rawlings ( 1997 ) , while our limit for ` b ' falls far below it and is a factor of @xmath84 below all other upper limits .
we therefore reach the opposite conclusion to eales & rawlings ( 1990 ) and lacy & rawlings ( 1994 ) , namely that ` a ' not ` b ' is more likely to be the jet - producing agn , although only a deeper radio observation revealing jets will provide a conclusive result .
if ` a ' is the core of the radio source , however , we are left with the problem of explaining the radio emission from galaxy ` b ' , which is several orders of magnitude more luminous than is typical of normal galaxies ( e.g.fabbiano et al .
a flat spectrum suggests optically - thin thermal emission from a starburst , but the brightness temperature at a rest - frame frequency , @xmath85 , of 10ghz is @xmath86k , which is too high ( cf .
condon et al . 1991 ) .
alternatively , we could be seeing the turnover in a ` normal ' starburst spectrum ( i.e. thermal plus non - thermal emission ) where the free free optical depth is unity at @xmath87ghz .
in such a case , the observed 8.4-ghz flux is approximately equal parts thermal and non - thermal emission ( equation 7 of condon et al.1991 ) , and the implied production rate of ionizing photons is therefore @xmath88s@xmath7 ( condon & yin 1990 ) .
this corresponds to a star formation rate of @xmath89yr@xmath7 ( kennicutt 1982 ) , far in excess of the limits from the h@xmath90 and submillimetre emission ( archibald et al .
a more plausible explanation is offered from the similarity between the ultraviolet optical spectrum of component ` b ' ( cimatti et al .
1997 plus our cgs4 data ) and that of the liner ngc 1052 ( gabel et al .
both are core - dominated , flat - spectrum radio sources ( wrobel 1984 ) , and ngc 1052 is an order of magnitude less luminous in both line and radio emission , suggesting that 3c 356b is simply a scaled - up version of the low - redshift object .
the implication is that the radio emission , and potentially the narrow line emission , from galaxy ` b ' is caused by a second , separate agn .
although it might seem perverse to demand the presence of a second agn , very similar systems are known to exist , such as the two flat - spectrum ellipticals near the centre of the low - redshift giant radio galaxy 3c 326 ( rawlings et al .
1990 ) , or the double - nucleus radio galaxy 3c 75 ( owen et al . 1985 ) and in the next section
we describe how they may have been synchronously triggered . our new deep near - infrared images of 3c 356 ( fig .
[ fig : closeup ] ) confirm the impression from the _ hst _ data ( e.g. best et al .
1997 ) that galaxy ` b ' is highly disturbed ; best et al . suggest that this disturbance is caused by an interaction with the southern radio jet from ` a ' .
we argue here that an attractive alternative explanation is that ` a ' and ` b ' suffered an interpenetrating collision around the time the jets from 3c 356 were first triggered .
visual inspection of fig .
[ fig : closeup ] shows that the 3c356 ` a ' and ` b ' system is very reminiscent of a pair of colliding galaxies ( see , e.g. , fig .
3 of struck 1999 ) .
a ` bridge ' can clearly be seen pointing from ` b ' towards ` a ' , and there is additional emission to the north and east of ` b ' , including an arc - like feature which is bright in the optical images and a filament arcing northwards .
component ` c ' ( detected in [ o ii ] by lacy & rawlings 1994 ) is also between the two main galaxies . in fig .
[ fig : debris ] we plot the four - colour broad - band seds of four regions of debris , plus the galaxies ` a ' and ` b ' .
these are compared to the seds ( from fioc & rocca - volmerange 1997 ) of a 4.3-gyr - old e galaxy ( a good fit to galaxy ` b ' ) combined with varying fractions of a 50-myr - old ( see section [ sec : timescales ] ) instantaneous starburst ( note that the starburst model has an essentially flat spectrum and could therefore be equally representative of scattered quasar light ) . a discrepancy between the observed @xmath45 colours and those of the models
is noted , but is hard to explain given the agreement in fig .
[ fig : colcol ] . with the exception of the very blue ( @xmath91 ) component ` c ' ,
the rest of the debris is well - modelled by a combination where only 520 per cent of the @xmath11-band light ( and hence 15 per cent of the stellar mass ) originates from young stars ( if the normalization is made to the @xmath10-band photometry , this mass fraction rises to @xmath92% for galaxy ` a ' and the region to the northeast of ` b ' ) .
it is clear , therefore , that the extended light in the system can not be due solely to recent star formation caused , for example , by the passage of the radio jets , but must be predominantly composed of old stars disturbed during the collision between ` a ' and ` b ' . while there is remarkably little discussion in the literature of what happens when two massive ellipticals interpenetrate , one important mode by which material can be ripped from a galaxy is mediated by tidal torques ( e.g. struck 1999 ) which obviously affects stars as well as gas , and so there seems to be no physical reason why the debris can not be material stripped from ` b ' ( indeed , the colours of ` b ' and the ` bridge ' are very similar ; fig . [ fig : debris ] ) .
it seems far harder to find a physical process by which the disturbed structures of ` b ' can be promoted by interactions with a radio jet .
lacy et al .
( 1998 ) present a comprehensive study of a probable jet
galaxy interaction involving the radio source 3c 441 , from which it seems clear that there are significant effects on gas in a galaxy struck by a jet , but little or no effect on old stars since there is no obvious way in which the jet power can be coupled effectively to their motions .
similarly , the red colours of component ` d ' ( which are very similar to those of ` b ' ) imply that it is an old galaxy and therefore its high velocity relative to ` a ' and ` b ' can not be explained by an interaction with the radio jet in the manner proposed by lacy & rawlings ( 1994 ) .
neither can it be explained as a result of a strong gravitational interaction with the two main galaxies , since the maximum velocity which can be attained in such a scenario is @xmath93kms@xmath7 , where @xmath94 is the combined mass of components ` a ' and ` b ' , and @xmath95 is the impact parameter .
neither does component ` d ' appear to be disturbed , although it does possess weak [ o ii ] emission .
we are therefore left with the rather unsatisfying conclusion that it is an unrelated background galaxy . adopting the physical model for frii radio sources of willott et al.(1999 )
, we can use the observed properties of 3c 356 to estimate its total jet power @xmath96 and the time @xmath97 since the jets were first triggered . if the jet axis makes an angle @xmath98 with respect to the line of sight , then equation ( 11 ) of willott et al .
yields @xmath99w , where we have adopted the following ` canonical ' values for the source and environmental parameters discussed by willott et al . : @xmath100 and @xmath101 ( see also blundell & rawlings 2000 ) ; @xmath102 and @xmath103 . equation ( 9 ) of willott et al . gives the age of the radio source as @xmath104yr .
so , if the merger axis of ` a ' and ` b ' is close in three dimensions to the jet axis ( e.g. west 1991 ; see also the discussion in the next section ) , then the point of closest approach happened @xmath105yr ago , where @xmath106kpc is the projected separation between ` a ' and ` b ' , and @xmath107kms@xmath7 is the line - of - sight velocity difference .
thus , accounting for uncertainties like small ( @xmath108 ) misalignments between the radio axis and the velocity vector of ` a ' , it looks highly plausible that the jets were triggered during the interpenetrating collision between ` a ' and ` b ' , a few tens of myr in the past .
this simultaneity requires a pre - existing supermassive black hole at the centre of ` a ' , in line with the suggestion of willott , rawlings & blundell ( 2001 ) that @xmath109 radio galaxies are undergoing a rebirth . since the fractional separation difference can not be larger than the expansion speed of the outer radio components , @xmath110 ( see blundell & rawlings 2000 for a recent discussion of expansion speeds )
, the large arm - length asymmetries can not be due to light travel time effects , and the most likely cause is an environmental asymmetry .
if so , the model of willott et al .
( 1999 ) implies that the mean density to the southeast of ` a ' is @xmath111 times higher than that to the northwest .
radio depolarization measurements ( pedelty et al . 1989 ) also indicate a denser medium to the southeast , and visual inspection of fig .
[ fig : kimage ] suggests that this may be linked to a larger concentration of companion galaxies to the southeast , presumably associated with the group containing ` b ' .
our study of 3c 356 has led us to a model in which a powerful jet - producing agn and a weaker , but possibly also jet - producing , agn are stimulated in the 3c 356 system as the result of an interpenetrating collision between two massive galaxies , which occurred along an axis closely aligned with the radio jet axis .
the high relative velocity of ` a ' and ` b ' indicate that they are not members of the same group , while the small number of companion galaxies suggests that the system will relax to a relatively poor cluster , no richer than abell class 0 .
this begs two questions which we address in this section .
first , how do we square these inferences with the x - ray properties of this system , which are more reminiscent of a richer cluster of galaxies ? and second , can the trigger mechanism inferred for 3c 356 be generalized to other objects with powerful radio jets ?
the _ rosat _ images of 3c 356 reveal the x - ray emission to be extended ; while there are no significant sources in the hri image ( crawford & fabian 1996 ) , the pspc image ( crawford & fabian 1993 ) tentatively suggests that the emission peaks to the north of the radio core and is extended in the direction of the radio jets .
the bolometric x - ray luminosity of 3c 356 is @xmath112w ( crawford & fabian 1993 , corrected to our cosmology ) , making it lie close to the luminosity temperature relation for nearby clusters ( fig .
[ fig : lxtx ] ) .
however , as we have seen in section [ sec : richness ] , the small number of galaxies around 3c 356 does not support the existence of a cluster with the richness implied by the x - ray temperature .
we consider the possibility that the x - ray emission from 3c 356 is a superposition of cooler , less luminous ( i.e. , consistent with the poorness of the optical / infrared environment ) intracluster gas with an attenuated power law from the agn in galaxy ` a ' .
crawford & fabian ( 1996 ) estimate that no more than one - third of the observed soft x - ray emission can arise from a point source , and this would reduce the bolometric luminosity of the thermal component ( which becomes cooler due to the addition of the hard non - thermal emission ) by a factor of @xmath12 , insufficient to invalidate the arguments made at the end of section [ sec : richness ] . )
clusters from donahue et al .
( 1999 ) , della ceca et al .
( 2000 ) , schindler ( 1999 ) , and fabian et al .
the star represents 3c 356 and the open circle 3c 294 .
the solid line is the relationship for nearby clusters from wu , xue & fang ( 1999 ) . ]
we therefore turn to a scenario of merging subclusters , such as that proposed by markevitch , sarazin & vikhlinin ( 1999 ) for the cluster around cygnus a. suggestions that radio galaxy activity and cluster mergers are related have also been advanced by burns et al .
( 1993 , 1994 ) .
markevitch et al.s model consists of two hot , massive clusters in the early stages of a merger , whereas the above arguments suggest that 3c 356 represents the late stages of a merger between two much less massive groups .
typical properties for such groups are @xmath113m@xmath114 , @xmath1152kev , and core radius @xmath116kpc ( e.g. kriss , cioffi & canizares 1983 ; mulchaey & zabludoff 1998 ) .
when two clusters , each of mass @xmath117 , fall from a large distance to a separation of @xmath118 ( which can be considered as the start of the merger ) , their relative velocity is @xmath119kms@xmath7 ( i.e. very similar to the current line - of - sight velocity difference between the two main galaxies ) . from equations 1 and 2 of markevitch
et al . ( 1999 ) , this results in post - shock gas with twice the temperature and slightly more than twice the density of the pre - shock material , and hence almost an order of magnitude higher emissivity ( these numbers are insensitive to the exact value of the initial gas temperature ) .
for a merger as advanced as that of 3c 356 , where the two central galaxies have already suffered an interpenetrating collision , it is reasonable to assume that approximately half the gas from each group has passed through the shock front , and hence the x - ray luminosity of 3c 356 will be about an order of magnitude more than that of the original galaxy groups .
since emission from thermal bremsstrahlung has a bolometric luminosity @xmath120 , the merger causes 3c 356 to simply move up the luminosity temperature relation , @xmath121 ( wu et al .
1999 ) , and its x - ray properties will therefore resemble those of a single , richer cluster ( this statement is true for a wide range of merger parameters ) .
the post - shock gas has a cooling time - scale , @xmath122yr , much longer than the shock crossing time , @xmath123yr , and it will therefore dominate the observed x - ray emission due to its much higher surface brightness .
higher - quality x - ray data could either spatially or spectrally separate the pre- and post - shock material and are obviously desirable .
this model can also explain the alignment effect in 3c 356 , as well as the tentative alignment between the x - ray - emitting gas and the radio axis .
these alignments are all secondary and arise because the radio emission , x - ray emission , and galaxy interaction direction are all aligned with the direction of the cluster merger .
the x - ray
merger alignment is the easiest to understand , since the x - rays arise from the gas travelling in the direction of the merger which has passed through the shock front .
the line joining ` a ' and ` b ' will also be aligned with this direction since the collision velocity between the two merging groups is larger than the velocity dispersion of either group and so the galaxies will approach each other along a vector which is biased towards the merger axis .
to explain the alignment of the radio jets , we turn to west ( 1994 ) , who suggested that supermassive black holes form with their spin axes aligned with the local large scale structure , as a result of a series of highly anisotropic mergers . since the interaction between ` a ' and
` b ' is unlikely to have caused any precession of the spin axis of the jet - producing black hole in ` a ' ( scheuer 1992 ) , when this dormant black hole is re - fuelled ( see section [ sec : timescales ] ) , the radio jets will align with the filament of large scale structure , along which the groups containing ` a ' and ` b ' are presumably merging .
we also consider the possibility that the x - ray emission is boosted in a more direct way by the radio source activity .
first , as fabian et al.(2001 ) note in the case of the @xmath4 radio galaxy 3c 294 , inverse - compton scattering of cmb photons by low-@xmath124 electrons can produce luminous x - rays .
this mechanism requires bubbles of electrons which are spatially separated from the ultra - relativistic material visible as high - frequency radio emission , but studies of m 87 ( owen , eilek & kassim 2000 ) show that this is plausible , at least in this object .
in addition , low-@xmath124 electrons are observed to be distributed on large scales in merging systems where it is plausible that they are re - energized by shocks which would extend their effective lifetimes ( rttgering et al.1997 ) . as also noted by fabian et al .
( 2001 ) , this scenario requires two bubbles of hot confining gas , which can be naturally associated with the merging subclusters in our model , and therefore also predicts an alignment between the x - ray and radio axes .
second , kaiser & alexander ( 1999 ) suggested that the x - ray emission from 3c 356 may arise from gas in the intergalactic medium which has been heated by the passage of the expanding radio source .
however , their fig .
10 indicates that the required luminosity of @xmath125w can not be achieved for a source with a mean arm length of @xmath126kpc unless the density profile index @xmath127 ( we know of no sources where such a flat profile has been observed ) and/or the igm density is larger than their assumed value , which is typical of 3c radio source environments at @xmath2 ( this same value is justified by willott et al . 1999 ) . since the environment of
3c 356 i somewhat poorer than this , a higher density is inconsistent with the observations . also , kaiser & alexander s model requires a higher density on the north side of the radio source since the x - ray emission peaks to the north of the core ( crawford & fabian 1993 ) , which is at odds with the longer arm length if the igm is static .
our model naturally explains this apparent discrepancy if the group approaching from the south were denser ( as we claim to be the case in section [ sec : timescales ] ) .
this gas will be the most luminous in x - rays after it has been compressed and heated by passing through the shock front to the north side .
on the other hand , the radio hotspots advance much more rapidly than the slow ( @xmath128kms@xmath7 ) igm and therefore work against the pre - shock material , which is denser to the south .
it seems plausible that some x - ray emission may arise from processes directly associated with the radio source , but a full consideration of all its observed properties seem to demand that 3c 356 is at the centre of a merger between two galaxy groups .
the time - scales involved suggest that the radio source was triggered by the close approach , or interpenetration , of the dominant galaxies from each group , which channelled fuel onto a dormant supermassive black hole .
our observations and model are at odds with the idea of bremer et al .
( 1997 ) that powerful radio galaxies lie at the centres of rich clusters with strong cooling flows .
however , the richness of the cluster around 3c 356 is fairly typical of powerful radio sources at high redshift , and it is therefore worth investigating whether our model for 3c 356 can be applied to other radio - loud agn .
the radio galaxy 3c 294 has recently been observed at high resolution in x - rays with _
( fabian et al . 2001 ) and at near - infrared wavelengths with adaptive optics on the canada
hawaii and keck telescopes ( stockton , canalizo & ridgway 1999 ; quirrenbach et al.2001 ) .
the excellent quality of these data make it suitable for testing the generality of our model .
quirrenbach et al.s observations reveal two galaxies separated by @xmath129kpc , the brighter of which has a diffuse , clumpy morphology , while the fainter is compact and is identified with the source of the radio jets . applying the model of willott et al .
( 1999 ) to 3c 294 , we derive a jet power @xmath130w , and an age of @xmath131yr .
this much younger age is consistent with the smaller separation of the two optical components in 3c 294 , and the ( rest - frame ) optical properties of this system are then completely analogous to those of 3c 356 at an earlier epoch .
it is also worth noting that in both 3c 356 and 3c 294 , the radio source is identified with the more compact object , while it is the other galaxy which is larger and more luminous , and apparently more like a brightest cluster galaxy .
this appears to be at odds with best et al.s ( 1998 ) suggestion that @xmath5 radio galaxies are ( the progenitors of ) cd galaxies unless the galaxy systems are always destined to merge .
the x - ray observations of 3c 294 also support a model of merging clusters .
the high resolution of _ chandra _ reveals double - lobed ( ` hour - glass ' shaped ) emission , not typical of a relaxed rich cluster , and more reminiscent of @xmath132-body simulations of merging clusters ( roettiger , burns & loken 1993 ) .
our interpretation is that we are observing gas which has been heated by two shocks propagating in opposite directions from the site of the collision . as discussed above , this produces hot , bright x - ray emission along the merger direction , surrounded by cooler , fainter , unshocked gas from the original clusters . as with 3c 356 , a collision velocity of @xmath1332000kms@xmath7 could raise the temperature from @xmath12kev to the observed 5kev , and would result in shocks propagating at @xmath134kms@xmath7 relative to the collision site .
if the two components observed by quirrenbach et al . ( 2001 ) were at the centres of the two groups , each shock would have travelled a distance of @xmath135kpc , consistent with the extent of the x - ray lobes . since the radio source is not triggered until much later
, it is possible for the shocked gas to have a larger extent than the more rapidly expanding radio lobes .
again , the major axis of the x - ray emission is broadly aligned with the radio axis , as predicted in our model ( see section [ sec : xray ] ) .
the lack of alignment between the radio axis and the line joining the two bright galaxies can be explained by the youth of the radio source , since we are seeing the two components soon after their time of closest approach .
we use the x - ray data of 3c 294 to draw two inferences which are relevant in our search for a single trigger mechanism for powerful radio sources at these intermediate epochs .
first , there is no evidence for an enhancement of the x - ray emission at the edge of the radio cocoon , as predicted by kaiser & alexander s ( 1999 ) model for heating of the igm by the radio source .
second , the highly - peaked x - ray surface brightness which is the signature of a cooling flow core is not present ( the weak x - ray core coincident with the radio galaxy is consistent with a heavily - absorbed power law ) , and therefore we do not believe that 3c 294 lies at the centre of an established cooling flow .
our model of merging clusters therefore provides the only way to explain the properties of both 3c 356 and 3c 294 .
our model may also be applicable to the most luminous @xmath136 radio source , cygnus a. cyg a is the brightest galaxy in a cluster which is merging with a richer cluster ( owen et al .
1997 ; markevitch et al .
1999 ) along an axis aligned with the radio jets .
this is a qualitatively similar situation to that we have proposed for 3c 356 and 3c 294 , and west ( 1994 ) noted this alignment as support for his model .
the quantitative differences can be easily explained .
first , the much richer environment is an inevitable consequence of hierarchical merging ( eke , navarro & frenk 1998 ) , since more massive clusters will have had time to form by the later epoch .
second , the earlier stage of the cluster merger at which the radio source is created can be explained by the higher galaxy density , since it is a galaxy galaxy interaction which provides the trigger .
while there is only tentative evidence for a galaxy merger in cyg a ( see stockton , ridgway & lilly 1994 ) , there is recent star formation ( jackson , tadhunter & sparks 1998 ) whose ( admittedly highly uncertain ) age is consistent with the age of the radio source ( @xmath137myr ) .
although very violent interactions are responsible for triggering the radio sources 3c 356 and 3c 294 , the weaker interaction in the case of cyg a may still be sufficient to trigger nuclear activity , since fuel is likely to be plentiful as a result of the massive cooling flow ( @xmath138yr@xmath7 ; reynolds & fabian 1996 ) . unlike at @xmath5 , where the interaction must provide fuel _ and _ drive it into the nucleus of the active galaxy , cyg a only requires a relatively small gravitational perturbation to disturb the reservoir of gas which is already present .
we note that the weaker , yet still extremely powerful , radio source 3c 295 also lies at the centre of a massive cooling flow ( allen et al .
2001 ) and appears to be undergoing an interaction ( smail et al .
1997 ) which could provide fuel to power the nuclear accretion .
we therefore offer support for the model of bremer et al .
( 1997 ) when applied to low - redshift sources where cooling flows have been set up , and suggest that the unspecified trigger mechanism is a galaxy galaxy interaction .
a powerful radio source requires ( i ) a supermassive , presumably spinning , black hole ; ( ii ) available fuel ; and ( iii ) a mechanism for delivering this fuel to an accretion disc around the black hole .
the formation of the black hole is likely to be accompanied by a period of agn activity , as fuel is plentiful .
this is believed to be the evolutionary state of the highest - redshift ( @xmath139 ) radio sources .
since black holes are not destroyed , lower - redshift radio sources require only fuel and a suitable delivery mechanism to be triggered in a suitable host galaxy . at intermediate redshifts ( @xmath109 ) ,
elliptical galaxies do not have large internal reservoirs of gas , and our study of 3c 356 and 3c 294 suggests that fuel can be made available and delivered by a violent galaxy galaxy interaction which is most likely to be orchestrated by the merger of two groups or clusters of galaxies .
we suggest that this is likely to be a very important , potentially dominant , trigger mechanism for powerful radio sources at these cosmological epochs . at even lower redshifts ( @xmath140 )
, fuel may become readily available from a cooling flow ( bremer et al .
1997 ) , and therefore a less violent trigger is required to perturb the gas onto the black hole . the steep decline in the number density of extragalactic radio sources from @xmath5 to the present epoch ( e.g. dunlop & peacock 1990 ; willott et al .
2001 ) can be understood in terms of the currently - favoured low matter - density universe as the end of the era of cluster formation via group
group mergers , and hence the end of the epoch of violent interactions which can fuel dormant black holes ( see also blundell & rawlings 1999 ) .
it is worth pointing out that this is precisely the opposite conclusion to that reached by fabian & crawford ( 1990 ) , who suggested that cluster mergers _ turned off _
quasars since it was then believed that @xmath141 and hence massive clusters formed at a much later epoch . our model
, if applicable to intermediate - redshift radio sources in general , has an important implication for the use of radio sources to locate clusters at @xmath54 , as it predicts that these radio - selected clusters are likely to be merging systems .
these systems are atypical and can not be used to infer details about the evolution of the cluster x - ray luminosity function .
however , proving that these systems are not simply relaxed clusters is likely to be difficult since the x - ray luminosity and temperature of these systems mimic those of a single , richer cluster ( see section [ sec : xray ] ) .
in addition , while the merger directions of radio galaxies lie close to the plane of the sky and therefore should produce elongated x - ray emission , radio - loud quasars lie in systems which are merging along axes close to the line of sight , and projection effects will make it much harder to observe evidence of a merger .
even some radio galaxies may be produced by mergers close to the line of sight , due to a combination of misalignments between the axes cluster and galaxy mergers , and between the axes of the galaxy merger and the radio emission .
we have demonstrated with new deep near - infrared imaging the absence of a rich cluster around the @xmath0 radio galaxy 3c 356 , apparently at odds with the observed x - ray emission .
we have proposed that the x - rays arise from shocked gas at the interface of two merging galaxy groups and that the two components of 3c 356 suffered a strong interaction during this merger , stripping material from one or both galaxies and funnelling it into the nucleus of ` a ' , which we believe is the source of the radio jets ( although we believe that ` b ' is also an agn ) .
we have proposed an identical scenario for the @xmath4 radio galaxy 3c 294 , and have suggested that this mechanism may be the dominant trigger for radio sources at @xmath142 . we have suggested that less violent galaxy interactions may trigger lower - redshift radio sources where a cooling flow has set up a reservoir of fuel in the host galaxy .
this model explains the rapid decline in the number density of these objects since @xmath5 , and also implies that galaxy clusters located by using radio sources will be an extremely biased sample inappropriate for learning about cluster formation and evolution .
it is vital to establish the ways in which the radio source can directly influence the observed x - ray properties , and also to understand how the triggering of a radio source may be intimately related to the evolution of its environment .
we wish to thank roberto de propris for giving up valuable observing time to obtain the cgs4 spectra , and adam stanford and curtis struck for useful discussions .
we are also grateful to katherine blundell for providing the radio map in fig . 1 , and the referee , paul alexander , for his comments .
this work is based in part on data collected at subaru telescope , which is operated by the national astronomical observatory of japan , and in part on observations with the nasa / esa _ hubble space telescope _
, obtained from the data archive at the space telescope science institute , which is operated by aura , inc . under nasa contract nas526555 .
the united kingdom infrared telescope is operated by the joint astronomy centre on behalf of the u. k. particle physics and astronomy research council . | we present deep near - infrared images , taken with subaru telescope , of the region around the @xmath0 radio source 3c 356 which show it to be associated with a poor cluster of galaxies .
we discuss evidence that this cluster comprises two subclusters traced by the two galaxies previously proposed as identifications for 3c 356 , which both seem to harbour agn , and which have the disturbed morphologies expected if they underwent an interpenetrating collision at the time the radio jets were triggered .
we explain the high luminosity and temperature of the diffuse x - ray emission from this system as the result of shock - heating of intracluster gas by the merger of two galaxy groups . taken together with the results on other well - studied powerful radio sources , we suggest that the key ingredient for triggering a powerful radio source , at least at epochs corresponding to @xmath1 , is a galaxy galaxy interaction which can be orchestrated by the merger of their parent subclusters .
this provides an explanation for the rapid decline in the number density of powerful radio sources since @xmath2 .
we argue that attempts to use distant radio - selected clusters to trace the formation and evolution of the general cluster population must address ways in which x - ray properties can be influenced by the radio source , both directly , by mechanisms such as inverse - compton scattering , and indirectly , by the fact that the radio source may be preferentially triggered at a specific time during the formation of the cluster .
galaxies : active galaxies : clusters : general galaxies : individual ( 3c 356 ) galaxies : interactions galaxies : photometry infrared : galaxies |
the work to be presented in this article lies at the boundary between physics and electrical engineering and aims at the experimental implementation of practical hardware technology for quantum computation @xcite .
at this point in time , one of the main objectives in the quantum computing community is to build and prototype tools for scalable architectures that may lead to the realization of a universal quantum computer @xcite . in this project
, we undertake the task of implementing an extensible wiring method for the operation of a quantum processor based on solid - state devices , e.g. , superconducting qubits @xcite .
possible experimental solutions based on wafer bonding techniques @xcite or coaxial through - silicon vias @xcite as well as theoretical proposals @xcite have recently addressed the wiring issue , highlighting it as a priority for quantum computing .
building a universal quantum computer @xcite will make it possible to execute quantum algorithms @xcite , which would have profound implications on scientific research and society . for a quantum computer to be competitive with the most advanced classical computer
, it is widely believed that the qubit operations will require error rates on the order of @xmath2 or less .
achieving such error rates is only possible by means of quantum error correction ( qec ) algorithms @xcite , which allow for the implementation of fault tolerant operations between logical qubits .
a logical qubit is realized as an ensemble of a large number of physical qubits ( on the order @xmath3 or larger ) , where each physical qubit behaves as an effective quantum - mechanical two - level system . among various qec algorithms ,
the most practical at present is the surface code algorithm @xcite .
the surface code requires only nearest - neighbor interactions between physical qubits on a two - dimensional lattice , one- and two - qubit physical gates , and qubit measurement with error rates below approximately @xmath4 @xcite .
both fast gates and measurements are indispensable to run quantum algorithms efficiently ; execution times on the order of tens of nanoseconds for physical operations ( e.g. , a gate or measurement operation ) are highly desirable and represent the state - of - the - art in qubit technology today . under these conditions , to factorize a @xmath5-bit number using shor s algorithm @xcite requires more than @xmath6 million physical qubits ( i.e. , approximately @xmath7 logical qubits storing meaningful data ; note that these logical qubits occupy less than of the quantum computer , the remainder of which is used to generate special states facilitating computation ) , with an overall computation time of @xmath0 @xcite .
quantum computing architectures can be implemented using photons @xcite , trapped ions @xcite , spins in molecules @xcite and quantum dots @xcite , spins in silicon @xcite , and superconducting quantum circuits @xcite .
the last are leading the way for the realization of the first surface code logical qubit , which is one of the priorities in the quantum computing community at present . recently
, several experiments based on superconducting quantum circuits have demonstrated the principles underlying the surface code .
these works have shown a complete set of physical gates with fidelities beyond the surface code threshold @xcite , the parity measurements necessary to detect quantum errors @xcite , and have realized a classical version of the surface code on a one - dimensional array of nine physical qubits @xcite .
notably , the planar design inherent to the superconducting qubit platform will make it possible to implement large two - dimensional qubit arrays , as required by the surface code . despite all these accomplishments ,
a truly scalable qubit architecture has yet to be demonstrated .
wiring is one of the most basic unsolved scalability issues common to most solid - state qubit implementations , where qubit arrays are fabricated on a chip .
the conventional wiring method based on wire bonding suffers from fundamental scaling limitations as well as mechanical and electrical restrictions .
wire bonding relies on bonding pads located at the edges of the chip .
given a two - dimensional lattice of @xmath8 physical qubits on a square chip , the number of wire bonds that can be placed scales approximately as @xmath9 ( @xmath10 bonds for each chip side ) .
wire bonding will thus never be able to reach the required @xmath11 law according to which physical qubits scale on a two - dimensional lattice .
furthermore , for large @xmath10 , wire bonding precludes the possibility of accessing physical qubits in the center region of the chip , which is unacceptable for a physical implementation of the surface code . in the case of superconducting qubits , for example , qubit control and measurement are typically realized by means of microwave pulses or , in general , pulses requiring large frequency bandwidths . by their nature
, these pulses can not be reliably transmitted through long sections of quasi - filiform wire bonds .
in fact , stray capacitances and inductances associated with wire bonds as well as the self - inductance of the bond itself limit the available frequency bandwidth , thus compromising the integrity of the control and measurement signals @xcite .
additionally , the placement of wire bonds is prone to errors and inconsistency in spacing . in this work
, we set out to solve the wiring bottleneck common to almost all solid - state qubit implementations .
our solution is based on suitably packaged _ three - dimensional micro wires _ that can reach any area on a given chip from above .
we define this wiring system as the _ quantum socket_. the wires are coaxial structures consisting of a spring - loaded inner and outer conductor with diameters of and , respectively , at the smallest point and with a maximum outer diameter of .
the movable section of the wire is characterized by a maximum stroke of approximately , allowing for a wide range of on - chip mechanical compression .
all wire components are non magnetic , thereby minimizing any interference with the qubits .
the three - dimensional wires work both at room temperature and at cryogenic temperatures as low as @xmath0 .
the wires test - retest reliability is excellent , with marginal variability over hundreds of measurements .
their electrical performance is good from dc to at least , with a contact resistance smaller than and an instantaneous impedance mismatch of approximately .
notably , the coaxial design of the wires strongly reduces unwanted crosstalk , which we measured to be at most for a realistic quantum computing application . in a recent work @xcite ,
seven sequential stages necessary to the development of a quantum computer were introduced . at this time , the next stage to be reached is the implementation of a single logical qubit characterized by an error rate that is at least one order of magnitude lower than that of the underlying physical qubits . in order to achieve this task , a two - dimensional lattice of @xmath1 physical qubits with an error rate of at most @xmath12 is required @xcite . in the case of superconducting qubits
such a lattice can be realized on a chip area of @xmath13 ( the largest square that can be diced from a standard @xmath14 inch wafer ) and wired by means of a quantum socket .
it is feasible to further miniaturize the three - dimensional wires so as to achieve a wire density of .
this would allow the manipulation of @xmath15 physical qubits and , possibly , the realization of simple fault tolerant operations @xcite .
furthermore , the wires could serve as interconnect between a quantum hardware layer fabricated on one chip and a classical hardware layer realized on a separate chip , just above the quantum layer .
the classical hardware would be used to manipulate the qubits and could be implemented by means of rapid single - flux - quantum ( rsfq ) digital circuitry @xcite .
the implications of our cryogenic micro - wiring method go beyond quantum computing applications , providing a useful addition to the packaging industry for research applications at low temperatures . with this work ,
we demonstrate that the laborious and error - prone wire bonding technique can be substituted by the simple procedure of inserting the chip into a sample box equipped with three - dimensional wires .
our concept will expedite sample packaging and , thus , experiment turn over even for research not directly related to quantum information .
this research article is organized as follows . in sec .
[ the : quantum : socket : design ] , we introduce the quantum socket design , with special focus on the three - dimensional wires ( cf . subsec .
[ three : dimensional : wires ] ) , the microwave package ( cf . subsec .
[ microwave : package ] ) , and the package holder ( cf .
[ package : holder ] ) . in the same section ,
we show microwave simulations of a bare three - dimensional wire , of a wire connected to a pad on a chip , and of an entire microwave package ( cf .
[ microwave : simulations ] ) . in sec .
[ the : quantum : socket : implementation ] , we show the physical implementation of all components described in sec .
[ the : quantum : socket : design ] . in this section ,
we describe the materials used in the quantum socket and show how the socket is assembled .
in particular , we describe the magnetic and thermal properties of the quantum socket components ( cf .
[ magnetic : properties ] and subsec .
[ thermal : properties ] , respectively ) as well as the spring characterization ( cf .
[ spring : characterization ] ) .
moreover , we discuss in detail the quantum socket alignment procedure ( cf . subsec .
[ alignment ] ) . in sec .
[ characterization ] , we present a variety of measurements used to characterize the quantum socket operation .
we first focus on a four - port measurement used to estimate a wire contact resistance ( cf .
[ four : point : measurements ] ) .
we then show a series of microwave measurements on samples with different geometries and materials , both at room temperature and at .
these measurements comprise two - port scattering parameter ( s - parameter ) experiments ( cf .
[ two : port : scattering : parameters ] ) , time - domain reflectometry ( tdr ) analysis ( cf . subsec .
[ time : domain : reflectometry ] ) , and signal crosstalk tests ( cf .
[ signal : crosstalk ] ) . in sec .
[ applications : to : superconducting : resonators ] , we show an application of the quantum socket relevant to superconducting quantum computing , where the socket is used to measure aluminum ( al ) superconducting resonators at a temperature of approximately . finally , in sec .
[ conclusions ] , we envision an extensible quantum computing architecture where a quantum socket is used to connect to a @xmath1 lattice of superconducting qubits and comment on the possibility to use the socket in conjunction with rsfq electronics .
the development of the quantum socket required a stage of meticulous micro - mechanical and microwave design and simulations .
it was determined that a spring - loaded interconnect the three - dimensional wire was the optimal method to electrically access devices lithographically fabricated on a chip and operated in a cryogenic environment .
an on - chip contact pad geometrically and electrically matched to the bottom interface of the wire can be placed easily at any desired location on the chip as part of the fabrication process , thus making it possible to reach any point on a two - dimensional lattice of qubits .
the contact pad is realized as a thin metallic film deposited on a dielectric substrate ; arbitrary film thicknesses can be deposited .
the rest of the sample is fabricated on chip following a similar process .
the coaxial design of the wire provides a wide operating frequency bandwidth , while the springs allow for mechanical stress relief during the cooling process .
the three - dimensional wires used in this work take advantage of the knowledge in the existing field of microwave circuit testing @xcite . however , reducing the wire dimensions to a few hundred micrometers and using it to connect to quantum - mechanical micro - fabricated circuits at low temperatures resulted in a significant extension of existing implementations and applications . in this section , we will describe the design of the three - dimensional wires , of the microwave package used to house the wires , and of the microwave package holder .
additionally , we will show a set of microwave simulations of the main components of the quantum socket .
figure [ figure01:bejanin ] shows the design of the quantum socket components .
figure [ figure01:bejanin ] ( a ) displays a model of a three - dimensional wire .
the coaxial design of the wire is visible from the image , which features a wire long when uncompressed .
the wire is characterized by an inner cylindrical pin of diameter and an outer cylindrical body ( the electrical ground ) of diameter at its narrowest region ; this region is the bottommost section of the wire and , hereafter , will be referred to as the wire _ contact head _ ( cf . the inset of fig . [ figure01:bejanin ] ( a ) , as well as the dashed box on the left of fig . [ figure02:bejanin ] ( a ) ) .
the contact head terminates at the wire _
bottom interface _ ; this interface is designed to mate with a pad on a chip ( cf .
[ figure02:bejanin ] ( b ) and ( c ) ) .
the outer body includes a rectangular aperture , the _ tunnel _ , to prevent shorting the inner conductor of an on - chip coplanar waveguide ( cpw ) transmission line @xcite ; the transmission line can connect the pad with any other structure on the chip .
two different tunnel dimensions were designed , with the largest one reducing potential alignment errors .
these errors can result in undesired short - circuit connections to ground .
the tunnel height was in both cases , with a width of or .
the internal spring mechanisms of the wire allow the contact head to be compressed ; the maximum stroke was designed to be , corresponding to a working stroke of .
the outer body of the three - dimensional wire is an m@xmath16 male thread used to fix the wire to the lid of the microwave package ( cf .
[ figure01:bejanin ] ( b ) and ( d ) ) .
the thread is split into two segments of length and that are separated by a constriction with outer diameter .
the constriction is necessary to assemble and maintain in place the inner components of the three - dimensional wire .
a laser - printed marker is engraved into the top of the outer body .
the marker is aligned with the center of the tunnel , making it possible to mate the wire bottom interface with a pad on the underlying chip with a high degree of angular precision .
the grooves partially visible on the bottom of fig .
[ figure01:bejanin ] ( a ) are used to torque the wire into a thread in the package s lid .
figure [ figure02:bejanin ] ( a ) shows a lateral two - dimensional cut view of the three - dimensional wire .
two of the main wire components are the inner and outer barrel , which compose part of the inner and outer conductor .
the inner conductor barrel is a hollow cylinder with outer and inner diameters of and ( indicated as part iv in fig . [ figure02:bejanin ] ( a ) ) , respectively .
this barrel encapsulates the inner conductor spring . the outer conductor barrel is a hollow cylinder as well , in this case with an inner diameter of ( parts ii and vii ) .
three polytetrafluoroethylene ( ptfe ) disks serve as spacers between the inner and outer conductor ; such disks contribute marginally to the wire dielectric volume , the majority of which is air or vacuum . the outer spring is housed within the outer barrel towards its back end , just before the last ptfe disk on the right - hand side of the wire .
the _ back end _ of the wire is a region comprising a female thread on the outer conductor and an inner conductor barrel ( cf .
dashed box on the right - hand side of fig .
[ figure02:bejanin ] ( a ) ) .
the inner conductor tip is characterized by a conical geometry with an opening angle of @xmath17 .
such a sharp design was chosen to ensure that the tip would pierce through any possible oxide layer forming on the contact pad metallic surface , thus allowing for a good electrical contact .
figure [ figure02:bejanin ] ( c ) shows the design of a typical on - chip pad used to make contact with the bottom interface of a three - dimensional wire .
the pad comprises an inner and outer conductor , with the outer conductor being grounded .
the pad in the figure was designed for a silver ( ag ) film of thickness .
a variety of similar pads were designed for gold ( au ) and al films with thickness ranging between approximately and .
the pad inner conductor is a circle with diameter that narrows to a linear trace ( i.e. , the inner conductor of a cpw transmission line ) by means of a raised - cosine taper .
the raised cosine makes it possible to maximize the pad area , while minimizing impedance mismatch .
as designed , the wire and pad allow for lateral and rotational misalignment of @xmath18 and @xmath19 , respectively .
the substrate underneath the pad is assumed to be silicon ( si ) with a relative permittivity @xmath20 .
the dielectric gap between the inner and outer conductor is in the circular region of the pad ; the outer edge of the dielectric gap then follows a similar raised - cosine taper as the inner conductor .
the pad characteristic impedance is designed to be @xmath21 .
the microwave package comprises three main parts : the lid ; the sample holder ; the grounding washer .
the package is a parallelepiped with a height of and with a square base of side length .
the chip is housed inside the sample holder .
all these components mate as shown in fig .
[ figure01:bejanin ] ( b ) and ( c ) . in order to connect a three - dimensional wire to a device on a chip , the wire is screwed into an m@xmath16 female thread that is tapped into the lid of the microwave package , as depicted in fig .
[ figure01:bejanin ] ( b ) .
the pressure applied by the wire to the chip is set by the depth of the wire in the package .
the wire stroke , thread pitch , and alignment constraints impose discrete pressure settings ( cf . appendix [ wire : compression ] ) . in the present implementation of the quantum socket ,
the lid is designed to hold a set of six three - dimensional wires , which are arranged in two parallel rows . in each row , the wires are spaced by from center to center , with the two rows being separated by a distance of .
a square chip of lateral dimensions @xmath22 is mounted in the sample holder in a similar fashion as in ref .
the outer edges of the chip rest on four protruding lips , which are wide .
hereafter , those lips will be referred to as the _ chip recess_. for design purposes , a chip thickness of is assumed .
correspondingly , the chip recess is designed so that the top of the chip is above the adjacent surface of the chip holder , i.e. , the depth of the recess is ( cf . fig .
[ figure01:bejanin ] ( c ) ) .
the outer edges of the chip are pushed on by a spring - loaded grounding washer .
the chip protrusion ensures a good electrical connection between chip and washer , as shown in fig .
[ figure01:bejanin ] ( c ) .
the grounding washer was designed to substitute the large number of lateral bonding wires that would otherwise be required to provide a good ground to the chip ( as shown , for example , in fig .
6 of ref .
the washer springs are visible in fig . [ figure01:bejanin ] ( b ) , which also shows a cut view of the washer .
the washer itself is electrically grounded by means of the springs as well as through galvanic connection to the surface of the lid .
the four feet of the washer , which can be seen in the cut view of fig .
[ figure01:bejanin ] ( b ) , can be designed to be shorter or longer .
this makes it possible to choose different pressure settings for the washer . after assembling the package ,
there exist two electrical cavities ( cf .
[ figure01:bejanin ] ( d ) ) : one above the chip , formed by the lid , washer , and metallic surface of the sample ( _ upper cavity _ ) , and one below the chip , formed by the sample holder and metallic surface of the sample ( _ lower cavity _ ) .
the hollow cavity above the sample surface has dimensions @xmath23 .
the dimensions of the cavity below the sample are @xmath24 .
the lower cavity helps mitigate any parasitic capacitance between the chip and the box ( ground ) .
additionally , it serves to lower the effective permittivity , increasing the frequency of the substrate modes ( cf .
[ microwave : simulations ] ) .
a pillar of square cross section with side length of is placed right below the chip at its center ; the pillar touches the bottom of the chip , thus providing mechanical support .
the impact of such a pillar on the microwave performance of the package will be described in subsec .
[ microwave : simulations ] .
a channel with a cross - sectional area of @xmath25 connects the inner cavities of the package to the outside , thus making it possible to evacuate the inner compartments of the package .
this channel meanders to prevent external electromagnetic radiation from interfering with the sample .
the three - dimensional wires , which are screwed into the microwave package , must be connected to the qubit control and measurement electronics .
in addition , for cryogenic applications , the package must be thermally anchored to a refrigeration system in order to be cooled to the desired temperature .
figure [ figure01:bejanin ] ( d ) shows the mechanical module we designed to perform both electrical and thermal connections . in this design
, each three - dimensional wire is connected to a _ screw - in micro connector _ , which is indicated by an arrow in fig .
[ figure01:bejanin ] ( b ) and is shown in detail in fig .
[ figure02:bejanin ] ( d ) .
one end of the micro connector comprises a male thread and an inner conductor pin that mate with the back end of the three - dimensional wire .
the other end of the micro connector is soldered to a coaxial cable .
the micro connector is necessary because the high temperatures generated by soldering a coaxial cable directly to the wire back end would damage some of the inner wire components .
the end of each coaxial cable opposite to the three - dimensional wire is soldered to a sub - miniature push - on ( smp ) connector .
the smp connectors are bolted to a horizontal plate attached to the microwave package by means of two vertical fixtures , as shown in fig .
[ figure01:bejanin ] ( d ) .
the vertical fixtures and the horizontal plate constitute the package holder .
the package holder and microwave package form an independent assembly .
a horizontal mounting plate , designed to interface with the package holder , houses a set of matching smp connectors .
the mounting plate is mechanically and , thus , thermally anchored to the mixing chamber ( mc ) stage of a dilution refrigerator ( dr ) .
this design significantly simplifies the typical mounting procedure of a sample box to a cryostat , since the package holder and microwave package can be conveniently assembled remotely from the dr and attached to it just prior to commencing an experiment .
the three - dimensional wires , the transition between the wire and the on - chip pad as well as the inner cavities of the fully - assembled microwave package were extensively simulated by means of the high frequency three - dimensional full - wave electromagnetic field simulation software ( hfss ) by ansys , inc .
the results for the electromagnetic field distribution at a frequency of approximately , which is a typical operation frequency for superconducting qubits , are shown in fig .
[ figure03:bejanin ] .
figure [ figure03:bejanin ] ( a ) shows the field behavior for a bare three - dimensional wire .
the field distribution resembles that of a coaxial transmission line except for noticeable perturbations at the dielectric ptfe spacers .
figure [ figure03:bejanin ] ( b ) shows the transition region .
this is a critical region for signal integrity since abrupt changes in physical geometry cause electrical reflections @xcite . in order to minimize such reflections ,
an impedance - matched pad was designed .
however , this leads to a large electromagnetic volume in proximity of the pad , as seen in fig .
[ figure03:bejanin ] ( b ) , possibly resulting in parasitic capacitance and crosstalk .
cccc [ 3mm][0 mm ] & @xmath26 & @xmath27 & @xmath28 + [ 0mm][2 mm ] & & & + [ 3mm][0mm]vacuum & @xmath29 & @xmath30 & @xmath30 + [ 3mm][0mm]vacuum with pillar & @xmath31 & @xmath32 & @xmath32 + [ 3mm][0mm]si & @xmath33 & @xmath34 & @xmath34 + [ 3mm][0mm]si with pillar & @xmath35 & @xmath36 & @xmath37 + [ table01:bejanin ] in addition to considering the wire and the transition region , the electrical behavior of the inner cavities of the package was studied analytically and simulated numerically . as described in subsec .
[ microwave : package ] , the metallic surface of the chip effectively divides the cavity of the sample holder into two regions : a vacuum cavity above the metal surface and a cavity partially filled with dielectric below the metal surface .
the latter is of greatest concern as the dielectric acts as a perturbation to the cavity vacuum , thus lowering the box modes . for a simple rectangular cavity
, the frequency @xmath38 of the first mode due to this perturbation can be found as @xcite , @xmath39 where @xmath40 is the frequency of the unperturbed mode , @xmath41 the relative permittivity of the dielectric , @xmath42 the substrate thickness , and @xmath43 the cavity height . from eq .
( [ equation:01 ] ) , we estimated this box mode to be . however , considering the presence of the pillar , the three - dimensional wires , etc . , we had to use numerical simulations to obtain a more accurate estimate of the lowest box modes .
the results for the first three modes are reported in table [ table01:bejanin ] .
discounting the pillar , the analytical and simulated values are in good agreement with each other .
the addition of the support pillar significantly lowers the frequency of the modes .
in fact , it increases the relative filling factor of the cavity by confining more of the electromagnetic field to the dielectric than to vacuum .
given the dimensions of this design , the pillar leads to a first mode which could interfere with typical qubit frequencies . in spite of this ,
the pillar was included in the final design in order provide a degree of mechanical support .
note that the pillar can alternatively be realized as a dielectric material , e.g. , ptfe ; a dielectric pillar would no longer cause field confinement between the top surface of the pillar and the metallic surface of the chip .
the physical implementation of the main components of the quantum socket is displayed in fig .
[ figure04:bejanin ] .
in particular , fig . [ figure04:bejanin ] ( a ) shows a macro photograph of a three - dimensional wire .
the inset shows a scanning electron microscope ( sem ) image of the wire contact head , featuring the version of the tunnel .
this wire was cycled approximately ten times ; as a consequence , the center conductor of the contact head , which had a conical , sharp shape originally , flattened at the top .
the metallic components of the wire were made from bronze and brass ( cf .
[ magnetic : properties ] ) , and all springs from hardened beryllium copper ( becu ) . except for the springs ,
all components were gold plated without any nickel ( ni ) adhesion underlayer .
the estimated mean number of cycles before failure for these wires is approximately @xmath44 .
figure [ figure04:bejanin ] ( b ) displays the entire microwave package in the process of locking the package lid and sample holder together , with a chip and grounding washer already installed .
as shown in the figure , two rows of three - dimensional wires , for a total number of six wires , are screwed into the lid with pressure settings as described in appendix [ wire : compression ] ; each wire is associated with one on - chip cpw pad .
the four springs that mate with the grounding washer feet are embedded in corresponding recesses in the lid ; the springs are glued in these recesses by way of vibra - tite @xmath45 ( from nd industries , inc . ) , a medium - strength thread locker that works well at low temperatures .
all package components were made from high - purity al .
figure [ figure04:bejanin ] ( c ) shows a picture of the assembled microwave package attached to the package holder ; the entire structure is attached to the mc stage of a dr .
all parts of the assembly were made from high thermal conductivity c10100 oxygen - free electrolytic ( ofe ) copper alloy . ] .
the parts were polished to a mirror finish before being gold plated .
the coaxial cables between the screw - in micro connectors and the smp connectors are from the ez form cable corporation , model ez 47-cu - tp ( ez 47 ) .
the smp connectors , also from ez form , are models smp bulkhead jack for @xmath46 inch coaxial cables ( smp 047 ; installed in the package holder horizontal plate ) and smp bulkhead plug with limited detent for @xmath47 inch cables ( smp 086 ; installed in the mounting plate attached to the mc stage of the dr ) .
all smp connectors were custom - made non - magnetic connectors . in the remainder of this section , we will discuss the magnetic and thermal properties of the materials used to implement the quantum socket as well as the spring characterization and the alignment procedure .
an important stage in the physical implementation of the quantum socket was the choice of materials to be used for the three - dimensional wires , the microwave package , and the package holder .
in fact , it has been shown that non - magnetic components in proximity of superconducting qubits are critical to preserve long qubit coherence @xcite .
the three - dimensional wires are the closest devices to the qubits .
for this reason , all their components should be made using non - magnetic materials .
due to machining constraints , however , alloys containing ferromagnetic impurities ( iron ( fe ) , cobalt ( co ) , and ni ) had to be used . for the outer conductor components we used brass , which is easy to thread ; the chosen grade was iso cuzn21si3p ( en cw724r ) . for the inner conductor components ,
brass cw724r did not meet the machining requirements .
consequently , we decided to use copper alloy ( phosphor bronze ) grade din 2.1030 - cusn8 ( en cw453k ) .
the chemical composition for these two materials is reported in table [ table04:bejanin ] of appendix [ the : quantum : socket : magnetism ] . the dielectric spacers were made from ptfe and the rest of the components from hardened becu ; both materials are non - magnetic .
the weight percentage of ferromagnetic materials is non - negligible for both cw453k and cw724r .
thus , we performed a series of tests using a zero gauss chamber ( zgc ) in order to ensure both materials were sufficiently non - magnetic .
the results are given in appendix [ the : quantum : socket : magnetism ] and show that the magnetic impurities should be small enough not to disturb the operation of superconducting quantum devices .
the microwave package and grounding washer were made from high - purity al alloy 5n5 ( purity ) provided by laurand associates , inc .
the very low level of impurities in this alloy assures minimal stray magnetic fields generated by the package itself , as confirmed by the magnetic tests discussed in appendix [ the : quantum : socket : magnetism ] .
the thermal conductance of the three - dimensional wires is a critical parameter to be analyzed for the interconnection with devices at cryogenic temperatures .
low thermal conductivity would result in poor cooling of the devices , which , in the case of qubits , may lead to an incoherent thermal mixture of the qubit ground state @xmath48 and excited state @xmath49 @xcite .
even a slightly mixed state would significantly deteriorate the fidelity of the operations required for qec @xcite .
it has been estimated that some of the qubits in the experiment of ref .
@xcite , which relies solely on al wire bonds as a means of thermalization , were characterized by an excited state population @xmath50 . among other possible factors
, it is believed that this population was due to the poor thermal conductance of the al wire bonds .
in fact , these bonds become superconductive at the desired qubit operation temperature of @xmath51 , preventing the qubits from thermalizing and , thus , from being initialized in @xmath48 with high fidelity . in order to compare the thermal performance of an al wire bond with that of a three - dimensional wire ,
we estimated the heat transfer rate per kelvin of a wire , @xmath52 , using a simplified coaxial geometry . at a temperature of , we calculated @xmath53 . at the same temperature , the heat transfer rate per kelvin of a typical al wire bound
was estimated to be @xmath54 ( cf .
appendix [ thermal : conductance : of : a : three : dimensional : wire ] for more details ) . a very large number of al wire bonds
would thus be required to obtain a thermal performance comparable to that of a single three - dimensional wire .
another critical step in the physical implementation of the quantum socket was to select springs that work at cryogenic temperatures .
in fact , the force that a wire applies to a chip depends on these springs . this force , in turn , determines the wire - chip contact resistance , which impacts the socket s dc and microwave performance . among various options , we chose custom springs made from hardened becu . to characterize the springs ,
their compression was assessed at room temperature , in liquid nitrogen ( i.e. , at a temperature @xmath55 ) , and in liquid helium ( @xmath56 ) .
note that a spring working at is expected to perform similarly at a temperature of .
a summary of the thermo - mechanical tests is reported in appendix [ thermo : mechanical : tests ] .
the main conclusion of the tests is that the springs do not break ( even after numerous temperature cycles ) and have similar spring constants at all measured temperatures . and
@xmath57 are indicated in ( a ) by means of magenta bars .
( c)-(d ) al pad before and after a cooling cycle to @xmath58 .
center conductor dragging due to cooling is indicated by a green bar .
the magenta dashed line in ( a ) indicates tunnel ( i.e. , rotational ) alignment for the ag pad . the images ( dark field )
were taken with an olympus mx61 microscope at @xmath59 magnification , manual exposure , and @xmath60 exposure time .
note that the geometries for the pads in panels ( a ) and ( b ) are optimized for a ag film and , thus , are slightly different than those for the pads in panels ( c ) and ( d ) , which are designed for a al film.,scaledwidth=49.0% ] in order to implement a quantum socket with excellent interconnectivity properties , it was imperative to minimize machining errors and mitigate the effects of any residual errors .
these errors are mainly due to : dicing tolerances ; tapping tolerances of the m@xmath16-threaded holes of the lid ; tolerances of the mating parts for the inner cavities of the lid and sample holder ; tolerances of the chip recess .
these errors can cause both lateral and rotational misalignment and become likely worse when cooling the quantum socket to low temperatures .
more details on alignment errors can be found in appendix [ alignment : errors ] .
the procedure to obtain an ideal and repeatable alignment comprises three main steps : optimization of the contact pad and tunnel geometry ; accurate and precise chip dicing ; accurate and precise package machining . for the quantum socket described in this work
, the optimal tunnel width was found to be .
this maintained reasonable impedance matching , while allowing greater cpw contact pad and tapering dimensions .
the contact pad width @xmath61 and taper length @xmath57 were chosen to be @xmath62 and @xmath63 .
these are the maximum dimensions allowable that accommodate the geometry of the wire bottom interface for a nominal lateral and rotational misalignment of @xmath64 and @xmath65 , respectively . in order to select the given pad dimensions
, we had to resort to a @xmath66 matched raised - cosine tapering .
the majority of the chips used in the experiments presented here was diced with a dicing saw from the disco corporation , model dad3240 . to obtain a desired die length ,
both the precision of the saw stage movement and the blade s kerf had to be considered .
for the dad3240 saw , the former is @xmath67 , whereas the latter changes with usage and materials . for the highest accuracy cut
, we measured the kerf on the same type of wafer just prior to cutting the actual die . in order to achieve maximum benefit from the saw ,
rotational and lateral alignment dicing markers were incorporated on the wafer .
such a meticulous chip dicing procedure is only effective in conjunction with a correspondingly high level of machining accuracy and precision .
we used standard computer numerical control ( cnc ) machining with a tolerance of @xmath68 thou ( ) , although electrical discharge machining can be pursued if more stringent tolerances are required .
following the aforementioned procedures we were able to achieve the desired wire - pad matching accuracy and precision , which resulted in a test - retest reliability ( repeatability ) of over @xmath69 instances .
these figures of merit were tested in two steps : first , by micro imaging several pads that were mated to a three - dimensional wire ( cf .
[ on : chip : pad : micro : imaging ] ) ; second , by means of dc resistance tests ( cf .
[ dc : resistance : tests ] ) .
micro imaging was performed on a variety of different samples , four of which are exemplified in fig .
[ figure05:bejanin ] .
the figure shows a set of micro images for ag and al pads ( details regarding the fabrication of these samples are available in appendix [ sample : fabrication ] ) .
figure [ figure05:bejanin ] ( a ) and ( b ) show two ag pads that were mated with the three - dimensional wires at room temperature .
panel ( a ) shows a mating instance where the wire bottom interface perfectly matched the on chip pad .
panel ( b ) shows two mating instances that , even though not perfectly matched , remained within the designed tolerances .
notably , simulations of imperfect mating instances revealed that an off - centered wire does not significantly affect the microwave performance of the quantum socket .
finally , panels ( c ) and ( d ) display two al pads which were both mated with a wire one time .
while the pad in ( c ) was operated only at room temperature , the pad in ( d ) was part of an assembly that was cooled to @xmath58 for approximately three months .
the image was taken after the assembly was cycled back to room temperature and shows dragging of the wire by a few tens of micrometers .
such a displacement can likely be attributed to the difference in the thermal expansion of si and al ( cf . appendix [ alignment : errors ] ) . as a diagnostic tool , micro images of a sample already mounted in the sample holder after a mating cycle can be obtained readily by means of a handheld digital microscope .
cccccccccccc [ 3mm][0mm]metal & @xmath70 & @xmath71 & @xmath72 & @xmath73 & @xmath74 & @xmath75 & @xmath10 & @xmath76 & @xmath77 & @xmath78 & @xmath79 + [ 0mm][2mm]- & & & & & & & & & & & + [ 3mm][0mm]au & @xmath80 & @xmath81 & @xmath82 & @xmath83 & @xmath84 & @xmath85 & @xmath86 & @xmath87 & @xmath88 & @xmath89 & @xmath89 + [ 3mm][0mm]au & @xmath80 & @xmath81 & @xmath82 & @xmath90 & @xmath91 & @xmath85 & @xmath92 & @xmath87 & @xmath93 & @xmath94 & @xmath94 + [ 3mm][0mm]au & @xmath80 & @xmath81 & @xmath82 & @xmath95 & @xmath96 & @xmath85 & @xmath92 & @xmath87 & @xmath97 & @xmath98 & @xmath98 + [ 3mm][0mm]au & @xmath80 & @xmath81 & @xmath82 & @xmath99 & @xmath100 & @xmath85 & @xmath92 & @xmath87 & @xmath101 & @xmath102 & @xmath103 + [ 3mm][0mm]au & @xmath80 & @xmath104 & @xmath82 & @xmath105 & @xmath106 & @xmath85 & @xmath80 & @xmath107 & @xmath108 & @xmath109 & @xmath109 + [ 3mm][0mm]au & @xmath80 & @xmath104 & @xmath110 & @xmath105at . ] & @xmath106 & @xmath111 & @xmath112 & @xmath113 & @xmath114 & @xmath115 & @xmath116 + [ 3mm][0mm]ag & @xmath86 & @xmath117 & @xmath118 & @xmath105 & @xmath106 & @xmath85 & @xmath112 & @xmath119 & @xmath120 & @xmath121 & @xmath121 + [ 3mm][0mm]al & @xmath122 & @xmath123 & @xmath124 & @xmath125 & @xmath126 & @xmath85 & @xmath127 & @xmath128 & @xmath129 & @xmath130 & @xmath130 + [ table02:bejanin ] in contrast to the micro imaging tests , which require the removal of the microwave package s lid , dc resistance tests can be performed _ in situ _ at room temperature after the package and package holder have been fully assembled .
these tests were performed on all devices presented in this work , including au , ag , and al samples .
the typical test setup comprises a microwave package with two three - dimensional wires each mating with an on - chip pad .
the two pads are connected by means of a cpw transmission line with series resistance @xmath76 .
the back end of the wires is connected to a coaxial cable ending in a microwave connector , similar to the setup in figs .
[ figure01:bejanin ] ( d ) and [ figure04:bejanin ] ( c )
. the dc equivalent circuit of this setup can be represented by way of a four - terminal pi network .
the circuit comprises an input `` @xmath131 '' and output `` @xmath132 '' terminal , two terminals connected to a common ground `` @xmath133 , '' an input - output resistor with resistance @xmath77 , and two resistors to ground with resistance @xmath78 and @xmath134 .
the @xmath131 and @xmath132 terminals correspond to the inner conductor of the two microwave connectors .
the outer conductor of both connectors is grounded .
the resistance @xmath77 is that of the center conductor of the cpw transmission line , including the contact resistance @xmath135 for each wire - pad interface and the series resistance @xmath136 of the wire s and coaxial cable s inner conductor , @xmath137 .
the resistances @xmath78 and @xmath79 are those of the path between each center conductor and ground and include the resistance of the inner and outer conductor of the various coaxial cables and wires as well as any wire - pad contact resistance . ideally , these ground resistances should be open circuits .
in reality , they are expected to have a finite but large value because of the intrinsic resistance of the si wafers used as a substrate .
the design parameters , electrical properties , measurement conditions as well as the measured values of @xmath77 , @xmath78 , and @xmath138 for various au , ag , and al samples are reported in table [ table02:bejanin ] .
the resistances were probed by means of a multimeter from the fluke corporation , model 289 .
measuring resistances significantly different from the expected values means that either a lateral or rotational misalignment occurred .
the resistances for some au samples were also measured at to verify whether a good room temperature alignment persisted in cryogenic conditions .
the cold measurements were realized by dunking the package holder into liquid nitrogen . the measured value of @xmath77 for the ag samples is larger than the estimated trace resistance by @xmath139 .
this simple result makes it possible to find an upper bound value for the contact resistance , @xmath140 .
a more accurate estimate of the contact resistance based on four - point measurements will be described in subsec.[four : point : measurements ] .
the dc resistance testing procedure presented here will be useful in integrated - circuit quantum information processing , where , for example , cpw transmission lines can serve as qubit readout lines @xcite .
these tests can be expanded to encompass different circuit structures such as the qubit control lines utilized in ref .
the three - dimensional wires are multipurpose interconnects that can be used to transmit signals over a wide frequency range , from dc to @xmath141
. these signals can be : the current bias used to tune the transition frequency of a superconducting qubit ; the gaussian - modulated sinusoidal or the rectangular pulses that , respectively , make it possible to perform xy and z control on a qubit ; the continuous monochromatic microwave tones used to read out a qubit state or to populate and measure a superconducting resonator @xcite . in general , the wires can be used to transmit any baseband modulated carrier signal within the specified frequency spectrum , at room and cryogenic temperatures . in this section ,
we report experimental results for a series of measurements aiming at a complete electrical characterization of the quantum socket at room temperature and at approximately ( i.e. , in liquid nitrogen ) .
first , we performed four - point measurements to estimate the contact resistance of a three - dimensional wire .
second , we measured the s - parameters of a wire at room temperature .
third , we measured the s - parameters of the quantum socket with an au sample at room temperature and at and an ag sample at room temperature .
fourth , we realized time - domain measurements of the quantum socket .
last , we performed four - port s - parameter measurements in order to assess the socket crosstalk properties .
the wire - pad contact resistance @xmath135 is an important property of the quantum socket .
in fact , a large @xmath135 would result in significant heating when applying dc bias signals and rectangular pulses , thus deteriorating qubit performance . in order to assess @xmath135 for the inner and outer conductor of a three - dimensional wire
, we performed four - point measurements using the setup shown in the inset of fig .
[ figure06:bejanin ] . using this setup
, we were able to measure both the series resistance of the wire @xmath142 and the contact resistance @xmath135 .
this allows us to estimate the overall heating that could be generated during a qubit experiment .
the setup comprises a microwave package with a chip entirely coated with a thick al film ; no grounding washer was used .
the package featured three three - dimensional wires , of which two were actually measured ; the third wire was included to provide mechanical stability .
the package was attached to the mc stage of a dr and connected to a set of phosphor bronze twisted pairs .
the twisted pairs were thermally anchored at all dr stages and connected at room temperature to a precision source - measure unit ( smu ) from keysight technologies inc . , model b2911a .
we measured the resistance between the inner conductor of a wire and ground , @xmath78 .
this resistance comprises the inner conductor wire resistance @xmath143 in series with the inner conductor contact resistance @xmath144 and any resistance to ground , @xmath145 . note that , at the operation temperature of the experiment ( @xmath58 ) , al is superconducting and , thus
, the metal resistance can be neglected .
figure [ figure06:bejanin ] shows the current - voltage ( i - v ) characteristic curve for @xmath78 . with increasing bias currents , the contact resistance results in hot - spot generation leading to a local breakdown of superconductivity . for sufficiently high bias currents , superconductivity breaks down completely . at such currents ,
the observed hysteretic behavior indicates the thermal limitations of our setup .
note , however , that these currents are at least one order of magnitude larger than the largest bias current required in typical superconducting qubit experiments @xcite . in order to estimate @xmath78 from the i - v characteristic curve , we selected the bias current region from @xmath146 to @xmath147 and fitted the corresponding slope .
we obtained @xmath148 .
this value , which represents an upper bound for the wire resistance and the wire - pad contact resistance , @xmath149 , is significantly larger than that associated with al wire bonds @xcite . in future versions of the three - dimensional wires we will attempt to reduce the wire - pad contact resistance by rounding the tip of the center conductor , stiffening the wire springs , using a thicker metal film for the pads , and plating the contact pads with au or titanium nitride .
we note , however , that even a large value of the wire and/or wire - pad contact resistance will not significantly impair the quantum socket microwave performance . .
the sweeps were conducted by both increasing ( red ) and decreasing ( blue ) the applied current between @xmath150 and @xmath151 .
the voltage measurements were delayed by and averaged over .
the displayed data is averaged over @xmath81 sweeps .
the shaded region indicates two standard deviations .
the dashed black lines indicate the region ( @xmath152 ) for which the resistance value was found using linear regression . the origin of the hystereses is explained in the main text .
the inset shows the circuit diagram of the device under test , including all resistors measured by means of the four - point measurement .
the position of the pad is indicated by an arrow.,scaledwidth=49.0% ] the s - parameter measurements of a bare three - dimensional wire were realized by means of the setup shown in the inset of fig .
[ figure07:bejanin ] ( a ) .
the device under test ( dut ) comprises a cable assembly attached to a three - dimensional wire by means of a screw - in micro connector .
the cable assembly is made of an approximately long semi - rigid coaxial cable ez 47 , which is soldered to an ez form custom - made sub - miniature type a ( sma ) male connector , model 705538 - 347 .
the other end of the coaxial cable is soldered to the screw - in micro connector .
the sma connector of the dut is connected to one port of a vector network analyzer ( vna ) from keysight , model pna - l n5230a by means of a flexible coaxial cable .
the bottom interface of the wire is connected to a 2.92 mm end launch connector from southwest microwave , inc . , model 1092 - 01a-5 , which then connects to the other port of the vna through a second flexible coaxial cable .
the 2.92 mm adapter is characterized by a flush coaxial back plane , which mates with the wire bottom interface well enough to allow for s - parameter measurements up to . in order to measure the s - parameters of the dut
, a two - tier calibration was performed .
first , a two - port electronic calibration module ( ecal ) from keysight , model n4691b , with 2.92 mm male connectors was used to set the measurement planes to the end of the flexible cables closer to the dut .
second , a port - extension routine was performed to correct for the insertion loss , phase , and delay of the 2.92 mm adapter .
this made it possible to set the measurement planes to the ports of the dut .
the magnitudes of the measured reflection and transmission s - parameters are displayed in fig .
[ figure07:bejanin ] ( a ) .
we performed microwave simulations of a three - dimensional wire for the same s - parameters ( cf .
[ microwave : simulations ] for the electric field distribution ) , the results of which are plotted in fig . [ figure07:bejanin ] ( b ) .
the s - parameters were measured and simulated between and .
the s - parameters @xmath153 and @xmath154 show a featureless microwave response , similar to that of a coaxial transmission line .
the attenuation at is @xmath155 and the magnitude of the reflection coefficients at the same frequency is @xmath156 and @xmath157 .
the phase of the various s - parameters ( not shown ) behaves as expected for a coaxial transmission line .
all measurements were performed at room temperature .
the s - parameter measurements of a three - dimensional wire indicate a very good microwave performance . however , these measurements alone are insufficient to fully characterize the quantum socket operation . a critical feature that deserves special attention is the transition region between the wire bottom interface and the on - chip cpw pad .
it is well - known that transitions can cause significant impedance mismatch and , thus , signal reflection @xcite . in quantum computing applications
, these reflections could degrade both the qubit control and readout fidelity .
figure [ figure08:bejanin ] shows a typical setup for the characterization of a wiring configuration analogous to that used for qubit operations .
the setup comprises a vna from keysight , model pna - x n5242a , with ports @xmath68 and @xmath92 connected to a pair of flexible coaxial cables from huber+suhner ag , model sucoflex 104-pe ( with sma male connectors ) . in order to calibrate the measurement ,
the flexible cables were first connected to a two - port ecal module from keysight , model n4691 - 60006 , featuring 3.5 mm female connectors .
these cables were then connected to the sma female bulkhead adapter at the input and output ports of the dut shown in fig .
[ figure08:bejanin ] .
the dut incorporates a microwave package with a pair of three - dimensional wires , which address one cpw transmission line on an au or ag chip .
the microwave package was attached to the package holder , as described in subsec .
[ package : holder ] and sec .
[ the : quantum : socket : implementation ] ( cf . also figs .
[ figure01:bejanin ] ( d ) and [ figure04:bejanin ] ( c ) ) . for the measurements in this section
, however , the smp adapters were substituted by sma female - female bulkhead adapters .
the transmission line geometrical dimensions and wire pressure settings are reported in table [ table02:bejanin ] ; only the au samples and the ag samples were characterized at microwave frequencies .
the back end of each three - dimensional wire is connected to one end of an ez 47 cable by means of the screw - in micro connector described in subsec .
[ package : holder ] ; the other end of the ez 47 cable is soldered to an sma male connector .
one of the ez cables is long and the other is long ; the longer cable ( output ) is connected directly to the sma bulkhead adapter , whereas the shorter cable ( input ) is prolonged using an sma female - male adapter and , then , connected to the bulkhead adapter .
these bulkhead adapters are the reference planes ii and xii associated with the input and output ports of the dut , respectively , as shown in fig .
[ figure08:bejanin ] .
we performed a two - port s - parameter measurement of the dut from to .
we selected an intermediate frequency ( if ) bandwidth @xmath158 , a constant excitation power @xmath159 ( dbm ) , and @xmath160 or @xmath161 measurement points for the au and ag samples , respectively .
the measurement results at room temperature for the au and ag samples are shown in figs .
[ figure09:bejanin ] ( a ) and [ figure10:bejanin ] ( a ) , respectively .
the results for the au sample at are shown in fig .
[ figure09:bejanin ] ( b ) .
the s - parameter measurements of the au sample show that the quantum socket functions well at microwave frequencies , both at room temperature and at .
since most of the mechanical shifts have already occurred when cooling to @xcite , this measurement allows us to deduce that the socket will continue functioning even at lower temperatures , e.g. , @xmath58 .
the au sample , however , is characterized by a large value of @xmath77 , which may conceal unwanted features both in the transmission and reflection measurements .
therefore , we prepared an ag sample that exhibits a much lower resistance even at room temperature .
the behavior of the ag s - parameters is similar to that of a transmission line or coaxial connector .
for example , @xmath162 is approximately ; as a reference , for a high - precision sma connector at the same frequency @xmath163 .
the presence of the screw - in micro connector can occasionally deteriorate the microwave performance of the quantum socket .
in fact , if the micro connector is not firmly tightened , a dip in the microwave transmission is observed . at room temperature ,
it is straightforward to remove the dip by simply re - tightening the connector when required . on the contrary , for the measurements at and for any other application in a cryogenic environment assuring that the micro connector is properly torqued at all times
can be challenging .
figure [ figure09:bejanin ] ( b ) , for example , shows the s - parameters for an au sample measured at .
a microwave dip appeared at approximately , with a bandwidth of approximately @xmath164 .
the inset in fig .
[ figure09:bejanin ] ( b ) displays the phase angle of @xmath165 between and , showing that the dip is unlikely a lorentzian - type resonance ( more details in the supplemental material at http://www.supplemental-material-bejanin ) .
note that the dip is far from the typical operation frequencies for superconducting qubits .
additionally , as briefly described in sec .
[ conclusions ] , we will remove the screw - in micro connector from future generations of the three - dimensional wires .
figure [ figure10:bejanin ] ( b ) shows a simulation of the s - parameters for the ag sample , for the same frequency range as the actual measurements . while there are visible discrepancies between the measured and simulated s - parameters , the latter capture well some of the characteristic features of the microwave response of the dut .
in particular , the measured and simulated reflection coefficients display a similar frequency dependence .
it is worth mentioning that we also simulated the case where the wire bottom interface is not perfectly aligned with the on - chip pad ( results not shown ) .
we considered lateral misalignments of and rotational misalignments of @xmath166 .
this allowed us to study more realistic scenarios , such as those shown in fig .
[ figure05:bejanin ] .
we found that the departure between the misaligned and the perfectly aligned simulations was marginal .
for example , the transmission s - parameters varied only by approximately @xmath167 . in appendix [
microwave : parameters ] , we show a set of microwave parameters obtained from the measured s - parameters for the au sample at room temperature and at and for the ag sample at room temperature .
these parameters make it possible to characterize the input and output impedance as well as the dispersion properties of the quantum socket . in tdr measurements , a rectangular pulse with fast rise time and fixed length
is applied to a dut ; the reflections ( and all re - reflections ) due to all reflection planes in the system ( i.e. , connectors , geometrical changes , etc . )
are then measured by way of a fast electrical sampling module .
the reflections are , in turn , related to the impedances of all of the system components .
thus , tdr makes it possible to estimate any impedance mismatch and its approximate spatial location in the system .
tdr measurements were performed on the dut shown in fig .
[ figure08:bejanin ] , with the same au or ag sample as for the measurements in subsec .
[ two : port : scattering : parameters ] .
as always , the au sample was measured both at room temperature and at , whereas the ag sample was measured only at room temperature .
the tdr setup is analogous to that used for the s - parameter measurements , with the following differences : the dut input and output reference planes were extended to include the sucoflex flexible coaxial cables ( i.e. , these cables were not calibrated out ) ; when testing the dut input port , the output port was terminated in a load with impedance @xmath168 and vice versa when testing the dut output port .
the tdr measurements were realized by means of a sampling oscilloscope from teledyne lecroy , model waveexpert 100h ; the oscilloscope features an electrical sampling module with bandwidth and a tdr step generator , model st-20 .
the generated signal is a voltage square wave characterized by a nominal pulse rise time of , amplitude of , pulse width of , and pulse repetition rate of .
the voltage reflected by the dut , @xmath169 , is acquired as a function of time @xmath170 by means of the sampling module .
this time is the round - trip interval necessary for the voltage pulse to reach a dut reflection plane and return back to the sampling module .
the measured quantity is given by into distance is only possible with detailed knowledge of geometries and materials for all regions of the dut . since this information
is not known to a high degree of accuracy , we prefer to express all measured quantities as a function of @xmath170 . ]
@xmath171 where @xmath172 is the amplitude of the incident voltage square wave . from eq .
( [ equation:02 ] ) , we can obtain the first - order instantaneous impedance as @xcite @xmath173 where @xmath174 .
figure [ figure11:bejanin ] shows @xmath175 for the dut with the au sample at room temperature and at ; the measurement refers to the input port of the dut , including a flexible cable .
the figure inset shows the room temperature data for a shorter time interval .
this corresponds to a space interval beginning at a point between planes iv and v and ending at a point between planes vii and viii ( cf .
[ figure08:bejanin ] ) . figure [ figure12:bejanin ] ( a ) shows @xmath175 for the ag sample at room temperature .
figure [ figure12:bejanin ] ( b ) displays the data in ( a ) for a time interval corresponding to a space interval beginning at a point between planes iv and v and ending at a point between planes x and xi ; as a reference , the au data are overlaid with the ag data . of the setup in fig .
[ figure08:bejanin ] .
( b ) zoomin of ( a ) addressing the three - dimensional wire and the transition region between the wire and the cpw transmission line ( blue ) .
the room temperature au data ( red ) is also displayed as a reference.,scaledwidth=49.0% ] for the au sample , the first main reflection plane ( plane ii ) is encountered at @xmath176 .
the second main reflection plane ( plane v ) appears after @xmath177 relative to the first plane , at @xmath178 . from that time instant and for a span of approximately , the tdr measurement corresponds to @xmath179 of the three - dimensional wire itself .
the maximum impedance mismatch between the ez form cable and the three - dimensional wire is approximately .
the third main reflection plane ( plane vii ) corresponds to the transition region ; for the au sample , it is impossible to identify features beyond this plane owing to the large series resistance of the on - chip cpw transmission line . from empirical evidence ,
the impedance @xmath175 of a lossy line with series resistivity @xmath72 increases linearly with the length of the line @xmath180 as @xmath181 .
in fact , for the au sample we measured an impedance step across the cpw transmission line of approximately @xmath182 at room temperature and at .
these steps are approximately the @xmath183 values reported in table [ table02:bejanin ] . in order to obtain a detailed measurement of the impedance mismatch beyond the transition region
, we resorted to the tdr measurements of the dut with the much less resistive ag sample .
first , we confirmed that @xmath179 of the input three - dimensional wire for the ag sample is consistent with the tdr measurements of the au sample ; this is readily verified by inspecting fig .
[ figure12:bejanin ] ( b ) .
the three - dimensional wire is the structure ending at the onset of the large impedance step shown by the au overlaid data .
the structure spanning the time interval from @xmath184 to @xmath185 is associated with the input transition region , the cpw transmission line , and the output transition region .
the output three - dimensional wire starts at @xmath186 , followed by the ez form coaxial cable , which finally ends at the sma bulkhead adapter at @xmath187 . the maximum impedance mismatch associated with the transition regions and the cpw transmission line is @xmath188 .
notably , this mismatch is smaller than the mismatch between the three - dimensional wire and the coaxial cable .
this is an important result .
in fact , while it would be hard to diminish the impedance mismatch due to the transition region , it is feasible to further minimize the wire mismatch by creating accurate lumped - element models of the wire and use them to minimize stray capacitances and/or inductances .
it is worth comparing @xmath175 of the quantum socket with that of a standard package for superconducting qubits , where wire bonds are used to make interconnections between a printed circuit board and the control and measurement lines of a qubit on a chip .
a detailed study of the impedance mismatch associated with wire bonds is found in ref .
@xcite , where the authors have shown that a long wire bond ( of length between @xmath189 and @xmath190 ; typical length in most applications ) can lead to an impedance mismatch larger than ( cf .
s3 in the supplementary information of ref .
@xcite ) ; on the contrary , a short wire bond ( between @xmath191 and @xmath192 ; less typical ) results in a much smaller mismatch , approximately . in terms of impedance mismatch the current implementation of the quantum socket , which is limited by the mismatch of the three - dimensional wires , lies in between these two extreme scenarios .
crosstalk is a phenomenon where a signal being transmitted through a channel generates an undesired signal in a different channel .
inter - channel isolation is the figure of merit that quantifies signal crosstalk and that has to be maximized to improve signal integrity .
crosstalk can be particularly large in systems operating at microwave frequencies , where , if not properly designed , physically adjacent channels can be significantly affected by coupling capacitances and/or inductances . in quantum computing implementations based on superconducting quantum circuits ,
signal crosstalk due to wire bonds has been identified to be an important source of errors and methods to mitigate it have been developed @xcite . however
, crosstalk remains an open challenge and isolations ( opposite of crosstalk ) lower than are routinely observed when using wire bonds .
the coaxial design of the three - dimensional wires represents an advantage over wire bonds .
the latter , being open structures , radiate more electromagnetic energy that is transferred to adjacent circuits .
the former , being enclosed by the outer conductor , limit crosstalk due to electromagnetic radiation . in realistic applications of the quantum socket
, the three - dimensional wires must land in close proximity of several on - chip transmission lines . in order to study inter - channel isolation in such scenarios
, we designed a special device comprising a pair of cpw transmission lines , as shown in the inset of fig .
[ figure13:bejanin ] ( a ) .
one transmission line connects two three - dimensional wires ( ports @xmath68 and @xmath92 ) , exactly as for the devices studied in subsecs .
[ two : port : scattering : parameters ] and [ time : domain : reflectometry ] ; the other line , which also connects two three - dimensional wires ( ports @xmath193 and @xmath14 ) , circumvents the wire at port @xmath68 by means of a cpw semicircle . the distance between the semicircle and the wire outer conductor is designed to be as short as possible , @xmath194 . the chip employed for the crosstalk tests is similar to the ag sample used for the socket microwave characterization and was part of a dut analogous to that shown in fig .
[ figure08:bejanin ] .
the dc resistances of the center trace of the @xmath195 and @xmath196 transmission lines were measured and found to be @xmath197 and @xmath198 , respectively ( note that the @xmath196 transmission line is @xmath199 long ) .
all dc resistances to ground and between the two transmission lines were found to be on the order of a few kilohms , demonstrating the absence of undesired short circuit paths . a four - port calibration and measurement of the dut
were conducted by means of the ecal module and pna - x .
we selected a frequency range from to , @xmath200 , @xmath201 dbm , and @xmath202 . among the @xmath118 s - parameters , fig .
[ figure13:bejanin ] ( a ) shows the magnitude of the transmission coefficients @xmath165 and @xmath203 , along with the magnitude of the crosstalk coefficients @xmath204 , and @xmath205 .
the results show that the isolation in the typical qubit operation bandwidth , between and , is larger than @xmath206 .
note that the crosstalk coefficients shown in fig .
[ figure13:bejanin ] ( a ) include attenuation owing to the series resistance of the ag transmission lines . the actual isolation ,
due only to spurious coupling , would thus be smaller by a few decibels .
figure [ figure13:bejanin ] ( b ) shows the microwave simulations of the crosstalk coefficients , which agree reasonably well with the experimental results .
these simulations are based on the models explained in subsec .
[ microwave : simulations ] . from simulations
, we believe the isolation is limited by the crosstalk between the cpw transmission lines , instead of the three - dimensional wires .
note that the peaks at approximately correspond to an enhanced crosstalk due to a box mode in the microwave package .
the peaks appear in the simulations , which are made for a highly conductive package , and may appear in measurements performed below @xmath207 , when the al package becomes superconductive .
for the room temperature measurements shown in fig .
[ figure13:bejanin ] ( a ) , these peaks are smeared out due to the highly lossy al package .
-wave resonators .
the grounding washer , with its four protruding feet , is placed above the chip covering the chip edges .
the marks imprinted by the bottom interface of the three - dimensional wires on the al pads are noticeable .
more detailed images of these marks are shown in fig .
[ figure05:bejanin ] . ]
thus far , we have shown a detailed characterization of the quantum socket in dc and at microwave frequencies , both at room temperature and at . in order to demonstrate the quantum socket operation in a realistic quantum computing scenario
, we used a socket to wire a set of superconducting cpw resonators cooled to approximately in a dr .
we were able to show an excellent performance in the frequency range from to , which is the bandwidth of our measurement apparatus .
the experimental setup is described in appendix [ dilution : refrigerator : setup ] and shown in fig .
[ figure19:bejanin ] .
figure [ figure14:bejanin ] shows a macro photograph of a @xmath208 chip housed in the sample holder ; the chip is the al sample described in subsec .
[ alignment ] , with geometrical and dc electrical parameters reported in table [ table02:bejanin ] .
the sample comprises a set of three cpw transmission lines , each connecting a pair of three - dimensional wire pads ; multiple shunted cpw resonators are coupled to each transmission line . in this section , we will focus only on transmission line three and its five resonators .
the transmission line has a center conductor width of and gap width of , resulting in a characteristic impedance of approximately .
the resonators are @xmath209-wave resonators , each characterized by a center conductor of width @xmath70 and a dielectric gap of width @xmath210 .
the open end of the resonators runs parallel to the transmission line for a length @xmath211 , providing a capacitive coupling ; a ground section separates the gaps of the transmission line and resonators ( cf .
s2 of the supplemental material at http://www.supplemental-material-bejanin ) .
the nominal resonance frequency @xmath212 as well as all the other resonator parameters are reported in table [ table03:bejanin ] .
cccccccccccccc [ 3mm][0mm]@xmath213 & @xmath212 & @xmath40 & @xmath70 & @xmath210 & @xmath214 & @xmath215 & @xmath216 + [ 0mm][2mm](- ) & & & & & & & + [ 3mm][0mm]1 & @xmath217 & @xmath218 & @xmath219 & @xmath220 & @xmath221 & @xmath222 & @xmath223 + [ 3mm][0mm]2 & @xmath224 & @xmath225 & @xmath122 & @xmath226 & @xmath85 & @xmath227 & @xmath228 + [ 3mm][0mm]3 & @xmath229 & @xmath230 & @xmath231 & @xmath122 & @xmath221 & @xmath232 & @xmath233 + [ 3mm][0mm]4 & @xmath234 & @xmath235 & @xmath122 & @xmath226 & @xmath85 & @xmath236 & @xmath237 + [ 3mm][0mm]5 & @xmath238 & @xmath239 & @xmath219 & @xmath220 & @xmath221 & @xmath240 & @xmath241 + [ table03:bejanin ] a typical dr experiment employing the quantum socket consists of the following steps .
first , the sample is mounted in the microwave package , which has already been attached to the package holder ( cf .
[ package : holder ] and sec .
[ the : quantum : socket : implementation ] ) .
second , a series of dc tests is performed at room temperature .
the results for a few al samples are reported in table [ table02:bejanin ] .
third , the package holder assembly is characterized at room temperature by measuring its s - parameters .
the results of such a measurement are shown in fig .
[ figure15:bejanin ] ( a ) .
fourth , the package holder is mounted by means of the smp connectors to the mc stage of the dr and an @xmath165 measurement is performed .
the results ( magnitude only ) are shown in fig .
[ figure15:bejanin ] ( b ) in the frequency range between and .
fifth , the various magnetic and radiation shields of the dr are closed and the dr is cooled down .
sixth , during cooldown the @xmath165 measurement is repeated first at @xmath242 and , then , at the dr base temperature of approximately .
the results are shown in figs .
[ figure15:bejanin ] ( b ) and ( c ) , respectively . at @xmath242
we note the appearance of a shallow dip at approximately , probably due to a screw - in micro connector becoming sightly loose while cooling ( cf .
[ two : port : scattering : parameters ] ) .
it is important to mention that in the next generation of three - dimensional wires we will eliminate the screw - in micro connector , since we believe we found a technique to overcome the soldering issues detailed in subsec .
[ package : holder ] ( cf .
[ conclusions ] for a brief description ) . at the base temperature ,
all five resonators are clearly distinguishable as sharp dips on the relatively flat microwave background of the measurement network .
we then select a narrower frequency range around each resonator and make a finer @xmath165 measurement .
for example , fig .
[ figure15:bejanin ] ( d ) shows the magnitude and phase of the resonance dip associated with resonator number @xmath193 .
the s - parameters of each resonator were measured with the pna - x power set to m. considering that the total input channel attenuation at room temperature is @xmath243 at , the power at the resonator input is approximately m ( a few higher when cold ) .
the normalized inverse transmission coefficient @xmath244 was fitted as in ref .
this procedure makes it possible to accurately estimate both the internal @xmath216 and the rescaled coupling @xmath215 quality factors of a resonator .
the fit results are shown in table [ table03:bejanin ] .
the plot of the fits for the magnitude and phase of @xmath165 for resonator @xmath193 are overlaid with the measured data in fig .
[ figure15:bejanin ] ( d ) .
the real and imaginary parts of @xmath244 for the same resonator , as well as the associated fit , are shown in fig .
s3 in the supplemental material at http://www.supplemental-material-bejanin .
figure [ figure16:bejanin ] shows an extensible quantum computing architecture where a two - dimensional square lattice of superconducting qubits is wired by means of a quantum socket analogous to that introduced in this work .
the architecture comprises three main layers : the quantum hardware ; the shielding interlayer ; the three - dimensional wiring mesh . as shown in fig .
[ figure16:bejanin ] ( a ) , the quantum hardware is realized as a two - dimensional lattice of superconducting qubits with nearest neighbor interactions .
the qubits are a modified version of the xmon presented in ref .
each qubit is characterized by seven arms that make it possible to connect it to one xy and one z control line as well as one measurement resonator and four inter - qubit coupling resonators .
we name this type of qubit the _ heptaton_. the inter - qubit coupling is mediated by means of superconducting cpw resonators that allow the implementation of control z ( cz ) gates between two neighboring qubits @xcite .
a set of four heptatons can be readout by way of a single cpw transmission line connected to four cpw resonators , each with a different resonant frequency .
figure [ figure16:bejanin ] also shows the on - chip pads associated with each three - dimensional wire . in the supplemental material at http://www.supplemental-material-bejanin , we propose a more general surface code architecture where each qubit can be measured by means of two different resonators , one with frequency above and the other with frequency below all coupling resonator frequencies . assuming a pitch between two adjacent three - dimensional wires of
, the lateral dimension of one square cell having four heptatons at its edges is .
the three distances @xmath245 , @xmath246 , and @xmath247 between wire pads and resonators leading to this quantity are indicated in fig .
[ figure16:bejanin ] ( b ) .
it is thus possible to construct a two - dimensional lattice of @xmath1 heptatons on a square chip with lateral dimension @xmath248 .
a @xmath249 square chip is the largest chip that can be diced from a standard @xmath14 inch wafer
. this will allow the implementation of a logical qubit based on the surface code , with at least distance five @xcite .
in this architecture , the coupling resonators act as a _ coherent spacer _ between pairs of qubits , i.e. , they allow a sufficient separation to accommodate the three - dimensional wires , while maintaining qubit coherence during the cz gates . additionally , these resonators will help mitigate qubit crosstalk compared to architectures based on direct capacitive coupling between adjacent qubits ( cf . ref .
in fact , they will suppress qubit - mediated coupling between neighboring control lines @xcite .
it is worth noting that adjacent coupling resonators can be suitably designed to be at different frequencies , thus further diminishing qubit - mediated crosstalk .
implementing a large qubit chip with a lateral dimension of presents significant challenges to the qubit operation at microwave frequencies .
a large chip must be housed in a large microwave package , causing the appearance of box modes that can interfere with the qubit control and measurement sequences @xcite . moreover
, a large chip will inevitably lead to floating ground planes that can generate unwanted slotline modes @xcite .
all these parasitic effects can be suppressed by means of the shielding interlayer , as shown in fig .
[ figure16:bejanin ] ( a ) .
this layer can be wafer bonded @xcite to the quantum layer . through holes and cavities on the bottom part of the layer
can be readily fabricated using standard si etching techniques .
the holes will house the three - dimensional wires whereas the cavities will accommodate the qubit and resonator structures on the quantum hardware .
large substrates also generate chip modes that , however , can be mitigated using buried metal layers and through vias @xcite .
the three - dimensional wires to be used for the @xmath1 qubit architecture will be an upgraded version of the wires used in this work . in particular , the m@xmath16 thread will be removed and the wires will be inserted in a dedicated substrate ( cf . fig .
[ figure16:bejanin ] ( a ) ) ; additionally , the screw - in micro connector will be substituted by a direct connection to a subminiature push - on sub - micro ( smps ) connector ( not shown in the figure ) . in future applications of the quantum socket
, we envision an architecture where the three - dimensional wires will be used as interconnect between the quantum layer and a classical control / measurement layer .
the classical layer could be realized using rsfq digital circuitry @xcite .
for example , high - sensitivity digital down - converters ( ddcs ) have been fabricated based on rsfq electronics @xcite .
such circuitry is operated at very low temperatures and can substitute the room temperature electronics used for qubit readout .
note that cryogenic ddc chips with dimensions of less than @xmath250 can perform the same operations presently carried out by room temperature microwave equipment with an overall footprint of @xmath251 .
recent interest in reducing dissipation in rsfq electronics will possibly enable the operation of the classical electronics in close proximity to the quantum hardware .
we also believe it is feasible to further miniaturize the three - dimensional wires so that the wire outer diameter would be on the order of . assuming a wire - wire pitch also of
, it will therefore be possible to realize a lattice of @xmath252 wires connecting to @xmath253 qubits arranged on a @xmath254 two - dimensional qubit grid with dimensions of @xmath255 .
this will allow the implementation of simple fault tolerant operations between a few tens of logical qubits @xcite .
matteo mariantoni and the digital quantum matter laboratory acknowledge the alfred p. sloan foundation as well as the natural sciences and engineering research council of canada ( nserc ) , the ministry of research and innovation ( mri ) of ontario , and the canadian microelectronics corporation ( cmc ) microsystems .
corey rae h. mcrae acknowledges a waterloo institute for nanotechnology ( win ) nanofellowship-2013 and jrmy h. bjanin a provost s entrance award of the university of waterloo ( uw ) .
we also acknowledge fruitful discussions with john m. martinis and the martinis group , eric bogatin , and william d. oliver , as well as the group of adrian lupacu for their assistance in the deposition of al films and nathan nelson - fitzpatrick and the uw s quantum nanofab facility for their support .
we thank dario mariantoni and alexandra v. bardysheva for help with a few figures .
in this appendix , we discuss the pressure settings of the three - dimensional wires . in the current implementation of the quantum socket
, the pressure exerted by the three - dimensional wires on the chip is controlled by the installation depth of the wire in the lid .
this depth depends on the number of rotations used to screw the wire into the m@xmath16-threaded hole of the lid .
since the wire s tunnel has to be aligned with the corresponding on - chip pad , a discrete number of wire pressure settings is allowed .
for the package shown in fig .
[ figure01:bejanin ] ( b ) and fig . [ figure04:bejanin ] ( b ) , the minimum length an unloaded wire has to protrude from the ceiling of the lid s internal cavity to touch the chip surface is @xmath256 ( cf .
[ figure01:bejanin ] ( c ) ) . for a maximum wire stroke @xmath257 ,
the maximum length an unloaded wire can protrude from the cavity ceiling without breaking when loaded is @xmath258 .
the first allowed pressure setting , with wire and pad perfectly aligned , is for @xmath259 .
the pitch for an m@xmath16 screw is .
hence , five pressure settings are nominally possible , for @xmath260 mm , with @xmath261 .
we found the ideal pressure setting to be for @xmath262 , corresponding to a nominal @xmath263 mm ; the actual average setting for @xmath264 wires was measured to be @xmath265 , with standard deviation due to the machining tolerances . for greater depths we experienced occasional wire damage ; lesser depths were not investigated .
possible effects on the electrical properties of the three - dimensional wires due to different pressure settings will be studied in a future work .
cccccccccc [ 3mm][0mm]material & cu & sn & zn & fe & ni & pb & p & si & others + [ 3mm][0mm]cw724r & @xmath266 & @xmath267 & rest & @xmath267 & @xmath268 & @xmath269 & @xmath270 & @xmath271 & al @xmath272 , manganese @xmath272 + [ 3mm][0mm]cw453k & rest & @xmath273 & @xmath274 & @xmath275 & @xmath274 & @xmath276 & @xmath277 & - & @xmath268 + [ table04:bejanin ] in this appendix , we describe the measurement setup used to characterize the magnetic properties of the materials used in the quantum socket and present the main results . additionally , we give an estimate of the strength of the magnetic field caused by one three - dimensional wire inside the microwave package .
the zgc used in our tests comprises three nested cylinders , each with a lid with a central circular hole ; the hole in the outermost lid is extended into a chimney that provides further magnetic shielding .
the walls of the zgc are made of an alloy of ni and fe ( or mu - metal alloy ) with a high relative permeability @xmath278 .
the alloy used for the chamber is a co - netic aa alloy and is characterized by a dc magnetic permeability at , @xmath279 , and an ac magnetic permeability at and at , @xmath280 . as a consequence , the nominal magnetic field attenuation lies between @xmath281 and @xmath282 .
the zgc used in our tests was manufactured by the magnetic shield corporation , model zg-209 .
the flux gate magnetometer used to measure the magnetic field @xmath283 is a three - axis dc milligauss meter from alphalab , inc .
, model mgm3axis .
its sensor is a @xmath284 parallelepiped at the end of a @xmath285 long cable ; the orientation of the sensor is calibrated to within and has a resolution of ( i.e. , ) over a range of @xmath286 ( i.e. , @xmath287 ) .
the actual attenuation of the chamber was tested by measuring the value of the earth s magnetic field with and without the chamber in two positions , vertical and horizontal ; inside the chamber the measurements were performed a few centimeters from the chamber base , approximately on the axis of the inner cylinder . in these and all subsequent tests , the magnetic sensor was kept in the same orientation and position .
the results are reported in table [ table05:bejanin ] , which shows the type of measurement performed , the magnitude of the measured magnetic field @xmath288 , and the attenuation ratio @xmath289 .
the maximum measured attenuation was @xmath290 in the horizontal position .
ccc [ 3mm][0mm]measurement & @xmath288 & @xmath289 + [ 0mm][2 mm ] & & + [ 3mm][0mm]vertical position , background field & @xmath291 & - + [ 3mm][0mm]vertical position , with zgc & @xmath292 & @xmath293 + [ 3mm][0mm]horizontal position , background field & @xmath294 & - + [ 3mm][0mm]horizontal position , with zgc & @xmath295 & @xmath296 + [ table05:bejanin ] the zgc characterization of table [ table05:bejanin ] also serves as a calibration for the measurements on the materials used for the quantum socket . in these measurements , each test sample was positioned approximately away from the magnetic sensor .
the results , which are reported in table [ table06:bejanin ] , were obtained by taking the magnitude of the calibrated field of each sample .
the calibrated field itself was calculated by subtracting the background field from the sample field , component by component .
note that the background and sample fields were on the same order of magnitude ( between and ) , with background fluctuations on the order of .
thus , we recorded the maximum value of each @xmath297 , @xmath298 , and @xmath299 component . considering that the volume of the measured samples is significantly larger than that of the actual quantum socket components , we are confident that the measured magnetic fields of the materials should be small enough not to significantly disturb the operation of superconducting quantum devices . as part of our magnetism tests , we measured a block of approximately of 5n5 al in the zgc ; as shown in table [ table06:bejanin ]
, the magnitude of the magnetic field was found to be within the noise floor of the measurement apparatus and a pull of .
we found magnetic fields with the same order of magnitude as in table [ table06:bejanin ] . ] .
cc [ 3mm][0mm]material & @xmath288 + [ 0mm][2 mm ] & + [ 3mm][0mm]cw724r & @xmath300 + [ 3mm][0mm]cw453k & @xmath301 + [ 3mm][0mm]al 5n5 & @xmath302 + [ table06:bejanin ] a simple geometric argument allows us to estimate the actual magnetic field due to one three - dimensional wire , without taking into account effects due to superconductivity ( most of the wire is embedded in an al package , which is superconductive at qubit operation temperatures ) .
we assume that one wire generates a magnetic field of @xmath303 mg ( i.e. , the maximum field value in table [ table06:bejanin ] ; this is a large overestimate considering the tested samples had volumes much larger than any component in the wires ) and is a magnetic dipole positioned @xmath122 mm away from a qubit . the field generated by the wire at the qubit
will then be @xmath304 , where @xmath305 is the distance at which the field was measured in the zgc ; thus , @xmath306 .
assuming an xmon qubit @xcite with a superconducting quantum interference device ( squid ) of dimensions @xmath307 , the estimated magnetic flux due to the wire threading the squid is @xmath308 .
this is approximately three orders of magnitude smaller than a flux quantum @xmath309 ; typical flux values for the xmon operation are on the order of @xmath310 .
in this appendix , we describe the method used to estimate the thermal performance of a three - dimensional wire and compare it to that of an al wire bond .
note that at very low temperature , thermal conductivities can vary by orders of magnitude between two different alloys of the same material .
the following estimate can thus only be considered correct to within approximately one order of magnitude .
thermal conductivity is a property intrinsic to a material . to characterize the cooling performance of a three - dimensional wire , we instead use the heat transfer rate ( power ) per kelvin difference , which depends on the conductivity .
the power transferred across an object with its two extremities at different temperatures depends on the cross - sectional area of the object , its length , and the temperature difference between the extremities . since the cross section of a three - dimensional wire is not uniform , we assume the wire is made of two concentric hollow cylinders .
the cross - sectional area of the two cylinders is calculated by using dimensions consistent with those of a three - dimensional wire .
the inner and outer hollow cylinders are assumed to be made of phosphor bronze and brass alloys , respectively .
the thermal conductivities of these materials at low temperatures are determined by extrapolating measured data to .
the al wire bonds are assumed to be solid cylinders with diameter . in the superconducting state
, the thermal conductivity of al can be estimated by extrapolating literature values @xcite .
the heat transfer rate per kelvin difference is calculated by multiplying the thermal conductivity @xmath311 with the cross - sectional area @xmath245 and dividing by the length of the thermal conductor @xmath312 .
the heat transfer rate per kelvin difference of a three - dimensional wire is calculated by summing the heat transfer rate per kelvin difference of the inner conductor to that of the outer conductor and is found to be @xmath53 at . at the same temperature , the heat transfer rate per kelvin of a typical al wire bond
is estimated to be @xmath313 ( cf .
table [ table07:bejanin ] ) , much lower than for a single three - dimensional wire .
note that , instead of al wire bonds , gold wire bonds can be used .
these are characterized by a higher thermal conductivity because they remain normal conductive also at very low temperatures .
however , al wire bonds remain the most common choice because easier to use .
ccccc [ 3mm][0 mm ] & @xmath314 & @xmath315 & @xmath245 & @xmath311 + [ 0mm][2 mm ] & & & & + [ 3.5mm][0mm]inner cond .
& @xmath316 & @xmath317 & @xmath318 & @xmath319 + [ 3.5mm][0mm](phos bronze ) & & & & + [ 3.5mm][0mm]outer cond . &
@xmath320 & @xmath321 & @xmath322 & @xmath323 + [ 3.5mm][0mm](brass ) & & & & + [ 3.5mm][0mm]wire bond & - & @xmath109 & @xmath324 & @xmath325 + [ 3.5mm][0mm](al ) & & & & + [ table07:bejanin ]
cccccc [ 3mm][0mm]spring type & @xmath326 & @xmath71 & @xmath327 & @xmath328 & @xmath329 + [ 0mm][2 mm ] & & & & & + [ 3mm][0mm]fe-113 225 & @xmath330 & @xmath331 & @xmath332 & @xmath333 & @xmath334 + [ 3mm][0mm]fe-112 157 & @xmath335 & @xmath336 & @xmath337 & @xmath338 & @xmath334 + [ 3mm][0mm]fe-50 15 & @xmath339 & @xmath340 & @xmath341 & @xmath342 & @xmath343 + [ table08:bejanin ] in this appendix , we discuss the performance of the springs used in three - dimensional wires at various temperatures .
the three types of tested springs are called fe-113 225 , fe-112 157 , and fe-50 15 and their geometric characteristics are reported in table [ table08:bejanin ] .
we ran temperature cycle tests by dunking the springs repeatedly in liquid nitrogen and then in liquid helium without any load . at the end of each cycle
, we attempted to compress them at room temperature .
we found no noticeable changes in mechanical performance after many cooling cycles .
subsequently , the springs were tested mechanically by compressing them while submerged in liquid nitrogen or helium .
the setup used for the compressive loading test of the springs is shown in movie 4 of the supplemental material at http://www.supplemental-material-bejanin , which also shows a properly functioning spring immediately after being cooled in liquid helium . in these tests , we only studied compression forces because in the actual experiments the three - dimensional wires are compressed and not elongated .
the compression force was assessed by means of loading the springs with a mass .
the weight of the mass that fully compressed the spring determined the spring compression force @xmath329 .
the compression force of each spring is reported in table [ table08:bejanin ] .
we observed through these tests that the compression force is nearly independent of the spring temperature , increasing only slightly when submerged in liquid helium .
assuming an operating compression @xmath344 , we expect a force between and for the inner conductor and between and for the outer conductor of a three - dimensional wire at a temperature of
. note that we chose spring model fe-113 225 for use with the grounding washer .
in this appendix , we provide more details about alignment errors .
figure [ figure17:bejanin ] shows a set of micro images for au and ag samples . the au pads in panels ( a ) and ( b ) were mated two times at room temperature ; the three - dimensional wires used to mate these pads featured the smaller tunnel ( width ) .
the pad dimensions were @xmath345 and @xmath346 .
noticeably , in panel ( a ) the wire bottom interface matched the contact pad in both mating instances , even though the matching was affected by a rotational misalignment of approximately @xmath347 with respect to the transmission line longitudinal axis .
in panel ( b ) the inner conductor landed on the dielectric gap in the second mating instance . in our initial design , a perfect match required that the die dimensions should be at most @xmath68 thou smaller than the dimensions of the chip recess , as machined . in the case of
the sample holder used to house the au samples , the chip recess side lengths were , , , and .
the au samples were diced from a si wafer using a dicing saw from disco , model dad-2h/6 , set to obtain a @xmath208 die . due to the saw inaccuracies ,
the actual die dimensions were @xmath348 , significantly smaller than the chip recess dimensions .
this caused the die to shift randomly between different mating instances , causing alignment errors .
as described in the main text , in order to minimize such errors a superior disco saw was used , in combination with a disco electroformed bond hub diamond blade model zh05-sd 2000-n1 - 50-f e ; this blade corresponds to a nominal kerf between and .
additionally , we used lateral markers spaced with increments of that allowed us to cut dies with dimensions ranging from to , well within the machining tolerances of the sample holder .
after machining , the actual inner dimensions of each sample holder were measured by means of a measuring microscope .
the wafers were then cut by selecting the lateral dicing markers associated with the die dimensions that fit best the holder being used .
figure [ figure17:bejanin ] ( c ) shows a successful alignment for six ag pads on the same chip ; the chip is mounted in a sample holder with grounding washer .
all three main steps for an ideal and repeatable alignment ( cf .
subsec . [ alignment ] ) were followed .
figure [ figure17:bejanin ] ( d ) shows the distinctive marks left by the grounding washer on an ag film .
the marks are localized towards the edge of the die ; the washer covered approximately of ag film .
this indicates a good electrical contact at the washer - film interface . in conclusion ,
it is worth commenting some of the features in fig .
[ figure05:bejanin ] ( d ) in the main text .
the figure clearly shows dragging of a three - dimensional wire due to cooling contractions .
in fact , for the al chip recess an estimate of the lateral contraction length from room temperature to @xmath349 can be obtained as @xmath350 , where @xmath351 is the integrated linear thermal expansion coefficient for al 6061-t6 at from refs .
@xcite and @xmath352 is the room temperature length of the recess side can be accurately estimated from the data at http://www.cryogenics.nist.gov/mpropsmay/6061%20aluminum/6061_t6aluminum_rev.htm . ] .
note that the sample holder is actually made from al alloy 5n5 ; however , different al alloys contract by approximately the same quantity . for the si sample substrate , the lateral contraction length from room temperature to @xmath349
is approximately given by @xmath353 , where the integrated linear thermal expansion coefficient at was found in table 2 of ref .
below , the thermal expansion of both materials is negligible for our purposes and , thus , the estimate can also be considered to be valid at @xmath58 .
in this appendix , we outline the fabrication processes for the samples used to test the quantum socket .
a set of samples was made by liftoff of a @xmath354 ag film , which was grown by means of electron beam physical vapor deposition ( ebpvd ; from intlvac canada inc .
, model nanochrome ii ) on a @xmath193 inch float - zone ( fz ) si @xmath355 wafer of thickness .
the superconducting samples were made by etching a @xmath356 al film that was deposited by ebpvd on a fz si wafer .
last , two sets of test samples were made by etching au films of thickness and with a ti adhesion underlayer in both sets .
the films were grown by ebpvd on a @xmath193 inch czochralski ( cz ) undoped si @xmath355 wafer of thickness .
the ag samples were required to reduce the series resistance of the cpw transmission lines ( cf .
subsecs .
[ two : port : scattering : parameters ] , [ time : domain : reflectometry ] , and [ signal : crosstalk ] ) .
fabricating such a relatively thick film necessitated a more complex process as compared to that used for the au and al samples .
the ag samples were fabricated with a thick resist tone reversal process .
the wafer was spun with an az p4620 positive tone resist to create a resist thickness of @xmath357 , then soft baked for at . because the resist layer is so thick , a rehydration step of was necessary before exposure .
optical exposure was performed for in a mask aligner from sss microtec ag , model ma6 , in soft contact with a photomask .
after exposure the sample was left resting for at least so that any nitrogen created by the exposure could dissipate .
the tone reversal bake was done for in an oven set to , filled with ammonia gas .
the sample then underwent a flood exposure for and was developed in az 400k for .
subsequently , of ag was deposited and liftoff of the resist was performed in acetone for with ultrasounds .
in this appendix , we present a set of microwave parameters that help further analyze the performance of the quantum socket . these parameters were obtained from the measured s - parameter data of figs .
[ figure09:bejanin ] and [ figure10:bejanin ] ( a ) and are shown in fig .
[ figure18:bejanin ] .
the complex input impedance can be obtained from the frequency dependent impedance matrix @xmath358$]as @xmath359 where @xmath360 is the load impedance .
the impedance matrix was obtained using the measured complex s - parameter matrix @xmath361 $ ] from @xmath362 the magnitude of @xmath363 is shown in fig .
[ figure18:bejanin ] ( a ) .
the input voltage standing wave ratio ( vswr ) was obtained from @xcite @xmath364 and is displayed in fig .
[ figure18:bejanin ] ( b ) .
the phase delay was calculated as @xmath365 and is displayed fig .
[ figure18:bejanin ] ( c ) .
finally , the group delay was obtained from @xcite @xmath366 and is displayed in fig .
[ figure18:bejanin ] ( d ) .
the derivative in eq .
( [ equation:08 ] ) was evaluated numerically by means of central finite differences with 6th order accuracy . the data in fig .
[ figure18:bejanin ] ( d ) were post - processed using @xmath367 smoothing .
note that the output impedance and vswr were also evaluated and resembled the corresponding input parameters .
the input and output impedances as well as the vswrs indicate a good impedance matching up to approximately .
the phase and group delays , which are directly related to the frequency dispersion associated with the quantum socket , indicate minimal dispersion .
this is expected for a combination of coaxial structures ( the three - dimensional wires ) and a cpw transmission line .
thus , we expect wideband control pulses to be transmitted without significant distortion in applications with superconducting qubits ( cf . supplemental material at http://www.supplemental-material-bejanin for further details about microwave pulse transmission ) .
the experimental setup used to measure the superconducting cpw resonators is shown in fig .
[ figure19:bejanin ] .
the low - temperature system is a cryogen - free dr from bluefors cryogenics ltd . , model bf - ld250 .
the dr comprises five main temperature stages , where microwave components and samples can be thermally anchored : the rt , , , still ( @xmath368 ) , cold plate ( cp ; @xmath369 ) , and mc stage .
we will describe the setup following the input signal through the various temperature stages , from port @xmath68 to the input port of the microwave package ( where the resonator sample is mounted ) and from the output port of the package to port @xmath92 .
the two ports are connected to the pna - x , which serves as both the microwave source and readout apparatus .
port @xmath68 is connected to the rt stage of the dr with sucoflex flexible cables followed by a series of two semi - rigid coaxial cables from ez form , model ez 86-cu - tp / m17 ( each approximately long , with silver - coated copper center conductor , solid ptfe dielectric , and tin - plated seamless copper outer conductor ) . except for the pna - x ports , which feature connectors , all the connectors and bulkhead adapters
are sma type .
in particular , the rt stage of the dr features a set of hermetic sma bulkhead adapters from huber+suhner , model 34_sma-50 - 0 - 3/111_n , with a tested leak rate for helium-4 lower than .
the dr stages , all the way to the mc stage , are connected by the series of five semi - rigid coaxial cables from coax co. , ltd .
, model sc-219/50-ss - ss ( with stainless steel ( sus304 ) center and outer conductor and solid ptfe dielectric ; the cable lengths from rt to mc are : , , , , and , respectively ) .
the cables are thermalized to the dr stages by way of cryogenic attenuators from the xma corporation - omni spectra , model 2082 - 6418-xx - cryo , where xx is the attenuation level in db ; for each stage between rt and mc , we chose @xmath370 , and @xmath371 , respectively .
the input signals are filtered by means of a low - pass filter from marki microwave , inc . , model flp-0960 - 2s , with bandpass from dc to .
the filter is heat sunk at the mc stage by anchoring it to a hardware module , which is bolted to the mc stage .
the filter module , and similarly all the other modules used to heat sink microwave components in the dr , are made from c10100 ofe copper alloy .
a non - magnetic semi - rigid coaxial cable ez 86-cu - tp / m17 connects the output port of the marki filter to an smp 086 connector on the mounting plate ; the cable is long and enters the a4k shield through one of the chimneys on the shield lid .
the shield , which is thermalized to the mc stage , is characterized by a dc relative permeability close to @xmath372 at .
the smp 086 connector is mated to the input port of the dut shown schematically in fig .
[ figure08:bejanin ] .
the dut used in the dr features smp 047 connectors in lieu of sma connectors .
the dut when connected to the mounting plate is shown in fig .
[ figure01:bejanin ] ( d ) and fig . [ figure04:bejanin ] ( c ) .
the output port of the dut is then connected to a series of two cryogenic circulators from raditek inc .
, model radc-4.0 - 8.0-cryo-4 - 77k - s3 - 1wr - b ( with magnetic shielding ) by means of a semi - rigid superconducting coaxial cable from coax co. , model sc-219/50-nb - nb , of length .
the circulators are thermalized to the mc stage and are connected to each other by means of a semi - rigid superconducting coaxial cable from coax co. , model sc-219/50-nb - nb , of length ; the spare port of each circulator is terminated with an xma cryogenic load , model 2001 - 7010 - 02-cryo , which is thermalized to the mc stage .
the output port of the second circulator is connected by way of a long sc-219/50-nb - nb cable to a third circulator at the still stage ( the spare port is terminated with a load thermalized to the still ) . a long sc-219/50-nb - nb cable connects the output port of the third circulator to a cryogenic microwave amplifier from low noise factory ab , model lnf - lnc1_12a .
the amplifier , which is thermalized to the stage , is characterized by a nominal gain of approximately and a noise temperature of at an operating temperature of in the to frequency range .
finally , the amplifier output port is connected to the and rt stages by a series of two sc-219/50-ss - ss cables of length and , respectively ; the cables are thermalized to the stage by means of a xma attenuator .
two ez form copper cables in series , followed by sucoflex flexible cables , complete the network to port @xmath92 .
the input channel described here is one of three equivalent channels dedicated to resonator measurements .
the three channels share the output line ; this is possible thanks to a microwave switch from radiall , model r573.423.605 , which is operated at the mc stage .
the switch is located after the dut but before the two mc circulators ( two of the three input channels and the switch are not shown in fig .
[ figure19:bejanin ] ) .
the switch has six inputs and one output , making it possible to further extend the number of input microwave channels .
84ifxundefined [ 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 https://arxiv.org/abs/1510.04375 [ `` , '' ] ( ) , link:\doibase 10.1098/rspa.1985.0070 [ * * , ( ) ]
link:\doibase 10.1038/nature07128 [ * * , ( ) ] link:\doibase 10.1126/science.1231930 [ * * , ( ) ] link:\doibase 10.1038/npjqi.2015.5 [ * * , ( ) ] in link:\doibase 10.1117/12.926491 [ _ _ ] , vol . , ( , ) p. https://www.google.com/patents/us20140264287 [ `` , '' ] ( ) , link:\doibase 10.1063/1.4935541 [ * * , ( ) ] http://meetings.aps.org/link/baps.2016.mar.x48.3 [ `` , '' ] ( ) , http://meetings.aps.org/link/baps.2016.mar.x48.4 [ `` , '' ] ( ) , link:\doibase 10.1038/npjqi.2016.2 [ * * , ( ) ]
link:\doibase 10.1126/science.270.5234.255 [ * * , ( ) ] link:\doibase 10.1017/cbo9780511976667 [ _ _ ] ( , , ) \doibase 10.1002/1521 - 3978(200009)48:9/11<771::aid - prop771>3.0.co;2-e [ * * , ( ) ] link:\doibase 10.1017/cbo9780511813870 [ _ _ ] ( , ) link:\doibase 10.1038/nature08812 [ * * , ( ) ] link:\doibase 10.1103/physreva.83.012303 [ * * , ( ) ] link:\doibase 10.1038/npjqi.2015.23 [ * * , ( ) ] link:\doibase 10.1090/psapm/068/2762145 [ `` , '' ] ( ) link:\doibase 10.1103/physrevlett.98.190504 [ * * , ( ) ] link:\doibase 10.1103/physreva.86.032324 [ * * , ( ) ] in link:\doibase 10.1109/sfcs.1994.365700 [ _ _ ] , ( , , ) pp .
link:\doibase 10.1103/revmodphys.79.135 [ * * , ( ) ] link:\doibase 10.1038/nphoton.2009.229 [ * * , ( ) ] link:\doibase 10.1126/science.1231298 [ * * , ( ) ] in link:\doibase 10.1364/ls.2006.ltui3 [ _ _ ] ( , ) p. link:\doibase 10.1103/physrevx.2.031007 [ * * , ( ) ] link:\doibase 10.1103/revmodphys.79.1217 [ * * , ( ) ] link:\doibase 10.1038/nature10707 [ * * , ( ) ] link:\doibase 10.1103/revmodphys.85.961 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.112.026801 [ * * , ( ) ] link:\doibase 10.1038/npjqi.2015.19 [ * * , ( ) ] link:\doibase 10.1038/nature13171 [ * * , ( ) ] link:\doibase 10.1038/ncomms7979 [ * * , ( ) ] link:\doibase 10.1038/ncomms7983 [ * * , ( ) ] link:\doibase 10.1038/nature14270 [ * * , ( ) ] \doibase http://dx.doi.org/10.1063/1.4886408 [ * * , ( ) ] link:\doibase 10.1109/6.887595 [ * * , ( ) ] link:\doibase 10.1109/tasc.2010.2096792 [ * * , ( ) ] in link:\doibase 10.1109/arftg74.2009.5439097 [ _ _ ] ( , ) pp .
link:\doibase 10.1109/9780470544662 [ _ _ ] ( , , ) http://ca.wiley.com/wileycda/wileytitle/productcd-ehep002016.html[__ ] ( , , ) link:\doibase 10.1088/0953 - 2048/24/6/065001 [ * * , ( ) ] link:\doibase 10.1002/0471224758 [ _ _ ] ( , , ) link:\doibase 10.1109/tasc.2005.850084 [ * * , ( ) ] link:\doibase 10.1103/physrevb.79.174512 [ * * , ( ) ] link:\doibase 10.1063/1.3271523 [ * * , ( ) ] link:\doibase 10.1063/1.3693409 [ * * , ( ) ] link:\doibase 10.1063/1.3658630 [ * * , ( ) ] link:\doibase 10.1103/physreva.89.032316 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.111.080502 [ * * , ( ) ] in link:\doibase 10.1109/relphy.2004.1315388 [ _ _ ] ( , ) pp . in link:\doibase 10.1007/0 - 306 - 47112 - 4_84
[ _ _ ] ( , ) pp .
@noop _ _ ( , , ) http://patents.justia.com/patent/20140274725 [ `` , '' ] ( ) https://www.google.com/patents/us20140266406 [ `` , '' ] ( ) , link:\doibase 10.1103/physrevb.82.024514 [ * * , ( ) ] link:\doibase 10.1126/science.1208517 [ * * , ( ) ] link:\doibase 10.1103/physrevb.78.104508 [ * * , ( ) ] link:\doibase 10.1109/tasc.2010.2095399 [ * * , ( ) ] link:\doibase 10.1063/1.1135452 [ * * , ( ) ] link:\doibase 10.1063/1.555681 [ * * , ( ) ] link:\doibase 10.1063/1.3638063 [ * * , ( ) ] * supplemental material for `` the quantum socket : three - dimensional wiring for extensible quantum computing '' *
this supplemental material is organized as follows . in sec . [ animated : electric : field : simulations ] , we show a set of simulated electric field animations . in sec .
[ screw : in : micro : connector : anomaly ] , we characterize the microwave anomaly due to the screw - in micro connector . in sec . [
superconducting : resonators : coupling : and : fit ] , we discuss in detail the geometry of the coupling region between a cpw transmission line and a superconducting resonator ; additionally , we show the fit required to extract the internal quality factor of the resonators . in sec .
[ spring : tests ] , we present a movie of the cryogenic spring tests . in sec .
[ quantum : socket : surface : code ] , we propose a surface code architecture that makes use of two readout resonators per qubit . finally , in sec .
[ microwave : pulse : transmission ] , we present the quantum socket behavior when fast microwave pulses are applied .
the electric fields discussed in subsec .
ii d of the main text were generated for multiple phases , enabling a time - domain analysis .
this made it possible to better determine the behavior of the simulated fields associated with the various components of the quantum socket .
the time - domain animations of the simulated electric field for the three - dimensional wire , the transition between the wire and the on - chip pad , and the first box mode are shown in movies 1 , 2 , and 3 , respectively .
the simulation of the three - dimensional wire shows a clear impact to the electric field at the dielectric spacers , giving an indication of where potential future improvements of the wire should focus .
the simulation of the transition reveals a surprisingly clean transmission , with no immediately apparent regions of poor performance .
the broad field distribution below the contact pad does suggest the appearance of a stray capacitance to the cavity below . however , the tdr results in subsec .
iv c indicate that such a capacitance must be small because of a very good impedance matching at the transition region .
the box mode animation shows the `` pinching '' effect to the electric field caused by the metallic pillar , leading to a significantly higher relative permittivity and , in turn , a much lower frequency of the first box mode ( @xmath373 ) .
some other regions of interest in movie 3 are the outer edges of the chip recess , where the field becomes fully confined within the substrate .
as the magnitude of the field in this region is quite low , this effect should have only a marginal impact on the performance of the quantum socket .
the animation also reveals some field interaction between the box mode and the contact pads ( which are characterized by a relatively large area ) , suggesting these as the main reason for any spurious coupling between on - chip structures and box modes .
; the black dashed lines delimit the frequency region between and .
the dip at approximately corresponds to the loose micro - connector . ] in order to gain further insight into the electrical behavior of the screw - in micro connector , we report here a set of room temperature measurements showing the appearance of a microwave anomaly associated with the micro connector .
these measurements complement the discussion associated with fig .
9 ( b ) in subsec .
iv b of the main text .
figure [ figure01s : bejanin ] shows the magnitude and unwrapped phase angle of the s - parameters as a function of frequency for an au sample in the case of a slightly loose screw - in micro connector ; the measurement was performed at room temperature ( cf .
iv b of the main text for details on the dut used for the measurements ) .
the microwave dip is centered at approximately and has a bandwidth of approximately .
the dip is likely not a lorentzian - type feature due to a resonance mode in the package .
this is clearly demonstrated by the phase angle of the coefficients @xmath165 and @xmath374 , which has the characteristic frequency dependence of a transmission line ( i.e. , no phase shift associated with a resonance ) .
note that a similar behavior is encountered when performing a microwave measurement with , e.g. , a not well torqued sma connector .
as expected from energy conservation , the dip in @xmath165 corresponds to a peak in @xmath375 ; however , the absence of a peak in @xmath376 indicates that only the micro connector at the dut input line was loose in this instance . as explained in the main text ,
curing the dip at room temperature is rather straightforward .
however , it is much harder at low temperatures .
hence , we decided to cool down a dut without curing the dip that appeared at room temperature .
a set of data at was taken immediately after the room temperature data , as shown in fig .
9 ( b ) in subsec .
iv b of the main text .
we monitored the s - parameters throughout the cooldown , observing how the dip shifted from approximately to while cooling .
it is reasonable to assume that , when cooling down , the micro connector became slightly more loose due to thermal contractions , thus giving rise to a microwave transmission dip at lower frequency .
figure [ figure02s : bejanin ] shows a detail of the coupling region between one of the cpw transmission lines and a @xmath209-wave resonator shown in fig .
14 of the main text . the width and gap of the cpw transmission line are indicated .
the resonator center conductor width @xmath70 and dielectric gap @xmath210 are also indicated , as well as the coupling length @xmath211 and the line - resonator ground separation ( ) . the length @xmath211 determines how strongly the resonator is coupled to the transmission line and , thus , the resonator external quality factor @xcite .
-wave resonator .
white surfaces represent conductive material and blue surfaces dielectrics . ]
figure [ figure03s : bejanin ] shows a polar plot of the real and imaginary part of the normalized inverse transmission coefficient @xmath244 of the data shown in fig .
15 ( d ) of the main text , along with the fit .
the equation used for the fit is @xcite @xmath377 where @xmath378 is an offset angle , @xmath379 , and @xmath380 .
a normalization is applied before the fit to set the off - resonance transmission magnitude and phase to and , respectively .
the fit parameters for the resonator in fig . 15 ( d ) of the main text were found to be @xmath381 , @xmath382 , @xmath383 , and @xmath384 . for superconducting resonator @xmath193 discussed in the main text .
data in blue and fit from eq .
( [ equation:01s ] ) in red . ]
as discussed in appendix d of the main text , movie 4 shows the setup used to test the springs at low temperature . in the movie ,
two springs mounted in series are compressed , after having been submerged in liquid helium .
this process was repeated multiple times , without any damage occurring to the spring or any measurable change in the spring constant .
the same setup also made it possible to preform compression measurements while the springs were submerged in a cryogenic liquid ( nitrogen or helium ) ; the compressive force was applied at the top of the rod , which always remained at room temperature . , and @xmath220 .
the cz gates follow the same numeration @xcite .
measurement qubits are indicated in red and blue and data qubits in black .
( b ) extended two - dimensional lattice . ]
figure [ figure04s : bejanin ] shows a more general version of the surface code architecture in fig .
16 of the main text . in panel
( a ) , numbers from @xmath385 to @xmath220 represent arbitrary frequencies from low to high ( @xmath385 means a lower frequency than the others ) .
every qubit is coupled to two measurement resonators ; note that label @xmath385 is associated with a cpw transmission line that is coupled to a set of four measurement resonators , all at slightly different frequencies ( to allow multiplexing ) centered around frequency @xmath385 .
similarly , label @xmath220 is associated with a cpw transmission line that is coupled to a set of four measurement resonators , all at slightly different frequencies centered around frequency @xmath220 . in this architecture ,
all qubits start at a low frequency then increase in frequency in unison to perform cz gates through the coupling resonators in the correct sequence ; in the figure , this sequence is indicated by the labels ( coupling frequencies ) @xmath386 , and @xmath14 .
finally , the measurement qubits ( red and blue ) are readout at frequency @xmath220 .
this is one surface code cycle @xcite .
the process is then reversed and during the downward uniform frequency sweep again an appropriate sequence of cz interactions is executed , followed by readout at frequency @xmath385 , implementing the next cycle .
this process avoids sweeping either measurement or data qubits past resonators during the cycles .
panel ( b ) shows an extended portion of the surface code lattice .
in appendix g of the main text , we have shown that the quantum socket is characterized by a small frequency dispersion . in this section ,
we describe a set of experiments where actual fast microwave pulses were applied to the quantum socket as well as a reference cable .
this makes it possible to assess the effects of dispersion directly on pulses similar to those that will be used for the manipulation of superconducting qubits .
figure [ figure05s : bejanin ] shows the magnitude of the fourier transform of a pulse transmitted through the reference cable ( panel ( a ) ) and the quantum socket ( panel ( b ) ) , both at room temperature .
the pulse was generated by mixing a sinusoidal carrier signal at with a gaussian - modulated sinusoidal pulse at .
the time length of the gaussian envelope was ( full width at half maximum ( fwhm ) ) .
the carrier signal was generated by means of a microwave source from keysight , model e8257d psg ; the mixer was an in - phase and quadrature ( iq ) mixer from marki , model iq-0307lxp ; the modulated gaussian was generated by means of a customized arbitrary waveform generator ( awg ) from tabor electronics ltd .
, model wx2184c+ ; all instruments were synchronized by way of a rubidium atomic timebase embedded within a digital delay generator from stanford research systems , inc .
, model dg645/15 , which was also used to trigger the awg through an srd1 module from stanford research systems .
the generated signal was split with a microwave power divider from krytar , inc .
, model 6005180 , with one output sent through a reference sucoflex flexible coaxial cable from huber+suhner of length and the other output through the quantum socket ( dut of fig . 8 of the main text ) .
the output of each path was acquired and digitized using a real - time oscilloscope from keysight , model dso91204a , also synchronized with the rubidium timebase .
the pulse going through the quantum socket is attenuated due to the resistance of the ag cpw transmission line , hence , the magnitude of the fourier transform is lower than of the reference signal .
however , the overall shape of the two pulses is very similar , with marginal distortions .
3ifxundefined [ 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 10.1109/tasc.2005.850084 [ * * , ( ) ] link:\doibase 10.1063/1.3693409 [ * * , ( ) ] link:\doibase 10.1103/physreva.86.032324 [ * * , ( ) ] | quantum computing architectures are on the verge of scalability , a key requirement for the implementation of a universal quantum computer .
the next stage in this quest is the realization of quantum error correction codes , which will mitigate the impact of faulty quantum information on a quantum computer .
architectures with ten or more quantum bits ( qubits ) have been realized using trapped ions and superconducting circuits . while these implementations are potentially scalable , true scalability will require systems engineering to combine quantum and classical hardware .
one technology demanding imminent efforts is the realization of a suitable wiring method for the control and measurement of a large number of qubits . in this work
, we introduce an interconnect solution for solid - state qubits : _ the quantum socket_. the quantum socket fully exploits the third dimension to connect classical electronics to qubits with higher density and better performance than two - dimensional methods based on wire bonding .
the quantum socket is based on spring - mounted micro wires _ the three - dimensional wires _
that push directly on a micro - fabricated chip , making electrical contact .
a small wire cross section ( @xmath0 ) , nearly non - magnetic components , and functionality at low temperatures make the quantum socket ideal to operate solid - state qubits .
the wires have a coaxial geometry and operate over a frequency range from dc to , with a contact resistance of @xmath0 , an impedance mismatch of @xmath0 , and minimal crosstalk . as a proof of principle , we fabricated and used a quantum socket to measure superconducting resonators at a temperature of @xmath0 .
quantum error correction codes such as the surface code will largely benefit from the quantum socket , which will make it possible to address qubits located on a two - dimensional lattice .
the present implementation of the socket can be readily extended to accommodate a quantum processor with a @xmath1 qubit lattice , which would allow the realization of a simple quantum memory . |
t tauri stars are considered to resemble our sun at an age of a few million years .
studies of their surrounding gas and dust can therefore provide important clues on the early evolution of the solar nebula .
it is well established through surveys at infrared and millimeter wavelengths that most t tauri stars have circumstellar disks with masses of @xmath11 m@xmath12 and sizes of @xmath13 au ( see overviews by @xcite , dutrey et al .
1996 , mundy et al .
in addition to serving as a conduit for mass accretion onto the young star , the disks also provide a reservoir of gas and dust for the formation of potential planetary systems ( shu et al .
theories of disk evolution depend strongly on the radial and vertical temperature structure of the disks ( e.g. , hartmann et al .
1998 ) , but these parameters are still poorly constrained by the available observations .
we report here the results of a deep survey for the lowest two pure rotational lines of h@xmath0 , the @xmath1=@xmath14 s(0 ) line at 28.218 @xmath3 m and the @xmath15 s(1 ) transition at 17.035 @xmath3 m , using the short wavelength spectrometer ( sws ) on board the _ infrared space observatory _ ( iso ) . in emission ,
the h@xmath0 lines originate from levels at 509.9 k and 1015.1 k above ground and are thus excellent tracers of the `` warm '' ( @xmath1680 k ) gas in disks , especially in the interesting inner part where jovian planets may form .
h@xmath0 has the advantage that it dominates the mass budget and that it does not deplete onto grains , contrary to co. moreover , the lines are optically thin up to very high column densities owing to the small einstein @xmath17 coefficients for electric quadrupole transitions , so that the modeling of the radiative transfer is simple . here we present h@xmath0 pure rotational line observations of the gg tau system , which is situated at the edge of the taurus - auriga cloud complex at a distance of approximately 140 pc ( @xcite ) .
gg tau consists of two close binary pairs separated by 10@xmath18 , or 1400 au .
the main binary gg tau a has a separation of @xmath19 au ( ghez et al .
1997 ) and is composed of a k7 and a m0.5 star ( white et al .
1999 ) , both classified as emission - line or `` classical '' t tauri stars by herbig & bell ( 1988 ) with an estimated accretion rate of 2 @xmath20 10@xmath21 m@xmath12 yr@xmath22 ( hartmann et al .
the gg tau b binary is comprised of an m5 and m7 star separated by @xmath23 au .
the age of the gg tau system is estimated to be @xmath24 myrs by white et al .
( 1999 ) , using the evolutionary models of baraffe et al .
( 1998 ) and assuming the four stars to be coeval .
high spatial resolution images taken in the near infrared show that each of the stars in the gg tau a system has associated circumstellar material within a radius of less than 10 au ( roddier et al .
these stars are located within a cavity of radius @xmath25 au cleared by the dynamical interaction of the binary ( ghez et al .
1997 ) , and a circumbinary disk extending up to @xmath26 au as imaged at millimeter wavelengths ( dutrey et al .
1994 , guilloteau et al . 1999 ) .
the circumbinary disk mass of @xmath27 m@xmath12 , deduced from the strong millimeter dust continuum and assuming a gas - to - dust ratio of 100:1 , is one of the largest observed to date ( guilloteau et al .
however , observations of @xmath28co and c@xmath29o indicate gas masses that are up to a factor of 100 lower , ( dutrey et al . 1994 ) . explanations for this discrepancy include the possible freeze - out of co in the cold outer part of the disk and/or a gas dissipation time scale that is shorter than that of the dust ( zuckerman et al .
the h@xmath0 observations presented here allow a direct measurement of the amount of warm gas in the disk .
the h@xmath0 data for gg tau form part of a survey of a larger number of t tauri and herbig ae stars with the iso - sws by thi et al .
( 1999a ; and in preparation ) .
an initial account of the results has been given in van dishoeck et al .
the low - lying pure - rotational h@xmath30 @xmath31 s(0 ) line at 28.218 @xmath3 m and the @xmath15 s(1 ) line at 17.035 @xmath3 m were observed with the iso - sws in the aot02 mode ( @xcite ) .
the observations were centered in the direction of gg tau a at ra(2000)=04@xmath3232@xmath3330@xmath34 , dec(2000)=17@xmath3531@xmath3642@xmath18 .
typical integration times were 6001000 s per line , in which the 12 detectors were scanned several times over the 28.0528.40 and 16.9617.11 @xmath3 m ranges around the lines . the @xmath37 s(3 )
9.66 @xmath3 m and @xmath38 s(5 ) 6.91 @xmath3 m lines were measured in parallel with the s(0 ) and s(1 ) lines , respectively , at virtually no extra time .
the spectral resolution for point sources is 2000 ( 150 km s@xmath22 ) at 28 @xmath3 m , 2400 ( 125 km s@xmath22 ) at 17 @xmath3 m , 2280 ( 130 km s@xmath22 ) at 9.7 @xmath3 m and 1550 ( 195 km s@xmath22 ) at 6.9 @xmath3 m .
the sws apertures are @xmath39 at s(0 ) , @xmath40 at s(1 ) , and @xmath41 at the s(3 ) and s(5 ) lines .
two independent sets of observations have been carried out in orbital revolutions 668 and 834 .
the expected peak fluxes of the h@xmath0 lines are close to the sensitivity limit of the instrument , and the raw data show a high level of noise induced by charged - particle impacts on the detectors . in order to extract the h@xmath0 lines , special software designed to handle weak signals was used for the data reduction in combination with the standard interactive analysis package .
the details and justification of the methods used in the software are described by valentijn & thi ( 1999 ) .
because the mid - infrared continuum emission from gg tau is weak , @xmath421 jy at @xmath4217 @xmath3 m and @xmath10 jy at 28 @xmath3 m , the data do not suffer from fringing effects caused by an inadequate responsivity function correction .
the h@xmath0 s(0 ) and s(1 ) lines are detected in both sets of observations ( see table 1 ) and the differences in fluxes between the two sets are @xmath43% .
this is well within the estimated total error of @xmath44% , which is mostly due to uncertainties in the flux calibration .
the s(3 ) and s(5 ) lines are not detected down to a limit of @xmath45 erg s@xmath22 @xmath46 ( @xmath47 ) .
the spectra of the h@xmath30 s(0 ) and s(1 ) lines are displayed in figure 1 for revolution 668 . since the turbulent and rotational velocities in disks are only a few km s@xmath22 , respectively , both lines are unresolved . after subtraction of the continuum ,
the lines are fitted by gaussians with widths fixed by the instrumental resolution and the line fluxes are computed from @xmath48 to @xmath49 hwhm . in the optically thin limit ,
the observed line fluxes are directly related to the populations in the h@xmath0 @xmath1=2 and 3 levels . the derived kinetic temperature assuming local thermodynamic equilibrium ( lte ) is @xmath50 k , where the error bar reflects the 30% uncertainty in the fluxes .
although the @xmath1=3 level has a factor of 40 lower population than the @xmath1=2 level in gas with a temperature near 100 k , the radiative transition probability of the s(1 ) line , @xmath51 s@xmath22 , is much larger than that of the s(0 ) transition , @xmath52 s@xmath22 .
in addition , the spectral resolution at 17 @xmath3 m is somewhat higher than that at 28 @xmath3 m , and the line - to - continuum ratio is correspondingly larger . both of these factors explain why the s(1 ) line is detectable as well .
the limits on the s(3 ) and s(5 ) lines imply temperatures less than @xmath53 k and @xmath54 k , respectively , neglecting any correction for differential extinction . since the iso beam is much larger than the size of the circumbinary disk , it is important to check whether any of the observed h@xmath0 emission may arise from residual extended envelope or cloud material .
h@xmath0 lines up to s(9 )
are readily detected toward embedded herbig ae stars with the iso - sws ( e.g. , van den ancker et al .
1998 ) . in these cases ,
the emission is dominated by the interaction of the young star with its surrounding envelope through shocks and ultraviolet photons .
the typical h@xmath0 excitation temperatures for these regions are @xmath55=500700 k , much larger than the value found for gg tau .
deep iso searches for the h@xmath0 s(0 ) and s(1 ) lines toward diffuse and translucent clouds have been performed by thi et al.(1999b ) .
the lines are not detected down to @xmath56 erg s@xmath22 @xmath46 ( @xmath57 ) in gas with densities of a few hundred to a few thousand @xmath58 exposed to the normal interstellar radiation field .
the @xmath59co 10 emission around gg tau is less than 50 mk ( 3@xmath60 ) ( @xcite ) , more than an order of magnitude lower than found for diffuse clouds such as that toward @xmath61 oph .
the corresponding mass in the sws beam is estimated to be less than a few @xmath62 m@xmath12 , significantly lower than the mass derived from the h@xmath0 lines ( see below ) .
keck images of gg tau in @xmath63 continuum and h@xmath0 emission in the @xmath64=1@xmath20 and @xmath64=2@xmath21 s(1 ) lines at 2.1250 and 2.2486 @xmath3 m were obtained on 03 and 06 november 1998 using the facility near infrared camera ( nirc ) and the appropriate filters .
no h@xmath0 emission down to @xmath65 @xmath3jy ( 2@xmath60 ) was detected in a 1.5 - 6@xmath18 ( i.e. 200 - 800 au ) radius around the stars , nor outside this region .
this translates to a limit on the intensity of @xmath66 erg s@xmath22 @xmath46 sr@xmath22 .
altogether , we are confident that the bulk of the h@xmath0 emission toward gg tau originates from the disk(s ) rather than interstellar material in the iso beam .
assuming no continuum extinction and lte excitation , the corresponding mass of warm gas is computed via the relation @xmath67 where @xmath68 is the integrated flux in erg s@xmath22 @xmath46 , @xmath69 is the distance of gg tau in pc ( taken to be 140 pc ) , @xmath70 is the frequency of the transition in hz , and @xmath71(@xmath16 ) is the fractional population in the upper rotational level @xmath72 .
the derived amount of warm gas is ( 3.6 @xmath732.0 ) @xmath20 10@xmath74 m@xmath12 ( 3@xmath60 ) , including the 30% uncertainty in the fluxes .
the derivation assumes that the ortho / para h@xmath0 ratio is @xmath75 , the lte value at 110 k. if the emission were affected by 30 magnitudes of visual extinction , the derived excitation temperature would increase to 121 k and the mass to @xmath76 10@xmath74 m@xmath12 .
the inferred warm h@xmath0 mass of @xmath77 10@xmath74 m@xmath12 is about 3% of the total gas + dust mass of 0.12 m@xmath12 derived from millimeter continuum observations of the circumbinary disk assuming a gas - to - dust ratio of 100:1 and a gas + dust absorption coefficient of 0.01 @xmath78 g@xmath22 at 2.6 mm ( guilloteau et al .
the temperature is much higher than that derived from the continuum spectral energy distribution and optically thick co emission , which give a temperature of only 34 k at the inner edge of the circumbinary disk at 180 au . where does the warm h@xmath0 emission originate and what is the heating mechanism ?
the bulk of the circumbinary disk is too cold to account for emission by gas at 100 k. moreover , the disk is optically thick in the mid - infrared continuum .
the h@xmath0 emission must therefore arise either from the circumstellar disk(s ) , or from the surface layers and inner edges of the circumbinary disk .
two kinds of heating mechanisms may be at work : heating by absorption of part of the stellar and accretion luminosity , and heating by dynamical processes including shocks and turbulent decay .
these possibilities are discussed in turn below .
several radiative processes must be examined .
the first possibility is that the h@xmath30 lines arise from material within @xmath79 au around the individual stars , where the gas and dust are heated by the ultraviolet radiation from the ysos . in general , a large fraction of this warm gas in the inner circumstellar disk(s ) may be hidden by the optically thick continuum of colder surrounding dust , especially if the disks are observed nearly edge on .
however , gg tau presents a special case , since the dynamical interaction of the binary has cleared the inner part of the circumbinary disk .
the near - infrared observations of roddier et al .
( 1996 ) suggest that at least some of the radiation from the inner disks can escape through holes in the circumbinary disk combined with favorable orientations of the material .
the masses of the inner circumstellar disks are estimated to be only @xmath80 m@xmath12 each , however , based on the millimeter continuum data ( guilloteau et al .
it therefore does not appear that there is sufficient mass in the inner disks to explain the warm h@xmath0 emission .
consider next the case of the more extended circumbinary disk . if this disk is flared , as is expected if hydrostatic equilibrium is approached , there exists a surface layer that is heated by radiation from the central star(s ) to temperatures near 100 k out to radii of @xmath81 au ( e.g. , chiang & goldreich 1997 , 1999 ) .
these disk models have a non - isothermal vertical temperature profile , and the warm gas is located in the near - surface regions of the disk .
thus , the emission arising from the heated material is not absorbed by cooler layers before reaching the earth . an h@xmath0 excitation calculation has been performed using the density and temperature structure of the standard model of chiang & goldreich ( 1997 ) , assuming a gas - to - dust mass ratio of 100:1 with @xmath82 .
typical s(0 ) and s(1 ) fluxes from a single face - on surface layer are @xmath83 and @xmath84 erg s@xmath22 @xmath46 , respectively .
these values depend sensitively on the adopted continuum opacities at mid - infrared wavelengths and the geometry , resulting in uncertainties of factors of 23 . even taking these factors into account , however ,
the model fluxes are a factor of 510 lower than the observed values . moreover , the model s(0)/s(1 ) ratio of 0.4 is lower than the observed ratio of @xmath85 , and the model s(3)/s(1 ) ratio of 0.4 is higher than the observed ratio of @xmath420.2 . a more detailed radiative transfer simulation including the disk inclination angle of @xmath86 ( roddier et al . 1996 , guilloteau et al .
1999 ) and possible velocity shifts between the warm and cold gas is needed to provide a more accurate assessment of this model , but is beyond the scope of this letter .
a third radiative mechanism may be provided by ultraviolet radiation from the star - inner disk boundary layer(s ) which can irradiate the inner edge of the circumbinary disk ( or any residual gas in the cavity ) and heat it to @xmath87 k. this situation has been described for circumstellar envelopes by spaans et al .
the intensity of the ultraviolet radiation from a 10,000 k boundary layer with a luminosity of 0.3@xmath88 is estimated to be up to a factor of 600 larger than the average interstellar radiation field at a distance of @xmath89 au .
if the density at the boundary is assumed to be a few@xmath90 @xmath58 , the computed h@xmath0 fluxes are @xmath91 and @xmath92 erg s@xmath22 @xmath46 for the s(0 ) and s(1 ) lines , respectively , an order of magnitude lower than observed .
similar discrepancies between models and h@xmath0 observations are found for dense interstellar clouds exposed to ultraviolet radiation ( e.g. , draine & bertoldi 1999 ) , indicating that the heating mechanisms are not fully understood .
the s(0)/s(1 ) ratio of @xmath24 and s(3)/s(1 ) ratio of @xmath93 in these models are consistent with the data within the errors , however .
further modeling of the radiative heating of the surface layer and inner edge of the circumbinary disk is needed to investigate whether a combination of these radiative mechanisms can reproduce the observations .
an alternative class of heating mechanisms involves shocks caused by infalling material at the inner disk surface(s ) or by the interaction between an outflowing supersonic wind and the surfaces of the circumstellar or circumbinary disk(s ) ( hartmann & raymond 1989 ) .
evidence for such winds comes from optical observations of atomic and ionic lines , from which hartigan et al .
( 1995 ) derive a mass loss rate of 7.9 @xmath20 10@xmath94 m@xmath12 yr@xmath22 for gg tau . since the masses of the inner circumstellar disk(s ) are too low to explain the h@xmath0 emission , only the interaction of the wind with the circumbinary disk at @xmath89 au needs to be considered .
the problem with these models is that shocks are expected to warm the surface layers to sufficiently high temperatures to emit strongly in the s(5 ) and s(3 ) lines and the 2 @xmath3 m vibration - rotation lines .
consider as an example the wind - disk models of hartmann & raymond ( 1989 ) .
for typical wind velocities of 200 km s@xmath22 , the estimated shock velocities along the disk surface range from 2040 km s@xmath22 at distances of 50200 au .
comparison with the @xmath1 and @xmath95shock models of burton , hollenbach & tielens ( 1992 ) and kaufman & neufeld ( 1996 ) shows that in virtually all models the flux in the s(3 ) line is predicted to be larger than that in the s(1 ) line , in contrast with the observations . using the s(3)/s(1 ) ratio as a constraint , at most 30% of the s(1 ) emission
could be contributed by shocks .
the lack of detected h@xmath0 @xmath64=1@xmath20 s(1 ) emission also indicates the absence of shocks faster than @xmath65 km s@xmath22 .
thus , the h@xmath0 s(3 ) and 2 @xmath3 m upper limits suggest that heating by shocks is unlikely to be the major contributor to the line emission
. most likely , a combination of heating by ultraviolet photons and dynamical processes is responsible for the warm molecular gas . in summary
, this work demonstrates that h@xmath0 pure rotational lines can be detected from disks around pre - main sequence stars and that they provide complementary information to sub - millimeter observations of co and other molecules .
the h@xmath0 observations are particularly sensitive to the warm gas in the disks . with current instrumentation , masses of warm h@xmath0
can be detected that are only a small fraction of the total gas + dust mass in circumstellar disks .
the line ratios provide important constraints on the heating mechanisms . in order for the h@xmath0 emission to escape , however
, the emission must arise either from the disk surface layers or requires the presence of gaps or holes in the disks . in the case of gg
tau , the binary nature of this system has cleared a large inner cavity in the circumbinary disk , which may have facilitated the detection of the lines .
future observations at higher spectral and spatial resolution such as provided by mid - infrared spectrometers on ground - based telescopes and aboard platforms such as the _ stratospheric observatory for infrared astronomy _ ( sofia ) and the _ next generation space telescope _ ( ngst ) , accompanied by more sophisticated modeling , should be able to clarify the origin of the h@xmath0 emission from disks around t tauri and herbig ae stars and allow much more sensitive searches .
thi , w.f .
, van dishoeck , e.f . , black , j.h . , jansen , d.j . , evans , n.j . , & jaffe , d.t .
1999b , to appear in the universe as seen by iso , eds .
p. cox & m.f .
kessler , esa - sp 427 ( noordwijk : estec ) , in press cccc 668 & h@xmath0 0 - 0 s(0 ) & 28.218 & 2.5 10@xmath96 834 & h@xmath0 0 - 0
s(0 ) & 28.218 & 2.3 10@xmath96 668 & h@xmath0 0 - 0 s(1 ) & 17.035 & 2.8 10@xmath96 834 & h@xmath0 0 - 0 s(1 ) & 17.035 & 2.9 10@xmath96 668 & h@xmath0 0 - 0 s(3 ) & 9.66492 & @xmath42 5.4 10@xmath97 668 & h@xmath0 0 - 0 s(5 ) & 6.90952 & @xmath42 7.1 10@xmath97 | we present the first detection of the low - lying pure rotational emission lines of h@xmath0 from circumstellar disks around t tauri stars , using the short wavelength spectrometer on the _ infrared space observatory_. these lines provide a direct measure of the total amount of warm molecular gas in disks .
the @xmath1=2@xmath20 s(0 ) line at 28.218 @xmath3 m and the @xmath1=3@xmath21 s(1 ) line at 17.035 @xmath3 m have been observed toward the double binary system gg tau . together with limits on the @xmath1=@xmath4 s(3 ) and @xmath5 s(5 ) lines , the data suggest the presence of gas at @xmath6 k with a mass of @xmath7 m@xmath8 ( @xmath9 ) .
this amounts to @xmath10 % of the total gas + dust mass of the circumbinary disk as imaged by millimeter interferometry , but is larger than the estimated mass of the circumstellar disk(s ) .
possible origins for the warm gas seen in h@xmath0 are discussed in terms of photon and wind - shock heating mechanisms of the circumbinary material , and comparisons with model calculations are made . |
liquid - liquid dispersions are encountered for instance in extraction or chemical engineering when contact between two liquid phases is needed . without surfactants ,
when two immiscible fluids are mechanically agitated , a dispersed state resulting from a dynamical equilibrium between break - up and coalescence of drops can be reached .
their modelling in turbulent flows is still limited ( portela and oliemans , 2006 ) .
some open questions we address are first the potential increase of turbulent dissipation due to the dispersion , the modification of the turbulent transport inside the flow , and the droplet size distribution , governed by the breakup and coalescence rates . finally , some concentrated liquid - liquid flows are also known to experience hysteretic phase inversion ( piela et al . ,
2006 ) . to answer these questions ,
we have chosen a model shear - flow : a taylor
couette flow where the fluids are stirred in the gap between two counterrotating coaxial cylinders .
we use low - viscous fluids and first check that the single - phase flows are strongly turbulent .
we measure the wall shear stress , and use refractive - index matched fluids in order to measure the phases repartition .
the flow is produced between two coaxial cylinders .
the inner one is of radius @xmath1 mm , and the outer one of radius @xmath2 mm , which gives a gap ratio @xmath3 .
the length of the cylinders is @xmath4 mm , i.e. the axial aspect ratio is @xmath5 .
the cylinders axis is vertical in order to keep axisymmetric conditions .
the system is closed at both ends , with top and bottom lids rotating with the outer cylinder .
there is a gap of @xmath6 mm between these lids and the flat ends of the inner cylinder .
the cylinders are driven by two independent brushless dc motors ( maxon , @xmath7w ) , at a speed up to @xmath8hz .
the torque @xmath9 exerted on the inner cylinder is measured with a hbm rotating torquemeter ( t20wn , @xmath10n.m ) .
a photograph of the setup is presented in figure [ fig : manip ] .
the optical measurements have been done with an outer glass cylinder , of radius @xmath11 mm ( @xmath12 ) . for a newtonian fluid of kinematic viscosity @xmath13
, we use the set of parameters defined by dubrulle et al .
( 2005 ) : a mean reynolds number @xmath14 based on the shear and on the gap @xmath15 ; and a rotation number @xmath16 which is zero in case of perfect counterrotation ( @xmath17 ) . at @xmath8hz in counterrotation ,
the typical shear rate is around @xmath18s@xmath19 and @xmath20 for pure water .
.physical properties of the fluids at @xmath21c : density @xmath22 ( kg.m@xmath23 ) , dynamic viscosity @xmath24 ( pa.s ) and kinematic viscosity @xmath13 ( [email protected]@xmath19 ) .
[ cols="<,^,^,^",options="header " , ] [ tab : fluids ] the physical properties of the fluids we use are given in table [ tab : fluids ] .
the oil is the shell macron edm110 ; it is a low - viscosity paraffinic hydrocarbons mix .
its refractive index measured with an abbe refractometer at @xmath21c is @xmath26 .
the interfacial tension between oil and water is @xmath27n.m@xmath19 . to have a refractive index matched dispersion ( budwig , 1994 ) , in addition to pure water
, we also use a sodium iodide solution as acqueous phase ( narrow et al . , 2000 ) .
the concentration of nai is @xmath28g.l@xmath19 , and traces of na@xmath29s@xmath29o@xmath30 are added to avoid yellowish coloration .
the viscosity ratio is then close to unity and the density ratio is close to @xmath10 . in the following ,
unless specified , dimensionless quantities are deduced from @xmath31 as unit length , @xmath32 as unit time , and @xmath33 as unit mass .
we first study the turbulent taylor - couette flow in exact counterrotation , and characterize its turbulent state .
though the taylor - couette flow has been widely studied , few experimental results and theories are available for @xmath34 ( dubrulle et al . , 2005 ) .
we report torque measurements on pure fluids in figure .
[ fig : pure ] .
they have been made in a first version of the experiment , with a free surface and a space of @xmath8 mm between the cylinders bottom ends .
the torque due to the bottom part has been removed by studying different filling levels .
the range of reynolds numbers we cover is @xmath35 . in figure
[ fig : pure]a , we plot the dimensional torque _ vs. _ speed for the three pure fluids .
the torques behave non - linearly with speed , as expected for turbulent regimes , and the oil and water are very close , the higher viscosity of oil balancing its lower density .
we plot in figure [ fig : pure]b the dimensionless torque @xmath36 normalised by the laminar dimensionless torque @xmath37 _ vs. _ @xmath38 .
the experimental data fall on the same curve , which is consistent with the formula proposed by dubrulle and hersant ( 2002 ) of the form @xmath39 . here
, a non - linear fit gives @xmath40 and @xmath41 . at the end of the range , this is close to a power - law behaviour in @xmath42 .
this scaling indicates that in counterrotation , a turbulent regime is easily reached ( let us remind that the scaling expected in the taylor - vortices regime is @xmath43 ) . moreover , esser and grossmann ( 1996 ) proposed a formula for the first instability threshold . here
it gives @xmath44 . at first glance
, this value seems to be consistent with our results ( see figure [ fig : pure]b ) . in forthcoming experiments , we will use glycerol to lower the reynolds number in order to check this prediction . ) , water ( @xmath45 ) and nai ( @xmath46 ) .
( b ) : dimensionless torque @xmath47 normalised by laminar torque @xmath48 _ vs. _ @xmath38 .
solid line is the fit proposed by dubrulle and hersant ( 2002 ) .
@xmath49 is the expected value for the first instability threshold , according to the theory of esser and grossmann ( 1996).,width=283 ] we measure the three components of the velocity in one plane by stereoscopic piv .
the plane , illuminated by a double - pulsed nd : yag laser is normal to the mean flow , i.e. vertical ( see figure [ fig : manip ] ) : the in - plane components are the radial ( @xmath50 ) and axial ( @xmath51 ) velocities , while the out - of - plane component is the azimuthal component ( @xmath52 ) .
it is imaged using two double - frame pco - cameras , mounted on scheimpflug adaptors , located at each side of the light sheet , with an angle of @xmath53 in the air .
the flow is seeded with tracer particles ( 8 micrometer sphericel ) .
the field of view measures @xmath54 mm @xmath55 @xmath56 mm , corresponding to a resolution of @xmath57 pixels .
special care have been taken concerning the calibration procedure on which the third azimuthal component rely .
we use a very thin transparent sheet with crosses printed on it , with a weight at its bottom , and attached to a rotating and translating micro - traverse ( see figure [ fig : manip ] ) .
it is first put into the light sheet and traversed perpendicularly to it . typically five pictures are taken with steps of @xmath58 mm .
the raw piv - images are processed with pivware ( westerweel , 1993 ) , with a last interrogation area of @xmath59 pixels with @xmath60 overlap , and normalised median filtering as post - processing .
then the vector mapping and the third component reconstruction are made with matlab .
the mapping function is a third - order polynomial , and interpolations are bilinear . .
solid line : measured mean azimuthal velocity .
dotted line : theoretical profile .
dashed line : fit of the form @xmath61 .
the radial component @xmath50 which should be zero is also shown.,width=283 ] in order to check the quality of the technique , we measured the basic laminar flow with the outer cylinder at rest , for @xmath62 .
the measurements are done in @xmath63 glycerol - water mixture .
we plot in figure [ fig : pivgly ] the velocity radial profiles , averaged over @xmath64 recordings .
the flow is stationary and laminar .
it is in good agreement with the theoretical profile : @xmath65 , except near the inner and outer walls , where some reflections disturb the measurement .
the error in the in - plane velocity is roughly constant , around @xmath66 . .
( a ) : dimensionless mean velocity profiles @xmath50 ( dotted line ) , @xmath51 ( dashed line ) , and @xmath52 ( solid line ) . the azimuthal profile is not measured at the walls where its magnitude is @xmath67 .
( b ) : dimensionless turbulent shear - stress @xmath68 ( solid line ) and dimensionless viscous shear - stress computed from the mean velocity profile .
the dash - dotted line connects the two known values of the shear - stress at the walls , deduced from torque measurements.,width=283 ] we then measured the counterrotating flow , at @xmath69 .
the results are presented in figure [ fig : pivcontra ] .
the measurements are triggered on the outer cylinder position , and are averaged over @xmath70 images . in the counterrotating case , for this large gap ratio and at this value of the reynolds number , the instantaneous velocity field is really desorganised and does not contain obvious structures like taylor - vortices , in contrast with other situations ( wang et al . , 2005 ) .
no peaks are present in the time spectra , and there is no axial - dependency of the time - averaged velocity field .
we thus average in the axial direction the different radial profiles .
the azimuthal velocity @xmath52 in the bulk is low , i.e. its magnitude is below @xmath71 between @xmath72 that is @xmath73 of the gap width .
the two other components are zero within @xmath74 .
the fluctuations of the velocity are not isotropic .
they are of the order of @xmath75 for the radial and axial components , and of the order of @xmath71 for the azimuthal components .
we can also compute some terms in the angular momentum balance equation ( mari and daviaud , 2004 ) . according to the torque measurements described in the previous paragraph
, the shear - stress at the inner wall is @xmath76pa , i.e. @xmath77 in dimensionless form .
the dimensionless shear - stress at the outer wall is @xmath78 .
we report these two values in figure [ fig : pivcontra]b , together with the viscous shear - stress @xmath79 ( dashed line ) and the turbulent shear stress @xmath80 ( solid line ) .
the viscous shear - stress is negligible in the bulk , and is dominant at the walls , where it should ensure all the transport of angular momentum .
though the boundary layer is very badly resolved , the rough estimate of the viscous stress at the walls corresponding to the tails of the dashed line is consistent with the expected values .
the measured turbulent shear - stress is roughly constant in the bulk , and has the good order of magnitude .
the correlation coefficient between @xmath50 and @xmath52 , defined as @xmath81 is @xmath82 .
the turbulent shear - stress represents @xmath83 of the expected shear - stress .
we conclude that this flow is dominated by turbulent transport , which is also consistent with the high scaling of the torque with @xmath38 ( remind that @xmath42 ) .
we now use two immiscible fluids in the taylor - couette facility .
dark region corresponds to the oil , and bright regions to the sodium iodide solution .
the outer cylinder is on the right side of each picture.,width=283 ] to look at the phases repartition , we perform visualisations with a laser induced fluorescence technique . we add a small amount of rhodamine b in the refractive - index matched sodium iodide solution .
the rhodamine is sensitive to the green light of the nd : yag laser ( @xmath84 nm ) , and emits a yellow fluorescent light .
we use melles - griot low - pass filters , with a cutoff wavelength of @xmath85 nm to only record the fluorescent light .
examples of visualisations are given in figure [ fig : lif ] .
starting from fluids at rest , the system is organised into two layers .
when the cylinders are rotating , there is a competition between centrifugal forces and gravity .
a froude number measuring the ratio of the centrifugal forces to buoyancy can be estimated as @xmath86 .
we indeed observe that for @xmath87hz the fluids stay separated , but that the interface is tilted , the heavier fluid being pushed outwards .
the interface is very weavy and non stationary .
the ratio of turbulent shear stress to buoyancy can be estimated as @xmath88 , with @xmath89 being the typical rms velocity measured by piv .
a weber number comparing the effects of turbulent stresses to interfacial tension can be defined as @xmath90 . at a speed of @xmath6hz ,
some oil droplets start to be entrained into the acqueous phase .
very few drops of acqueous phase into oil are present .
increasing further the speed , more and more droplets of oil are entraped into nai , and there start to be acqueous drops into the oil , mainly visible as films , i.e. drops into drops , as can be seen in figure [ fig : lif ] .
then , there seems to be three regions in the fluids , from top to bottom : an oil - continuous region with nai drops , a foam of possibly multiple droplets , and a nai - continuous region with a lot of oil droplets . finally , for @xmath91hz , the fluids are fully dispersed .
one of the open question we adress is whether the turbulent dissipation is changed with respect to a monophasic case , and in that case , how much is that change .
there are models for the effective viscosity of emulsions ( pal , 2001 ) , but what does happen in a turbulent dispersion where molecular viscosity is usually negligible ? ) and @xmath92 water in oil ( @xmath46 ) .
speed is increased by steps of @xmath93hz every @xmath94s , up to @xmath95hz and then decreased back to @xmath96 .
the points clouds are parametric plots from experiments ( b ) and ( c ) .
( b ) : from a two - layer system at rest , the speed is set to @xmath97hz .
( c ) : from this speed , after 20 minutes , the speed is decreased to @xmath98hz.,width=283 ] we plot in figure [ fig : dispersion]a the torque _ vs. _ the frequency for a water and oil system .
the volumic fraction of water is @xmath92 .
the experiment have been automatised and we proceed as follows for this experiment : starting from rest we record the torque during @xmath94s and then increase the speed by @xmath93hz , record one minute and so forth until @xmath95hz , which is reached after @xmath99 minutes .
we then decrease the speed until @xmath96 , by steps of @xmath93hz .
the torque for the water and oil system ( @xmath46 ) is at low speed very close to the single - phase case ( plotted in the same figure for pure oil ( @xmath100 ) ) .
this is consistent with a sum of two layers weakly interacting ( small viscosity difference ) , which basically give the same dimensionnal torque as discussed in figure [ fig : pure ] . increasing the speed
, the torque shows up a fast increase around @xmath101hz ( fig .
[ fig : dispersion]a ) , corresponding to the point at which the liquids start to fully disperse . at the speed of @xmath95hz ,
the torque per unit mass of fluid to stirr is @xmath102n.m.kg@xmath19 whereas it is @xmath103n.m.kg@xmath19 for pure oil .
this corresponds to an increase of @xmath104 .
when decreasing the speed , there seems to be hysteresis in the system .
the torques on the return path are slightly higher than during the increase .
the end of the hysteresis loop seems to be at @xmath105hz .
indeed , visual observation confirms that the fluids are still dispersed , even below @xmath10hz .
similar behaviours are present for other concentrations we studied ( see figure [ fig : tableau ] ) , with less hysteresis for the lowest oil concentration . to better understand this hysteresis , we performed the following long - time experiments .
starting from a two - layer system at rest , we increase quickly the rotation up to @xmath106hz and record then the torque ( figure [ fig : dispersion]b ) . the torque gradualy increase during the @xmath107s of the experiments .
the final stationary state is not reached before the end of these @xmath107s .
we plot also this experiment in figure [ fig : dispersion]a as a parametric points cloud with time as parameter .
one can notice that the torque at the beginning is the one for oil , and gradually reaches the upper branch .
the complementary experiment ( figure [ fig : dispersion]c ) consists of lowering the speed to @xmath108hz after @xmath99 minutes .
it shows a slow decrease and the final state is not on the lower branch ( figure [ fig : dispersion]a ) : the two - layers system is not the one stable above @xmath108hz .
the time scales involved in the dispersion and separation processes are thus very long , the latter being even longer ( figure [ fig : dispersion]b - c ) .
complicated behaviours including periodic oscillations ( with a period of @xmath109min ) have been observed in between and requires further investigation .
hz , normalised by the pure oil case _ vs. _ volume fraction of water ( @xmath45 ) and nai ( @xmath46).,width=283 ] we report in figure [ fig : tableau ] the evolution with the volume fraction of the acqueous phase of the torque per unit mass of fluid to stirr at @xmath95hz .
the data have been normalised by that of pure oil ( @xmath103n.m.kg@xmath19 ) .
this quantity is maximum for a volume fraction of the acqueous phase around @xmath92 and reaches a factor @xmath10 .
similar behaviours have been evidenced recently in liquid - liquid turbulent pipe flow experiments ( piela et al . , 2006 ) .
we have built a facility with well - controlled parameters to study liquid - liquid dispersions in a turbulent flow . in a first step ,
the single - phase flow has been characterized in terms of scaling of the dissipation , and also in terms of velocity and reynolds stress profiles .
we measured the turbulent fluctuations intensity who are a key feature for the dispersion mechanisms .
further investigations on the scaling and threshold for first instability for the counterrotating monophasic flow will be performed , on a wider reynolds number range , using glycerol . ) and @xmath60 water in oil ( @xmath46 ) .
( b ) : dimensionless torque @xmath47 _ vs. _ @xmath38 , molecular value for pure water ( @xmath45 ) , and effective values proposed by pal ( 2001 ) for @xmath60 water in oil ( @xmath46).,width=283 ] a global measurement such as in - line torque measurement is able to characterize the state of the system in terms of whether it is dispersed or not .
data presented here for the torques in the dispersions were still limited by old motors with low power .
further work have to be done to explore the scaling of the torque for turbulent dispersions .
we plot in figure [ fig : zob ] a first result obtained for @xmath60 water in oil .
the torque in the dispersed state seems to follow a nice scaling at higher velocities . in order to check the idea of a proper `` effective molecular viscosity ''
, we try the one proposed by pal ( 2001 ) : for viscosity ratio 3 ( oil drops in water ) and for fraction 0.5 , the effective viscosity is @[email protected]@xmath19 .
it seems then possible to collapse the points onto a single curve when one uses this effective viscosity to compute the reynolds number and to normalise the torque in the dispersed state ( figure [ fig : zob]b ) . notice that in fact we do not have enough points at low values of reynolds numbers for newtonian fluids to really compare this rescaling with consistent data . in order to understand the interaction between the different scales , we are now working on edge detection algorithm to study the drop size distribution with the lif images
budwig , r. , 1994,``refracive index matching methods for liquid flow investigations '' , _ exp . in fluids _ , vol .
350 - 355 .
dubrulle , b. , dauchot , o. , daviaud , f. , longaretti , p .- y . ,
richard , d. and zahn , j .-
, 2005 , `` stability and turbulent transport in taylor couette flow from analysis of experimental data '' , _ phys . fluids _ ,
095103 .
piela , k. , delfos , r. , ooms , g. , westerweel , j. , oliemans , r. v. a. , and mudde , r. f. , 2006 , `` experimental investigation of phase inversion in an oil - water flow through a horizontal pipe loop '' , _ int .
j. multiphase flows _ , vol .
32 , pp . 1087 - 1099 .
portela , l. m. , and oliemans , r. v. a. , 2006 , `` possibilities and limitations of computer simulations of industrial turbulent dispersed multiphase flows '' , _ flow turbulence combust .
_ , vol . 77 , pp .
381 - 403 . | we study liquid - liquid dispersions in a turbulent taylor couette flow , produced between two counterrotating coaxial cylinders . in pure water and in counterrotation , reynolds numbers up to @xmath0 are reached .
the liquids we use are a low - viscous oil and pure water or a sodium iodide solution with a refractive index matched to that of oil , in order to get transparent dispersions .
we first characterize the single - phase flow , in terms of threshold for transition to turbulence , scaling of the torque and measurements of the mean flow and of the reynolds stress by stereoscopic piv .
we then study the increase of the dissipation in the two - phase flows and find that the torque per unit mass can be twice the torque for a single - phase flow .
long - time behaviours are also reported . |
the electron flow in semiconductor nanostructures containing a two - dimensional electron gas ( 2deg ) buried shallow beneath the surface of the structure can be probed by a charged tip of an atomic force microscope .
the technique , known as the scanning gate microscopy @xcite ( sgm ) , gathers conductance maps as functions of the position of the tip that is capacitively coupled to the electron gas . for 2deg confined in bulk semiconductors the maps are used to extract electron trajectories @xcite , branching of the electron flow via quantum point contacts ( qpcs ) @xcite , many - body phenomena @xcite , aharonov - bohm effects @xcite and scarred wave functions in quantum billiards @xcite as well as the interference effects @xcite due to backscattering by the tip . the sgm technique has also been used @xcite for graphene with the electron gas remaining strictly on the surface of the structure . in particular , sgm technique was applied in studies of charge inhomogeneities@xcite , charging localized states formed within a widening of the nanoribbons @xcite and at potential constrictions @xcite , universal conductance fluctuations @xcite and weak localization effects @xcite .
very recently a sgm study of the graphene quantum rings was reported @xcite with conductance fringes as due to resonant localized states .
quantum point contacts are basic elements of the quantum transport circuitry allowing for the current injection and detection , control of the number of conducting of modes , etc .
the conductance quantization for qpcs defined within the bulk semiconductors is usually observed for electrostatic potentials @xcite controlling the width of the constriction . for graphene systems
this approach is rather excluded , since the klein tunneling @xcite and perfectly conducting channels @xcite in graphene nanoribbons lead to a low effectiveness of electrostatic confinement and backscattering , allowing the dirac electrons to pass across potential barriers @xcite .
the qpcs are therefore formed by tailoring constrictions @xcite of the nanoribbons @xcite .
the ineffectiveness of backscattering in graphene seems reflected by the results of the experimental sgm studies of the quantum point contacts ( qpc ) @xcite which find flat conductance maps outside the qpc @xcite in a distinct contrast to the results obtained for bulk semiconductor in which interference fringes due to the backscattering by the tip and the resulting formation of standing wave between the qpc and the probe are clearly observed @xcite .
the scanning gate microscopy was used @xcite for detection of spontaneous quantum dots formed along the disordered ribbon within the transport gap @xcite .
the purpose of the present paper is to determine the response of graphene quantum point contacts formed by constrictions of the nanoribbons to the scanning probe [ fig .
1(c ) ] . as a starting point of the study we consider reaction of homogenous graphene nanoribbons [ fig .
1(a ) ] to the perturbation by an external potential and asymmetric narrowing [ fig . 1 ( b ) ] of the channels @xcite .
the qpc [ fig .
1(c ) ] is formed by two asymmetric connections [ fig .
1(b ) ] , whose scattering properties and response to the probe needs to be determined in order to separate the effects of the resonances formed within the qpc .
we consider systems of perfect armchair or zigzag edges and systems varying with atomic - step edges , as can be produced with the joule heating technique @xcite .
for pristine graphene ribbons @xcite the edge determines the character of the ribbon ( semiconducting or semimetallic ) and the ribbons with boundaries other than armchair have the transport properties similar to the zigzag edges @xcite . for the purpose of the present study
we develop an implementation of the quantum transmitting boundary @xcite method for the solution of the coherent transport problem set by scattering solution of the schrdinger equation for the tight - binding hamiltonian .
we study both the effects of the local potential variation induced by the tip on edge - localized states and the effects of backscattering within the constrictions .
for qpcs the scattering induced by disorder @xcite and the inhomogeneity of the edges @xcite destroy the conductance quantization which can still be observed @xcite for mesoscopically smooth boundaries @xcite of the constriction . as compared to qpcs , we find that an asymmetric narrowing , i.e. a contact between nanoribbons of varied width , exhibits a very clear conductance quantization as a function of the fermi energy , also when the edges are not smooth at the mesoscopic scale .
we study quantum point contacts and find that for short constrictions there is a flat minimum of conductance for the tip above the constriction in agreement with the results of ref .
generally , for any tip potential we find that outside the graphene constrictions the variation of the conductance by the tip potential is not pronounced as in the experiments of refs .
@xcite . for longer constrictions
however a regular formation of the standing waves occurs which according to the present study
should be resolved by the scanning gate microscopy .
the resonant states localized at armchair - zigzag connections at ribbon constrictions formed by atomic steps @xcite induce a strong backscattering .
these resonant states quench the conductance to zero at low energy .
we find that the charge transport can be unblocked by the potential of the probe which neutralizes the scattering centers , with conductance maps forming halos similar to those found in the coulomb blockade experiment of ref .
the response for homogenous nanoribbons to the probe potential at low fermi energies is determined by the existence of perfectly conducting channels ( metallic armchair and zigzag ribbons ) , as well as by formation of local n - p junctions at the edges ( zigzag ribbons ) .
this paper is organized as follows : in section ii we outline the calculation method which is described in detail in the appendix , section iii , iv and v describe the response of the nanoribbons , their narrowing and constrictions , respectively ( see fig .
[ schema ] ) .
summary and conclusions are given in section vi .
numbers the elementary cells , @xmath0 the ions within the cell @xmath1 nm is the nearest neighbor distance , @xmath2 nm is the lattice constant , and @xmath3 is the period of the considered nanoribbon.,scaledwidth=30.0% ] we use the tight binding hamiltonian for @xmath4 electrons with the nearest neighbor hopping @xmath5 and the hopping energy of @xmath6 ev . for the description of the transport at the fermi level we prepared a tight - binding version of the quantum transmitting boundary method @xcite ( for the finite difference approach see ref .
the studied system consists of an input and output channels and a scattering region .
we assume that the channels are infinite and uniform , i.e. periodic in the atomic scale .
the electron eigenstates in the nanoribbons are determined in the bloch form , @xmath7 where @xmath8 is the wave vector , @xmath9 numbers the elementary cells of the ribbon ( see fig .
[ ryszu ] ) , @xmath0 the atoms within the elementary cell , and @xmath10 is the periodic function , assuming the same values within each elementary cell of the ribbon . the detailed procedure for determining the dispersion relation and bloch functions for the channels
is given in the appendix .
generally , in the hamiltonian eigenstates within the channels the scattering wave function is a linear combination of the bloch functions for a number of subbands at the fermi level @xmath11 for the electron incident to the scattering region from the left in the input ( left ) lead far away from the scattering region the electron wave function takes the form of a superposition of incoming ( @xmath12 ) and backscattered ( @xmath13 ) bloch waves , @xmath14 where @xmath15 are the amplitudes of the incoming states and @xmath16 - of the outgoing ones .
the sum in eq .
( [ eq : left1 ] ) runs over the current carrying states since at a large distance from the scatterers the contribution of the evanescent modes can be neglected . in the right
lead the wave function is a superposition of bloch waves traveling from the scattering region , @xmath17 for solution of the scattering problem we assume that the electron is incident in a single subband @xmath18 , @xmath19 .
then , the amplitudes @xmath20 and @xmath21 are to be determined . at the boundaries between the leads and the scattering region which we cover within the computational box we require continuity of both of the wave function and its derivative . the numerical procedure for determination of the scattering amplitudes
is given in the appendix .
once the scattering amplitudes are determined we proceed to calculation of transmission @xmath22 and reflection @xmath23 probabilities for the @xmath24th incident mode according to the landauer approach @xcite , @xmath25 and @xmath26 where @xmath27 and @xmath28 , are the probabilities of the electron transfer from incident subband @xmath24 to @xmath29 subband of the output and input channels , respectively , and @xmath30 [ @xmath31 is the probability current flux of the @xmath24th mode wave function traveling in the right [ left ] direction . the currents in the tight binding approach flow along the @xmath4 interatomic bonds , and the formula for the current flowing from atom @xmath32 to atom @xmath33 as derived @xcite from the schrdinger equation is @xmath34,\ ] ] where @xmath35 is the wave function at the @xmath32-th ion .
the flux is evaluated as @xmath36 where the sum runs over the atoms across the end of the channel @xmath18 and their neighbors @xmath37 within the interior of the computational box .
finally , the conductance is evaluated from the landauer formula as @xmath38 . since the formula involves sum over incident subbands , for the calculations it is sufficient to consider the solution of the schrdinger equation for a single input channel only .
note , that our choice of the quantum transparent boundary conditions implies working with the wave functions which is an alternative to the greens function method @xcite , that deals with hamiltonian operators mainly .
the choice between the two is a matter of taste .
the effective potential of the charged tip as seen by the fermi level electrons is a result of the screening of its coulomb potential by the electron gas .
the effective potential as obtained by the schdinger - poisson modeling @xcite can be approximated by a lorentz function @xmath39 where @xmath40 are the coordinates of the tip position , @xmath41 the width of the effective tip potential and @xmath42 its height .
the height of the tip potential is determined by the charge accumulated by the tip , and @xmath41 is of the range of the tip - electron gas distance @xcite .
a discussion of the results for varied @xmath42 and @xmath41 parameters is provided below .
we consider first the zigzag ribbons - labeled by zz in the figures , and next proceed to armchair : semiconducting ( as ) and metallic ( am ) ribbons .
figure [ disprel](a ) shows the dispersion relation for a zigzag nanoribbon with 102 atoms across the channel .
the zigzag edge does not couple the states of the @xmath43 and @xmath44 valleys of graphene which are thus present in the dispersion relation ( @xmath45 ) .
the dispersion relation is given in figs .
[ disprel](a - c ) . for fermi energy @xmath46
ev there are five eigenstates with current flowing to the right ( @xmath47 ) and to the left ( @xmath48 ) .
let us consider the ribbons conductance response to the tip potential for a low energy of the incident electron .
figures [ lowestsb](a - b ) show the transfer probability as a function of the tip position across the ribbon for two values of @xmath49 corresponding to the lowest subband transport ( electron incoming with wave vector @xmath50 , that can be backscattered only to @xmath51 see fig .
[ disprel](b , c ) ) .
for both the considered fermi energies the conductance gets low when the tip is located near the edge , in spite of the fact that the current for the unperturbed zigzag ribbon vanishes at the edges [ figure [ lowestsb](d - e ) ] : at the zigzag edges only the ions of a single sublattice are occupied in the hamiltonian eigenstates [ figure [ lowestsb](g - h ) ] and the current flows between the ions belonging to different sublattices @xcite . for the central position of the tip we have either @xmath52 for larger @xmath53
[ lowestsb](b ) ] or a rapid variation of @xmath22 between 0 and 1 for the lower @xmath54 ev [ fig .
[ lowestsb](a ) ] . in the plot of @xmath55 as a function of the tip position and the fermi energy given in fig .
[ zigzagtip ] we find a wedge - shaped flat region with @xmath52 .
this region is related to the perfectly conducting channel @xcite in the zigzag ribbon .
perturbation by the tip even if large does not induce scattering between the @xmath43 and @xmath44 valleys as long as the tip potential is slowly varying at the atomic scale . since the tip potential has a long range character there is no intervalley scattering , and in consequence in the lowest subband transport conditions the backscattering is absent , unless the tip is located near the edge of the ribbon [ cf .
[ sztip ] ] .
the tip placed in the center of the ribbon is not exactly transparent for the electron flow , since the current makes its way between the edge [ figure [ prondy](a ) ] , but anyway no conductance response is observed . for the tip near the edge ,
the current circulates around the defect before it is backscattered ( see fig .
[ prondy](b ) the illustrated case corresponds to @xmath56 ) .
the edge mediates in the intervalley scattering provided that its potential is raised to the fermi energy @xcite by e.g. the tip inducing formation of a local n - p junction . in fig .
[ zigzagtip ] we marked with the black line the position of the tip for which its potential at the lower ribbon edge is raised to the fermi energy , with a perfect agreement with the boundaries of @xmath52 wedge region .
the transfer probability as a function of the tip position plotted in fig .
[ lowestsb](a ) for low @xmath49 exhibits a rapid variation between 0 and 1 which corresponds to the case when the tip potential at _ both _ edges exceeds @xmath49 , hence multiple intervalley scattering events occur , resulting in either @xmath56 or @xmath52 .
the @xmath52 wedge region continues smoothly into the regime of a few transport subbands at the fermi level [ see fig .
[ zigzagtip ] ] . for larger energies , in the right ( @xmath43 ) valley of the dispersion relation [ fig .
[ disprel](a , c ) ] the number of modes carrying the current to the right is by 1 larger than the number of modes carrying the current to the left [ fig .
[ disprel](c ) ] . hence , even for a strong but long - range perturbation the ribbon carries current at least in a single mode @xcite unless the tip raises the edge potential to the fermi level as it is still the case in fig .
[ zigzagtip](a , b ) also beyond the single subbands transport conditions . in figure [ zzsb ]
we plotted the backscattering probabilities for the parameters of the tip of fig .
[ zigzagtip](a ) ( large and wide tip potential @xmath57 ev , @xmath58 nm ) for the @xmath49 range with three subbands or six wave vectors at the fermi energy [ fig .
[ disprel](b - c ) ] .
the intervalley scattering in the lowest subband ( of probability @xmath59 corresponding to the change of the wave vector from @xmath60 to @xmath61 see fig .
[ disprel](b , c ) ) occurs only for the tip near the edge of the ribbon [ fig .
[ zzsb](a ) ] , but when it does , it is nearly complete ( @xmath62 ) .
the other intervalley scattering paths @xmath63 [ fig . [ zzsb](c ) ] , @xmath64 [ fig . [ zzsb](e ) ] , @xmath65 [ fig .
[ zzsb](f ) ] occur also for the tip near the edge but with smaller probabilities .
the effective backscattering channels for the tip in the center of the ribbon are @xmath66 [ fig . [ zzsb](d ) ] that appear within the same parabolic subbands [ fig .
[ disprel](b , c ) ] .
the backscattering by the tip in the @xmath67 channel governs the current and scattering density pattern that is displayed in fig .
[ prondy](c - f ) .
the oscillations at the left to the tip are due to the interference between @xmath68 and @xmath69 eigenstates and correspond to the period of @xmath70 , which is equal to @xmath71 and @xmath72 for @xmath73 and 0.4 ev , respectively .
the conductance of the ribbon raises from @xmath74 to @xmath75 for @xmath49 above 0.2 ev [ see the dashed lines in fig .
[ zigzagtip ] ] with backscattering by the tip in the @xmath76 channel disappearing at higher fermi energies , [ cf .
[ zigzagtip](e ) ] .
+ + [ cols="^,^,^ " , ] ev and @xmath77 at @xmath78 mev and the tip near the lower edge of the ribbon ( cross ) .
the circle corresponds to potential introduced by the tip equal to @xmath49 . ]
ev , @xmath77 for energy @xmath78 mev in the lowest subband transport and the tip in the center of the ribbon ( a ) for which @xmath52 and near the lower edge ( b ) of the ribbon [ @xmath56 ] .
( c - f ) results for three subbands at the fermi level for the amplitude of the current ( c , e ) and the scattering density ( d , f ) for @xmath73 ev and @xmath80 ev . the summed transfer probability is 1 at ( c , d ) and 3 at ( e , f).,scaledwidth=50.0% ] ( a ) a ) @xmath59 , b ) @xmath81 , c ) @xmath63 , d ) @xmath82 , e ) @xmath64 , f ) @xmath65 for @xmath83 ev , @xmath77 . , scaledwidth=50.0% ] and tip position for a metallic ( a ) and semiconducting ( b ) nanoribbons for @xmath84 ev , and @xmath77 .
the dashed lines separate the regions with varied number of subbands at the fermi level . ,
scaledwidth=35.0% ] ev , @xmath77 .
the number of subbands at the fermi level is given above the plots .
for the metallic ribbon ( a - c ) @xmath85 is zero . ,
scaledwidth=46.0% ] the armchair edges of the ribbons strongly mix the valleys and the zero point energy appears only for a single wave vector in the dispersion relation for metallic channels [ fig . [ disprel](d ) ] or a single extremum per energy band is observed [ fig .
[ disprel](g ) ] for the semiconducting channel . for the semiconducting armchair ribbon
we find that the tip strictly blocks the current for any low @xmath49 [ region 1 in fig .
[ armchairskan](b ) ] while the metallic ribbon ignores the presence of the tip in the single - subband transport conditions [ region 1 in fig .
[ armchairskan](a ) ] .
in contrast to the case of the zigzag ribbon , the conductance is a smooth function of both @xmath49 and the tip position .
the low - energy results are given in fig .
[ lowestsb ] with @xmath52 for the metallic ribbon for both @xmath54 ev and @xmath53 ev , irrelevant of the tip position .
the metallic armchair ribbon possesses a perfectly conducting channel due to the pseudovalley symmetry @xcite , conserved at low energy , hence the absence of backscattering in the lowest subband transport . for the semiconducting armchair ribbon
the tip blocks the current for any tip position above the ribbon [ @xmath54 ev in fig .
[ lowestsb](a ) ] or reduces it strongly , particularly near the center of the ribbon [ @xmath86 ev in fig . [ lowestsb](b ) ] . in general ,
the most effective backscattering channel is the one within the same subband [ @xmath87 , @xmath88 and for the tip above the center of the ribbon . at higher energies , for the semiconducting ribbon , only the backscattering within the same subband is non - negligible [ fig .
[ armchair](d - f ) ] , including the non - zero channel in the lowest subband . for the metallic ribbon the backscattering within the lowest subband
is absent due to the pseudovalley symmetry @xcite . scattering in the same subband [ fig .
[ armchair](b - c ) ] is the strongest as in the semiconducting case and it is most effective for the tip above the center of the ribbon [ fig .
[ armchair](b - f ) ] . for the metallic ribbon a non - zero scattering between the subbands
is also found [ fig . [ armchair](a ) ] but only for the tip near the edges of the ribbons . in fig . [ lowestsb](c ) we plotted a cross section of fig .
[ zigzagtip](c ) and fig .
[ armchairskan ] for fermi energies in multiple subband transport regime . for higher fermi energies
, the tip can no longer form the n - p junction at the edge , and the characteristic backscattering of the zigzag ribbon is not observed .
summarizing , for higher fermi energies the overall response of the ribbon conductance to the tip position is qualitatively similar for all ribbon types , i.e. it is smooth and maximal at the center of the ribbon .
the @xmath89 dependence can be approximately put in a @xmath90 form with @xmath91 that strongly varies with the energy for any type of the edge , @xmath92 being the width of the ribbon . for higher fermi energies
the response of the ribbon to the tip scanning its surface is qualitatively independent of the type of the boundary , in spite of the fact that the current and density distributions are distinctly different [ fig .
[ lowestsb](f , i ) ] . )
armchair edges of the channels .
the number of atoms across the narrow ( wide ) channels is 188 ( 564 ) for the zigzag system , 161 ( 363 ) for metallic armchair channels and 163 ( 365 ) for semiconducting armchair edges .
the dots on the lines indicate the points for which the sgm scans of figs .
[ zwezigmaps ] and [ zweziarm ] were calculated .
, scaledwidth=42.0% ] ev and @xmath93 ( a - d ) and @xmath84 ev and @xmath77 ( e - h ) . panels ( i - l ) show the scattering probability density for the electron incident from the left in the absence of the tip . by arrows in ( k , l )
we indicated the period of the probability density oscillation which corresponds closely to the values expected for the backscattering within the last subband ( see text ) .
the inset to fig .
12(i ) shows the atomic structure of the turn of the edge with a short armchair segment that induces backscattering at low fermi energy .
, scaledwidth=52.0% ]
as an intermediate step towards description of qpcs let us consider the channel of varied width that is depicted in the inset to fig .
[ zwezene ] .
the channels at both sides of connection ( @xmath94 and @xmath95 in fig .
[ zwezene ] ) as well as the linking edges ( @xmath96 in fig .
[ zwezene ] ) are assumed of the same type ( zigzag or armchair ) .
short segments of varied type of the edge appear at the connections between the horizontal and slanted ends of the ribbon ( see the inset in fig . [ zwezigmaps](i ) ) .
figure [ zwezene ] shows the conductance of the contact as a function of the fermi energy for both zigzag and armchair edges .
the conductance exhibits a very neat quantization as a function of the fermi energy .
a pronounced quantization of conductance @xcite is generally difficult to obtain in graphene quantum point contacts ( see introduction ) .
according to the results of fig .
[ zwezene ] a channel containing a narrowing is a good candidate for observation of conductance quantization in graphene systems .
let us consider the narrowing of the ribbon with zigzag edges .
at low fermi energy of 50 mev the connection of zigzag edges has a very low conductance of @xmath97 [ see the orange line in fig . [ zwezene ] ] .
the ribbons at both sides of connection contain a perfectly conducting channel , but the inhomogeneity of the width leads to a very strong backscattering . at the connection of the zigzag edges a short sequence of an armchair form appears [ see the arrow in fig .
[ zwezigmaps](i ) ] , which opens the intervalley backscattering channel .
the tip potential above any of the two points closes the intervalley backscattering channel , allowing for the electron passage to the wider ribbon [ fig .
[ zwezigmaps](a , b , i , j ) ] . when the tip is close to the zigzag edges , it produces a rapid variation of conductance which is related to formation of the n - p connection at the edge as discussed above .
note , that this applies not only for the narrower ribbon , but also at the widening and at the wider ribbon .
thus the probe resolves the zigzag edges in the entire system .
the strong conductance response near the scattering centers localized at the edges are the counterparts of the conductance halos found in sgm mapping of graphene constrictions @xcite . for the tip above the center of the narrower ribbon in fig .
[ zwezigmaps](b ) the potential at the edge stays below the fermi level and the presence of the tip is not resolved in the map which is flat in the interior of the narrow ribbon [ fig .
[ zwezigmaps](c , f ) ] . the insensitiveness of conductance to the tip position in the center of the narrow channel and the rapid variation at the zigzag edge are consistent with the properties of the zigzag ribbons for low fermi energies discussed above . for higher energies the effects of the edges are no longer present in the conductance maps [ fig .
[ zwezigmaps](c , d , g , h ) ] .
instead we observe particularly for the lower potential at the tip [ fig .
[ zwezigmaps](c , d ) ] an appearance of a periodic oscillation of conductance .
for instance the @xmath22 maxima in the sgm maps are spaced by @xmath98 for @xmath99 ev and coincide with the minima of @xmath66 backscattering .
the @xmath22 maxima are spaced by a characteristic wavelength for the intersubband scattering @xmath100 .
the wave vectors are equal to @xmath101 and @xmath102 for @xmath99 ev , producing spacing @xmath3 equal to @xmath103 , in a perfect agreement with the oscillation period found in the sgm map . for larger @xmath42 and wider @xmath41 a deviation from this perfect agreement
is found due to variation in the wavelengths induced by the external potential of the tip and backscattering involving a larger number of subbands . in particular in fig .
[ zwezigmaps](g ) for @xmath84 ev the transport is reduced to the lowest subband ( @xmath52 ) for the tip above the channel and the oscillations disappear in the sgm map which takes the form similar to the one found for lower fermi energy fig .
[ zwezigmaps](a , b ) .
the oscillations of the scattering density observed above for the uniform ribbons [ see fig .
[ prondy ] ] resulted from the presence of the tip but were not resolved by the conductance mapping .
for the non - uniform ribbons the oscillations appear in the scattering wave functions due to the inhomogeneity of the channels already in the absence of the probe [ fig .
[ zwezigmaps](k , l ) ] and according to the present results they can be resolved by the scanning gate microscopy . the number of subbands that appear at the fermi level increases with the energy and several backscattering channels appear . for @xmath104
mev [ fig .
[ zwezigmaps](d ) ] the spacing between the extrema of @xmath22 map are @xmath105 .
the backscattering probabilities are displayed in fig .
[ zwezigbac ] .
the sgm map @xmath22 for @xmath106 mev inside the channel [ see fig . [ zwezigmaps](d ) ] is very well correlated to spatial variation of backscattering @xmath107 .
the oscillation period resulting from the difference of the wave vectors is @xmath108 with a perfect agreement with the spacing of @xmath22 extrema in the sgm map .
the intervalley scattering [ @xmath109 in fig .
[ zwezigbac](b ) ] gives only an enhanced backscattering at the exit from the thinner channel with no pronounced oscillation within .
we find as a general rule , that at least for a moderate tip potential [ @xmath110 ev , @xmath93 ] the oscillations of conductance are observed as a function of the tip position inside the zigzag thinner channel and have periodicity corresponding to the backscattering within the highest subband .
figure [ zweziarm ] displays the @xmath22 maps for the armchair edges of the narrowing [ see the inset to fig .
[ zwezene ] ] . at low energy
the conductance of both metallic and semiconducting ribbons is nearly zero [ fig .
[ zwezene ] ] .
figure [ zweziarm](a , d ) indicates that similarly as for the zigzag system at low energy [ fig .
[ zwezigmaps](a , b ) ] the conductance is raised when the tip is located at the turns of the edges : at connection of the slanted edge to the thinner ( for the metallic ribbon fig .
[ zweziarm](a ) ) or wider ( for semiconducting ribbons fig . [ zweziarm](d ) ) armchair channel .
for @xmath112 mev and the thin semiconducting ribbon we have already two subbands at the fermi level [ fig . [ zwezene ] ] and
the @xmath22 map contains a clear oscillation [ fig . [ zweziarm](e ) ] of periodicity characteristic to the backscattering in the uppermost subband . for the metallic [ fig . [ zweziarm](b ) ] ribbon at this energy the thinner ribbon is still in the single subband transport conditions , no oscillations occur exactly as for the zigzag ribbon for low fermi energy [ see fig .
[ zwezigmaps](a - b , e - f ) ] .
for armchair systems at @xmath112 mev no edge effects are observed and the backscattering occurs only for the tip near the exit from the thin channel .
the oscillations for the metallic ribbon occur when the next subbands enter below the fermi energy [ fig .
[ zwezene ] ] see fig .
[ zweziarm](c ) , with the periodicity characteristic generally to backscattering within the highest subband with the period of @xmath113 for @xmath18 being the index of the last subband .
there are exceptions to this rule , in particular when the subband is very close to the fermi energy the backscattering of the subband can be nearly complete , then the one which is observed in the oscillation is not the highest but the next lower one .
ev , @xmath93.,scaledwidth=42.0% ] maps for a narrowing with armchair edges : metallic ( a - c ) and semiconducting ( d - f ) for @xmath110 ev and @xmath93 .
the insets in a ) and d ) display the fragments of the edges near the exit from the narrower channel and the entrance to the wider channel , respectively where a short segment of zigzag edge appears , leading to the strong backscattering which is neutralized by the probe potential as in fig .
[ zwezigmaps ] above.,scaledwidth=42.0% ]
, @xmath114 and @xmath115 for the armchair semiconducting ribbon , @xmath116 , @xmath117 and @xmath118 for the armchair metallic ribbon and @xmath119 , @xmath120 and @xmath121 for the zigzag ribbon .
, scaledwidth=48.0% ] we are now ready to proceed to description of sgm maps for quantum point contacts formed within the graphene ribbons .
we first consider a system based on two narrowings of the channels as described above , in the form given in the inset to fig .
[ qpcdlugie ] with a finite length @xmath122 of the narrower ribbon .
we assume that the narrower ribbon has a width @xmath123 20 nm and consider @xmath122 between 32 nm and 58 nm , for the transverse size of the wider ribbon between 44 nm and 60 nm .
the conductance of all the considered ribbons
zigzag and armchair
both semiconducting and metallic drops to zero for low @xmath49 .
the inhomogeneity of the ribbon leads in this way to appearance of the transport gap @xcite for any type of the edge of the wider ribbon .
the conductance of the qpc with zigzag edges is displayed in fig .
[ qpcdlugie ] with the solid orange line , while the dashed orange line indicates the conductance of the narrow part of the system ( the number of subbands carrying the current to the right at the fermi level ) .
the conductance of the qpc is generally lower than the conductance of the narrower part with exceptions of a series of conductance peaks . in fig .
[ dlzig ] we plotted the scattering probability density [ fig .
[ dlzig](ai - ap ) ] and the sgm maps [ fig . [ dlzig](aa - ah ) ] for the zigzag qpc and a number of chosen fermi energies marked in fig .
[ qpcdlugie ] with points . in the @xmath22 maps we find a pronounced variation of conductance maps within the narrower part of the ribbon . outside the constriction the conductance maps are almost flat .
no conductance variation of visible in the scale of the variations within the narrower channel .
for the low - energy peak @xmath124 mev within a single subband of the narrow channel
the conductance map of the system drops when the probe is located within the constriction [ fig .
[ dlzig](aa ) ] . in the scattering densities and conductance maps
we find signatures of backscattering at the turns of the edges as discussed above for the asymmetric narrowing of the channel [ fig .
[ dlzig](aa ) ] . for higher energies [ fig .
[ dlzig](ab - ah ) ] the corners correspond to high values of the probability densities but with no counterparts in the sgm map again as in the asymmetric narrowing of fig .
[ zwezigmaps ] . for higher fermi energies , within each segment of @xmath49 corresponding to a fixed number of subbands of the narrow channel ( dashed orange line in fig .
[ qpcdlugie ] we observe that 1 ) at the first maximum of @xmath125 at the low energy side the conductance map contains a single minimum along the channel , 2 ) each subsequent @xmath22 peak that is higher at the @xmath49 scale corresponds to conductance map with a number of minima increased by 1 .
a representative example is given in fig .
[ dlzig](ab - ad ) for the @xmath22 maps for the first , second and the third @xmath22 maxima of conductance [ see the numbers near the orange line in fig .
[ qpcdlugie ] ] .
the conductance resonances correspond to an integer number of the scattering wavelengths for the highest subband within the constriction with maxima observed in the scattering probability density [ cf .
[ dlzig](aj - al ) ] .
once another subband falls below the fermi energy ( for instance above @xmath126 ev ) another sequence of conductance peaks is formed with the number of maxima within the constriction starting from 1 again ( see the peaks labeled by primed numbers and maps of fig .
[ dlzig](ae , af ) ] . beyond the resonances this regular correspondence between the scattering density and the conductance map
is lost [ cf .
[ dlzig](ag , ah ) for energies @xmath127 mev and @xmath128 mev ] .
the properties of the qpcs with armchair edges and the width corresponding to a semiconducting dispersion relation turn out to be similar to those of the zigzag system , with pronounced @xmath125 peaks [ cf .
blue solid line in fig .
[ qpcdlugie ] ] corresponding to a determined number of maxima of the probability density correlated with the conductance maps . for the metallic armchair system @xmath125 dependence is more complex [ red line in [ qpcdlugie ] ] .
in particular we find that the peaks corresponding to a fixed number of extrema of the probability density and sgm maps appear in pairs [ cf .
[ dlzig](l , m ) for @xmath129 mev and @xmath130 mev for a single extremum , and fig .
[ dlzig](n , o ) for @xmath131 mev and @xmath132 mev for the double extremum , etc . ] .
the difference in the properties of semiconducting and metallic armchair systems has its source in the details of the dispersion relation .
when a subsequent subband for the electron transport is opened within the narrow channel , the subbands of the metallic ribbon are nearly degenerate [ see fig .
[ disprel](d ) for low @xmath8 ] while for the armchair ribbon the subbands are split near zero wave vector [ see fig .
[ disprel](g ) ] . thus ,
for the metallic ribbon we have two close highest subbands of similar fermi wavelengths , inducing formation of separate but close resonances for both .
the neighborhood of the wavelengths enhances the intersubband scattering leading to a complex form of @xmath125 at higher energies .
concluding , for qpcs with a long constriction the conductance peaks of @xmath125 dependence are related to resonances of the highest subband , and the resonant density within the constriction is resolved by the sgm conductance maps .
the tip potential enhances the backscattering within the highest nearly parabolic subband .
the tip - induced backscattering is most effective in the region where the scattering probability density is maximal .
the zigzag and semiconducting narrowing display a distinct regularity in the subsequent number of peaks and the form of the sgm map . for the metallic armchair ribbon
the @xmath125 dependence is more complex due to the near degeneracy of the highest subbands near their bottom and pairs of resonances with similar conductance maps are found .
for each system , outside the constriction the conductance map is almost flat .
, @xmath134 .[twe],scaledwidth=40.0% ] let us consider the qpc with a short constriction .
we studied first a wedge shaped constriction of a zigzag ribbon depicted in the inset to fig .
[ twe ] , with the ribbon narrowing from about 60 nm to about 6 nm .
the conductance of the qpc is displayed in the main panel of fig .
the short constriction can not support any localized resonances in the form discussed above , and the corresponding series of @xmath125 peaks are not observed . for any fermi energy a minimum is found in the sgm map within the constriction and only a low variation of conductance outside the constriction is found see the maps displayed in fig .
some variation outside the constriction is only observed for low @xmath49 [ fig .
[ twem](a ) ] and it disappears at higher energies in the scale of the variation in the narrow part .
[ twem](b , c ) ] .
the results found for this short , narrow and abrupt qpc are reproduced also for a large smooth at the atomic scale constriction of a cosine profile @xcite .
the system exhibits @xcite a detectable smooth steps of conductance @xcite also when the constriction remains quite wide . the @xmath125 result is presented in fig .
[ tcos ] , where we considered wide ( @xmath135 nm to 150 nm ) zigzag and semiconducting armchair edges that at the narrowest part have width of @xmath136 nm to 50 nm .
the conductance maps for the zigzag ribbon are displayed in fig .
[ dlmaps ] . in the scattering probability density maps we find maxima along the constriction [ fig . [ dlmaps](e - h ) ] .
the atomic steps forming constrictions contain multiple of zigzag - armchair short segments of the edges [ fig .
[ dlmaps](i ) ] which as discussed above
lead to an effective backscattering . at low energy
[ @xmath137 mev in fig .
[ dlmaps](a ) ] the tip neutralizes the backscattering when located above these parts of the edges in consistence with the results for asymmetric narrowing [ cf . fig .
[ zwezigmaps](a , b ) ] . at higher energies [ cf .
[ dlmaps](b - d ) ] we are left with a minimum of conductance at the center of the constriction and a low variation of conductance in the outside .
, @xmath138 for zigzag ribbons and @xmath139 , @xmath140 for armchair ribbons.[tcos],scaledwidth=40.0% ]
we have performed an analysis of the conductance response of graphene nanoribbons and their narrowing to a perturbation introduced by potential of a scanning probe .
the study was based on the tight binding implementation of the quantum transmitting boundary method .
we find that at low fermi energies the response of the zigzag ribbons to the scanning probe occurs only for the tip above the edges of the channels due to formation of the local n - p junctions allowing for a very effective intervalley backscattering and producing rapidly varying conductance maps .
for the probe above the interior of the zigzag ribbon the conductance does not react to the perturbation .
the finding is due to existence of the perfectly conducting channel and is also observed for metallic armchair ribbon irrespective of the position of the probe . for higher fermi energies the response of the zizgag and armchair nanoribbons is similar producing smooth conductance maps as functions of the tip position with the strongest response in the center of the ribbon . for the asymmetric narrowing of the ribbon which exhibits a very distinct conductance quantization the conductance maps at low energy resolve the zigzag edges and scattering centers near the zigzag - armchair segments which necessarily appear in atomic steps that form the narrowing . for higher fermi energy an oscillation of the scattering probability density
is observed , which in general corresponds to backscattering within the highest subband of the narrower channel .
according to the present results the oscillation can be resolved in the maps of conductance response to the probe . for quantum point contacts similar oscillations of conductance maps
are observed for longer constrictions that are connected with resonant peaks of @xmath125 dependence .
formation of conductance peaks have a regular reentrant character in the energy with subsequent subbands appearing at the fermi level .
the peaks correspond to scattering probability density and conductance map with a determined number of extrema along the constriction .
shorter constrictions that do not support the resonances within the narrowing produce a smooth @xmath125 dependence with sgm maps producing a minima of conductance within the contrictions . for the asymmetric narrowing at the wider part of the ribbon and for qpc outside the constriction no conductance response of the system is observed in the response maps
. 1 h. sellier , b. hackens , m.g .
pala , f. martins , s. baltazar , x. wallart , l. desplanque , v. bayot , and s. huant , sem .
sci . tech .
* 26 * , 064008 ( 2011 ) ; d.k .
ferry , a.m. burke , r. akis , r. brunner , t.e .
day , r. meisels , f. kuchar , j.p .
bird , and b.r .
bennett , sem .
sci . tech .
* 26 * , 043001 ( 2011 ) .
k. e. aidala , r. e. parrott , t. kramer , e. j. heller , r. m. westervelt , m. p. hanson , and a. c. gossard , nat . phys . * 3 * , 464 ( 2007 ) .
b. j. leroy , a. c. bleszynski , k. e. aidala , r. m. westervelt , a. kalben , e. j. heller , s. e. j. shaw , k. d. maranowski , and a. c. gossard , phys .
* 94 * , 126801 ( 2005 ) ; m. p. jura , m. a. topinka , m. grobis , l. n. pfeiffer , k. w. west , and d. goldhaber - gordon , phys . rev .
b * 80 * , 041303(r ) ( 2009 ) ; m. p. jura , m. grobis , m. a. topinka , l. n. pfeiffer , k. w. west , and d. goldhaber - gordon , phys . rev .
b * 82 * , 155328 ( 2010 ) .
n. paradiso , s. heun , s. roddaro , l.n .
pfeiffer , k.w .
west , l. sorba , g. biasiol , f. beltram , physica e * 42 * , 1038 ( 2010 ) .
b. brun , f. martins , s. faniel , b. hackens , g. bachelier , a. cavanna , c. ulysse , a. ouerghi , u. gennser , d. mailly , s. huant , v. bayot , m. sanquer , and h. sellier , nature communications * 5 * , 4290 ( 2014 ) .
f. martins , b. hackens , m. g. pala , t. ouisse , h. sellier , x. wallart , s. bollaert , a. cappy , j. chevrier , v. bayot , and s. huant , phys .
lett . * 99 * , 136807 ( 2007 ) ; m. g. pala , b. hackens , f. martins , h. sellier , v. bayot , s. huant , and t. ouisse , phys .
b * 77 * , 125310 ( 2008 )
. r. crook , c. g. smith , a. c. graham , i. farrer , h. e. beere , and d. a. ritchie , phys .
. lett . * 91 * , 246803 ( 2003 ) ; a. m. burke , r. akis , t. e. day , g. speyer , d. k. ferry , and b. r. bennett , phys .
. lett . * 104 * , 176801 ( 2010 ) .
r. jalilian , l. a. jauregui , g. lopez , j. tian , c. roecker , m.m .
yazdanpanah , r.w .
cohn , i. jovanovic , and y.p .
chen , nanotechnology * 22 * , 295705 ( 2011 ) .
garcia , m. knig , d. goldhaber - gordon , k. todd , phys .
b * 87 * , 085446 ( 2013 ) .
s. schnez , j. gttinger , m. huefner , c. stamper , k. ensslin , and t. ihn , phys . rev .
b * 82 * , 165445 ( 2010 ) . m.r .
connoly , k.l .
chiu , a. lombardo , a. fasoli , a.c .
ferrari , d. anderson , g.a.c .
jones , and c.g .
smith , phys .
b * 83 * , 115441 ( 2011 ) .
j. berezovsky , m.f .
borunda , e.j .
heller , and r.m .
westervelt , nanotechnology * 21 * , 274013 ( 2010 ) .
j. berezovsky and r.m .
westervelt , nanotechnology * 21 * , 274014 ( 2010 ) .
d. cabosart , s. faniel , f. martins , b. brun , a. felten , v. bayot , and b. hackens , phys .
b * 90 * , 205433 ( 2014 ) .
van wees , h. van houten , c.w.j .
beenakker , j.g .
wiliamson , l.p .
kouwenhoven , d. van der marel , c.t .
foxton , phys .
* 60 * , 848 ( 1988 ) .
t. anto , t. nakanishi , and r.saito , j. phys .
. jpn . * 67 * , 2857 ( 1998 ) ; v.v . cheianov and v.i .
falko , phys . rev .
b * 74 * , 041403(r ) ( 2006 ) ; m.i .
katsnelson , k.s .
novoselov , and a.k .
geim , nature phys .
* 2 * , 620 ( 2006 ) .
beenakker , rev .
phys . * 80 * , 1337 ( 2008 ) .
k. wakabayashi , y. takane , and m. sigrist , phys .
99 * , 036601 ( 2007 ) .
k. wakabayashi , y. takane , m. yamamoto , and m. sigrist , carbon * 47 * , 124 ( 2009 ) .
r. yang , l. huang , y .- c .
lai , and c. grebogi , phys .
b * 84 * , 035426 ( 2011 ) . y.m .
lin , v. perebeinos , z.h .
chen , and p. avouris , phys .
b * 78 * , 161409(r ) ( 2008 ) .
n. tombros , a. veligura , j. junesch , m.h.d .
guimares , i.j .
vera - marun , h.t .
jonkman , and b.j .
van wees , nat .
* 7 * , 697 ( 2011 ) .
s. ihnatsenka and g. kirczenow , phys .
b * 85 * , 121407(r ) ( 2012 ) . m.y .
han , b. zyilmaz , y. zhang , p. kim , phys .
lett . * 98 * , 206805 ( 2007 ) .
z.h . chen , y.m .
lin , m.j .
rooks , p. avouris , physica e * 40 * , 228 ( 2007 ) .
f. molitor , a. jacobsen , c. stampfer , j. gttinger , t. ihn , k. ensslin , phys .
b * 79 * 075426 ( 2009 ) .
l. brey , and h.a .
fertig , phys .
b * 73 * , 235411 ( 2006 ) .
areshkin , d. gunlycke , and c.t .
white , nano lett .
* 7 * , 204 ( 2007 ) .
k. wakabayashi , y .
takane , m. yamamoto , and m. sigrist , new j. phys * 11 * , 095016 ( 2009 ) .
y.w.son , m.l .
cohen . and s.g .
louie , phys .
97 * , 216803 ( 2006 ) .
k. nakada , m. fujita , g. dresselhaus , and m. s. dresselhaus , phys .
b * 54 * , 17954 ( 1996 ) .
petrovic and f.m .
peeters , phys .
b * 91 * , 035444 ( 2015 ) .
s. neubeck , l.a .
ponomarenko , a.s .
mayorov , s.v .
morozov , r. yang , and k.s .
novoselov , physica e * 44 * , 1002 ( 2012 ) .
n. pascher , d. bischoff , t. ihn , and k. ensslin , appl .
* 101 * , 063101 ( 2012 ) .
j. wurm , m. wimmer , i. adagidelli , k. richter , and h.u .
baranger , new j. phys . * 11 * , 095022 ( 2009 ) .
x. jua , m. hofmann , v. meunier , b.g .
sumpter , j. campos - delgado , j. m. romo - herrera , h. son , y .- p .
hsieh , a. reina , j. kong , m. terrones , m.s .
dresselhaus , science * 323 * , 1701 ( 2009 ) .
a.r > akhmerov , and c.w.j .
beenakker , phys .
b * 77 * , 085423 ( 2008 )
lent and d.j .
kirkner , j. appl .
67 , 6353 ( 1990 ) . m. wimmer and k. richter , j. comput .
* 228 * , 8548 ( 2009 ) .
f. muoz - rojas , d. jacob , j. fernndez - rossier , and j. j. palacios , phys .
b 74 , 195417 ( 2006 ) .
m. evaldsson , i. v. zozoulenko , hengyi xu , and t. heinzel , phys . rev .
b 78 , 161407(r ) ( 2008 ) .
k. kolasinski and b. szafran , phys .
b 89 , 165306 ( 2014 ) .
j. tworzydlo , b. trauzettel , m. titov , a. rycerz , and c. w. j. beenakker , phys .
96 * , 246802 ( 2006 ) ; a. rycerz , j. tworzydlo , and c.w.j .
beenakker , nature physics * 3 * , 172 - 175 ( 2007 ) .
peres , a.h .
castro neto and f. guinea , phys .
b * 73 * , 195411 ( 2006 ) . k. wakabayashi , phys .
b * 64 * , 125428 ( 2001 ) .
k. kolasinski and b. szafran , phys . rev .
b 88 , 165306 ( 2013 ) .
j. wurm , m. wimmer , and k. richter , phys .
b * 85 * , 245418 ( 2012 ) .
m. zwierzycki , p. a. khomyakov , a. a. starikov , k. xia , m. talanana , p. x. xu , v. m. karpan , i. marushchenko , i. turek , g. e. w. bauer , g. brocks , and p. j. kelly , phys .
status solidi b 245 , 623 ( 2008 ) .
the procedure for determining the dispersion relation and the hamiltonian eigenstates for the nanoribbon channels is described here . for an infinite , uniform lead the tight - binding hamiltonian
can be put in a matrix@xcite @xmath141 where matrix @xmath142 concerns a single elementary cell @xmath9 [ see fig .
[ ryszu ] ] and @xmath143 describes the connection between cells @xmath9 and @xmath144 . the @xmath142 and @xmath143 matrices have dimensions @xmath145 where @xmath33 is number of atomic orbitals within a single elementary cell . for ideally periodic nanoribbons forming the channels the matrices @xmath142 , @xmath143 are the same for all elementary cells , and we skip the @xmath9 index in the following . the hamiltonian eigenfunction inside the channel can be written in a form divided on separate cells @xmath146 where @xmath147 is a vector of size @xmath33 .
the wavefunction satisfies , @xmath148 based on the bloch form of the wave function [ eq . [ blw ] ] we make the following subsitution : @xmath149 for which @xmath150 with @xmath151 the eigenproblem can be put in form @xmath152 \left [ \begin{array}{c } \mathbf{\chi } \\
\mathbf{\eta } \\
\end{array}\right ] = 0.\ ] ] the generalized @xmath153 eigenproblem has @xmath154 solutions : @xmath33 left - going and @xmath33 right - going modes propagating or decaying .
@xcite for evanescent modes it is straighforward to identify right and left going modes . for right - going evanescent modes the eigenvalue satisfies @xmath155 and for left - going evanescent modes
the propagating modes for the bloch waves of eq .
( [ blw ] ) have the form @xmath157 , with a real @xmath8 , hence @xmath158 . for a given @xmath159
we look for the values of @xmath160 and determine the corresponding wave vectors and the periodic functions @xmath161 .
finally , we evaluate the current flux ( [ cflux ] ) and determine the direction of propagation the current flux is positive ( negative ) for right ( left ) going modes .
the scattering amplitudes of eq .
( [ eq : left1 ] ) and eq .
( [ eq : right ] ) are determined in the following manner .
we evaluate the coefficients @xmath16 multiplying ( [ eq : left1 ] ) by a complex conjugate of @xmath162 : @xmath163 here we denote the inner product in discrete form @xmath164 . defining vector *
a * with @xmath165 and matrices @xmath166 , @xmath167 eq .
( [ eeq ] ) can be written in a matrix form @xmath168 which implies @xmath169 from the asymptotic conditions for the wave functions ( [ eq : left1],[eq : right ] ) we derive the boundary conditions for the algebraic set of equations for @xmath21 and @xmath20 .
let us consider derivative of the wave function [ eq .
( [ eq : left1 ] ) ] at the left end of the computational box similarly , for the other end of the computational box we multiply equation ( [ eq : right ] ) by the complex conjugate of @xmath173 : @xmath174 hence @xmath175 the derivative at the end of the computational box is @xmath176 then , using ( [ eq : cout ] ) we obtain : @xmath177 the boundary conditions given by ( [ eq : psi_lewy ] ) and ( [ eq : psi_prawy ] ) are used as expressions for the wave functions outside of the computational box in the schrdinger equation @xmath178 .
having calculated the wavefunction @xmath179 , we can calculate the coefficients @xmath16 from ( [ eq : din ] ) and @xmath180 from ( [ eq : cout ] ) .
the results are next used for evaluation of the scattering probabilities and conductance in the landauer approach . | we provide a theoretical study of the conductance response of systems based on graphene nanoribbon to the potential of a scanning probe .
the study is based on the landauer approach for the tight - binding hamiltonian with an implementation of the quantum transmitting boundary method and covers homogenous nanoribbons , their asymmetric narrowing and quantum point contacts of various profiles .
the response maps at low fermi energies resolve formation of n - p junctions induced by the probe potential and a presence of zigzag - armchair segments of the edges for inhomogeneous ribbons . for an asymmetric narrowing of the nanoribbons the scanning probe resolves formation of standing waves related to backscattering within the highest subband of the narrower part of the system .
the qpcs containing a long constriction support formation of localized resonances which induce a system of conductance peaks that is reentrant in the fermi energy , with the form of the probability density that can be resolved by the conductance mapping . for shorter constrictions
the probe induces smooth conductance minima within the constrictions . in general , besides the low - energy transport gap , in the wider parts of the ribbon the variation of the conductance of the system is low compared to the narrower part . |
the x - ray nova and black hole candidate xte j1550564 was discovered ( smith et al .
1998 ) with the all sky monitor ( asm ; levine et al . 1996 ) aboard the _ rossi x - ray timing explorer _ ( rxte ) just after the outburst began on 1998 september 6 .
the position quickly led to the identification of the counterparts in the optical ( orosz , bailyn , & jain 1998 ) and radio ( campbell - wilson et al .
1998 ) bands . early observations of the source with batse ( wilson et al .
1998 ) revealed an x - ray photon index in the range of 2.12.7 ; on one occasion the source was detected at energies above 200 kev .
the x - ray light curve of xte j1550564 from the asm is shown in figure 1 .
the source is the brightest transient yet observed with rxte : the flare of 1998 september 19 - 20 reached 6.8 crab ( or 1.6 @xmath0 erg s@xmath1 @xmath2 ) at 210 kev .
the overall profile of the outburst , with its slow 10-day rise , dominant x - ray flare , 30-day intensity plateau , and relatively rapid @xmath310-day decay timescale ( after mjd 51110 ) , is different from the outbursts of classical x - ray novae like a062000 ( see chen , shrader , & livio 1997 ) .
the apparent optical and x - ray brightness of xte j1550564 , and even its x - ray light curve , are roughly similar to that of x - ray nova oph 1977 , a black hole binary , for which there is a distance estimate of @xmath3 6 kpc ( watson , ricketts , & griffiths 1978 ; remillard et al .
1996 ) . the discovery of xte j1550564 prompted a series of pointed rxte observations with the proportional counter array ( pca ; jahoda et al
. 1996 ) and the high - energy x - ray timing experiment ( hexte ; rothschild et al .
1998 ) instruments .
these were scheduled almost daily for the first 50 days of the outburst and roughly every two days during the following two months .
the first 14 rxte observations were part of a guest observer program with results reported by cui et al .
they found that during the initial x - ray rise ( 0.72.4 crab at 210 kev ) , the source exhibited very strong qpos in the range 0.088 hz .
the rapid variability and the characteristics of the x - ray spectrum suggested that xte j1550564 is powered by an episode of accretion in a black hole binary system ( see tanaka & lewin 1995 ) .
the possible presence of a @xmath4 mag counterpart ( jain et al .
1999 ) is especially important since this may allow radial velocity studies in quiescence that could confirm the black hole nature of the primary .
herein we present 60 x - ray spectral observations spanning 71 days of the 1998 outburst of xte j1550564 .
these observations include all of our rxte guest observer program ( # 30191 ) and the first five public observations of this source .
a timing study based on these same rxte observations and observations of the optical counterpart are presented in companion papers ( remillard et al .
1999 and jain et al . 1999
; hereafter paper ii and paper iii , respectively ) .
we present 60 observations of xte j1550564 ( see fig .
1 ) obtained using the pca instrument onboard rxte .
the pca consists of five xenon - filled detector units ( pcus ) with a total effective area of @xmath3 6200 @xmath2 at 5 kev .
the pca is sensitive in the range 260 kev , the energy resolution is @xmath317% at 5 kev , and the time resolution capability is 1 @xmath5sec .
the hexte data are not presented here due to uncertainty in the pca / hexte cross calibration .
the pca data were taken in the `` standard 2 '' format and the response matrix for each pcu was obtained from the 1998 january distribution of response files .
the pulse height spectrum from each pcu was fit over the energy range 2.520 kev , using a systematic error in the countrates of 1% , and background subtracted using the standard background models .
only pcus 0 & 1 were used for the spectral fitting reported here and both pcus were fit simultaneously using xspec .
the pca spectral data were fit to the widely used model consisting of a multicolor blackbody accretion disk plus power - law ( tanaka & lewin 1995 ; mitsuda et al . 1984 ; makishima et al . 1986 ) .
the fits were significantly improved by including a smeared fe aborption edge near 8 kev ( ebisawa et al .
1994 ; inoue 1991 ) and an fe emission line with a central energy around 6.5 kev and a fixed width of 1.2 kev ( fwhm )
. the fitted equivalent width of the fe emission line was @xmath6 ev .
interstellar absorption was modeled using the wisconsin cross - sections ( morrison & mccammon 1983 ) .
the fitted hydrogen column density varied from 1.7 to 2.2 @xmath7 @xmath2 and was fixed at @xmath8 @xmath2 in the analysis presented here , resulting in a total of eight free parameters .
three representative spectra are shown in figures 2a c .
these spectra illustrate the range of x - ray spectra for xte j1550564 , in which an intense power - law component dominates a hot disk ( a ) , a strong power - law dominates a warm disk ( b ) , or the disk dominates a weak power - law ( c ) .
the addition of the fe emission & absorption components is motivated in figures 3a & b , which show the ratio of a typical spectrum to the model without and with the fe emission & absorption .
the addition of the fe emission & absorption components reduces the @xmath9 from 7.9 to 0.9 in this example .
the fitted temperature and radius of the inner accretion disk presented here ( @xmath10 & @xmath11 ) are actually the color temperature and radius of the inner disk , which are affected by spectral hardening due to comptonization of the emergent spectrum ( shakura & sunyaev 1973 ) .
the corrections for constant spectral hardening are discussed in shimura & takahara ( 1995 ) , ebisawa et al .
( 1994 ) , and sobczak et al .
( 1999 ) , but are not applied to the spectral parameters presented here .
the physical interpretation of these parameters remains highly uncertain .
the model parameters and component fluxes ( see table 1 ) are plotted in figures 4a e . all uncertainties are given at the @xmath12 level .
the spectra from mjd 51074 to 51113 are dominated by the power - law component which has photon index @xmath13 2.352.86 ( fig . 4c , table 1 ) .
the source also displays strong 313 hz qpos during this time ( paper ii ) , whenever the power - law contributes more than 60% of the observed x - ray flux .
this behavior is consistent with the _ very high state _ of black hole x - ray novae ( bhxn ) ( see tanaka & lewin ( 1995 ) and references therein for further details on the spectral states of bhxn ) .
after mjd 51115 , the power - law component weakens rapidly , with @xmath14 2.02.4 , and the disk component begins to dominate the spectrum ( see fig .
the source generally shows little temporal variability during this time ( paper ii ) .
we identify this period with the _ high / soft state_. however , during this time the source occasionally exhibits qpos at @xmath3 10 hz ( table 1 ) and we identify those observations with the _ intermediate state_. the _ low state _ was not observed and the intensity increased again after mjd 51150 . from figure 4b
, it appears that the inner radius of the disk does not remain fixed at the last stable orbit throughout the outburst cycle . from table 1 , we see that the intense flare on mjd 51075 is accompanied by a dramatic decrease in the inner disk radius from 33 to 2 km ( for zero inclination ( @xmath15 ) and @xmath16 kpc ) over one day . similar behavior was observed for gro j165540 during its 1996 - 97 outburst : the observed inner disk radius decreased by almost a factor of four during periods of increased power - law emission in the very high state and was generally larger in the high / soft state ( sobczak et al .
1999 ) .
the physical radius of the inner disk may vary in these systems , by as much as a factor of 16 in the case of xte j1550564 .
another possibility , however , may be that the apparent decrease of the inner disk radius observed during intense flares from these two sources is caused by the failure of the multicolor disk model at these times .
this failure could occur when spectral hardening becomes significant in the inner disk , causing the color temperature to assume a steep radial profile ( shimura & takahara 1995 ) .
in such a case , fitting the multicolor disk model ( which assumes @xmath17 ) to the resulting spectrum yields an inner disk radius which is smaller than the physical value ( sobczak et al .
thus the actual physical radius of the inner disk may remain fairly constant in the presence of these intense flares .
the peak luminosity ( bolometric disk luminosity plus 2100 kev power - law luminosity ) observed during the flare on mjd 51075 is @xmath18 erg s@xmath1 , which corresponds to the eddington luminosity for @xmath19 at 6 kpc .
we have analyzed rxte data obtained for the x - ray nova xte j1550564 .
satisfactory fits to all the pca data were obtained with a model consisting of a multicolor disk , a power - law , and fe emission and absorption components .
xte j1550564 is observed in the very high , high / soft , and intermediate canonical outburst states of bhxn .
the source exhibited an intense ( 6.8 crab ) flare on mjd 51075 , during which the inner disk radius appears to have decreased dramatically from 33 to 2 km ( for zero inclination and @xmath16 kpc ) .
however , the apparent decrease of the inner disk radius observed during periods of increased power - law emission may be caused by the failure of the multicolor disk and the actual physical radius of the inner disk may remain fairly constant .
this work was supported , in part , by nasa grant nag5 - 3680 .
partial support for j.m . and
g.s . was provided by the smithsonian institution scholarly studies program .
c.b . acknowledges support from an nsf national young investigator award .
campbell - wilson , d. , mcintyre , v. , hunstead , r. , & green , a. 1998 , 7010 chen , w. , shrader , c. r. , & livio , m. 1997 , , 491 312 cui , w. , zhang , s. n. , chen , w. , & morgan , e. h. 1999 , , 512 , l43 ebisawa , k. , ogawa , m. , aoki , t. , dotani , t. , takizawa , m. , tanaka , y. , yoshida , k. , miyamoto , s. , iga , s. , hayashida , k. , kitamoto , s. , & terada , k. 1994 , , 46 , 375 inoue , h. 1991 , in frontiers of x - ray astronomy , ed .
y. tanaka & k. koyama ( tokyo : universal academy press ) , 291 jahoda , k. , swank , j. h. , giles , a. b. , stark , m. j. , strohmayer , t. , zhang , w. , & morgan , e. h. 1996 , proc .
spie 2808 , `` euv and gamma ray instumentation for astronomy '' vii , 59 jain , r. , bailyn , c. d. , orosz , j. a. , remillard r. a. , & mcclintock , j. e. 1999 , , in press levine , a. m. , bradt , h. , cui , w. , jernigan , j. g. , morgan , e. h. , remillard , r. , shirey , r. e. , & smith , d. a. 1996 , , 469 , 33 makishima , k. , maejima , y. , mitsuda , k. , bradt , h. v. , remillard , r. a. , tuohy , i. r. , hoshi , r. , & nakagawa , m. 1986 , , 308 , 635 mitsuda , k. , et al .
1984 , pasj , 36 , 741 morrison , r. & mccammon , d. 1983 , , 270 , 119 orosz , j. , bailyn , c. , & jain , r. 1998 , 7009 remillard , r. a. , orosz , j. a. , mcclintock , j. e. , & bailyn , c. d. 1996 , , 459 , 226 remillard , r. a. , mcclintock , j. e. , sobczak , g. j. , bailyn , c. d. , orosz , j. a. , morgan , e. h. , & levine , a. m. 1999 , , in press rothschild , r. e. , blanco , p. r. , gruber , d. e. , heindl , w. a. , macdonald , d. r. , marsden , d. c. , pelling , m. r. , wayne , l. r. , & hink , p. l. 1998 , , 496 , 538 shakura , n. i. & sunyaev , r. a. 1973 , , 24 , 337 shimura , t. & takahara , f. 1995 , , 445 , 780 smith , d. a. & rxte / asm teams 1998 , 7008 sobczak , g. j. , mcclintock , j. e. , remillard , r. a. , bailyn , c. d. , & orosz , j. a. 1999 , , 520 , in press tanaka , y. & lewin , w. h. g. 1995 , in x - ray binaries , ed . w. h. g. lewin , j. van paradijs , & e. p. j. van den heuvel ( cambridge : cambridge univ .
press ) watson , m. g. , ricketts , m. j. , & griffiths , r. e. 1978 , , 221 , l69 wilson , c. a. , harmon , b. a. , paciesas , w. s. , & mccollough , m. l. 1998 , 7010 | excellent coverage of the 1998 outburst of the x - ray nova xte j1550564 was provided by the _ rossi x - ray timing explorer_. xte j1550564 exhibited an intense ( 6.8 crab ) flare on 1998 september 19 ( ut ) , making it the brightest new x - ray source observed with rxte .
we present a spectral analysis utilizing 60 proportional counter array spectra from 2.520 kev spanning 71 days , and a nearly continuous all sky monitor light curve .
the spectra were fit to a model including multicolor blackbody disk and power - law components .
xte j1550564 is observed in the very high , high / soft , and intermediate canonical outburst states of black hole x - ray novae . |
the understanding and control of the decoherence of small quantum systems is central to a lot of recent developments in the fields of nanotechnology and quantum computers @xcite .
for example , the efficacy of qubits , the basic building block of a quantum computer , which can be spin qubits , josephson junction qubits , charge or flux qubits depends largely on the environments , or baths , to which they are often weakly coupled .
the effect of environments on the coherence of the qubit has been studied in various contexts with particular emphasis on bosonic baths @xcite . in the past few years
, the realization of solid state qubits in semiconducting heterojunctions has also resulted in the study of spin baths constituted of spins . at very low temperatures ,
the decoherence engendered by a spin environment is expected to dominate the loss of the coherence arising from a coupling to phononic degrees of freedom . in the limit of weak coupling between the qubit and the bath , the effect of intrabath interactions
have been explored earlier in other works @xcite . in some cases
, interactions were found to decrease the rate of decoherence , though the exact opposite was seen in cases where the spin bath was on the verge of a standard magnetic quantum phase transition @xcite . on the other hand , the qualitative nature of the decoherence i.e. , whether it is markovian or non - markovian depends largely on the nature of the initial state of the bath@xcite . for baths at zero temperature , the decoherence is often non - markovian .
examples include the well known spin - boson model @xcite , and models of interacting spin environments that we already mentioned above @xcite .
it has also been found in the case of a qubit interacting via the fermi contact hyperfine coupling with a bath of polarized nuclear spins @xcite .
for baths in thermal equilibrium i.e. , at _ finite temperatures _ , the decoherence in the weak coupling limit is expected to be markovian as in the case of the spin - boson model .
this aspect was also seen in the case of interacting spin baths at finite temperatures@xcite .
the associated markovian decay rate was also found to increase with temperature and , unlike the case of bosons , was found to saturate at high enough temperatures .
since interactions between spins in condensed matter systems are known to generate a whole range of complex thermodynamic behaviours , it is interesting to ask whether interactions between the bath spins generate novel behaviour for the coherence of a qubit coupled to such a bath . in this general context ,
various questions arise naturally : does the decoherence contain clear signals about the underlying thermodynamic phase of the bath ?
is markovian decay to be expected for all interacting spin environments at _ finite temperatures _
, where the bath is in a statistical mixture at temperature @xmath0 , and not in a pure state @xcite ? and can decoherence or others physical quantities asociated with the qubit be used as a sensitive probe of the dynamics of the underlying spin environment ? here , we shall explore some of these issues by considering the problem of the decoherence of a qubit induced by weak coupling to a model spin - bath with non - trivial dynamics arising from strong frustration and prepared in an initial state which is in thermal equilibrium .
highly frustrated or disordered baths were partially explored in [ ] where complete predictions for the decoherence could be made only for the case of a one dimensional ising bath or for spin shards in the infinite temperature limit . though these works illustrate the potential richness of interacting environments they also highlight the difficulty and limitations of using analytical methods to study these problems .
we adopt a disordered spin bath described by the meanfield random ising model in a transverse magnetic field .
this environment is characterized by a rich thermodynamic phase diagram which shows a spin - glass to paramagnetic transition as a function of temperature and external magnetic field @xcite .
the total hamiltonian for the system is given by @xmath1 where @xmath2 is the bath hamiltonian and @xmath3 denotes the spin - bath interaction .
the pauli matrices @xmath4 and @xmath5 denote the spin of the qubit and the @xmath6-th spin of the bath respectively .
the exchange energies @xmath7 are quenched random variables with a probability distribution @xmath8 , where the standard deviation of the distribution @xmath9 sets the unit of energy .
the qubit couples to the bath operator @xmath10 , with @xmath11 representing either the @xmath12 or the @xmath13 component of the spins .
the coupling constants @xmath14 are also taken to be quenched random variables .
the physics of the bath is described by @xmath2 , the random transverse ising model ( rtim ) which is known to have the following phase diagram : at zero temperature , for transverse fields smaller than a critical field @xmath15 , i.e. , @xmath16 , the system is in a spin glass phase @xcite . for @xmath17
the system is magnetized with a gap in its spectrum . at finite temperatures ,
the spin glass order survives up to a critical temperature @xmath18@xcite . a physical system which is expected to be reasonably described by the above model is @xmath19 , where the ho concentration , x , tunes the system between different physical regimes @xcite .
it has been proposed as a minimal toy model to investigate the effects of quantum entanglement of spins @xcite .
the critical temperature @xmath20 , that sets the value of the coefficient @xmath9 , is of the order of @xmath21 @xcite , in the compounds .
consequently , the effective high temperature paramagnetic phases we study in this paper correspond to rather low real temperatures , where we expect the spin environment to dominate and phononic contributions to the decoherence can be neglected .
there have also been propositions for engineering model magnetic environments with very small energy scales using cold atoms @xcite .
we note that in the spin glass phase , only the correlations involving the @xmath13-component of the spin exhibit spin glass features whereas the correlation of the @xmath22 components merely exhibit a gap .
clearly , it would be interesting to explore the effect of these phases on the decoherence of the qubit .
moreover , due to the inherent anisotropy of the bath , we expect the decoherence to be dependent on whether the qubit couples to the @xmath13 or @xmath12 components of the bath spins . in this paper
, we address these questions in the limit of weak qubit - spinbath coupling .
since this problem can not be studied analytically , we use numerical exact diagonalization methods to calculate the bath eigenstates and eigenvalues and consequently , the resulting decoherence .
we obtain a rich spectrum of results for the coherence of the qubit and in particular , a non - markovian i.e. , a non - exponential decay of the coherence at finite temperatures .
the paper is organized as follows : in sec .
[ sec : formalism ] , we present the weak coupling formalism used to calculate the decoherence of the qubit .
we present our numerical method in sec .
[ sec : numerical ] and use it to study the analytically tractable case of decoherence induced by an ising chain spin - bath so as to benchmark our method .
we compare our results for the markovian decoherence rate with known analytical results for an infinitely long chain , so to establish the importance of the finite size effects introduced by our numerical method . in sec .
[ sec : results ] , we present our results for the decoherence in the weak coupling regime for the long range sherrington - kirkpatrick model which has an ising spin glass phase at low enough temperatures and present a comparison of our results with some analytical results which are known in the high temperature paramagnetic phase . in sec . [
sec : tim ] , we present in detail our main results for the decoherence induced by the rtim spin - bath for two different couplings of the qubit to the bath . we conclude with a discussion of our results in sec . [
sec : conclusions ] .
in this section , we summarize the weak coupling formalism @xcite used to calculate the decoherence .
this approach is valid provided the energy associated with the qubit - bath coupling is smaller than all the scales of the bath .
we first assume that at time @xmath23 , the combined system of the central spin and the bath have a factorizable initial density matrix : @xmath24 . depending on the coupling of the central spin to the bath operator @xmath25 , the qubit is in a pure state whose basis vectors
are defined by @xmath26 ( @xmath27 ) where @xmath28 , @xmath29 and @xmath30 , @xmath31 .
the bath is chosen to be at thermal equilibrium with temperature @xmath32 leading to a density matrix @xmath33 where @xmath34 is the bath partition function .
the time evolved reduced density matrix is given by @xmath35 where the factor @xmath36 is a measure of the decoherence induced by the bath at time @xmath37@xcite .
note that @xmath38 denotes the usual trace as @xmath2 and @xmath25 are operators in the bath hilbert space . for weak coupling to the environment i.e. , for small @xmath14
, we can use the super - operator formalism@xcite to obtain the following form for the decoherence@xcite , being the laplace transform of @xmath39 @xmath40 where @xmath12 is a complex variable with @xmath41 . as shown in [ ] , this laplace transform can be written as @xmath42^{-1 } \label{eq : lt}\ ] ] where the self - energy @xmath43 up to second order is given by @xmath44 \label{eq : sosigma}\end{aligned}\ ] ] where the connected correlation functions are defined as @xmath45 the coherence @xmath46 can thus be written in terms of the real functions @xmath47 and @xmath48 defined by @xmath49 where @xmath50 is real .
this then leads to the weak coupling result @xmath51 where @xmath52 is the heaviside step function and @xmath47 and @xmath48 satisfy standard kramers - kronig relations . in general , @xmath53 might have analytic as well as non - analytic parts .
it is interesting to note that in the weak coupling limit , the decoherence is essentially dictated by @xmath53 which as can be seen from ( [ eq : sosigma ] ) is proportional to the symmetrized dynamic structure factor of the bath . typically for spin baths , we encounter situations where
, one can have a net magnetization along a certain spin direction @xmath54 , and/or magnetic ordering where @xmath55 with @xmath56 and @xmath57 being constants independent of time , and @xmath58 a time - varying function .
the resulting self energy in the laplacian variables then takes the form @xmath59 where @xmath60 is some analytic function of @xmath12 . for the asymptotic decoherence ,
the first two terms generate oscillations resulting in a decoherence of the form @xmath61 where @xcite @xmath62 and @xmath63\ ] ] being @xmath64 the fourier transform of the function @xmath58 .
( [ eq : mtilde ] ) is also known as the time - convolutionless projection operator ( tcl ) approximation @xcite .
the true intermediate time decoherence is expected to be lightly modified with respect to the result predicted by , but it is expected to be qualitatively similar to that predicted by , which is valid for all @xmath37 . in the absence of any intrinsic dynamics of the central spin , the second order aproximation used here , leads to an equation for the decoherence which resembles that obtained for the spin - boson model @xcite . though is exact for the spin - boson case , here it is valid only in the limit of weak qubit - spinbath coupling .
morevoer , not all spectral densities @xmath53 obtained from interacting systems can be simulated by the usual non - interacting boson baths . for the disordered systems studied here , the correlation functions ( [ eq : conncorr ] ) and ( [ eq : sosigma ] )
need to be averaged over the probability distributions of both the exchange interactions @xmath7 and the coupling constants @xmath65 . in the rest of the paper , the coupling constants @xmath65 in
are chosen to have the following disorder averages : @xmath66 and @xmath67 where @xmath68 is the total number of spins in the bath .
this choice leads to the vanishing of the first order correction to the self energy since the disorder average of @xmath56 is zero .
the disorder averaged @xmath48 , is then directly related to the connected local spin correlation functions @xmath69 , and it is evaluated using the spectral representation for the dynamical structure factor @xmath70 where @xmath46 is the number of realizations of disorder for the exchange interactions @xmath7 and @xmath71 is the disordered average of the square of the local magnetic moment .
our method for evaluating ( [ eq : spectral ] ) is presented below .
to compute @xmath72 for the rtim spin - bath we use exact diagonalization methods .
the advantage of the full diagonalization is that it permits us to calculate finite temperature quantities as well .
we first study the coherence of the qubit coupled to a random ising chain bath described @xmath2 with the external field set to @xmath73 , where known analytical results can be used to benchmark the implemented numerical method .
the interactions in the bath are confined to nearest neighbors on the chain and the spin - bath coupling is taken to be in the @xmath74 direction .
we consider baths with the number of spins @xmath68 varying from 3 to 10 spins . for a given set of exchange interactions , the resulting hamiltonian which is a @xmath75 matrix is exactly diagonalized and
the eigenenergies and functions are obtained .
these are then used to compute the dynamical structure factor and @xmath53 .
since the interactions are quenched random variables , we repeat the above procedure for several thousands of realization of disorder and we obtain the disorder averaged @xmath48 . in fig . [ fig : sizeeffects ] , we plot @xmath53 for the case of the ising bath for different sizes of the bath , and fixed temperature @xmath76 . the analytical results for the thermodynamic case are also shown in the graph .
( color online ) @xmath77 as a function of @xmath50 for @xmath76 and different @xmath68 .
note that as @xmath68 increases , the numerical results approach the analytical result ( an . ) .
a similar result is found for the sk long range system . ]
note that as the number of spins @xmath68 in the chain increases , the numerical results converge to the analytical curve obtained for a thermodynamic bath . at high @xmath50
, @xmath53 is almost independent of the size of the bath .
this is due to the fact that the typical energy bonds are of order @xmath7 ( @xmath78 ) ( independent of the size of the bath ) , which are sufficiently small to produce any modification in the high energy spectrum . at low energies ,
size effects are most significant , since at finite sizes there are no exact cancellations that lead to the vanishing of @xmath53 at @xmath79 .
nevertheless , the spectra shows good uniform convergence as @xmath68 increases . at higher
@xmath0 , when the thermal energy is higher than the typical energy of the bonds , the convergence is even better and @xmath53 is roughly independent of the size of the bath . in fig .
[ fig : tempeffects ] , we plot the variation of @xmath53 for different temperatures . as shown in ref . , it is the low frequency part that is mainly affected by the temperature effects .
@xmath77 as a function of @xmath50 for different @xmath0 for @xmath80 .
] at high @xmath0 , @xmath53 is basically a broad peak of width @xmath9 , centered at @xmath79 .
this is due to the thermal fluctuations which result in uncorrelated spins , and @xmath53 merely reflects the distribution of the bonds @xmath7 and the structure of the energy levels .
both temperature and size effects are seen to have an impact on the low frequency part of @xmath53 with the consequence that the asymptotic decoherence is more sensitive to thermal fluctuations and finite size effects .
we use our numerical results for @xmath53 to calculate the decoherence @xmath39 given by .
a comparison of our results with the analytical results obtained for the infinite chain @xcite permits us to gauge the importance of finite size effects . as for the infinite chain case of ref .
, the decoherence for the finite chain can also be analyzed in terms of three regimes described by the two characteristic times : @xmath81 and @xmath82 . at short times , the coherence is given by the universal gaussian , @xmath83 , given by the sum rule of @xmath53@xmath84 the asymptotic markovian regime , where the coherence decays exponentially is characterized by the values of @xmath53 at @xmath79 ( for @xmath85 , @xmath86 ) . at high @xmath0 , @xmath39 remains practically constant up to @xmath87@xmath88 , where a final markovian regime arises .
when @xmath89 , the gaussian regime is followed by an intermediate time regime where the coherence decay as a power law , before reaching the markovian regime .
a careful inspection of our numerical data shows that , @xmath90 scales as @xmath91 ( fig .
[ fig : cadgamma0 ] ) .
the extrapolated values of @xmath90 for @xmath92 as a function of @xmath0 are plotted in fig .
[ fig : cadgamma0 t ] , where it can be seen that there is an excellent agreement between the analytical values given by@xcite @xmath93 , and the extrapolated values except for @xmath94 ( inset of fig .
[ fig : cadgamma0 t ] ) . at
very small @xmath0 , since both size effects and temperature effects are non negligible at low energies , we surmise that the accessible system sizes are not sufficient to reach the scaling regime .
@xmath95 as a function of @xmath91 .
note the excellent agreement of the extrapolated value of @xmath90and the analytical one , except at low @xmath0 ( @xmath94 ) . ]
@xmath95 ( extrapolated to the thermodynamic limit ) as a function of @xmath0 .
good agreement with analytical results ( dotted curve ) except at @xmath94 ( inset ) . ] to summarize , we see that our numerical method does reproduce the expected analytical behavior for the spin chain bath except at low temperatures where , the markovian rate is overestimated in our approach .
nonetheless , our results reproduce the existence of the different regimes .
we would like to draw attention to the fact that the benchmark case of the chain is in a sense a worst case scenario .
in fact , for a 1-d spin - bath one expects to have stronger finite size effects compared to the models to be studied in the rest of the paper : i.e. , infinite - range models which are known to have much reduced finite size effects @xcite . before leaving this numerical method section
we would like to note that our method differs substantially from others that are based on the solution of finite size clusters and cluster expansions .
those methods usually deal with a non - disordered model hamiltonian , which is solved for a small system size .
the finite size effects are dealt with by implementing various cluster expansion schemes . in those cases ,
the finite nature of the cluster provides a discrete pole structure with a finite low energy cut - off and hence a finite recursion time that provides a long time cut - off for the study of the decoherence effects . in our case
the situation is different .
the effect of the bath enters through the spin susceptibility , and this quantity in a disordered model is computed through the disorder average . since we have a continuous ( gaussian ) distribution of couplings , the disorder - averaged susceptibility , though computed on finite size clusters , does not have a finite low - energy cut - off .
therefore , the decoherence does not have a finite recursion time .
all the systematic errors in our approach are due to the second order approximation , which is safe so long the coupling between the qubit and the bath is small , and to the finite size of the clusters that are diagonalized to compute the susceptibility of the bath s hamiltonian .
this latter quantity can be , nevertheless , extrapolated to large system sizes , as we shall see latter .
this is validated by the benchmark of the method against an analytically solvable case that we describe in the beginning of next section .
in this section , we apply our numerical method to study the decoherence induced by interacting quantum spin - baths which do not have conventional magnetic order like ferromagnetism or antiferromagnetism .
the models studied all have infinite range interactions : i.e. , all spins interact with each other which leads to the presence of a significant geometric frustration .
similar models have been studied in ref . , where the focus was on the nature of the distribution of the energy level spacings and their impact on the decoherence .
a standard lore is that thermal fluctuations in the weak coupling regime always leads to a markovian i.e. , exponential decoherence . here
, we will discuss cases , where the decoherence remains highly non - markovian in certain finite temperature paramagnets .
we study the decoherence induced by the infinite - range random ising bath also known as the sherrington - kirkpatrick ( sk ) model @xcite . for a gaussian distribution with zero mean of the exchange interactions between the spins , the system is paramagnetic except below the spin - glass transition temperature @xmath96 where the system develops spin glass order . note that for the infinite range model , extensiveness of the free energy requires as to scale the interactions @xmath97 . in this case
, there is no first order contribution to the self energy and the second order contribution is given directly by the full correlation function as opposed to the connected correlation function of ( [ eq : conncorr ] ) i.e. , both @xmath98 .
moreover , @xmath48 is proportional to the probability distribution of the local magnetic fields as discussed in ref . .
this distribution is straightforward to evaluate analytically in the paramagnetic phase , but numerical methods are required to obtain the same in the spin glass phase .
earlier work only discussed the evolution of the markovian rate close to the spin glass transition temperature @xcite . here , we compute the coherence for all times and find an asymptotic markovian regime at all finite temperatures . our results are shown in fig .
[ fig : csg97n9 ] and we see that they qualitatively resemble the results obtained for the ising chain .
moreover , our method reproduces known analytical results in the paramagnetic phase and also approaches the expected linear behavior of @xmath99 at low frequencies as @xmath100 @xcite .
( color online ) @xmath77 for different values of @xmath0 and @xmath101 , for the sk bath .
the results are qualitatively similar to that of the chain . ]
( color online ) @xmath102 for the sk bath and the comparison with the free spin - bath ( fsb ) ; ( a ) for short times , where it can be seen that at @xmath103 , there s a change of the curvature in @xmath104 ; ( b ) for long times . ] the numerical results for @xmath104 are shown in figure [ fig : csgmt ] . at short times , @xmath105 ,
the sum rule obeyed by @xmath48 ensures that @xmath39 decays as a gaussian akin to the free spin bath . for times
@xmath106 , temperature starts playing a relevant role . at high @xmath107 , since @xmath53 is a peak of width @xmath78 cf .
[ fig : csg97n9 ] , @xmath39 decays as a markovian for all times @xmath108 . for @xmath109 ,
the asymptotic markovian regime is preceded by a power law regime resulting from the partial linear behavior of @xmath53 for small @xmath50 . as @xmath110 , this intermediate power law regime
is expected to extend to the asymptotic regime .
this is however , hard to infer from the numerics , since finite size effects smear the value of @xmath90 and hence , the linear behavior of @xmath53 predicted by the tap - method calculation [ ] .
nevertheless , we see that as @xmath0 decreases from the paramagnetic phase , @xmath90 decreases and the linear contribution to @xmath53 become more significant . extrapolating the results of @xmath90 to the limit of @xmath92 to obtain the markovian decay rate in the thermodynamic limit
, we find that @xmath90 scales @xmath111 cf .
[ fig : csggamma0 ] ( instead of the @xmath91 scaling in the ising chain ) .
the resulting markovian rate is plotted as a function of @xmath0 in fig .
[ fig : csggamma0 t ] , and perfectly matches the analytical values predicted in [ ] in the pm - phase . as before , there is a mismatch of the results in the very low temperature regime , probably because our system sizes do not access the scaling regime . as in the case of the ising chain , we find that interactions between the spins do lead to longer coherence times as opposed to the case of a free spin bath
. the only visible effect of the spin glass phase transition is the point of inflexion in the curvature of @xmath112 for @xmath113 when the asymptotic markovian regime takes over .
this curvature is negative for @xmath114 and positive for @xmath115 .
the absence of a radical change in the decoherence is not surprising since the qubit does not couple to the operator which corresponds to the spin glass order parameter . scaling plot for @xmath95 as a function of @xmath111 for the sk bath . ]
@xmath95 ( extrapolated to the thermodynamic limit ) as a function of @xmath0 for the sk bath .
the dashed curve represents the analytical results in the paramagnetic phase . ] in the following section , we study the rtim which is known to have both , a quantum phase transition at @xmath116 , and a finite temperature classical phase transition .
we study two different couplings between the qubit and the bath spin operators and show that the nature of these couplings has highly non - trivial consequences for the decoherence .
the sherrington - kirkpatrick model in a transverse field ( rtim ) is a more interesting case than the previously studied @xmath73 case . at @xmath116 ,
the system undergoes a phase transition from a spin glass state for @xmath118 to a gapped phase for @xmath119 . as temperature
is increased , the spin glass phase disappears at a finite temperature which depends on the value of the magnetic field .
the magnetic field @xmath120 is therefore , expected to deeply influence the behavior of @xmath53 , and hence the coherence of the central spin .
since the model is not isotropic , it is important to note that the spin glass order exists only along the @xmath13 component of the spin . to probe the physical ramifications of this anisotropy
, we examine two different couplings of the central spin to the bath : a coupling of the spin operators in the direction of the field ( @xmath74 ) and a coupling perpendicular to the field ( @xmath121 ) . as we will show below , these lead to radically different predictions for the decoherence .
previous numerical studies @xcite of this model have shown finite size effects to be rather minimal .
given the numerical complexities in the vicinity of the phase transition , we limit the scope of the present study to consider two values of the field @xmath122 and @xmath123 , that set the system in qualitatively different regimes .
since a magnetic moment is present in the @xmath12 direction for @xmath124 , the @xmath53 is given by the full connected correlation function .
as shown in ( [ eq : spectral ] ) , this contributes a singular term @xmath125 to @xmath53 which then leads to oscillations in @xmath39 ( equation [ eq : mt - osc ] ) .
an unambiguous way to extract the coefficient @xmath57 is via the sum rule satisfied by the dynamical structure factor @xcite .
this , however , is numerically cumbersome and in the rest of the paper , we do not deal explicitly with these singular terms , since they only induce oscillations and instead concentrate on the non singular part of @xmath53 that leads to a decay of the coherence .
the @xmath116 behavior of @xmath53 when a small field @xmath120@xmath126@xmath15 is applied is shown in figure [ fig : csgtrans97n9h0.1hcn9dift ] .
we note that even for such small fields , @xmath53 is radically different from the earlier case of @xmath73 .
( color online ) @xmath77 for different @xmath0 , with @xmath101 and magnetic field set to @xmath120@xmath126@xmath15 .
note the completely different behavior from the case @xmath73 ( fig .
[ fig : csg97n9 ] , at small @xmath0 , @xmath53 is gapped ( right inset , @xmath53 in a logarithmical scale ) . for higher @xmath0 ( left inset ) , @xmath53 is linear at low energies .
we note that the @xmath116 curve is indistinguishable from the one at @xmath127 . ]
our results for @xmath53 obtained for @xmath101 are shown in figs .
[ fig : csgtrans97n9h0.1hcn9dift ] and [ fig : csgtrans97h2hcn9dift ] , for the small and large field case respectively . in the former , with @xmath122 , @xmath53 exhibits a gap of order @xmath128 at @xmath116 ( numerically , the @xmath116 curve is indistinguishable from the @xmath127 curve shown in fig . [
fig : csgtrans97n9h0.1hcn9dift ] ) . at high
@xmath0 , @xmath53 shows a pronounced peak around @xmath129 and broader peaks of width @xmath130 in the background .
similar features are seen at high magnetic fields ( fig .
[ fig : csgtrans97h2hcn9dift ] ) : existence of a gap @xmath128 at @xmath116 and a two peak structure of width @xmath131 at high temperatures . for high enough @xmath0 , the susceptibility at low frequencies becomes linear with a temperature dependent slope , @xmath132 as @xmath133 .
note however , that in the high field case the @xmath134 term arising from moment formation carries most of the spectral weight at @xmath116 , as the spectral weight of the regular part of @xmath53 decreases dramatically .
the principal difference between the high and low field case is the presence of the sharp peak around @xmath135 in the low field case .
we have studied the finite size scaling of @xmath53 and we find that the features mentioned above are robust to finite size effects . moreover , contrary to the previous cases , here we obtain a good convergence with system size in the low frequency regime .
interestingly , our results bely the naive expectation that the behavior in the high temperature paramagnetic phase be independent of the values of @xmath120 considered here .
this can be attributed to the fundamental difference between the structure of eigenfunctions and eigenvalues in the two cases : for @xmath122 , they correspond to that of a spinglass system , while the high field case essentially correspond to spins in a strong field where the interactions between the spins can be viewed as a perturbation . at high enough temperatures , all eigenfunctions and eigenvalues contribute equally to the @xmath53
thus the large @xmath0 regime actually reveals the `` geometric '' underlying structure of the hamiltonian , rather than a naively expected universal paramagnetic regime of free spins .
this observation remains relevant for all the different spin - models analyzed in this work .
( color online ) @xmath77 for different @xmath0 , with fixed @xmath101 and field @xmath120@xmath136@xmath15(the @xmath116 curve is not plotted here since is indistinguishable from that of @xmath127 ) .
right inset : behavior for low @xmath50 . left inset : high frequency peaks .
the missing spectral weight is carried by the @xmath134 term described in the text . ]
the results for @xmath53 indicate a very rich evolution of the coherence of the central spin , particularly at long times .
we first analyze the results for @xmath39 plotted in figs .
[ fig : csgtrans70h0.1hctodast ] ( high @xmath0 ) and [ fig : csgtrans70h0.1hct04y07 ] ( low @xmath0 ) , for @xmath122 . as before , for @xmath137@xmath88 ,
the sum rule obeyed by @xmath53 dictates an universal gaussian decay for the decoherence , independently of @xmath120 , @xmath9 and @xmath0 .
this is indeed verified by our results .
( color online ) @xmath102 with @xmath138@xmath15 .
the upper graph ( linear scale ) , shows an intermediate time markovian regime ( @xmath139 ) , while the lower graph ( logarithmic scale ) shows a final power law regime @xmath140 ) .
insets : @xmath104 for low @xmath0 . ]
( color online ) @xmath102 with @xmath138@xmath15 .
the upper graph ( linear scale ) , in contrary to the high @xmath0 coherence of fig .
[ fig : csgtrans70h0.1hctodast ] , does not show an intermediate markovian regime , while the lower graph ( logarithmic scale ) shows a final power law regime in which @xmath141 increases with @xmath0 . ] for @xmath116 ( insets of fig .
[ fig : csgtrans70h0.1hctodast ] ) , the decoherence is only partial due to the presence of a gap in @xmath53and shows oscillations with a frequency proportional to the size of the gap .
similar oscillations are also seen at very low temperatures .
the short time regime @xmath142 is followed by a power - law regime for @xmath143 which saturates to a finite value at long times . as @xmath0 increases
, the short time and asymptotic power law regime are separated by an intermediate markovian regime for @xmath144 . for comparison ,
the coherence for a bath with @xmath145 spins is plotted in figure [ fig : csgtransmth0.1hcdiftn10z ] various @xmath0 . as in the case for @xmath73 , not dramatic change is seen as one traverses the spin glass transition temperature .
the existence of an asymptotic power law regime at finite temperatures completely defies the conventional lore that the asymptotic decoherence at finite temperatures is markovian @xcite .
a systematic analysis of our numerical results indicates that for low fields , @xmath146 is an increasing function of temperature .
this implies that in the high temperature paramagnetic phase , the power law regime is pushed to ultra long times and the decoherence is markovian for realistic times .
though the exact values of @xmath146 and @xmath147 are subject to finite size effects , we have studied various system sizes and find that the intermediate markovian and power law regimes will survive in the thermodynamic limit .
( color online ) coherence of the central spin as a function of time , for a bath of @xmath145 spins for @xmath122 and a bath - qubit coupling @xmath148 .
the free spin case ( fsb ) is also plotted for reference . ] before moving on to the high magnetic field case , we shall present an intuitive physical picture of the very low field case .
we first recall the physical picture of the spin - glass ground state at @xmath116 and @xmath73 , that was discussed in the numerical investigation of ref . .
there it was argued that the spin system can be viewed as a coexisting collection of large non - frustrated clusters of spins with few remnant strongly frustrated `` dangling '' spins .
thus , the dangling spins experience a distribution of effective magnetic fields @xmath149 which , from the work of sherrington and kirkpatrick is known to have a linear distribution .
this linear is directly reflected in the @xmath53 @xmath150 that we discussed before . now ,
when a small external magnetic field is turned on , the unfrustrated spins of the clusters will remain essentially unaffected .
in contrast , the dangling spins will align in the direction of @xmath151 .
this means that the finite external field @xmath120 will act as a low frequency cut - off , pushing spectral weight towards higher frequencies and thus opening an @xmath120-controlled gap in @xmath53 around @xmath79 ( see the low-@xmath0 curve of the right inset of fig .
[ fig : csgtrans97n9h0.1hcn9dift ] ) .
this behavior is in stark contrast with the reference case @xmath73 presented in section [ sec : results ] where @xmath53 remains always gapless ( fig .
[ fig : csg97n9 ] ) .
the dramatic change in the low @xmath50 regime implies also the qualitative change in the long time behavior of the decoherence that we described in the previous section .
this line of argumentation also allows us to qualitatively understand the origin of the linearity of @xmath53 at high @xmath0 . in this limit , from ( [ eq : spectral ] ) we see that all eigenvalues and eigenfunctions contribute to @xmath53 . on the other hand ,
when @xmath73 , the data of fig .
[ fig : csg97n9 ] shows that @xmath53 is finite as @xmath133 , so there are plenty of contributions at arbitrary low frequencies , which originate in pairs of quasi - degenerate eigenstates @xmath152 and @xmath153 with eigenenergies @xmath154 , which also have a non - vanishing matrix element @xmath155 .
when a small external field ( @xmath120 @xmath156 @xmath15 ) is turned on we may consider it as a perturbation an perform the following qualitative analysis : the ( 2x2 ) block of @xmath157 in the subspace of @xmath158 and @xmath159 will read : @xmath160 where @xmath161 is the quasi - degenerate energy of the states .
the effect of the small @xmath120 is to lift the degeneracy of the pair of levels resulting in a transfer of spectral weight from @xmath162 to higher frequencies . extending the analysis to all pairs of quasi - degenerate states contributing to @xmath53 implies the immediate collapse of the finite value of @xmath163 to zero when @xmath120@xmath164 , as is seen in our results of fig .
[ fig : csgtransdifh ] .
moreover , the new eigenvalues of @xmath165 are @xmath166 , thus with the reasonable assumption of a featureless distribution of the @xmath167 matrix elements , we conclude that the degeneracy lifting is uniformly distributed and directly proportional to @xmath120 .
this implies that the behavior of @xmath53 at very low frequencies should be linear in @xmath50 with a slope roughly given by @xmath168 .
this analysis is in qualitative agreement with the low field data shown in fig .
[ fig : csgtransdifh ] . at high magnetic fields ( fig .
[ fig : csgtrans70h2hctodast ] ) , the coherence presents a different qualitative behavior . at low @xmath0 ,
since @xmath53 is nearly gapped , and most of the spectral weight is in the @xmath134 term , the central spin decoherence is very weak and shows oscillations . at high @xmath0 ,
in contrast to the low @xmath120behavior , there is no well defined intermediate markovian regime , but we recover a power - law decay ( @xmath169 ) at asymptotic times .
this coefficient @xmath170 , which characterizes the power law decay , increases with @xmath0 and saturates to a finite value at very high temperatures , proving that the qubit decoheres faster as temperature increases .
the value of the exponent @xmath170 extrapolated to the thermodynamic limit , as a function of @xmath0 , is shown in figure [ fig : csgtransslopeth2hc ] .
( color online ) @xmath77 as a function of @xmath50 for different values of the field @xmath120 with @xmath101 and @xmath171 .
inset : linear behavior of @xmath53 for low energies . ]
( color online ) @xmath102 as a function of time for @xmath120@xmath136@xmath15 . for low @xmath0 ,
since @xmath53 is gaped , the decoherence is very weak and shows oscillations ( inset ) .
as @xmath0 increases , the coherence decays asymptotically as a power law . ]
the exponent @xmath170 ( @xmath169 ) extrapolated to the thermodynamic limit , as a function of @xmath0 for @xmath172 . ]
we now explore the case where the qubit couples directly to the bath operator which is related to the spin glass order parameter . in the spin - glass part of the phase diagram ( @xmath173 ) , we now have @xmath174 and @xmath175 @xmath176 .
as in sec .
[ sec : sigmaz ] , @xmath53 is again a sum of a regular and a singular contribution .
the singular part is a @xmath134 contribution whose strength is related to the spin glass order parameter or edwards anderson parameter @xmath177 , @xmath178@xmath177@xmath179 .
the non - regular part induces oscillations in @xmath39 and the regular part results in the decay of @xmath39 .
the estimation of @xmath177 is a central problem in spin - glasses and as expected our numerics do not produce well converged results of the values of this parameter .
@xmath177 is typically estimated by systematically studying @xmath53 for different sizes of the system , as in ref . .
though it is reasonably straightforward to estimate @xmath177 at @xmath116 , this is not the case at finite temperatures where the delta peak is broadened by both finite size effects and thermal excitations.@xcite when @xmath73 , the operator @xmath180 is constant of motion since @xmath181=0 $ ] resulting in a temperature independent @xmath182 .
as discussed in sec .
[ sec : formalism ] , this results in a purely oscillatory behavior of the form @xmath183.when the field @xmath120 is turned on , spin flips are allowed , @xmath181 \neq 0 $ ] and @xmath53 has a richer structure . for small @xmath120 ( @xmath184 ) and @xmath116 , in addition to a @xmath134 term @xmath53 has a regular part @xmath185 as @xmath133 and has a peak of width @xmath131 , centered around @xmath186@xcite . as temperature is increased , since @xmath177 is expected to decrease , a part of the spectral weight of the @xmath134 term is transferred to the regular part @xmath187 which is now nonlinear at low energies ( fig . [
fig : csgtrans87h0.1hcn9dift ] ) .
( color online ) @xmath77 for n=9 and field @xmath120@xmath126@xmath15for different @xmath0 . for @xmath116 , @xmath53
increase linearly for low @xmath50 and has a maximum at @xmath186 . as @xmath0 increases , spectral weight from the @xmath134 ( discussed in text ) is transferred to low energies .
inset : @xmath53 for @xmath188@xmath15 showing thermal smearing of the gap . ]
we can now construct a picture of the coherence in the spin - glass phase based on the above description of @xmath53 .
the short time behavior is governed by a gaussian . due to the afore - mentioned problem with the broadening of the delta term
, we have not been able to obtain clear predictions for the intermediate and asymptotic decoherence for low @xmath120 and low @xmath0 .
this is intimately linked to the underlying spin - glass order since in this case some of the contribution to the singular part exists only in the thermodynamic limit , as opposed to the preceding section , where the singular term arises from a straightforward magnetization of the underlying system and hence one did not have to deal with broadening induced by finite size .
nonetheless , for @xmath116 and @xmath184 it is possible to have a qualitative picture of the coherence . as in the zero - field case , the linearity of @xmath53 as @xmath133 leads to a power - law asymptotic decay of the coherence . at finite temperatures
, we expect the asymptotic decay to be markovian .
( color online ) @xmath189 for n=9 and field @xmath120@xmath136@xmath15 for different @xmath0 . for @xmath116 , since @xmath53 is gapped , the decoherence is partial but asymptotic markovian decoherence is recovered at finite @xmath0 . ] for high @xmath120 ( @xmath190 ) , there is no spin glass order and hence no singular contribution to @xmath53 . at @xmath116 ,
the function @xmath53 is gaped , and it is centered in @xmath191 , with a width of @xmath192 . as @xmath0 is increased , thermal excitations appear within the gap ( fig .
[ fig : csgtrans87h0.1hcn9dift ] ) .
thus , in the former case the decoherence is partial ( fig .
[ fig : csgtrans80h2hcn9dift ] ) and we see oscillations with a frequency given by the gap size . as @xmath0 increases ,
thermal excitations start to fill the gap and @xmath39 decays in a markovian way given by the value of @xmath90 .
this behavior conforms to the usual expectations of asymptotic markovian decay at finite temperatures and is very different from the small @xmath120 case .
in this paper , we have used exact diagonalization methods to study the decoherence of a central spin induced by spinbaths with random interactions .
we find that the asymptotic decoherence is intricately linked with the nature of the interactions in the bath .
moreover , for a given set of bath parameters , the decoherence crucially depends on the nature of the coupling between the qubit and the bath spins . for the cases of the random transverse ising model studied here , we find that the underlying nature of the eigenstates of the bath hamiltonian play a preponderant role in determining the decoherence even in the finite temperature paramagnetic phase .
more precisely , we find that the decoherence in some of the finite temperature paramagnetic phases is strongly non - markovian .
we emphasize that standard markovian approximations used to obtain the density matrix and the decoherence at finite temperatures should not be used blindly as they can lead to highly misleading results as in the present case of disordered interacting systems .
unfortunately , we have not been able to study the decoherence in the vicinity of the phase transitions in the bath ( both at zero and finite temperatures ) .
this requires the use of other numerical and analytical methods which is beyond the scope of the present work
. it will be interesting to study if these features are seen in other highly frustrated spin baths and whether this non - markovian behavior survives when order corrections to the self - energy are taken into account .
these questions are left for future work . | we study the decoherence of a qubit weakly coupled to frustrated spinbaths .
we focus on spin - baths described by the classical ising spin glass and the quantum random transverse ising model which are known to have complex thermodynamic phase diagrams as a function of an external magnetic field and temperature . using a combination of numerical and analytical methods ,
we show that for baths initally in thermal equilibrium , the resulting decoherence is highly sensitive to the nature of the coupling to the environment and is qualitatively different in different parts of the phase diagram .
we find an unexpected strong non - markovian decay of the coherence when the random transverse ising model bath is prepared in an initial state characterized by a finite temperature paramagnet .
this is contrary to the usual case of exponential decay ( markovian ) expected for spin baths in finite temperature paramagnetic phases , thereby illustrating the importance of the underlying non - trivial dynamics of interacting quantum spinbaths . |
the luminosity density of the universe in the range from about 0.2@xmath1 m to 2.2@xmath1 m , mid - uv to near - ir , is dominated by stellar emission .
this wavelength range spans the peak in the spectra ( @xmath19 ) of stars with effective temperatures from about 20000k down to 2000k .
therefore , measurements of the luminosity density in various broadbands across this range provide a powerful constraint on cosmic star - formation history ( sfh ) and/or a universal stellar initial mass function ( imf ) .
the stellar imf describes the relative probability of stars of different masses forming ( see * ? ? ?
* for recent reviews and analyses ) .
its importance crosses many fields of astronomy from , for example , star formation ( testing theoretical models ) to cosmic chemical evolution ( heavy metal production from high mass stars ) .
it is widely used in the study of the sfh of galaxies from their integrated spectra .
generally , an imf is assumed and used as an input to evolutionary , stellar population , synthesis models , and these models are fitted to integrated spectra .
the first calculation of an imf was made by @xcite based on the observed luminosity function of solar - neighborhood stars , converting to mass , correcting for main - sequence lifetimes and assuming that the star - formation rate ( sfr ) has been constant for the last 5gyr . despite the uncertainties in mass - to - light ratios , stellar lifetimes and the sfr , this result ( a power - law slope of @xmath20 measured from about 0.3 to 15 )
is still commonly used today .
logarithmic mass bins and @xmath21 wrt . linear mass bins . ] measurements of the solar neighborhood imf were reviewed by @xcite producing an imf with a mass fraction peak around 0.51 ( fig .
[ fig : imfs ] ) .
when applied to galaxy populations , this imf is unable to reproduce h@xmath8 luminosities @xcite , and applied to cosmic evolution , is unable to match observed mean galaxy colors @xcite due to a too low fraction of high - mass stars ( @xmath22 ) . in a more recent review by @xcite , he concluded that the field star imf was of questionable use for a number of reasons ( e.g. the derived imf in the range 0.91.4 depends strongly on the assumed solar - neighborhood sfh ) .
instead , he summarized the results from studying star clusters in a triple - index power - law imf as an estimate of an average imf ( fig .
[ fig : imfs ] ) .
he also noted that `` if the existing empirical estimates of the imf are taken at face value , they present strong evidence for variations , and these variations do not seem to depend systematically on physical variables such as metallicity or stellar density ''
. the observed imf variations from stellar counts could be largely due to statistical fluctuations and/or observational biases such as mass segregation within a star cluster @xcite .
these points , if correct , mean that the concept of a universal imf is highly useful in the studies of integrated spectra of galaxies ( since they are mostly the result of many star - formation regions and episodes ) but the imf is still uncertain at the level of 0.5 in the power - law slopes . by ` universal imf ' , we mean an imf that represents the average in a significant majority of galaxies and over a significant majority of cosmic time .
measurements of the cosmic luminosity densities represent the @xmath23 components of the `` cosmic spectrum '' which is the luminosity - weighted , average galaxy spectrum @xcite .
thus , if there is a universal imf , it should certainly apply to this spectrum and constraints on the imf can be obtained by fitting population synthesis models .
these constraints are degenerate with the assumed cosmic sfh .
however , cosmic sfh is generally more accurately quantified than the sfh of any individual galaxy other than in the local group .
for example , star - forming galaxies have been observed as far back as @xmath24 @xcite , and there is evidence for reionization as early as @xmath25 @xcite , which for a universe age of @xmath26gyr @xcite gives a time of 12.313.7gyr since the onset of a significant rate of star formation in the universe .
the results for individual galaxies are often limited by age - metallicity degeneracy @xcite and recent bursts of star formation which can disguise their underlying age @xcite . given our knowledge of cosmic sfh , we can then place constraints on a universal imf by limiting the sfh we consider .
the primary knowledge of cosmic sfh comes from measuring the comoving density of sfr indicators at various redshifts @xcite .
these sfr indicators include uv luminosities and emission - line luminosities that are dominated by light from short - lived high - mass stars .
the conversion to sfr depends on the imf .
however , the _ relative _ cosmic sfh is well defined if the same indicator is used at each redshift regardless of the assumed imf . even if the indicator varies ( e.g. 0.2 to 0.3@xmath1 m rest - frame uv ) , the derived cosmic sfh is less sensitive to the slope of the imf than the local luminosity densities ( 0.2 to 2.2@xmath1 m ) .
it is this principal that enables constraints on a universal imf .
an analysis of this type was applied by @xcite using various luminosity densities ( 0.15 to 2.2@xmath1 m ) spread over a range of redshifts ( 04 ) .
they fitted cosmic sfhs for three different imfs . here
, we use a more quantitative approach to constraining a universal imf slope and use more accurate , recent , local ( @xmath0 ) luminosity density measurements . note that we assume that there is a universal imf and do not constrain any variation in the imf between different galaxies ( see e.g. * ? ? ?
* ; * ? ? ?
* for evidence for an invariant imf ) .
even if there is some variation , the results presented here could be regarded as constraints on a luminosity - weighted , average imf .
to summarize , if the cosmic sfh is assumed to be known , based on measurements of various sfr indicators with redshift , then local luminosity density measurements provide a constraint on a universal stellar imf . the other principal factors to consider are chemical evolution ( metallicity ) and dust attenuation
the plan of the paper is as follows . in section
[ sec : lum - dens ] , we describe details of recent luminosity density measurements . the measurements are illustrated in figure [ fig : lum - dens ] and summarized in table [ tab : lum - dens ] . in section [ sec : modeling ] , we describe our modeling and fitting procedure .
a summary of the parameters used in the modeling is given in section [ sec : summary - paras ] . in sections [ sec : results ] and
[ sec : conclusions ] , we present our results and conclusions . [ cols="<,<,^,^ , < " , ] three attenuation - corrected h@xmath8 luminosity density measurements are summarized in table [ tab : h - alpha ] .
we do not quote error bars as the uncertainties are dominated by systematics .
these include ( i ) subtraction of the stellar contamination which , in particular , affects the measurement of h@xmath27 and thus the estimate of the attenuation ( @xcite @xcite included this uncertainty which amounted to 0.10.15 in the @xmath28 result ) , ( ii ) uncertainties in the attenuation curve ( e.g. * ? ? ?
* ) which affect the conversion from a reddening measurement to absolute attenuation , and ( iii ) agn contamination .
both the @xcite and @xcite results are based on obtaining the emission - line luminosity function from spectra . however , the selection criteria of the surveys are different , an emission - line objective - prism survey and @xmath29-band photometry , respectively .
@xcite took a different approach of summing sdss spectra to form a cosmic optical spectrum before calculating the h@xmath8 and h@xmath27 luminosity densities .
we take the mean of these three results for our fitting : @xmath30 this is appropriate since the @xcite and @xcite results may be too high due to agn contamination which are enhanced by aperture affects .
the spectra taken through a fiber are normalized to a broadband filter and since agn are centrally concentrated , any emission line luminosity due to them will be over enhanced .
however , @xcite did measure the weak oi 6300 line , suggesting that the agn contribution was only a few percent at most . the @xcite result is probably too low because of the small survey ( i.e. large scale structure ) and/or because it only includes ew@xmath3110 ( h@xmath8 , emission positive ) galaxies . to compare this average measurement with model predictions
, we use the output from pegase for the h@xmath8 flux without dust attenuation . as with the earlier fitting , the synthetic spectra are normalized to the @xmath32 band so we are really comparing the ratio of h@xmath8 and @xmath32 band fluxes between the models and the data .
a correction is made for the fact that the measured @xmath32 band flux includes dust attenuation because we are fitting to an attenuation - corrected h@xmath8 measurement .
we use the fiducial attenuation of @xmath33 .
in other words , unlike the earlier fitting we now normalize the synthetic spectra to an attenuation - corrected luminosity density in the @xmath32 band .
figure [ fig : cont - halpha ] shows the results from fitting to the average attenuation - corrected h@xmath8 luminosity density .
the degeneracy in @xmath27 versus @xmath34 is similar to that for the broadband luminosity densities .
in addition , the best - fit imf slope , @xmath34 in the range 0.91.5 ( 68% conf . assuming @xmath35 and averaging over the metallicity approximations ) , is in good agreement with the those results
in this paper , we present the results of fitting spectral synthesis models with varying imfs to local luminosity densities . * a good fit is obtained to the measurements of @xcite , @xcite and @xcite @xmath36-band ( compilation sbc@xmath37 ) with a best - fit metallicity of around solar .
if we fit to the measurements of the first two papers and @xcite ( compilation sbh ) , the best fit metallicity is greater than twice solar and the results are inconsistent with solar metallicity at the 99.7% confidence level .
* the data can be well fit by a universal imf and , therefore , there is no need to invoke imf variations .
however , this provides only a weak constraint on the invariance of the imf because of significant degeneracies associated with this type of modeling . *
the best - fit universal imf slope marginalized over a significant range of cosmic sfh ( @xmath38 , @xmath39 ) is @xmath2 based on an average between a closed - box approximation and a constant - metallicity approximation around solar ( using compilation sbav ) .
our results are in good agreement with the salpeter imf slope . * a strong upper limit of @xmath7
is obtained with 99.7% or 95% confidence depending on the metallicity approximation .
this rules out the @xcite imf for a universal imf since the mass fraction of stars above 10 is similar to our parameterization with @xmath40 ( fig .
[ fig : imfs ] , see also table [ tab : publ - imfs ] ) .
a similar conclusion was reached by @xcite from fitting to luminosity densities over a range of redshifts and by @xcite from fitting to h@xmath8 ew - color relations for galaxies .
@xcite found a salpeter imf or an imf with @xmath41 provided adequate fits . here , we find that the latter slope does not provide a good fit to luminosity densities .
this is principally because we have used more accurate local luminosity density measurements .
if we increase our @xmath42 uncertainties by 0.05 at all wavelengths then we obtain @xmath7 with 80% confidence .
* the stellar mass density of the universe is in the range 1.12.0 @xmath10 @xmath11 based on marginalizing over cosmic sfh and our parameterization of the imf .
the current sfr density is in the range 0.74.1 @xmath13 @xmath14 .
the total bolometric stellar emission ( 0.095@xmath1 m ) is known more accurately , naturally because we observe light and not mass , and is in the range 1.21.7 @xmath16 @xmath17 derived from the fitted pegase model spectra ( the mass - to - light ratio is @xmath431.4 / ) .
we find that our dust model can reproduce the estimated total dust emission ( 0.30.7 @xmath16 @xmath17 , @xmath4451000@xmath1 m ) scaled from the far - ir luminosity density @xcite if we have @xmath45 ( @xmath4660% ) .
note that this represents a limit on the cosmic spectrum attenuation , i.e. , a luminosity - weighted average by flux ( not magnitudes ) .
* fitting to the local h@xmath8 luminosity density provides a similar result for the imf slope ( @xmath47 ) .
this provides some evidence that our upper mass cutoff of 120 is a reasonable approximation because the sensitivity of the h@xmath8 flux to massive stars is different to that of the mid - uv to optical fluxes .
more accurate measurements could test this upper mass limit . *
the quantitative results on @xmath34 rely on the accuracy of the population synthesis model ( pegase ) .
an alternative , qualitative result would be that there is consistency between the theory of evolutionary population synthesis ( evolutionary tracks and stellar spectra ) and the measurements of luminosity densities , cosmic sfh and an average mw imf derived from stellar counts ( e.g. @xmath48 from * ? ? ?
greater constraints can be placed on a universal imf both by improved accuracy in local luminosity density measurements ( in particular , @xmath49 to match the multi - wavelength sdss mgs ) and by improved accuracy of direct measures of cosmic sfh with redshift ( in particular , @xmath50 to @xmath512 ) . for the direct tracing of cosmic sfh ,
it is important that the uv is measured at the same rest - frame wavelength in order to avoid imf dependency . in other words , we need an imf - independent cosmic sfh in order for the local luminosity densities to accurately constrain a universal imf or the imf - dependency should be quantified . | we obtain constraints on the slope of a universal stellar initial mass function ( imf ) over a range of cosmic star - formation histories ( sfh ) using @xmath0 luminosity densities in the range from 0.2@xmath1 m to 2.2@xmath1 m .
the age - imf degeneracy of integrated spectra of stellar populations can be broken for the universe as a whole by using direct measurements of ( relative ) cosmic sfh from high - redshift observations .
these have only marginal dependence on uncertainties in the imf , whereas , fitting to local luminosity densities depends strongly on both cosmic sfh and the imf .
we fit to these measurements using population synthesis and find the best - fit imf power - law slope to be @xmath2 ( assuming @xmath3 for 0.5120 and @xmath4 for 0.10.5 ) .
this @xmath5 slope is in good agreement with the salpeter imf slope ( @xmath6 ) .
a strong upper limit of @xmath7 is obtained which effectively rules out the scalo imf due to its too low fraction of high - mass stars .
this upper limit is at the 99.7% confidence level if we assume a closed - box chemical evolution scenario and 95% if we assume constant solar metallicity .
fitting to the h@xmath8 line luminosity density , we obtain a best - fit imf slope in good agreement with that derived from broadband measurements .
marginalizing over cosmic sfh and imf slope , we obtain ( 95% conf .
ranges ) : @xmath9 = 1.12.0 @xmath10 @xmath11 for the stellar mass density ; @xmath12 = 0.74.1 @xmath13 @xmath14 for the star - formation rate density , and ; @xmath15 = 1.21.7 @xmath16 @xmath17 for the bolometric , attenuated , stellar , luminosity density ( 0.095@xmath1 m ) . comparing this total stellar emission with an estimate of the total dust emission implies a relatively modest average attenuation in the uv ( @xmath18 magnitude at 0.2@xmath1 m ) . |
the @xmath0 antiferromagnetic ( af ) heisenberg model on a square lattice has attracted a lot of attention due to its close relation to the disappearance of the af order in high - t@xmath4 superconducting material @xcite and its possibility of realizing the so called spin liquid state @xcite .
the spin liquids , defined as lacking long range order and supporting strong quantum fluctuation , have all the key properties of the insulating state `` nearby '' to a superconducting phase .
hence a simple example of spin liquids , the resonating valence bond ( rvb ) state was proposed to describe high - t@xmath4 superconductivity @xcite .
spin liquids have become the focus of research in modern condensed matter physics since the discovery of different kinds of spin liquids in theory @xcite .
although the existence and stability of spin liquids in realistic models and real material are still in question @xcite . the nature of the ground state of the @xmath0 af heisenberg model in the intermediate coupling region has been debated @xcite for decades due to the highly frustrated nature of this model and is still a open question .
recently , several numerical works revisited this model using novel concept , such as the topological entanglement entropy ( tee ) @xcite , and new numerical tools , such as the projected entangled pair states ( pepss ) @xcite .
amazingly many of them have reached a consistent and the lowest ground state energy ever @xcite , however their resulting ground states behave diversely .
a density matrix renormalization group ( dmrg ) study on long cylinders suggested a gaped @xmath5 spin liquids from the evidences of non - vanishing singlet and triplet gaps in the paramagnetic phase and a nonzero tee @xcite .
the gutzwiller - projected bcs wavefunction study indicated a gapless spin liquid state supported by the existence of gapless triplet excitations at momenta @xmath6 and @xmath7 @xcite .
most recently , the dmrg study with @xmath8 symmetry imposed on the states on long cylinders found a diverging dimer correlation length , and claimed a plaquette valence bond solid ( vbs ) state as the ground state @xcite .
these controversial results pose the question about the numerical convergence and the system size dependence @xcite . in this paper , we propose a novel wavefunction and use the variational monte carlo ( vmc ) method to tackle this system .
as one knows , the key to a meaningful variational method is to have a good trial wavefunction . here , we take a resonating valence bond state approach .
the reasons for such a choice are the following , first of all , rvb states can describe a variety of phases including the spin liquids , the af long range ordered states and the valence bond solids @xcite ; second of all , recent development in pepss showed that a family of one parameter pepss with bond dimension @xmath9 , which described the rvb state beyond the nearest neighbor pairing , can greatly lower the variational energy of this model compared to that of the short range rvb state @xcite .
we can imagine that with more build - in correlations and hence more variational parameters , the rvb states should approach the true ground state very well .
the simplest rvb ansatz on a bipartite lattice is the valence bond amplitude product ( ap ) state , where the probability of having a valence bond of separation @xmath10 is independent of each other and is denoted as @xmath11 @xcite .
consequently the bond amplitudes @xmath11 become the variational parameters for the valence bond ap states .
simulations based on this variational ansatz has been done @xcite : it was found that the amplitude @xmath12 tends to be negative for @xmath13 in order to minimize the energy within this variational subspace .
the place where @xmath12 changes sign signals the break down of the marshall s sign rule @xcite and is very close to the proposed critical point where the af long range order disappears .
it is interesting to ask the question that does the onset of the negative amplitude in the valence bond ap state indicate a phase transition to new states ? in this paper , we will give our answer to this question from the correlated valence bond point of view .
prior to our investigation , lin et al . introduced the so called correlated amplitude product ( cap ) state , where an extra weight factor has been given to a pair of valence bonds @xcite in order to build the bond - bond correlations into the wavefunction , however they have not yet applied the caps to the @xmath0 af heienberg model .
the goal of this paper is to simulate the ground state of the @xmath0 af heisenberg model on square lattice using the correlated valence bond states , whose definition will be clear in sec [ wf ] .
the rest of this paper is arranged as following : in sec .
[ wf ] , we define the correlated valence bond states , which have some similarity but are essentially different from the caps @xcite .
we introduce the variational monte carlo method and the optimization strategy for this wavefunction in sec .
[ methods ] . upon optimization of the correlated valence bond states by minimizing the ground state energies , we arrive at the optimal states . in sec . [ results ]
, we present the variational ground state energy and the ground state correlations for the optimized states . in sec .
[ discussion ] , we discuss the advantages and limitation of this ansatz and conclude for the nature of the intermediate phase of the @xmath0 af heisenberg model on square lattice .
the hamiltonian of the @xmath0 af heisenberg model on a square lattice is given by @xmath14 where the first summation runs over the nn pair @xmath15 and the second summation runs over the next nn ( nnn ) pair @xmath16 .
an equal weight superposition of the short range rvb state has been applied as a trial wavefunction to the ground state of hamiltonian eq .
( [ afheisenberg ] ) @xcite , but the variational energy is not good and it is nowhere close to the true ground state at any coupling ratio @xmath17 .
one step forward was made by introducing a positive amplitude to an arbitrary ranged bipartite valence bond in the frame work of the valence bond ap states @xcite ; the reported best thermodynamic energy at @xmath18 is @xmath19 per site , which is still much higher than the dmrg studies @xcite and has left a lot of room for improvement in the trial wavefunction . here
we speculate that the bond - bond correlations may play a crucial role than the individual long range bonds @xcite .
therefore , we define a class of ansatze called the correlated valence bond states by a set of statistical rules : 1 . considering without the presence of the correlated valence bond pairs ,
the ansatz is reduced to a short - bond amplitude product state : the bond amplitude @xmath20 and otherwise 0 . 2 .
if two short individual bonds are sitting next to a common edge of the lattice , they are allowed to quantum fluctuate to any other valence bond tilings on these four sites , and _ vice versa _ , as illustrated in fig .
each valence bond tiling of these four sites is associated with a amplitude marked as either 1 , @xmath21 or @xmath22 , where @xmath23 ( @xmath24 ) are the variational parameters .
the two short bonds on the left hand side of the reversible - arrow in fig . [ cap ] are two individual bonds ( uncorrelated ) , which can be randomly rearranged with all other uncorrelated short bonds into any allowed short bond product state .
however the two correlated bonds together with the lattice edge ( in blue dash line ) on the right hand side of the reversible - arrow form a single object such that the two bonds there in can not be treated independently .
the correlated - bond - pair can only be flipped back to the two short bonds exactly shown on the left hand side of the reversible - arrow .
the above three rules together form a statistical mechanic ensemble . before discussing the variational monte carlo algorithm ,
several comments are along the line .
we choose such a statistical ensemble as the trial wavefunction in order to address bond - bond correlations .
the amplitudes of the correlated - bond - pairs give extra factors to the wavefunction coefficients , and this ansatz goes beyond the ap state .
the correlated - bond - pair amplitudes serve as the variational parameters of the trial wavefunction .
we can easily translate this ansatz to a peps wavefunction that describes qualitatively the same statistical ensemble .
we impose the translational and rotational symmetries to the wavefunction to reduce the number of variational parameters .
this restriction is perfectly valid as long as we work with finite size systems .
there are four distinct arrangements of two short bonds that are connected by a common lattice edge , as illuminated in fig .
[ cap ] , therefore eight different correlated - bond - pairs whose amplitudes are denoted by @xmath25 and @xmath26 .
hereafter , we use the word `` correlated - bond - pair '' and its amplitude interchangeably . given these clarifications ,
we can write the trial wavefunction as following @xmath27 is a compact pack of individual short bonds and correlated - bond - pairs on a lattice which produce a valence bond tiling configuration @xmath28 , @xmath29 ( @xmath30 ) is the number of @xmath21 ( @xmath22 ) in a compact pack @xmath31 .
note that different packing @xmath31 can generate the same valence bond tiling configuration @xmath28 . .
the arrows marked for @xmath28 ( in black lines ) align with the sign convention , whereas for @xmath32 ( in gray lines ) anti - align with the sign convention .
@xmath33 , where @xmath34 is the number of arrows that violate the direction of the flow and @xmath35 is the number of loops in the transition graph .
in this section , we first briefly describe the variational valence bond monte carlo method , followed by the optimization method used to determine the optimal variational parameters . given a variational wavefunction @xmath36 ,
the expectation value of any operator @xmath37 is written as @xmath38 with the importance sampling weight defined as @xmath39 and the operator matrix element defined as @xmath40 here @xmath41 is the overlap matrix element . if both the coefficients @xmath42 and the overlap matrix elements @xmath43 are positive definite , we could use the standard valence bond mc method @xcite
. however our valence bond configuration @xmath28 allows valence bonds of the same sublattice pairing , therefore the overlap matrix elements can be negative ; in addition , the correlated - bond - pair amplitudes can take negative values , which means the coefficients @xmath42 can be negative too .
thus our wavefunction encounters a negative sign problem .
we can treat it in a standard way by rewriting eq .
( [ exp ] ) as @xmath44 where @xmath45 denotes the sign of the weight eq .
( [ wt ] ) .
we sample using the absolute value @xmath46 ( abbreviated as @xmath47 ) as the importance weight .
the operator expectation value can be obtained by taking the ratio of @xmath48 and @xmath49 .
we now explain how to calculate the overlap matrix element @xmath43 and how to count the sign in @xmath45 .
we fix the sign convention for any valence bond in @xmath28 : for the ab sublattice pairing , the direction of the valence bond is always pointing from the a sublattice to the b sublattice ; whereas for the same sublattice pairing , we refer to the signs drawn in fig .
[ sign](a ) as the convention . given @xmath28 and @xmath32 ,
the overlap matrix element for a transition graph is @xmath50 , where @xmath35 is the number of loops in the transition graph and @xmath34 is the number of singlets that violate the direction of flow that is arbitrarily chosen for each loop in the transition graph .
an example of the transition graph is shown in fig .
[ sign](b ) . here
@xmath51 is a normalization constant , since the maximum number of loops in a transition graph is @xmath52 .
the total number of minus sign factors in the coefficient @xmath42 depends on the sign convention and the signs of the variational parameters @xmath23 .
putting all signs together we have @xmath53 .
next , we explain how to calculate the operator matrix element @xmath54 .
since the singlet product states @xmath55 appear on both the numerator and the denominator , we can choose any singlet sign convention we want ( those sign factors will cancel if otherwise choose a difference sign convention ) .
for any transition graph , we choose one such that every loop has @xmath56 structure , _
i.e. _ we choose the sign to be always pointing from @xmath57 to @xmath58 . here
we use @xmath59 to differentiate the @xmath60 sublattices of the original lattice bipartition , and @xmath57 and @xmath58 only have a relative meaning within each loop . therefore all operator expectation values can be evaluated using the same formulas as in the ref .
@xcite with the replacement of @xmath61 ( @xmath62 ) by @xmath57 ( @xmath58 ) . with coupling strength @xmath1 $ ] .
parmeters not shown here are optimized close to 0 . ]
the mc sampling of wavefunction eq .
( [ crvb ] ) takes two types of update .
the first type of update is the local update .
we begin by randomly selecting a link from the lattice , one of the following cases will happen : 1 .
if the link is not occupied by any singlets and it sits beside two individual short bonds , propose to flip randomly to one of the correlated - bond - pairs or stay unchanged with the corresponding probability ; 2 .
if the link belongs to a correlated - bond - pair and is exactly sitting on the reference edge in blue dashed lines in fig .
[ cap ] , propose to flip to the two individual short bonds or stay unchanged with the corresponding probability ; 3 .
for any other cases , abandon such a choice and iterate .
if a local update is proposed , we accept or reject this local update with probability @xmath63 $ ] , where @xmath64 is the number of loop in the trial transition graph .
the second type of update is the loop update .
we randomly construct an allowed loop of alternating individual short bonds and vacant lattice links , then we shift all individual short bonds on that loop by one lattice space . this rearrangement can always be proposed with probability 1 , since all individual short bond have equal amplitudes , and it will keep all the correlated - bond - pairs untouched . the loop update is accepted or rejected with probability @xmath63 $ ] defined above .
we use variational mc to calculate the derivative of the energy with respect to a variational parameter @xmath65 ( @xmath66 ) as @xmath67 and update them according to @xcite @xmath68 where @xmath69 is the iteration index , @xmath70 is a random number @xmath71 , and @xmath72 until the energy converges . the function form @xmath73
is chosen heuristically @xcite .
the optimization results and correlation functions will be presented in the next section .
( green circle ) , @xmath74 ( green dots ) and @xmath75 ( blue triangles ) lattices , here the reference ground state energies for the @xmath75 lattice are taken from the dmrg results in ref .
red square is a comparison to the projected fermionic bcs wavefunction for a @xmath74 lattice without lanczos projection step @xcite . ]
we present the well optimized variational parameters for the wavefunction eq .
( [ crvb ] ) for system sizes @xmath2 at @xmath76 $ ] in fig . [ para ] .
we find strong size dependence for the optimized variational parameters @xmath77 .
however , parameters @xmath78 have less size dependence
. other parameters that are not shown in fig .
[ para ] are optimized close to zero .
the absolute energy error per site @xmath79 for sizes @xmath80 compared with the exact diagonalization ( ed ) results @xcite and for size @xmath3 compared with the dmrg results @xcite are presented in fig .
the absolute errors are small , _ e.g. _ , @xmath81 and @xmath82 . for comparison ,
we draw the gutzwiller - projected bcs wavefunction results without lanczos projection @xcite in the same frame : at @xmath83 , our absolute energy error is lower by about @xmath84 .
we take the optimized parameters from size @xmath3 to calculate the order parameters and the correlation functions for system sizes @xmath85 until the average signs are no longer manageable , because optimizing variational parameters for @xmath85 becomes not feasible .
we assume that the correlation functions at larger sizes will not be too sensitive to the variational parameters as long as they are within a reasonable range from the optimal values . in a log - log plot . power law functions @xmath86
, @xmath87 have been fitted to the points for @xmath88 . ]
let us first define a set of order parameters .
the sublattice magnetization is written as @xmath89 where @xmath90 , @xmath91 is the spin correlation function , and it is defined as @xmath92 the dimer order parameter is defined as @xmath93 where @xmath94 and @xmath95 , @xmath96 . the sublattice magnetization and the dimer order parameter are presented in log - log plots in fig .
[ ordpara ] .
the sublattice magnetization at couplings @xmath97 scale as @xmath98 , which is expected for the exponentially decaying spin correlation function eq .
( [ ssc ] ) .
the dimer order parameter follows a power law decay with system size as @xmath99 where @xmath100 .
therefore , we find a critical phase that has gapless singlet excitations and gaped triplet excitations within the range of @xmath1 $ ] .
we next make some connections to the results given by previous work .
a large optimal parameter @xmath101 is consistent with the results of an enhanced spin - spin correlation along the @xmath102 and @xmath103 axes shown in ref .
a negative optimal parameter @xmath104 is consistent with the result of having negative @xmath12 in the simulation using an ap product state @xcite .
our variational term @xmath105 generates configuration of parallel diagonal bond pair .
the parallel diagonal bond pairs together with the short bonds , if contribute with equal weights to the resonating wavefunction on a square lattice , will give a @xmath5 spin liquid state @xcite .
our optimized wavefunction , which contains both the parallel diagonal bonds and the short bonds , does not behave like a @xmath5 spin liquid , although it deviates from the special point of equal weights superposition defined in ref .
we need to do further investigation to answer why it fails to be a @xmath5 spin liquid .
the optimal parameter @xmath106 at system size @xmath3 grows rapidly as a function of @xmath107 , whose effect is to increase the dimer - dimer correlation .
however given this optimized parameter strength , the effect of @xmath106 will not induce a quantum phase transition from a critical phase to the vbs phase as predicted in ref @xcite , because the latter phase requires a very large value in @xmath106 @xcite .
let us turn to the question that we asked earlier : does a negative amplitude @xmath108 in an ap state indicate a phase transition ?
from our example , the answer seems to be no .
the critical phase presented in our work is simply a result of muting all the long range bonds in the ap state , as one increase the weights of @xmath104 and @xmath105 from 0 , there are no signs of a phase transition from the critical phase to either @xmath5 or vbs states . therefore we conclude that the negative amplitude @xmath108 along in an ap state could not trigger a phase transition .
using the correlated valence bond state , we minimized the ground state energy of the @xmath0 antiferromagnetic ( af ) heisenberg model on a square lattice for a coupling ratio @xmath1 $ ] by turning a few correlated - bond - pair amplitudes .
the energies are consistent with the exact diagonalization ( ed ) results on the @xmath109 and @xmath74 tori @xcite and the density matrix renormalization group ( dmrg ) results on the @xmath75 torus @xcite .
we applied the optimal variational parameters from the @xmath75 system to the larger tori and studied their correlation functions using the valence bond monte carlo ( mc ) sampling method .
we found that within the optimized phase , the neel order parameter scales as the inverse of the volume , and the dimer order parameter describing the columnar or plaquette valence bond solid ( vbs ) phases follows a power law decay with the system size @xmath110 approximately as @xmath111 .
these correlation functions indicate a critical phase with gapless singlet excitations and gaped triplet excitations in the entire range of @xmath1 $ ] . due to the negative sign problem
, we can not optimize even larger systems or further increase the number of variational parameters , such as turn on the individual long range bonds .
our simulation provide insights of how and when the critical phase can turn into a vbs or a @xmath5 spin liquid .
however with the current results , we can not conclude which one is the true ground state with the intermediate coupling strength . _
we thank j. richter for providing us the ed results .
we would like to thank o. motrunich , f. verstraete , k. beach , a. sandvik , z .- c .
gu , s .- s .
gong , w .- j .
hu for useful discussion .
the author especially thanks o. motrunich for reading and giving comments on the manuscript .
this work was supported by the institute for quantum information and matter , an nsf physics frontiers center with support of the gordon and betty moore foundation through grant gbmf1250 . | we propose a class of variational wavefunctions , namely the correlated valence bond states , for the frustrated hamiltonians in the paramagnetic phase .
this class of wavefunctions admits negative amplitude and the same sublattice pairing when a bipartite lattice is considered , thus suffers from the negative sign problem .
however if applied to small systems , the sign problem is manageable using the standard variational monte carlo method .
we optimize the wavefunctions for the @xmath0 antiferromagnetic heisenberg model on a square lattice in the coupling region @xmath1 $ ] for system sizes @xmath2 . to calculate the correlation functions and the order parameters for larger systems
, we make the extensive monte carlo samplings using the variational parameters optimized at system size @xmath3 .
we find that the paramagnetic phase is a gapless spin liquid in the entire range of @xmath1 $ ] with a gapless singlet excitation and a gaped triplet excitation . |
the lifetime of disks sets the time available for the planet formation process . in the solar system , chronology of the oldest solid components based on radionuclides ( e.g. , al@xmath4 )
indicates that the duration timescale of the solar nebula was short , of order 23 myr .
direct imaging of a @xmath5 m@xmath2 planet @xcite inside the debris disk of @xmath6 pic proves that large gas giants can form by 12@xmath7 myr . in protostellar systems , by 5 - 10 myr no gas is left @xcite .
the evidence points at early giant planet formation , as in the candidate massive protoplanets lkca 15 b @xcite and t cha b ( although both remain to be confirmed , e.g. by direct imaging ) .
radial gaps in gas - rich protoplanetary disks are currently thought of as possible signposts of planet formation .
previous @xmath8 and @xmath9 band coronographic images of hd 142527 reveal a hole in the disk , about 100 au in radius @xcite .
this inner cavity is in fact a gap with an inner boundary , abutting on an inner disk @xcite , roughly 10 au in radius ( anthonioz et al .
, in preparation ) , and up to 30 au , that accounts for the large near - ir excess of hd 142527 @xcite .
radiative transfer modelling of the spectral - energy distribution ( sed ) and nir images are consistent with an inclination angle of @xmath020 deg .
the youth of hd 142527a , at @xmath10 pc0.98 marcsec lower - limits its distance to 138 pc , at 3 @xmath11 .
] , is evident from the copious amounts of gas @xcite in its @xmath00.1 m@xmath12 circumstellar disk .
@xcite estimate a stellar age of @xmath13myr and a mass of [email protected] m@xmath12 .
here we report on deep adaptive optics near - ir imaging of the disk and gap of hd 142527 .
[ sec : observations ] describes our observational setups ,
[ sec : obsresults ] emphasizes our main observational results , [ sec : analysis ] interprets , and
[ sec : conclusion ] concludes .
_ nici adaptive optics imaging_. we conducted ir adaptive optics observations of hd 142527 with the nici coronograph @xcite on the gemini telescope .
nici is a dual - channel imager , the images were acquired simultaneously , by pairs , a different filter in each channel . here
we report on two pairs of filters : 1- broadband @xmath15 ( 2.15 @xmath16 m ) with h@xmath17(1 - 0)s(1 ) ( 2.12 @xmath16 m ) , both with a 0.22@xmath18 mask occulting 95% of the stellar light , and both exposed to 2 s and 30 s , for a total of 1 h ; 2- broadband @xmath19 ( 3.8 @xmath16 m ) with br@xmath20 ( 2.17 @xmath16 m ) , both maskless and exposed to 0.38s , for a total of 1 h. all setups were acquired in stare mode , compensating for parallactic angle rotation .
the result of our nici imaging is shown in fig .
[ fig : nici ] , after psf subtraction .
_ sinfoni integral field spectra_. we also conducted complementary integral - field - unit spectroscopy with the sinfoni instrument on the vlt in order to search for possible planetary objects in the gap . in our setup
we chose a very narrow jittering throw that samples the inner @xmath21 arcsec@xmath22 only .
we processed the eso pipeline datacubes using spectral deconvolution techniques @xcite , and also through conventional psf subtraction ( using a psf standard star ) .
the detection limits we achieve are consistent with those achieved by @xcite in ab dor with the same instrument ( @xmath09 mag contrast at 0.2 @xmath18 separation ) .
the nici images clearly reveal the non axi - symmetric shape of the cavity .
this is the highlight of our observations .
the shape is roughly elliptical , albeit with significant departures , with a representative eccentricity of @xmath23 ( where @xmath24 and @xmath25 are the ellipse axes ) .
but , as illustrated in fig . [ fig : nici_annot ] , the exterior ring can be described as being composed of four arms overlapping in their extremities , except for two small , @xmath00.2 arcsec gaps or intensity nulls along the ring , due north and south - south - east from the star .
the observation of these morphological feature is based on the shape of the inner rim of the ring seen in direct images ( even without subtraction of the psf glare ) .
additionally , an unresolved brightness increment , or knot , can be seen along the north - western segment ( which is also the brightest segment ) - the alignment of this unresolved feature in all filters is illustrated in the rgb image , although we can not determine from these data alone if this knot is real or perhaps only a ghost or a chance alignment of speckles .
the south - western spiral arm segment ( labelled arm 2 on fig .
[ fig : nici_annot ] ) sprouts away from the outer ring at a pa of -45 deg ( east of north ) , where it extends into the outer arm seen by @xcite .
this outer arm is at much fainter levels than the outer ring - our dataset has finer resolution but is shallower , so that only its root is seen in our figures .
as discussed in @xcite , this outer arm could have been triggered by a stellar encounter , but the putative partner remains to be identified .
a nearby point source turned out to be a background star @xcite - in our nici images this source is located at a separation of 5.44 from hd 142527 , and a pa of 220 deg ( east of north ) .
the inner radius of the ring varies from 0.7@xmath18 to the north , to about 1.0@xmath18 to the south , and a projection effect is discarded since the central star is clearly offset by @xmath1 from the centre of this approximate ellipse .
this offset was first noted by @xcite , albeit at lower angular resolution . in @xmath9-band
the west side of the rim we see is brighter , in agreement with @xcite .
@xcite and already noted that in the thermal ir , @xmath26 and @xmath27 bands , it is the _ eastern _ side that is brightest , which is consistent with a projection effect @xcite . at the shorter wavelength @xmath9-band ,
the front ( i.e. nearest ) side of the disk appears brighter because of forward scattering at the surface of the disk . on the opposite ,
the thermal emission from the larger projected surface of the rim of the back side of the disk appears brighter than the slimmer ( in projection ) rim of the near side .
this implies a major axis of the disk in the north - south direction and a minor axis in the e - w direction ( as observed here ) .
the emission null we detect due north is seen on both sets of subaru images , at @xmath9-band and in the mid - ir as well .
no companion is clearly detected in our nici images .
an upper limit on a binary companion is difficult to obtain because of the speckle noise , and because of the lack of accurate nici zero points .
however , the sinfoni observations allow us to place a lower limit flux ratio at 0.20.3@xmath18 separation of 1905 at 3 @xmath11 in @xmath9 ; fig . [
fig : sinfo ] illustrates this upper limit . modeling the sed we estimate that half of the total flux density of hd 142527 in k can be assigned to the disk . in that case , the sinfoni flux ratio limit of 1905 corresponds to an absolute magnitude @xmath28 at 3 @xmath11 . as a comparison point
we refer to the cond03 tracks at 1 myr of age , which would imply a mass of less than 12 m@xmath2 given this @xmath29 limit .
however the age of hd 142527 and the models are not precise enough to ascertain this upper limit mass value .
we stress again that this limit constrains the presence of massive gaseous giants at stellocentric radii of 14 to 35 au .
the contribution from photoevaporation to cavity clearing can be neglected in this case .
the inner disk , the large gap and the vigorous accretion rate of @xmath30 m@xmath31 yr @xmath32 argue against photoevaporation .
an x - ray driven wind can indeed explain a subset of the observed accreting disks with inner holes , but only those with radii smaller than 20 au and accretion rates less than 10@xmath33 m@xmath31 yr @xmath32 @xcite .
gap clearing by single or multiple planets could potentially explain wide gaps observed in transitional disks @xcite .
we thus focus on the simplest possible scenario to interpret the morphology of the cavity in hd 142527 s disk , i.e. dynamical clearing by one single embedded and massive - jupiter - size planet opening a gap in its gaseous disk .
the maximum gap opening width will depend on the mass and the orbital parameters of this planet @xcite .
the cavity forms due to the torque exerted by the star
planet binary @xcite .
elliptical cavities have been predicted with @xmath34 in the case of _ circular _ binaries with mass ratios @xmath35 , in which case the gap becomes large enough to deplete the region in which the eccentricity - damping outer 1:3 lindblad resonance is produced .
further numerical work @xcite shows that , for eccentric binaries , the outer edge of the cavity becomes even more eccentric .
the observed eccentric cavity is then a recurrent feature of gap - clearing models . in order to test further the single - planet origin for the wide gap in the gaseous disk of hd 142527
, we conducted hydrodynamic simulations with fargo .
our goal here is not to fit the data but to verify qualitatively the tenability of this scenario for hd 142527 .
fargo is a dedicated grid - based 2d code publicly available on the web , and specifically designed for planet disk interactions .
the run presented in this paper corresponds to a system evolution equivalent to 1.7 myr .
we picked common fiducial parameters for the model , as detailed below .
the model disk is initially axisymmetric , its surface density is initially a power law of radius , @xmath36 , where @xmath37 ( @xmath38 m@xmath12 au@xmath22 ) at @xmath39 au , so slightly smaller than the value assumed by @xcite . the model disk has a constant pressure scale height @xmath8 over radius @xmath40 with value @xmath41 . for the disk calculations the boundaries are 4 and 300 au for the inner and outer radii , respectively , giving a total mass of 0.1 m@xmath12 .
the inner boundary is open , allowing mass to fall onto the star , while the outer boundary is such that we allow for a steady state mass transfer into the disk .
accretion across the disk is modeled by using an @xmath42 prescription .
we adopted @xmath43 following @xcite . from the sinfoni limits described above we know that any planet present in the inner 14 to 35 au can not have a mass exceeding @xmath012 jupiter masses .
we assume no such planet is present further out in the disk either , and chose a planet mass of 10 @xmath44 .
we placed a 10 @xmath44 planet in a fixed circular orbit at 90 au in the model disk .
this radius was chosen so that the planet s sphere of influence , given by a few times its hill radius ( e.g. * ? ? ?
* ) , reaches the observed gap size of @xmath0130 au .
migration was inhibited until a sizeable gap was carved by the planet .
this artificial constraint stems from the lack of heat or momentum diffusion in fargo @xcite . the resulting gas surface density field is shown in fig .
[ fig : fargo ] .
this specific dynamically perturbed morphology will persist over a timescale of a few tens of thousand years ( i.e , around 100 orbits ) but is not in steady state .
the rim will remain asymmetric over time , and is composed of tightly wound spiral arms , whose superposition forms a roughly elliptical cavity .
the cavity centroid is persistently offset from the star , but any particular morphology of the gap is transient and lasts about 100 orbits . in this snapshot view ( fig .
[ fig : fargo ] ) , the outer edges of the cavity are most reminiscent of those seen in the nici images .
the cavity eccentricity is @xmath00.6 , and its centroid is offset from the star by @xmath016 au . as seen in fig .
[ fig : fargo ] , this model produces a massive inner disc , extending out to 50 - 60 au in radius , which is not consistent with the observations .
additional ingredients are required to deplete that region in the framework of a dynamical clearing model , e.g. , additional planets closer to the star ( as in * ? ? ?
fitting the inner disc is left for future work .
we stress that the purpose of these simulations is to provide a plausible explanation of the nici observations of the inner edge of the outer disk only .
our observations of hd 142527 indicate that the inner rim of its outer disk is roughly elliptical , and composed of an overlapping set of spiral arm segments .
we confirm that the star is offset by @xmath1 from the centroid of the cavity .
no companion is detected in the range of 14 to 35 au , with a lower limit @xmath28 at 3 @xmath11 ( for a distance of 140 pc ) .
the observed morphology of the disk in hd 142527 suggests a dynamically perturbed state
. the segmented spirals , and the offset between the cavity centroid and the star , can be seen among the varied morphologies predicted by hydrodynamic simulations of hd 142527 .
to explore , at least qualitatively , the origin of the cavity and its peculiar shape , we reported on a simple case . a massive protoplanet on a circular orbit whose radius is comparable to the disk s outer edge perturbs the entire disk , which does not reach steady state even after thousands of orbits .
the global shape of the cavity is a long - standing feature ; the general asymmetries ( off - center star , eccentricity of the cavity ) are imprinted early and remain for the duration of the calculations .
however , the specific configuration of the model run , position of the streamers , and position and number of spiral arms , form a specific morphology that is only a transient phase .
a particular morphology may recur , but lasts for a limited number of orbits .
that phase which qualitatively matches hd 142527 lasts for a hundred orbits or so .
it is tempting to suggest that the observed shape of the disk rim of hd 142527 today will also be brief compared to its age , with features that are unlikely to be found in other systems that are also undergoing dynamical clearing . in this scenario ,
snapshot imaging of transition disks are likely to look different in their fine details
. however , there should be common features like off - centering of the central star , non axi - symmetry of the disk rim , and presence of spiral arms .
current imaging campaigns are indeed revealing these structures , as in hd 100546 @xcite ; hd 135344b @xcite ; ab aur @xcite .
interstingly , the list of known transition disks is rapdily increasing with improving observational capabilities both in direct imaging at optical / near - ir ( see e.g. , the citations above ) and in the sub - mm regime @xcite
. we should be able to verify these ideas in the near future , specifically that the observed morphological variety of transition disks derives from a common dynamical history in this evolutionnary stage when there is large - scale feedback between the planet formation process and its parent disk .
a key to test this transient scenario for hd 142527 and transition disks in general would be to detect the gap - crossing planetary accretion flow and link it to the outer - disk asymmetries .
meanwhile , a quantitative comparison with hydrodynamic models requires radiative transfer calculations on the model density fields .
while this article was under review @xcite reported on a tentative detection of a low - mass stellar companion to hd 142527 , at @xmath013 au , so abutting on the inner disk .
this companion is inferred through parametrised - modeling of visibility data .
if real , this stellar companion will strongly impact on the inner disk , and its inclusion in simulation may bring closer agreement with the observed inner disk properties
. this would be particularly interesting given that our one - planet scenario currently fails to reproduce the properties of the inner disk .
it is currently too large in radius and mass , compared with observations . including a second body in the simulation would produce a larger gap with a smaller inner disk ( e.g , * ? ? ?
however , hd 142527b alone probably can not explain the extent of the cavity .
a single body can not clear a gap wider than @xmath05 times its tidal radius ( e.g. * ? ? ? * ) , given by @xmath45 , where @xmath46 is known as the hill radius .
mass and semi - major axis estimates for the companion star detected by @xcite give a hill radius @xmath47 au .
sc , sp , aj , fm , ms and ah acknowledge support from the millennium science initiative ( chilean ministry of economy ) , through grant `` nucleus p10 - 022-f '' .
we thank the eso director general for discretionary time that allowed the k - band limit .
sc , aj and jc thank the cata ( `` fondo basal pfb-06 , conicyt '' ) .
sc acknowledges fondecyt grant 1100221 .
aj acknowledges additional support from anillo act-086 and fondecyt project 1095213 .
fm acknowledges support from pnps of cnrs / insu and from eu fp7 - 2011 under grant agreement no .
284405 ( diana ) .
jc acknowledges vri - puc ( inicio 16/2010 ) and fondecyt 11100240 . | the vestiges of planet formation have been observed in debris disks harboring young and massive gaseous giants . the process of giant planet formation is terminated by the dissipation of gas in the protoplanetary disk .
the gas - rich disk around hd 142527 features a small inner disk , a large gap from @xmath010 to @xmath0140 au , and a massive outer disk extending out to @xmath0300 au .
the gap could have been carved - out by a giant planet .
we have imaged the outer regions of this gap using the adaptive - optics camera nici on gemini south .
our images reveal that the disk is dynamically perturbed .
the outer boundary of the roughly elliptical gap appears to be composed of several segments of spiral arms .
the stellar position is offset by @xmath1 from the centroid of the cavity , consistent with earlier imaging at coarser resolutions .
these transient morphological features are expected in the context of disk evolution in the presence of a perturbing body located inside the cavity .
we perform hydro - dynamical simulations of the dynamical clearing of a gap in a disk . a 10 m@xmath2 body in a circular orbit at @xmath3 au , perturbs the whole disks , even after thousands of orbits .
by then the model disk has an eccentric and irregular cavity , flanked by tightly wound spiral arms , but it is still evolving far from steady state .
a particular transient configuration that is a qualitative match to hd 142527 is seen at 1.7 myr . |
in this paper we are concerned with a percolation model , first introduced in @xcite , which is known as mandelbrot s fractal percolation process and which can be informally described as follows . for any integers @xmath1 and @xmath2
we start by dividing the unit cube @xmath3^d \subset { \mathbb r}^d$ ] into @xmath4 closed subcubes of equal size @xmath5 .
given @xmath6 $ ] and a subcube , we retain the subcube with probability @xmath7 and discard it with probability @xmath8 .
this is done independently for every subcube of the partition .
sometimes we adopt the terminology calling retained cubes _ black _ and deleted cubes _
white_. we define the random set @xmath9^d$ ] as the union of all retained subcubes .
next consider any retained ( assuming that @xmath10 ) subcube @xmath11 in @xmath12 we repeat the described procedure on a smaller scale by dividing @xmath11 into @xmath4 further subcubes , discarding or retaining them as above .
we do this for every retained subcube of @xmath12 this yields a new random set @xmath13 iterating the procedure on every smaller scale yields an infinite sequence of random sets @xmath3^d \supset { \mathcal c}_n^{1 } \supset { \mathcal c}_n^{2 } \supset \cdots $ ] and we define the limiting set @xmath14 we will hereafter suppress the @xmath15 in our notation and simply write @xmath16 for @xmath17 . we will need a more formal definition of the model as well . let @xmath18,\ ] ] where @xmath19 and @xmath20 . for @xmath21 consider the subcube @xmath22 of @xmath3^d$ ] defined by @xmath23 , and let @xmath24 .
a cube @xmath22 will sometimes be called a _ level_-@xmath25 cube .
we define the sample space by @xmath26 and denote an element of @xmath27 by @xmath28 .
we let @xmath29 be the borel @xmath30-algebra on @xmath27 generated by the cylinders and let @xmath31 denote the product measure on @xmath29 with density @xmath32 $ ] , that is , we let @xmath33 independently for every @xmath34 .
the limiting set is then defined to be the intersection of all @xmath35 such that @xmath36 .
let @xmath37^d)$ ] denote the event that @xmath16 contains a connected component which intersects the left hand side @xmath38^{d-1}$ ] of the unit cube and also intersects the right hand side @xmath39^{d-1}$ ] .
in this case we say that a _ left - right crossing _ of the unit cube occurs .
we define the _ percolation function _ @xmath40 by @xmath41^d)).\ ] ] the _ critical value _ is defined as @xmath42 it has been shown in @xcite that the phase transition in mandelbrot fractal percolation is non - trivial , i.e. @xmath43 . furthermore it was discovered in @xcite that for @xmath44 , @xmath45 is discontinuous at @xmath46 ( see @xcite for an easy proof ) .
this was generalised in @xcite to all @xmath47 and @xmath15 large enough but the result is conjectured to hold for all @xmath15 in any dimension .
( at this point we remark that as a corollary to the proof of theorem [ thm - main3 ] below , we obtain an explicit bound for the size of the discontinuity at the critical value @xmath48 , defined below and conjectured to coincide with @xmath46 , in terms of the hausdorff dimension of the set @xmath49 , also to be defined below . ) in @xmath44 , the set @xmath16 is a.s . totally disconnected for @xmath50 .
this is also known to be true in higher dimensions for the same set of @xmath15 for which it is known that @xmath45 is discontinuous at @xmath46 ( see @xcite ) and is conjectured to be true for all @xmath51 and @xmath15 .
it is therefore natural to work with the following critical value : @xmath52 it is known ( see @xcite ) that for any @xmath53 and @xmath54 , @xmath55 if @xmath56 . given this , it is an easy exercise to show that for @xmath57 , @xmath58 implies that the set @xmath59 consisting of the union of all connected components larger than one point is a.s . not empty .
we now fix @xmath60 and assume that @xmath61 .
for any point @xmath62 , let @xmath63 be the set of points @xmath64 that are connected to @xmath65 in @xmath66 .
we call @xmath67 the _ connected component _ of @xmath65 .
it is known ( see @xcite ) that for @xmath68 there exist a.s .
uncountably many @xmath62 such that @xmath69 .
we partition @xmath66 into two sets , @xmath70 and the aforementioned @xmath71 .
( to understand the notation : _ d _ is short for dust " , and _
c _ is short for connected " . ) before we can state our results we will need some more definitions .
the reader is referred to @xcite for a general overview of the subject of fractal sets .
a countable collection @xmath72 of subsets of @xmath73 with diameter at most @xmath74 is called an @xmath74-cover of @xmath75 if @xmath76 .
define the _
@xmath77-dimensional hausdorff measure _ of @xmath75 as follows : @xmath78 the _ hausdorff dimension _
@xmath79 of @xmath75 is defined as @xmath80 which also turns out to be equal to @xmath81 .
the hausdorff dimension of the limiting set in fractal percolation is a.s . given by the following equation
, whose proof can be found in @xcite or @xcite , proposition 15.4 : @xmath82 there are many other concepts of dimensionality and we will in particular use the following . for
a bounded set @xmath83 let @xmath84 be the minimal number of closed cubes of side length @xmath85 that is needed to cover @xmath75 .
the _ lower box counting dimension _ of @xmath83 is given by @xmath86 while the _ upper box counting dimension _ of @xmath83 is given by @xmath87 if @xmath88
then the common value is denoted @xmath89 and called the _ box counting dimension _ of @xmath75 .
it is known ( see e.g. @xcite ) that for any bounded set @xmath83 @xmath90 the next two theorems contain our dimension results for fractal percolation .
[ thm - main3 ] for @xmath91 , we a.s .
have @xmath92 if @xmath93 then a.s .
@xmath94 from which it easily follows that a.s .
@xmath95 note that if @xmath96 , @xmath97 a.s .
and so equation ( [ eqn - main3.2 ] ) still holds as long as @xmath98 .
[ prop1 ] for every @xmath7 there exists @xmath99 such that @xmath100 for @xmath101 , let @xmath102 be the union of the connected components of diameter at least @xmath74 .
the following result suggests that the small components " of @xmath49 are the ones which actually determine its box counting dimension .
[ thm - main4 ] if @xmath103 , then @xmath104\leq d \dim_b ( { \mathcal c}^c),\ ] ] where @xmath105 is independent of @xmath74 .
when @xmath106 , it is natural to ask about the nature of the left - right crossings of the unit cube .
for @xmath44 , it was shown in @xcite that @xmath66 contains at least one continuous curve crossing the square as soon as a connected component crossing the square exists .
it was later established in @xcite ( again for @xmath44 ) that any curve in @xmath66 must have hausdorff dimension strictly larger than 1 . in this paper ,
focusing again on the two - dimensional version of the model , we take the issue of the existence of continuous curves in @xmath66 much further , using the sophisticated machinery of aizenman and burchard @xcite .
their paper deals with scaling limits of systems of random curves , but we will show how their results can be useful in the context of fractal percolation as well . this is perhaps somewhat surprising , since the scaling limits in @xcite deal with convergence in distribution , whereas in the fractal context , the fractal limiting set is an a.s .
the key will be a very careful comparison between convergence in the weak sense of curves in an appropriate topology , and convergence in the a.s .
sense of compact sets in another topology .
from such a comparison , one can obtain information about the compact sets that make up the a.s .
limit of the fractal construction . in order to state our results , we need some definitions .
first of all , we define _ interface curves _ in the fractal process .
the complement @xmath107 consist of a finite number of connected components , exactly one of which is unbounded .
the boundary of any such connected component can be split into closed curves ( loops ) .
we call such loops _ interface curves _ and denote by @xmath108 the collection of interface curves after @xmath25 iterations of the fractal process . in order for our interface curves to be uniquely defined , we orient them in such a way that they have black ( retained ) squares on the left and white ( discarded ) squares on the right , and assume that they turn to the right at corners where two white and two black squares meet in a checkerboard configuration , see figure [ figuur1 ] . [ figuur1 ] a connected subset of an interface curve delimited by a starting and an ending point will be called an _ interface segment_. we continue with some general definitions concerning curves , mostly taken from @xcite ( see also @xcite ) .
we regard curves in @xmath3 ^ 2 $ ] as equivalence classes of continuous functions from @xmath3 $ ] to @xmath3 ^ 2 $ ] modulo strictly monotonic re - parametrizations .
below , @xmath109 will represent a particular curve and @xmath110 a particular parametrization of @xmath109 .
denote by @xmath111 the complete separable metric space of curves in @xmath3 ^ 2 $ ] with metric @xmath112 } |\gamma_1(t)-\gamma_2(t)|,\ ] ] where the infimum is over all parametrizations of @xmath113 and @xmath114 .
the distance between two sets @xmath115 and @xmath116 of curves is defined by the hausdorff metric induced by d , that is , @xmath117 if and only if @xmath118 the space @xmath119 of closed subsets of @xmath111 with the metric dist is also a complete separable metric space .
the fractal process induces a probability measure @xmath120 on @xmath119 , where @xmath121 denotes the distribution of @xmath108 . with this notation
, we can present our main result on continuous curves in @xmath66 .
[ thmnew1 ] the sequence of measures @xmath122 has subsequential weak limits . any such weak limit @xmath123 assigns probability 1 to curve configurations in which all curves are hlder continuous with the same exponent .
the limiting set @xmath66 has the same distribution as @xmath124 , where @xmath125 is a random set of curves distributed as @xmath123 .
in other words , @xmath66 is distributed as the union of the images of the curves in a sample from a weak limit of the @xmath122 . * remark * note that , although theorem [ thmnew1 ] only claims the existence of subsequential weak limits for @xmath121 as @xmath126 , theorem [ thmnew1 ] combined with the existence of a unique a.s .
limit for @xmath127 implies that @xmath128 has a unique weak limit .
we now briefly discuss this result .
since a single point is of course a hlder continuous curve , the bare statement that a set is the union of hlder continuous curves is in itself close to being an empty statement .
however , the curves in @xmath66 mentioned in theorem [ thmnew1 ] can not be exclusively curves whose image is one point .
one way to see this is to rephrase the notion of weak convergence as follows ( see also @xcite ) .
a sequence of probability measures @xmath122 on @xmath119 converges weakly to a probability @xmath123 measure on @xmath119 if and only if there exists a family of probability measures @xmath129 on @xmath130 such that the first marginal of @xmath129 is @xmath121 , the second marginal of @xmath129 is @xmath123 ( for all @xmath25 ) , and @xmath131 as @xmath126 .
it is now also clear what happens to the points in the dust set " @xmath132 : these are accounted for as well in the theorem , since any point @xmath133 can be approximated by curves in @xmath108 whose diameter and distance to @xmath65 converge to 0 .
it is also possible to specialise to certain particular curves . as an example
, we discuss the _ lowest crossing _ in @xmath66 which we will first properly define .
condition on the existence in @xmath134 of a left - right crossing of the unit square for all @xmath25 , and consider the lowest interface segment @xmath135 in @xmath136 connecting the left and right side of the unit square .
the closure of the region in the unit square above @xmath135 is a compact set ( in the euclidean topology ) which decreases in @xmath25 and which therefore converges as @xmath126 .
the lowest crossing in @xmath66 is defined as the boundary of this limiting set .
[ thm : lowest - crossing ] if @xmath66 contains a left - right crossing of the unit square , then the lowest crossing in @xmath66 is a hlder continuous curve .
the machinery of aizenman and burchard also allows for a quick proof , given in section [ sec - hausdorff ] , of the following result , first proved in @xcite .
[ thmnew2 ] in two dimensions , there exists a constant @xmath137 such that all continuous curves in @xmath49 have hausdorff dimension at least @xmath138 .
_ proof of theorem [ thm - main3 ] . _
we start with the second equality of .
let @xmath139 be the number of cubes of side length @xmath140 that are retained in @xmath141 .
first observe that @xmath142 it is well known from the theory of branching processes ( see e.g. @xcite ) that @xmath143 a.s . on the event @xmath58 .
therefore , for a.e .
@xmath28 such that @xmath144 , @xmath145 and hence @xmath146 . equations ( [ eqn : haus - dim - retain - set ] ) and ( [ eqn4 ] ) then imply that @xmath147 .
for the first equality of , let @xmath148 be a cover of @xmath59 using the minimal number @xmath149 of closed cubes of side length @xmath85 . for @xmath150
, define @xmath151 , with @xmath152 denoting euclidean distance .
assume that there exists @xmath153 such that @xmath154 .
then there must exist some @xmath22 such that @xmath155 and @xmath156 , which implies that @xmath157 ( i.e. , @xmath158 is retained ) . however , because of the scale invariant construction of @xmath16 and the fact that , for @xmath159 , a.s .
@xmath16 is either empty or contains connected components larger than one point ( see @xcite ) , @xmath160 must contain connected components larger than one point .
this contradicts the fact that @xmath148 is a cover of @xmath59 and shows that such an @xmath65 can not exist .
furthermore , since the union @xmath161 is closed , it follows that , if @xmath162 for @xmath62 , then @xmath163 .
therefore , @xmath149 must be the minimal number of closed cubes of side length @xmath85 that covers @xmath16 .
this concludes the proof of . since @xmath164
, follows from and the fact that ( see @xcite ) @xmath165 we proceed therefore by proving , inspired by an argument suggested by lincoln chayes to the second author .
recall that @xmath166 is the union of the connected components of diameter at least @xmath74 . for @xmath57
, we have that @xmath167 which is an easy consequence of the definition of hausdorff dimension ( see @xcite ) .
therefore , it suffices to find an upper bound of @xmath168 which is uniform in @xmath74 and strictly smaller than @xmath169 .
we will now assume that @xmath170 is odd .
this assumption will make certain definitions easier to write down : the reader can check that the squares @xmath22 below would not be uniquely defined for even @xmath15 .
we leave it to the reader to adapt the proof for all cases @xmath171 . for @xmath172 such that @xmath173 , let @xmath174^d$ ] and @xmath175 be the two cubes concentric to @xmath176 with side length @xmath177 and @xmath178 respectively .
let @xmath179 be the probability that there exists a connected component of @xmath16 that crosses the `` shell '' @xmath180 .
it follows from @xcite that @xmath181 whenever @xmath57 . assuming that @xmath103 , fix @xmath101 such that @xmath59 contains at least one component of diameter larger than @xmath74 and let @xmath182 be such that @xmath183 .
consider a cube @xmath22 for @xmath184 which is intersected by a component of @xmath59 of diameter larger than @xmath74 .
let @xmath185 and @xmath186 be two cubes which are concentric to @xmath22 and have side length @xmath187 and @xmath188 respectively .
obviously , for @xmath22 to be intersected by a connected component of diameter larger than @xmath74 , there must be a crossing of the shell @xmath189 .
( note that , depending on the position of @xmath22 , it is possible that @xmath185 and/or @xmath190 are only partially contained in @xmath3^d$ ] . ) we will now construct a specific cover of @xmath102 which we will use in our estimate for its hausdorff dimension .
let @xmath191 denote the set of cubes @xmath22 with the following two properties : * the intersection of all retained cubes of level @xmath25 and higher contains a crossing of the shell @xmath192 .
in other words , if we would make all cubes black until level @xmath193 ( inclusive ) , then there would be a connected component in @xmath66 crossing the shell @xmath192 , * @xmath22 is retained , that is , @xmath36 . by scale invariance and independence between the two conditions , we have that @xmath194 .
the inequality is due to a boundary effect since , as mentioned earlier , @xmath185 and/or @xmath186 need not be completely contained in @xmath3^d$ ] . for a given cube @xmath22 ,
let @xmath195 denote the level-@xmath196 cube which contains @xmath22 , where @xmath197 ( with @xmath198 ) .
we make two observations : 1 .
if @xmath22 has a non - empty intersection with @xmath102 , then we have that @xmath199 , for all @xmath200 .
2 . the events @xmath201
form a collection of independent events ; see figure [ figuur2 ] .
this motivates us to define @xmath202 as the collection of cubes @xmath22 for which the corresponding cubes @xmath195 are in @xmath203 , for all @xmath200 . from observation 1 , we have that the collection @xmath202 forms a cover of @xmath102
. we can now write , using observation 2 , @xmath204 [ figuur2 ] is in @xmath205 and @xmath206 , drawn with broken lines , belongs to @xmath207 .
note that the corresponding shells are disjoint.,width=302 ] using fatou s lemma and the fact that the collection of cubes in @xmath202 covers @xmath102 , we obtain ( writing @xmath208 for the number of cubes in @xmath202 ) @xmath209 the limit in is finite if and only if @xmath210 showing that @xmath211 it follows from ( [ eqn5 ] ) that @xmath212 and since @xmath213 the result follows from this and ( [ eqn : haus - dim - retain - set ] ) .
@xmath214 _ proof of theorem [ prop1 ] . _ if @xmath215 then there is nothing to prove , so we assume that @xmath216 .
let @xmath139 be the number of retained cubes after @xmath25 steps of the fractal construction procedure and let @xmath217 denote the retained cubes .
if the event @xmath218 occurs , then @xmath219 for at least one @xmath220 ( see , e.g. , @xcite ) .
therefore , @xmath221 however , because of scale invariance , @xmath222 , conditioned on the event that @xmath223 , must have the same distribution as @xmath224 and so in fact @xmath225 we can now write @xmath226{\ensuremath{\mathbb{p}}}_p(z_n\geq n).\end{aligned}\ ] ] this last quantity is bounded below by @xmath227{\ensuremath{\mathbb{p}}}_p(z_n\geq n \big{| } { \ensuremath{\mathcal{c}}}\neq \emptyset ) { \ensuremath{\mathbb{p}}}_p({\ensuremath{\mathcal{c}}}\neq \emptyset )
. \label{eqn11}\ ] ] as mentioned above , a.s . @xmath228
as @xmath229 if @xmath230 .
therefore , if @xmath231 , by taking the limit of as @xmath126 , we conclude that @xmath232 .
if @xmath233 , it follows that in fact @xmath234 . letting @xmath235 , we have that @xmath236 and either @xmath237 or @xmath238 when @xmath233 .
now observe that @xmath103 implies @xmath239 ( see , e.g. , proposition 4.1 of @xcite ) . since the inverse implication is obvious ,
we conclude that @xmath240 ( where the last inequality follows from the assumption made at the beginning of the proof ) and that @xmath241 .
hence , since @xmath242 , we obtain that @xmath243 up to a set of probability 0 .
moreover , since @xmath238 for at least @xmath244 , and @xmath245 for @xmath233 , we also obtain that @xmath246 , up to a set of probability 0 , for every @xmath233 such that @xmath247 .
writing @xmath248 we conclude from the above observations that , up to a set of probability 0 , if @xmath103 then @xmath249 , with @xmath250 as a consequence of the assumption that @xmath251 and the fact that @xmath239 as soon as @xmath103 .
@xmath214 theorem [ prop1 ] and the proof of theorem [ thm - main3 ] have an interesting corollary which links @xmath224 to the discontinuity at the critical point @xmath252 .
( this is the corollary that was announced in the introduction . )
[ cor1 ] let @xmath253 denote the a.s .
hausdorff dimension of @xmath59 when @xmath56 and @xmath103 .
then , @xmath254 _ proof .
_ let @xmath255 .
theorem 4.1 of @xcite and the observation preceding it show that @xmath256 .
moreover , it follows from @xcite that @xmath257 . combining these observations with the last line of the proof of theorem [ thm - main3 ]
, we obtain that , for @xmath159 , @xmath258 @xmath214 * remark * for theorem [ thm - main3 ] we use the result from @xcite that for @xmath56 , @xmath259 .
it is possible to prove the result without this prior knowledge , as follows .
we can start with the observation from @xcite that @xmath260
. we can then prove theorem [ thm - main3 ] in the case of @xmath261 and from the last line of that proof we get that @xmath262 , using corollary [ cor1 ] and the fact that the hausdorff - dimension of a connected set consisting of more than one point is at least 1 ( see , e.g. , proposition 4.1 of @xcite ) . using this uniform bound and a right - continuity argument similar to the ones in @xcite , we conclude that in fact @xmath263
. hence we can _ conclude _ that for @xmath56 , @xmath264 , and go through the proof once more to obtain the same result as above .
_ proof of proposition [ thm - main4 ] .
_ we again use the fact that @xmath166 is contained in the union of the cubes in @xmath202 , defined in the proof of theorem [ thm - main3 ] , and therefore , @xmath265 . first observe that @xmath266 by fatou s lemma and jensen s inequality we get along the same lines as in the last part of the proof of theorem [ thm - main3 ] ( using the same @xmath182 as in that proof ) that @xmath267 since @xmath268 it follows that @xmath269 . @xmath214
in this section and the next one we work in two dimensions , that is , @xmath44 .
we will be using the machinery in @xcite concerning scaling limits of systems of random curves .
one part of the argument is to apply some of the machinery developed in @xcite in order to show that the sequence of measures @xmath120 defined in the introduction has subsequential weak limits , and we first deal with this issue .
after that , we combine the existence of weak limits with the a.s .
limit behaviour of the fractal process in order to draw the final conclusions .
let @xmath270 denote a closed square centered at @xmath65 with side length @xmath271 and @xmath272 its interior .
for @xmath273 , let @xmath274 be an annulus .
the basic estimate is the following , from which everything else will follow .
[ lem - disjoint - crossings ] let @xmath275 .
there exists a sequence @xmath276 with @xmath277 and finite constants @xmath278 such that the following bound holds uniformly for all @xmath279 with @xmath271 small enough , and all @xmath65 : @xmath280 _ proof .
_ we are looking for a collection of mutually disjoint annuli ( all contained in @xmath281 and surrounding " @xmath270 ) of the form @xmath282 , where @xmath283 is a corner point of some square @xmath158 of the fractal construction .
it is not hard to see that for any @xmath284 ^ 2 $ ] , we can find such a collection with at least @xmath285 elements , for a suitable [ uniform ? ? ? ] positive constant @xmath286 .
for small enough @xmath271 , @xmath287 is at least 1 . for a given @xmath65
, we denote these annuli by @xmath288 , where the indices @xmath289 refer to the @xmath25 associated with the annuli .
the idea is that if there are @xmath290 disjoint crossings of @xmath291 , then each annulus @xmath292 must also be crossed by @xmath290 disjoint crossings in @xmath108 .
this is exponentially unlikely in the number of annuli , as we will show now .
we first consider the annulus @xmath293 .
let @xmath294 , and perform the fractal process until level @xmath295 ( inclusive ) .
the annulus @xmath293 consists of 12 level-@xmath295 squares ( figure [ figuurtje ] ) , some of which are in @xmath296 and some of which are not . now observe that the collection of black ( retained ) squares of level @xmath295 in @xmath293 is partitioned into at most 6 level-@xmath295 components " , where two neighbouring retained level-@xmath295 squares are in the same component if they share an edge : see figure [ figuurtje ] for an illustration of the event that the annulus contains 3 such level-@xmath295 components .
[ figuurtje ] with 3 level-@xmath295 components , drawn in dark grey.,width=302 ] now let @xmath290 be large enough ( exactly how large will become clear soon ) .
if @xmath293 is , after @xmath25 iterations , crossed by @xmath290 disjoint interface segments , then at least one of the level-@xmath295 components must be crossed by at least @xmath297 such segments , since there are at most 6 components to accomodate these crossings ( we should write @xmath298 but we ignore these details for the sake of notational convenience ) . between the interface crossings , one has alternating black ( retained ) and white ( discarded ) crossings of the annulus .
this implies that the component with at least @xmath297 interface crossings is , after @xmath25 iterations , also crossed by at least @xmath299 white crossings which are between " the first " and last " of the @xmath297 interface crossings of said component .
the point of considering the level-@xmath295 components introduced above is that they are disjoint and separated by vacant squares , so that the fractal constructions inside them from level @xmath295 on are independent of each other , and none of the interface segments crossing @xmath293 can intersect more than one of them . therefore , adding extra retained level-@xmath295 squares in @xmath293 does not affect the @xmath299 white crossings in the component with at least @xmath297 interface crossings .
this implies that the probability of having at least @xmath290 interface crossings in @xmath293 after @xmath25 iterations is bounded above by the probability of having at least @xmath299 white crossings after @xmath25 iterations , conditioned on having full retention in @xmath293 up to level @xmath295 .
however , scaling tells us that , if we condition on retention until level @xmath295 , the probability in question is the same as the probability of having at least @xmath299 disjoint white components crossing @xmath300 when we perform @xmath301 iterations of the fractal process in @xmath302 ^ 2 $ ] rather than in @xmath3 ^ 2 $ ] , seen as the union of 16 independent fractal processes on the 16 unit squares making up @xmath302 ^ 2 $ ] .
for these white crossing components we can use the bk inequality ( see @xcite - a similar bk inequality for black crossing is not available ) and deduce that the probability of having @xmath299 of such white crossing components is bounded above by the probability of having at least one , raised to the power @xmath299 .
this then finally leads to the estimate that @xmath303 is bounded above by @xmath304 if there is a white component crossing @xmath305 , then there is no black circuit surrounding the origin in @xmath305 .
the probability of having such a black circuit after @xmath301 iterations is at least as large as the probability to have such a circuit in the limit , and by the weak rsw theorem for fractal percolation in @xcite and the fkg inequality , we have that for @xmath275 this probability is strictly positive .
it follows that there exists @xmath306 such that ( [ mooi ] ) is bounded above by @xmath307 , uniformly in @xmath294 .
next we consider @xmath293 and @xmath308 simultaneously .
take @xmath309 .
the probability to have @xmath290 interface crossings in @xmath293 and also in @xmath308 is the probability that this happens in @xmath293 multiplied by the probability that this happens in @xmath308 conditioned on the fact that it happens in @xmath293 .
we can treat this conditional probability exactly as above : we can change the conditioning into one involving complete retention inside @xmath308 until level @xmath310 , to get an upper bound .
it follows that the probability that both @xmath293 and @xmath308 have @xmath290 interface crossings is bounded above by the square of the individual bounds , that is , by @xmath311 .
we continue in the obvious way now , leading to the conclusion that for @xmath312 , the probability that _ all _ annuli @xmath313 are crossed by @xmath290 interface crossings , is bounded above by @xmath314 a little algebra shows that this is equal to @xmath315 and this is a bound of the required form , with @xmath316 .
@xmath214 we now describe how the existence of subsequential weak limits of the sequence @xmath120 follows from this lemma .
( this is well known but perhaps not immediately obvious from the literature , hence our summary for the convenience of the reader . ) for @xmath317 and positive integers @xmath318 define @xmath319 ^ 2 , \text { is}\\ \text{crossed by } k \text { disjoint crossings in } { \ensuremath{{\cal{f}}}}_{n } \end{array } \right\ } , 1 \right ) .\ ] ] it follows exactly as in @xcite or @xcite that , as a consequence of lemma [ lem - disjoint - crossings ] , for any @xmath101 , for large enough @xmath290 the random variables @xmath320 are stochastically bounded away from zero as @xmath126 , that is @xmath321 note that in lemma [ lem - disjoint - crossings ] we have the result only for @xmath271 small enough , while the corresponding hypothesis h1 in @xcite is stated without that restriction .
our lemma [ lem - disjoint - crossings ] however is sufficient to prove ( [ bnd - stoch - away ] ) , since the latter concerns the behaviour of @xmath322 as @xmath323 .
note also that hypothesis h1 is used in equation ( 3.4 ) in @xcite only for annuli whose inner radius equals @xmath324 , with @xmath325 for @xmath326 . by taking @xmath327 sufficiently small in the proof of lemma 3.1 in @xcite , also the inner radius @xmath328 becomes sufficiently small and we can apply our lemma [ lem - disjoint - crossings ] . hence , ( [ bnd - stoch - away ] ) follows from our lemma [ lem - disjoint - crossings ] by the same arguments used in @xcite or @xcite . as shown in @xcite ( see in particular their proof of theorem 1.1 and equation ( 1.7 ) in their remark ( ii ) following the statement of theorem 1.1 , but note the typo in equation ( 1.7 ) , where @xmath329 should be @xmath330 ) , equation ( [ bnd - stoch - away ] ) implies the following result .
[ thm - regularity ] for any @xmath331 , all curves @xmath332 can be parametrized by continuous functions @xmath333 \to [ 0,1]^2 $ ] such that for each curve , for all @xmath334 @xmath335 where the random variables @xmath336 are stochastically bounded as @xmath126 , that is , @xmath337 once we have this result , we use theorem 5.7 in @xcite and theorem 1.1 in @xcite to conclude that the sequence of measures @xmath338 is tight . since @xmath119 is separable , it then follows from prohorov s theorem that for every sequence @xmath339 there exists a subsequence @xmath340 such that @xmath341 converges weakly to a probability measure on @xmath119 .
finally , from the fact that the @xmath336 are stochastically bounded and the fact that the collection of curves with a given hlder exponent is compact , we have ( see also @xcite and @xcite ) that if we sample from any such weak limit , all curves @xmath109 in the sample can be parametrized in such a way that latexmath:[\[\label{eqnholder } @xmath287 is a random number common to all curves in the same sample , and @xmath343 is a ( non - random ) constant .
next we combine the above weak convergence with the a.s.convergence of the retained squares ( as compact sets ) in the fractal process .
let @xmath344 denote the metric space of compact subsets of @xmath3^d$ ] with the hausdorff distance @xmath345 and let @xmath346 be as defined after equation ( [ hausdorff - d ] ) .
furthermore let the function @xmath347 be defined by @xmath348 and define @xmath349 .
our next result concerns weak convergence of @xmath350 where we use the product topology on @xmath351 .
[ lem - joint - weak - conv ] the distribution of @xmath350 converges weakly along a subsequence .
furthermore , any pair @xmath352 of random variables sampled from any such weak limit a.s .
satisfies @xmath353 . _ proof .
_ we already know that the distribution @xmath121 of @xmath136 converges weakly along a subsequence . using ( [ distance ] ) and ( [ hausdorff - d ] ) , if @xmath354 , this immediately implies that @xmath355 since the images of any two curves
@xmath356 such that @xmath357 are within hausdorff distance @xmath85 .
this proves that @xmath358 is continuous .
the convergence in distribution of @xmath136 along some subsequence @xmath359 to a limit @xmath360 implies the existence of coupled versions @xmath361 and @xmath362 of @xmath363 and @xmath360 respectively , such that @xmath361 converges to @xmath362 in probability as @xmath364 ( see , e.g. corollary 1 in @xcite ) .
moreover , since @xmath358 is continuous , @xmath365 converges in probability to @xmath366 as @xmath364 .
this implies convergence in probability of the vector @xmath367 to @xmath368 , which yields the joint convergence in distribution of @xmath369 to some limit @xmath352 with @xmath353 a.s .
@xmath214 _ proof of theorem [ thmnew1 ] .
_ according to lemma [ lem - joint - weak - conv ] there exists a subsequence @xmath370 such that @xmath371 converges weakly to some limit @xmath352 where @xmath75 is a.s . the union of the images of @xmath115 .
furthermore , we claim that a.s .
@xmath372 where @xmath373 . to see this ,
let @xmath374 denote the @xmath74-neighbourhood of @xmath375 and note that @xmath376 implies that there exists an @xmath377 such that @xmath378 and so @xmath379 , otherwise an interface curve would be closer to @xmath65 than @xmath74 .
it is however easy to prove that the probability that there exists a ball of radius @xmath74 in @xmath141 goes to 0 . for the other direction ,
let @xmath380 be the complement of the open @xmath74-neighbourhood of @xmath16 .
obviously @xmath381 is a compact set
( if nonempty ) and furthermore @xmath382 for every @xmath25 . therefore , by compactness , if @xmath383 for every @xmath25 then @xmath384 and so there are points in @xmath16 that are also in @xmath385 which is a contradiction .
therefore , the open @xmath74-neighbourhood of @xmath16 will eventually contain @xmath141 and hence also @xmath375 .
since @xmath386 converges weakly to @xmath75 and a.s .
to @xmath16 ( because of ( [ as - convergence ] ) ) , we conclude that @xmath75 and @xmath16 have the same distribution , and hence @xmath16 has the same distribution as @xmath387 .
the fact that ( as noted above ) a.s .
, a realisation from @xmath125 only contains hlder continuous curves satisfying ( [ eqnholder ] ) finishes the proof .
@xmath214 _ proof of theorem [ thm : lowest - crossing ] .
_ let @xmath388 be the distribution of @xmath135 .
we repeat the proof of theorem [ thmnew1 ] , with the distance ( [ hausdorff - d ] ) replaced by ( [ distance ] ) .
this shows that conditioned on the existence of a left - right crossing for all @xmath25 , @xmath388 has subsequential weak limits as @xmath126 and in addition that any such limit assigns probability 1 to hlder continuous curves . @xmath214
we will start by showing that the fractal percolation process satisfies hypothesis h2 in @xcite , from which theorem [ thmnew2 ] follows .
the hypothesis concerns probabilistic bounds on crossings in the long direction of certain rectangles .
a collection of sets @xmath389 is called _ well - separated _ if for all @xmath390 the distance of each set @xmath391 to the other sets @xmath392 is at least as large as the diameter of @xmath391 .
hypothesis h2 in @xcite reads as follows .
[ hyp-2 ] there exist @xmath393 and some @xmath394 such that for every collection of @xmath290 well - separated rectangles @xmath395 of width @xmath396 and length @xmath397 , the following inequality holds : @xmath398 _ proof .
_ we assume without loss of generality that @xmath399 . let @xmath400 be the smallest integer @xmath25 for which _ all _ rectangles of dimensions @xmath401 and @xmath402 contain an @xmath25-level square of the fractal construction , and define @xmath403 as the event of complete retention until iteration step @xmath400 .
let @xmath404 denote the event that the closed rectangle @xmath391 is crossed in the long direction by an interface segment after @xmath25 iterations of the fractal process , and define @xmath405 similarly for a black crossing .
note that by the fact that @xmath406 are well - separated and by the choice of @xmath295 , given @xmath407 , the event @xmath408 is , when @xmath294 , conditionally independent of the events @xmath409 , since no level-@xmath25 square intersects more than one rectangle .
note also that , if a rectangle is crossed in the long direction by an interface segment , then it also contains a black crossing in the long direction .
using these two facts and the fkg inequality gives , for @xmath294 , @xmath410 since @xmath399 , we have @xmath411 and hence @xmath412 .
it follows as before that for @xmath309 we have @xmath413 we repeat this procedure @xmath414 more times , finally obtaining , for @xmath415 , that @xmath416 it remains to show that @xmath417 is uniformly bounded above by some @xmath418 .
this can be seen as follows .
let , for each @xmath390 , @xmath419 be a smallest collection of level-@xmath420 squares in the fractal process with the property that if each of the squares in @xmath419 is white , then @xmath391 can not be crossed by a black path in the long direction .
it is easy to see from the choice of @xmath400 that the cardinality of @xmath419 is uniformly bounded above and hence that the probability that all squares in @xmath419 are white is uniformly bounded below by some positive number .
it follows that the probability of a long black crossing in @xmath391 is uniformly bounded above by some number strictly smaller than 1 .
@xmath214 e.i .
broman , f. camia , large-@xmath15 limit of crossing probabilities , discontinuity , and asymptotic behavior of threshold values in mandelbrot s fractal percolation process , _ electr . j. prob . _
* 13 * , 980 - 999 ( 2008 ) . | we relate various concepts of fractal dimension of the limiting set @xmath0 in fractal percolation to the dimensions of the set consisting of connected components larger than one point and its complement in @xmath0 ( the `` dust '' ) . in two dimensions ,
we also show that the set consisting of connected components larger than one point is a.s .
the union of non - trivial hlder continuous curves , all with the same exponent . finally , we give a short proof of the fact that in two dimensions , any curve in the limiting set must have hausdorff dimension strictly larger than 1 . _ ams subject classification _ : 60k35 , 28a80 , 37f35 , 54c05 . _ key words and phrases _ : fractal percolation , hausdorff dimension , box counting dimension , hlder continuous curves , subsequential weak limits . |
hot cores ( hcs ) are the suspected birthplaces of high - mass stars ( m@xmath1m@xmath2 ) @xcite . based on mm
/ submm interferometric results which provide sufficient angular resolution to separate hcs from neighbouring ultracompact hii regions ( s ) , first attempts were made to derive properties of the embedded high - mass stars as well as the surrounding envelope using infall models in combination with 1d radiative codes @xcite .
however , these results relied on the mm / submm part of the spectral energy distribution ( sed ) only .
we performed infrared ( ir ) observations of hcs to provide flux densities ( or at least upper limits ) for the atmospheric 10 and 20 spectral windows in order to complete the coverage of their seds with sub - arcsecond beam sizes and to investigate their morphology .
the observations were carried out with spectrocam10 ( sc10 ) on the 5-m hale and timmi2 on the eso 3.6-m telescopes .
sc10 is the cornell - built 813 @xmath3 m spectrograph / camera @xcite which utilizes a rockwell 128@xmath4128 si : as bib array , providing a circular field of view ( fov ) in imaging mode of 16 ( pixel size 025 ) .
the sc10 imaging was performed in december 1998 and june 1999 .
the images were filtered using a wavelet algorithm @xcite to enhance the signal - to - noise ratio ( snr ) .
timmi2 is the new eso thermal imaging multi - mode instrument covering the wavelength range from 520 @xcite .
its 320@xmath4240 si : as bib array manufactured by raytheon provides an unprecedented fov in imaging mode of @xmath5 ( pixel scale 02 ) at 10 and 20 and @xmath6 ( pixel scale 03 ) at 4.7 .
the observations were carried out in march 2001 .
o ) is located at the offset position [ + 36,-10 ] . the inset image of @xmath7 tau indicates the beam size .
] [ eps1 ] the hc w3(h@xmath0o ) is located @xmath8 east of the w3(oh ) at a distance of 2.2kpc .
the proper motion of its h@xmath0o masers @xcite and the associated radio continuum jet @xcite suggest recent outflow activity .
interferometric radio observations of w3(h@xmath0o ) @xcite yielded h@xmath0 column densities of up to 1.5@xmath9@xmath10 and rotation temperatures in the range of 160200k .
the detection of the hc at mid - infrared ( mir ) wavelengths was claimed by @xcite .
1 shows the sc10 image of w3(h@xmath0o)/w3(oh ) with black contours delineating the 3.6 cm radio continuum @xcite .
w3(h@xmath0o ) is at the offset position [ + 36,-10 ] and traced by its radio continuum jet , but is not detected at 11.7 .
the feature identified as w3(h@xmath0o ) by @xcite is presumably the weak cometary northeast of w3(oh ) .
the radio position of the w3(oh ) served as as astrometric reference .
the comparison of the spatial extent of w3(oh ) with the standard star @xmath7 tau ( inset of fig .
1 ) shows that it is clearly resolved .
+ the temperature and size derived from molecular line interferometry for w3(h@xmath0o ) suggest that it should be a bright object in the mir , with a predicted 11.7 flux density of @xmath11jy . our failure to detect it despite a much better sensitivity ( 3@xmath12 point source detection limit of 6mjy ) than @xcite implies a large amount of cold dust in front of the hot core . a detailed analysis of the seds of w3(h@xmath0o)/w3(oh )
is given in @xcite .
[ eps2 ] the object g9.62 + 0.19 ( iras18032 - 2032 ) comprises several s at a distance of 5.7kpc .
recently , weak radio continuum emission has been detected @xcite at the location of the hc ( component f ) as well as outflow activity @xcite .
remarkably , this object seems to be associated with 2.2 emission @xcite , a fact which can be hardly reconciled with spherical models of hcs .
such models imply extinction values which should prevent the detection of the hc at near - infrared ( nir ) wavelengths .
2 shows the 11.7 sc10 image of g9.62 + 0.19 with contours of the 1.3 cm radio continuum @xcite .
the bulk of the 11.7 emission stems from the extended radio component b which was used as astrometric reference .
surprisingly , no thermal emission was seen from the d which again implies a considerable optical depth even at mir wavelengths for this source .
this result concerning component d contradicts that of @xcite due to differing astrometry .
a pointlike source is obvious close to the location of the hc ( offset position [ -20,-10 ] ) .
this emission is almost coincident with the 2.2 feature .
the poster contribution of linz et al .
( this volume ) discusses this source and the relation of the observed ir emission to the recently detected molecular outflow @xcite in more detail . possibly , ir radiation arising from the hc escapes in the outflow lobe inclined towards the observer where the optical depth is lower . line ( black , from @xcite ) . ] according to the radial brightness temperature profile based on interferometric maps of the nh@xmath13(4,4 ) line , the hc g10.47 + 0.03 ( iras18056 - 1952 ) seemed to be one of the rare cases with _ established _ evidence for internal heating by embedded ob stars contrary to external heating by adjacent s @xcite .
two stars from the corresponding 2mass images of the region were also detected with sc10 at 11.7 .
they provide astrometric reference with sub - arcsecond precision .
mir emission from two sites in the g10.47 + 0.03 region was detected .
these sources are not seen on the 2.2 2mass image ( 10@xmath12 detection limit of 1.2mjy ) .
[ eps3 ] displays the sc10 11.7 image with contours of the 6 cm radio continuum ( grey , from @xcite ) and of the nh@xmath14 emission ( black , from @xcite ) .
the hc is located at the offset position [ 00,16 ] . to the northwest of the hc
, the nh@xmath13 emission is diminished due to absorption by the s in the foreground .
the centroid of the 11.7 radiation almost coincides with the center of the nh@xmath13 emission . a possible configuration which explains this morphology might be a cometary , similar to g29.960.02 and its adjacent hc ( which are seen side - on ) , with the line of sight slightly inclined with respect to the symmetry axes of the hii region .
the fact that there is only one mir peak associated with the hc suggests that the multiple structure in the radio continuum ( 3 components are present in the map of @xcite ) is presumably due to locally enhanced plasma density and not caused by a few widely distributed high - mass stars .
the lack of nh@xmath13 emission for the second mir source ( offset position [ + 15,+30 ] ) indicates that this is presumably a more evolved .
the g232.620 + 0.996 ( iras07299 - 1651 ) is associated with oh and ch@xmath13oh masers @xcite,@xcite which are signs of newly formed massive stars . the is 15 northwest of the maser position @xcite
. these objects are located at the southern border of a dense core seen in the 1.3 mm map taken with the sest @xcite .
our 2.2 polarimetric map obtained with sofi at the eso - ntt revealed that the bipolar - like extended emission ( see fig .
4 ) at the southern rim of the dense core is dominated by scattered light .
the location of the illuminating source was derived by minimizing the sum of the scalar products between polarisation and radius vector as a function of source position .
it is marked in fig . 4 by the black asterisk , and situated very close to the masers and the .
the large fov of timmi2 covered the target and hd60068 ( situated 47 northwest of it ) simultaneously which allowed to establish accurate astrometry .
a resolved ir source was found close to the maser position .
4 also displays contours of the 4.7 and 11.9 image taken with timmi2 .
it can be noticed that the emission peaks of both wavelengths are spatially offset , indicating a strong gradient in the optical depth in the immediate neighbourhood of the illuminating source .
this work was supported by dfg grants ste 605/17 - 1 and ste 605/18 - 1 .
j. alcolea , k.m .
menten , j.m .
moran , m.j .
reid : in : _ astrophysical masers _
, ed . by a.w .
clegg & g.e .
nedoluha ( berlin : springer ) , 225 ( 1993 ) j.l .
caswell : mnras * 297 * , 215 ( 1998 ) r. cesaroni , e. churchwell , p. hofner , c.m .
walmsley : a&a * 288 * , 903 ( 1994 ) r. cesaroni , p. hofner , c.m .
walmsley , e. churchwell : a&a * 331 * , 709 ( 1998 ) j.m .
debuizer , r.k .
pia , c.m .
telesco : apjs * 130 * , 437 ( 2000 ) t.l .
hayward , j.w .
miles , j.r .
houck , g.e .
gull , j. schoenwald j. : in : _ infrared detectors and instrumentation _ , ed . by a.w .
fowler , proc .
spie , 1946 , 334 ( 1993 ) p. hofner , h. wiesemeyer , th .
henning : apj * 549 * , 425 ( 2001 ) e. keto , d. proctor , r. ball , j. arens , g. jernigan : apj * 401 * , l113 ( 1992 ) m.j .
reid , a.l .
argon , c.r .
masson , k.m .
menten , j.m .
moran : apj * 443 * , 238 ( 1995 ) h.g .
reimann , h. linz , r. wagner , h. relke , h.u .
kufl , e. dietzsch , m. sperl , j. hron : in : _ optical and ir telescope instrumentation and detectors _ , ed . by m. iye & a.f .
moorwood , proc .
spie * 4008 * , 1132 ( 2000 ) b. stecklum , b. brandl , t.l .
hayward , th .
henning , i. pascucci , j. wilson : a&a submitted , ( 2001 ) l. testi , m. felli , p. persi , m. roth : a&a * 329 * , 233 ( 1998 ) l. testi , p. hofner , s. kurtz , m. rupen : a&a * 359 * , l5 ( 2000 ) a.j .
walsh , m.g .
burton , a.r .
hyland , g. robinson : mnras * 301 * , 640 ( 1998 ) d.j .
wilner d.j .
welch , j.r .
forster : apj * 449 * , l73 ( 1995 ) d.o.s .
wood , e. churchwell : apjs * 69 * , 831 ( 1989 ) f. wyrowski , p. schilke , c.m .
walmsley , k.m .
menten : apj * 514 * , l43 ( 1999 ) | we report on mid - infrared imaging of hot cores performed with spectrocam10 and timmi2 .
the observations aimed at the detection of thermal emission presumably associated with the hot cores .
mid - infrared flux measurements are required to improve the luminosity and optical depth estimates for these sources .
results are presented for w3(h@xmath0o ) , g9.62 + 0.19 , g10.47 + 0.03 , and the possible hot core candidate g232.620 + 0.996 .
they illustrate that the morphology of these sources can not be described by simple geometries .
therefore , line - of - sight effects and considerable extinction even at mid - infrared wavelengths must not be neglected . |
the response of nanotubes to external electric fields is of interest both for transport devices@xcite and for nano electromechanical systems @xcite . when the transverse field strength is large enough to couple neighboring subbands , it is not appropriate to use the conventional 1d approximation , which assumes a uniform charge and potential distribution along the nt circumference . in this article
, we address the issue of how the nt material properties are modified when the electronic potential along the circumference is no longer uniform .
we use a single @xmath0 orbital tight - binding model in a self - consistent way to account for screening effect caused by the non - uniform charge distribution along the nt circumference . to our knowledge , this is the first study that considers a wide range of applied transverse electric field conditions . within the limit of relatively weak electric fields ,
we obtain different behaviors for semiconducting and metallic nanotubes , either intrinsic or doped .
the weak field condition can be expressed as @xmath1(@xmath2 for @xmath3 ) , where @xmath4 and @xmath5 are the hopping integral and lattice constant of two - dimensional graphite respectively , as defined for example in the book @xcite .
one may safely use a rigid bond approximation in this field limit , without severely perturbing the original bandstructure .
we also investigate the regime of relatively strong fields , @xmath6 , where bandstructure modifications need to be considered and subband mixing can not be neglected , while a tb approach can still be used . also for this regime , qualitatively different behaviors of semiconducting and metallic nts are found .
the paper is organized as follows . in sec.[sec : polarizability ] we calculate the transverse polarizability for nanotubes of various symmetry , considering a linear approximation and no phonon contributions .
the dependence of the polarizability on tube radius , symmetry and fermi level position are discussed .
the role of the fermi level is not related to intra - band free carrier transitions , but rather to inter - band transitions between neighboring conduction / valence bands .
we then exam the effect of a uniform transverse transverse field on the bandstructure of nanotubes in sec.[sec : bandstructure ] .
because of the symmetry of the nanotube , the subbands of the tube are mixed according to certain selection rules and the bandstructure and bandgap are correspondingly modified .
several interesting phenomena are predicted to happen in the transverse field : bandgap opening / closing , energy subband flattening and over - bending , lifting of subband degeneracy , and generation of multiple valleys .
these effects can be used to modulate the bandgap , effective mass and carrier densities of nts and to enhance the density of states(dos ) at the fermi level .
for example , it has been suggested that opening of energy gaps of metalic nanotubes under a local transverse electric field can be used for designing quantum switches@xcite .
our calculation shows that there exists a critical magnitude of the field strength beyond which the gap starts to decrease and oscillates further on . for simplicity
, only zigzag ( both metallic and semiconducting ) and armchair tubes are considered in the full - band tb calculations . the extension of the numerical results to chiral tubes is confirmed by the analytical results within the @xmath7 method .
an external electrical field creates an induced dipole moment @xmath8 on the nanotube , which can be estimated through the calculation of the polarizability of @xmath0 electrons @xcite .
the unscreened linear polarizability @xmath9 relates the dipole moment @xmath10 to the total electrical field @xmath11 with @xmath12 , which accounts for the contribution from the single particle excitation .
the dielectric function , defined as @xmath13 , and the actual ( or screened ) polarizability @xmath14 can be easily retrieved from @xmath15 as @xmath16 as discussed above , when the field strength is less than @xmath17 , one can use wavefunctions and energies of the electron states obtained without external perturbation .
results of the tb calculation in the static limit are shown in figure [ fig : a0 ] , where metallic and semiconducting tubes are compared .
the unscreened polarizability @xmath15 is quadratically proportional to @xmath18 and is dependent on the conducting properties ( symmetry of the tube ) . as the radius of nts increases ,
@xmath15 can be fitted by an universal expression : @xmath19 the corresponding dielectric function @xmath20 is @xmath21 which agrees very well with previous results @xcite . for comparison
, we have calculated the transverse polarizability using the @xmath22 scheme @xcite .
this approximation is equivalent to the linerization of the electron energy dispersion in the vicinity of the fermi points @xmath23 .
the cutoff of @xmath24 is used for the calculation and only the lowest subbands are included . with these approximations ,
the analytical expression for the polarizability is @xmath25 here , @xmath26 is the fermi velocity , with @xmath27 taken to be unity
. we also have for the integration constant , @xmath28 . the value of @xmath29 , calculated with the @xmath22 scheme in eq .
( [ a0_kp ] ) , is smaller than the numerical result obtained from by the full - band calculation , eq .
( [ a0_fullband ] ) , but it provides a more intuitive understanding of the physical factors which affect the polarizability .
the independence of the dielectric function on the geometry ( nt radius ) holds only as long as one can verify the condition that the nt bands are half - filled .
however , we notice that some nanotubes are naturally doped during the growth process , in which case the fermi level @xmath30 would shift away from the charge neutrality level , and similarly under conditions of charge injection or application of an external bias to the nanotube . in these cases ,
the polarizability depends on the occupation or depletion of new states .
let s assume @xmath31 ( note that the sign of @xmath30 does not affect the results ) and define the unscreened polarizability for a metallic tube as : @xmath32 where @xmath33 accounts for the contribution due to the occupation of @xmath34th conduction subband .
at low fermi energy , the @xmath22 approximation gives : @xmath35\left(e_f\sqrt{e_f^2-e_{b2}^2 } -e_{b2}^2\mathrm{arcsinh}\left[\frac{\sqrt{e_f^2-e_{b2}^2}}{e_{b2}}\right]\right)\end{aligned}\ ] ] here , @xmath36 is the unit step function ( @xmath37=1 $ ] when @xmath38 and @xmath39 otherwise ) , and @xmath40 is the energy of the bottom of the @xmath41 subband .
if one considers the low energy properties of nanotubes , the @xmath22 method provides a reasonable approximation .
when the fermi level is within the first subband , the unscreened polarizability and the dielectric function for a metallic tube of radius r can be written as : @xmath42 @xmath43 where the prefactor @xmath44 is the dimensionless density of states of the first subband .
as shown above , the shift of fermi level away from the charge neutrality level changes considerably the polarizability and , consequently , the nanotube dielectric function .
the dielectric screening increases with the radius of the tube , according to eq .
( [ epsilon_kp_ef_met ] ) , in contrast to the screening of a neutral tube , which is independent of the radius .
this is because of the strong dependence on @xmath45 found in the transition energy of the main terms in the expressions for @xmath20 , @xmath46 .
one must conclude that the dielectric properties of a nanotube depend not only on radius and type of bandstructure ( metal or semiconductor ) , but also on the charge carrier density in the tube @xcite .
this charge density varies with applied field if the tube is connected to electron reservoirs and charge flow through the tube is allowed .
thus , the calculation of the polarizability of the nanotube in a device environment is a complicated problem which has to be solved self - consistently .
when the applied field is strong enough to mix neighboring subbands , that is @xmath47 , the bandstructure of a nanotube is considerably modified .
we have calculated the energy bands of a [ 10,10 ] armchair tube in electrical fields of different strength .
since we are interested in the low energy region of nanotubes , only subbands in the energy range @xmath48 are shown in figure [ fig : bs_dos_a10 ] .
we emphasize that the two lowest subbands always cross , even at a very large field , although the bandstructure has been noticeably modified @xcite . at @xmath49 ( see figure [ fig : bs_dos_a10](b ) ) , the fermi points shift toward the @xmath50 point ( @xmath51 ) and the two lowest subbands are flattened near the fermi points . at the same time , all states which were degenerate with respect to the magnetic number , @xmath52 , split .
the splitting becomes more obvious closer to the lowest subbands .
the large degeneracy at the first brillouin zone ( fbz ) boundary @xmath53 is also lifted and bending is observed for all subbands at this point . as the field strength increases
, the two lowest subbands show oscillatory bends with multiple nodes generated , while the first node moves even closer to @xmath51 ( figure [ fig : bs_dos_a10](c ) ) . for other subbands ,
the splitting of @xmath52 subbands become more significant .
this bandstructure modification is clearly seen in the density of states of the nanotube , as shown in figure [ fig : bs_dos_a10](d f ) . as the field
is applied , the low energy plateau displays a bump which increases with field .
the enhanced dos near @xmath54 is due to the flattening and bending of the two lowest subbands . on the other hand , the lifting of @xmath52 degeneracy of all doublets and the bending at the fbz boundary split the single vhs peaks into multiple ones . although the dos structure ( figure [ fig : bs_dos_a10 ] ) changes considerably as compared to the case @xmath55 , all dos features may be attributed to specific symmetry of the states .
we stress that many experimental techniques , ranging from raman scattering to scanning tunneling spectroscopy use high electric fields to probe the electronic properties of a nanotube , which may perturb the underlying electronic structure .
our theoretical results may help to understand disagreements between experimental measurements and predictions for dos , effective masses and the locations of vhs peaks . for example
, we obtained different shifts of optical transition energy @xmath56 and @xmath57 . for quasi - metallic zigzag tubes ,
the physics is quite different . at weak fields ,
a bandgap opens at the fermi point @xmath51 .
when the total field is smaller than a critical field @xmath58 , the gap is quadratically proportional to the product of the field strength @xmath59 and the radius @xmath45 , as shown in figure [ fig : bgc_zmet_r ] . the analytical expression for the gap can be obtained within a degenerate perturbation theory : @xmath60 when the field increases beyond @xmath58 , the minimum gap shifts away from the original fermi point @xmath51 and the gap starts to decrease .
the value of the critical field depends on the nanotube radius ( see figure [ fig : bgc_zmet_r ] inset ) and is fitted to be : @xmath61 thus , the degenerate perturbation theory which is used in deriving eq .
( [ perturbation ] ) is no longer valid , when the external potential exceeds the energy distance between neighboring subbands , @xmath62 .
it may be interesting for electronics applications to be able to modulate locally the gap of the nanotube .
our study shows that for metallic zigzag nt one can , indeed , open the gap .
however the gap can not exceed some critical value beyond which a further increase of the field begins to close the gap . from eqs .
( [ perturbation ] ) and ( [ fit : ec ] ) , the critical bandgap for a metallic zigzag tube @xmath63 $ ] is approximately : @xmath64 which is almost negligible for nanotubes with large radius ( @xmath65 ) .
although larger radius nts may seem more attractive due to the smaller critical field is smaller , in these structures the maximum gap will also be smaller .
on the other hand , we notice that for very narrow nts , the @xmath66 mixing may results in the opening of secondary gaps , which may prohibit using very narrow nts for band modulations . the transverse field effect discussed above , including gap opening or preserved subband crossing as well as degeneracy lifting ,
can be explained by using methods of group theory . for a detailed description of a group theory technique for nanotubes ,
we refer the reader to @xcite . for armchair and zigzag tubes , the full set of quantum numbers includes the longitudinal momentum @xmath67 , the angular momentum @xmath68 , and parities with respect to the vertical mirror reflection @xmath69 and the horizontal mirror reflection @xmath70 @xcite . of the @xmath71 bands ( for a @xmath72 $ ] or @xmath73 $ ] tube ) , only those with @xmath74 have definite even or odd parity about @xmath69 and they are non - degenerate ( spin degeneracy is not relevant here and is not discussed ) .
all the other bands are doubly degenerate at zero field with respect to @xmath75 and @xmath76 indices .
we assume that the components of the electron state on different sub - lattices are not mixed when the magnitude of the field satisfies @xmath77 , where @xmath78 is the nearest neighbor distance in the cnt .
thus , a uniform transverse electric field possesses an odd parity with respect to a vertical mirror plane of the nanotube . due to the commutation of a perturbed hamiltonian @xmath79 and the vertical parity operator @xmath69 , parities with respect to @xmath69
remain good quantum numbers . for an armchair tube ,
the two crossing subbands ( @xmath80 ) have opposite @xmath69 parities . in the presence of a transverse field ,
the degeneracy of these two subband at @xmath81 is lifted in the second order of perturbation theory , but since they have different @xmath69 parities , the crossing at the new fermi points is not prohibited .
in contrast , for metallic zigzag tubes , electron states in the four lowest subbands @xmath82 do not have definite @xmath69 parities and can be mixed by the external field in a high order perturbation theory by coupling to the states with @xmath83 .
two pairs of new states with opposite parities with respect to @xmath69 are the result .
according to the anti crossing rule for nts subbands noticed in @xcite , the degeneracy at @xmath84 will be lifted .
similarly , the splitting of the degenerate subbands @xmath52 is explained by their mixing to the other subbands by the transverse field .
because of the selection rule @xmath85 , the two degenerate states @xmath86 and @xmath87 can only be mixed and split at a high order of perturbation theory .
thus , the coupling strength between @xmath88 states depends on min@xmath89 $ ] , which is a `` distance '' between these two states and the non - degenerate states @xmath90 and @xmath91 .
this explains why , at low energy , the splitting is more prominent for subbands in armchair tubes , i.e. near @xmath92 and almost indiscernible for subbands in zigzag tubes , i.e. away from @xmath74 . in summary
, we studied for the first time the effect of doping and/or charge injection on the transverse polarizability of nanotubes .
we found that for metallic tubes , the polarizability grows quadratically with @xmath30 and scales as @xmath93 at low fermi levels , leading to an enhancement of the dielectric function for the doped nanotubes . with an increase of the the applied field ,
the bandstructure is considerably modified due to the lowering of symmetry .
the zero - gap structure of armchair tubes is always preserved while gap opening and closing occur in metallic and semiconducting zigzag tubes respectively .
degeneracy lifting of @xmath52 subbands and the flattening of lowest subbands are predicted , which changes the shape and magnitude of the peaks in the density of states.thus we conclude that formation of multiple valleys in the bandstructure , enhancement of dos at the fermi level and engineering of subbands with required effective mass of charge carriers are possible with a transverse electric field .
we also predict a maximum bandgap which opens in a metalic tube and give an optimal range of nanotube radius for bandgap engineering .
this work was partial supported by the national science foundation , grant itr / sy 0121616 .
one of us ( svr ) acknowledges support through a cri grant of uiuc , doe grant de - fg02 - 01er45932 , nsf grant ecs-0210495 and beckman fellowship from the arnold and mabel beckman foundation 9 j. appenzeller , j. knoch , v. derycke , et al , phys.rev.lett . *
89 * , 126801 ( 2002 ) p. kim and c.m .
lieber , science * 286 * , 2148 ( 1999 ) ; s. akita , y. nakayama , s. mizooka , et al .
lett . * 79 * , 1691 ( 2001 ) .
m. dequesnes , s.v .
rotkin and n.r .
aluru , nanotechnology * 13 * , no 1 , 120 ( 2002 ) .
r. saito , g. dresselhaus , m.s .
dresselhaus , _ physical properties of carbon nanotubes _ , imperial college press ( 1998 ) yh.kim and k.j.chang , phys.rev.b * 64 * , 153404 ( 2001 ) l.x .
benedict , s.g .
louie , and m.l .
cohen , phys.rev.b .
* 52 * , 8541 ( 1995 ) .
novikov and l.s .
levitov , cond - mat/0204499 ( 2002 ) t. ando , t. nakanishi , and r. saito , j.phys.soc.jpn . * 67 * , 2857 ( 1998 ) ; p.l .
mceuen et al . , phys.rev.lett . , * 83 * , 5098 ( 1999 ) .
our calculation shows that the temperature contribution adds to the dielectric function in eq .
( [ epsilon_kp_ef_met ] ) as @xmath94 .
this correction is smaller than @xmath95% at room temperature for tubes with a moderate radius ( @xmath96 ) .
this phenomenon is in contradiction to the results of x. zhou , h. chen , zc .
ou - yang , j.phys.cond.mat .
* 13 * , l635 ( 2001 ) .
they found that for a [ 10,10 ] nanotube in a transverse field , a gap will open and vary quadratically with the strength of the field . | the electronic properties of carbon nanotubes ( nts ) in a uniform transverse field are investigated within a single orbital tight - binding ( tb ) model . for doped nanotubes ,
the dielectric function is found to depend not only on symmetry of the tube , but also on radius and fermi level position .
bandgap opening / closing is predicted for zigzag tubes , while it is found that armchair tubes always remain metallic , which is explained by the symmetry in their configuration .
the bandstructures for both types are considerably modified when the field strength is large enough to mix neighboring subbands .
* electronic response and bandstructure modulation of carbon nanotubes in a transverse electrical field * + + * keyword : * theoretical , nanotubes , polarizability , dielectric function , electronic bandstructure , bandgap engineering |
recently the possibility of extracting information on the symmetry energy and the isospin ( neutron - to - proton ratio ) of the fragments in a multifragmentation reaction has gained tremendous importance @xcite .
such information is of importance for understanding some of the key problems in astrophysics @xcite , and various aspects of nuclear physics such as the structure of exotic nuclei ( the binding energy and rms radii ) @xcite and the dynamics of heavy ion collisions @xcite .
traditionally , the symmetry energy of nuclei has been extracted by fitting the binding energy in their ground state with various versions of the liquid drop mass formula @xcite .
the properties of nuclear matter are then determined by theoretically extrapolating the nuclear models designed to study the structure of real nuclei
. however , real nuclei are cold , nearly symmetric ( @xmath3 ) and found at equilibrium density .
it is not known how the symmetry energy behaves at temperatures , isospin ( neutron - to - proton ratio ) and densities away from the normal nuclear matter .
theoretical many - body calculations @xcite and those from the empirical liquid drop mass formula @xcite predict symmetry energy near normal nuclear density ( @xmath4 0.17 @xmath5 ) and temperature ( @xmath6 @xmath4 0 mev ) , to be around 28 - 32 mev . in a multifragmentation reaction
, an excited nucleus expands to a sub - nuclear density and disintegrates into various light and heavy fragments @xcite .
the fragments are highly excited and neutron - rich ; their yields depend on the available free energy which in turn depends on the strength of the symmetry energy and the extent to which the fragments expand . by studying the isotopic yield distribution of these fragments
, one can extract important information about the symmetry energy and the properties of the fragments at densities , excitation energies and isospin away from those of ground state nuclei .
experimentally , the determination of fragment isotopic yield distribution is not straight forward .
it is influenced by the complex de - excitation of the hot ( primary ) fragments into observed cold ( secondary ) fragments .
theoretical calculations of the secondary de - excitations require accurate accounting of the feeding from the particle unstable states and are subject to uncertainties in the levels that can be excited , and the structure effects that govern their decay @xcite . from various statistical model approaches @xcite
, it has been shown that the ratio of primary fragment yield for a given isotope or isotone produced in two different reactions with similar temperature , exhibit an exponential dependence on proton and neutron number , an observation known as isoscaling .
the dependence has been interpreted in terms of a scaling parameter that is related to the symmetry energy of the primary fragment binding energy .
the scaling parameter has been shown to be independent of the complex nature of the secondary de - excitation and is thus a robust observable for studying the fragment isotopic yield distribution . in a recent work @xcite
, it was shown that the symmetry energy of the primary fragments deduced from the reduced neutron density is significantly lower than that for the normal nuclei at saturation density .
le fevre _ et al . _
@xcite , in their recent work studied the fragmentation of excited target residues following collisions of @xmath7c on @xmath8sn , at incident energies of 300 and 600 mev / nucleon .
they observed that the symmetry energy co - efficient deduced from the data are near 25 mev for peripheral collisions and lower than 15 mev for the central collisions .
@xcite , studied the fragments produced in the multifragmentation of @xmath9xe projectiles in mid - peripheral collisions with a lead target at 1 gev / nucleon .
they used both , the @xmath10 ratio and the isoscaling of the fragments and found that the experimentally determined value of the scaling parameter can be reproduced within the statistical model framework by lowering the symmetry energy to as low as 11 - 12 mev .
the @xmath10 ratios of the fragments on the other hand , could be reproduced with symmetry energy co - efficient as low as 14 - 15 mev . in this work ,
we study the primary fragment yield distribution in a number of reactions at various excitation energies and isospins using the isoscaling approach and the equilibrium statistical multifragmentation model .
it is observed that the isoscaling parameter @xmath2 for the hot fragments decreases with increasing excitation energy and decreasing symmetry energy
. the alpha values also increase with increasing difference in the isospin ( neutron - to - proton asymmetry ) of the fragmenting system .
a similar behavior is also observed for the cold secondary fragments .
the secondary de - excitation is found to have very little influence on the isoscaling parameter at lower excitation energies and isospin of the fragmenting system .
the symmetry energy however , strongly influence the properties of the hot fragments .
the experimentally determined isoscaling parameters can be explained by symmetry energy that is significantly lower than that for the normal ( cold ) nuclei at saturation density , indicating that the properties of nuclei at high excitation energy , isospin and reduced density are sensitive to the symmetry energy .
the paper is organized as follows . in sec .
ii , we describe the experiment in detail .
the experimental results are presented and discussed in sec .
section iv contains a brief description of the statistical multifragmentation model used in the analysis .
section v contains a comparison between the experimental data and the statistical multifragmentation model .
finally , a summary and conclusion is given in section vi .
the experiments were carried out at the cyclotron institute of texas a@xmath11 m university ( tamu ) using the k500 superconducting cyclotron where targets of @xmath1fe ( 2.3 mg/@xmath12 ) and @xmath1ni ( 1.75 mg/@xmath12 ) were bombarded with beams of @xmath0ar and @xmath0ca at 25 , 33 , 45 and 53 mev / nucleon @xcite .
the various combinations of target and projectile nuclei allowed for a range of @xmath13 ( neutron - to - proton ratio ) ( 1.04 - 1.23 ) of the system to be studied , while keeping the total mass constant ( @xmath14 = 98 ) .
the beams were fully stripped by allowing them to pass through a thin aluminum foil before being hit at the center of the target inside the tamu 4@xmath15 neutron ball @xcite .
light charged particles ( @xmath16 @xmath17 2 ) and intermediate mass fragments ( @xmath16 @xmath18 2 ) were detected using six discrete telescopes placed inside the scattering chamber of the neutron ball at angles of 10@xmath19 , 44@xmath19 , 72@xmath19 , 100@xmath19 , 128@xmath19 and 148@xmath19 . a schematic diagram of the setup showing placement of the telescopes within the scattering chamber of the neutron ball is shown in figure .
each telescope consisted of a gas ionization chamber ( ic ) followed by a pair of silicon detectors ( si - si ) and a csi scintillator detector , providing three distinct detector pairs ( ic - si , si - si , and si - csi ) for fragment identification .
the ionization chamber was of axial field design and was operated with cf@xmath20 gas at a pressure of 50 torr .
the gaseous medium was 6 cm thick with a typical threshold of @xmath21 0.5 mev / nucleon for intermediate mass fragments .
the silicon detectors had an active area of 5 cm @xmath22 5 cm and were each subdivided into four quadrants .
the first and second silicon detectors in the stack were 0.14 mm and 1 mm thick , respectively .
the dynamical energy range of the silicon pair was @xmath21 16 - 50 mev for @xmath23he and @xmath21 90 - 270 mev for @xmath7c .
the csi scintillator crystals that followed the silicon detector pair were 2.54 cm in thickness and were read out by photodiodes .
good elemental ( @xmath16 ) identification was achieved for fragments that punched through the ic detector and stopped in the first silicon detector .
fragments measured in the si - si detector pair also had good isotopic separation .
the trigger for the data acquisition was generated by requiring a valid hit in one of the silicon detectors .
the calibration of the ic - si detectors were carried out using the standard alpha sources and by operating the ic at various gas pressures .
the si - si detectors were calibrated by measuring the energy deposited by the alpha particles in the thin silicon and the punch - through energies of different isotopes in the thick silicon .
the si - csi detectors were calibrated by selecting points along the different light charged isotopes and determining the energy deposited in the csi crystal from the energy loss in the calibrated si detector .
neutrons were measured with the 4@xmath15 neutron ball that surrounded the detector assembly .
the neutron ball consisted of eleven scintillator tanks segmented in its median plane and surrounding the vacuum chamber .
the upper and the lower tank were 1.5 m diameter hemispheres .
nine wedge - shaped detectors were sandwiched between the hemispheres .
all the wedges subtended 40@xmath19 in the horizontal plane .
the neutron ball were filled with a pseudocumene based liquid scintillator mixed with 0.3 weight percent of gd salt ( gd 2-ethyl hexanoate ) .
scintillations from thermal neutrons captured by gd were detected by twenty 5-in phototubes : five in each hemisphere , one on each of the identical 40@xmath19 wedges and two on the forward edges .
the efficiency with which the neutrons could be detected is about 83@xmath24 , as measured with a @xmath25cf source .
the detected neutrons were used to differentiate between the central and peripheral collisions . to understand the effectiveness of neutron multiplicity as a centrality trigger ,
simulations were carried out using a hybrid buu - gemini calculations at various impact parameters for the @xmath0ca + @xmath1fe reaction at 33 mev / nucleon .
the simulated neutron multiplicity distribution was compared with the experimentally measured distribution . the multiplicity of neutron for the impact parameter _
b _ = 0 collisions was found to be higher than the _ b _ = 5 collision . by gating on the 10@xmath24 highest neutron multiplicity events
, one could clearly discriminate against the peripheral events . to determine the contributions from noncentral impact parameter collisions , neutrons emitted in coincidence with fragments at 44@xmath19 and 152@xmath19 were calculated at _
b _ = 0 fm and _ b _ = 5 fm .
the number of events were adjusted for geometrical cross sectional differences .
a ratio was made between the number of events with a neutron multiplicity of at least six , calculated at _
b _ = 0 fm , and the number of events with the same neutron multiplicity at _ b _ = 5 fm .
the ratios were observed to ca + @xmath1ni ( stars and solid lines ) , @xmath0ar + @xmath1ni ( circles and dashed lines ) and @xmath0ar + @xmath1fe ( squares and dotted lines ) reactions at various beam energies.,scaledwidth=50.0% ] be 19.0 and 11.1 at 44@xmath19 and 1.3 and 2.2 at 152@xmath19 for 33 and 45 mev / nucleon respectively . at intermediate angles , high neutron multiplicities were observed to be outside the region in which _ b _ = 5 fm contributes significantly . at backward angles the collisions at
_ b _ = 5 fm made a larger contribution to the neutron multiplicity .
in addition to the neutron multiplicity distribution , the charge distribution of the fragments was also used to investigate the contributions from central and mid - impact parameter collisions .
the _ b _ = 5 collisions produced essentially no fragments with charge greater than three in the 44@xmath19 telescope . in an earlier work @xcite
, some analysis of the fragment kinetic energy and charge distributions were presented .
it was shown that at a laboratory angle of 44@xmath19 the kinetic energy and the charge distributions are well reproduced by the statistical model calculation .
using a moving source analysis of the fragment energy spectra , it was also shown that the fragments emitted at backward angles originate from a target - like source , while those emitted at 44@xmath19 originate primarily from a composite source . in this work
, we will concentrate exclusively on data from the laboratory angle of 44@xmath19 to study the symmetry energy and the isoscaling properties of the fragments produced .
the choice of this angle enables one to select events which are predominantly central and undergo bulk multifragmentation .
the contributions to the intermediate mass fragments from the projectile - like and target - like sources can therefore be assumed to be minimal and the use of equilibrium statistical model appropriately justified .
the experimentally measured relative isotopic yield distributions for the lithium ( left ) , berillium ( center ) and carbon ( right ) elements , in @xmath0ca + @xmath1ni ( star symbols ) , @xmath0ar + @xmath1ni ( circle symbols ) and @xmath0ar + @xmath1fe ( square symbols ) reactions , are shown in figure 2 for the beam energies of 25 , 33 and 45 mev / nucleon .
the distributions for each element show higher fragment yield for the neutron rich isotopes in @xmath0ar + @xmath1fe reaction ( squares ) , which has the largest neutron - to - proton ratio ( @xmath13 ) , in comparison to the @xmath0ca + @xmath1ni reaction ( stars ) , which has the smallest neutron - to - proton ratio . the yields for the reaction , @xmath0ar + @xmath1fe ( circles ) , which has an intermediate value of the neutron - to - proton ratio , are in between those of the other two reactions .
the figure thus shows the isospin dependence of the composite system on the properties of the fragments produced in the multifragmentation reaction .
one also observes that the relative difference in the yield distribution between the three reactions decreases with increasing beam energy .
this is due to the secondary de - excitation of the primary fragments , a process that becomes important for systems with increasing neutron - to - proton ratio and excitation energy . in the following subsections
, we will utilize the experimentally determined isotopic yield distributions to establish the isoscaling properties of the produced fragments before comparing them with the statistical multifragmentation model in section v. ca + @xmath1ni reaction , square symbols to @xmath0ar + @xmath1ni reaction , and triangle symbols to @xmath0ar + @xmath1fe reaction .
the solid lines are the best fit to the data.,scaledwidth=47.0% ] , for various beam energies .
the left column correspond to @xmath0ar + @xmath1fe and @xmath0ca + @xmath1ni pair of reactions .
the right column correspond to @xmath0ar + @xmath1ni and @xmath0ca + @xmath1ni pair of reactions .
the different symbols correspond to @xmath16 = 3 ( circles ) , @xmath16 = 4 ( open stars ) , @xmath16 = 5 ( triangles ) , @xmath16 = 6 ( squares ) and @xmath16 = 7 ( filled stars ) elements .
the lines are the exponential fits to the data as explained in the text.,scaledwidth=47.0% ] , for various beam energies .
the left column correspond to @xmath0ar + @xmath1fe and @xmath0ca + @xmath1ni pair of reaction .
the right column correspond to @xmath0ar + @xmath1ni and @xmath0ca + @xmath1ni pair of reaction .
the different symbols correspond to @xmath26 = 3 ( circles ) , @xmath26 = 5 ( triangles ) , @xmath26 = 6 ( squares ) and @xmath26 = 7 ( stars ) elements .
the lines are the exponential fits to the data as explained in the text .
, scaledwidth=47.0% ] ( solid symbols ) and @xmath27 ( open symbols ) as a function of the beam energy .
the solid circles and open squares correspond to @xmath0ar + @xmath1fe and @xmath0ca + @xmath1ni pair of reactions .
the solid stars and open triangles correspond to @xmath0ar + @xmath1ni and @xmath0ca + @xmath1ni pair of reactions .
the lines are the exponential fits to the data .
the error bars are of the size of the symbols.,scaledwidth=49.0% ] , as a function of the neutron number @xmath26 , for the 25 and 45 mev / nucleon beam energies .
the top panel is for the @xmath0ar + @xmath1fe/@xmath0ca + @xmath1ni pair , and the bottom panel is for the @xmath0ar + @xmath1ni/@xmath0ca + @xmath1ni pair of reactions .
the symbols correspond to @xmath28 obtained from various elements ( @xmath16 ) .
the lines are the best fits to the data.,scaledwidth=46.0% ] of the systems for the beam energies of 25 , 33 and 45 mev / nucleon.,scaledwidth=47.0% ] in a multifragmentation reactions , the ratio of isotope yields in two different systems , 1 and 2 , @xmath29 , has been shown to obey an exponential dependence on the neutron number ( @xmath26 ) and the proton number ( @xmath16 ) of the isotopes , an observation known as isoscaling @xcite .
the dependence is characterized by a simple relation , @xmath30 where , @xmath31 and @xmath32 are the yields from the neutron - rich and neutron - deficient systems , respectively .
@xmath33 , is an overall normalization factor and @xmath2 and @xmath27 are the parameters characterizing the isoscaling behavior .
the necessary condition for observing isoscaling in multifragmentation reaction is the near equality of temperature for systems chosen in the above scaling relation @xcite . for the present work
, this condition was tested by determining the temperature using the double isotope ratio method of albergo _ et al .
the albergo method relates the apparent temperature @xmath6 of the system at the freeze - out to the double isotope ratio , @xmath34/[y(n , z_{2})/y(n + 1 , z_{2})]$ ] , through a relation @xmath35 where @xmath36 is a factor that depends on the statistical weights of the ground state nuclear spins , and @xmath37 - [ b(n , z_{2 } ) - b(n + 1 , z_{2})]$ ] is the difference in the binding energy .
the method was applied to all the three systems and beam energies studied .
figure 3 shows the double isotope ratios obtained using the various combinations of isotopes as a function of the difference in the binding energy .
the top panel correspond to those determined for the beam energies of 25 mev / nucleon , the center for the 45 mev / nucleon and the bottom for the 53 mev / nucleon .
the different symbols correspond to the three different reactions studied .
the isotope ratios for the three reactions at each beam energy overlap nicely indicating formation of composite systems with similar temperature / excitation energy .
the slopes of the exponential fits to the data ( shown by solid line ) , which correspond to the apparent temperatures of the systems , show a gradual decrease with increasing beam energy indicating an increase in the excitation energy / temperature of the system .
having satisfied the necessary condition for isoscaling , the isotopic yield ratio as a function of neutron number @xmath26 , for the beam energies of 25 , 33 , 45 and 53 mev / nucleon is plotted in fig 4 .
the left column shows the ratio for the @xmath0ar + @xmath1fe and @xmath0ca + @xmath1ni pair of reaction and the right column shows the ratio for the @xmath0ar + @xmath1ni and @xmath0ca + @xmath1ni pair of reaction .
one observes that the ratio for each element lies along a straight line in the logarithmic plot and align with the neighboring element quite well .
this feature is observed for all the beam energies and both pairs of reaction studied .
one observes that the alignment of the data points varies with beam energies as well as the pairs of reaction . to have a quantitative estimate of this variation
, the ratio for each elements ( @xmath16 ) were simultaneously fitted using an exponential relation ( shown by the solid lines ) to obtain the slope parameter @xmath2 .
the values of the parameters are shown at the top of each panel in the figure .
the value of the slope parameter @xmath2 is larger for the @xmath0ar + @xmath1fe and @xmath0ca + @xmath1ni reactions , which has a larger difference in the @xmath13 of the systems in the pair , compared to the @xmath0ar + @xmath1ni and @xmath0ca + @xmath1ni reactions , which has a smaller difference in the corresponding @xmath13 .
the @xmath2 value furthermore decreases with increasing beam energy .
figure 5 . shows the isotonic yield ratio as a function of atomic number @xmath16 , for the same beam energies and pairs of systems as shown in figure 4 . once again
, one observes the ratios for each isobar to align nicely with each other at all beam energies .
the scaling parameters @xmath27 in this case , shows an increase with increasing beam energy .
the values of the @xmath27 parameter are larger for the @xmath0ar + @xmath1ni and @xmath0ca + @xmath1ni reaction pair compared to the @xmath0ar + @xmath1fe and @xmath0ca + @xmath1ni reaction pair .
figure 6 shows a relative comparison of how the @xmath2 and @xmath27 parameters evolve as a function of beam energy and the isospin of the system .
the temperature / excitation energy dependence of the isoscaling properties can be further studied by constructing the scaling factor , @xmath28 = @xmath38 .
the scaling factor is known to be a robust feature over a large range of data , from deep inelastic heavy ion reactions at lowest energies , through evaporation reactions induced by light - ion and heavy - ion projectiles , to high energy heavy ion reactions characterized by intermediate mass fragments and multifragmentation @xcite . with a single value for the @xmath27 parameter ,
all the isotopes should fall along a single line in a plot of @xmath28 vs @xmath26 .
this is shown in fig .
7 , where the @xmath28 from two different reaction pairs are plotted as a function of neutron number @xmath26 for beam energies of 25 mev / nucleon and 45 mev / nucleon .
the parameter @xmath27 , was taken from the fit to the isotonic yield ratio shown in figure 5 .
as shown in figure 7 , the values of @xmath28 obtained from various elements ( @xmath16 ) cluster and scale along a single line .
the figure shows a significant difference in the scaling for the two beam energies , indicating the influence of temperature on the isotopic yields of the light clusters .
the present observation alternatively demonstrates the role played by the temperature in the distillation of nuclear matter into a neutron - rich gas and a symmetric liquid phase .
we illustrate this further in the following section . in grand - canonical approach of the multifragmentation process ( see e.g. @xcite ) , the fragment yield with neutron number @xmath26 , and proton number @xmath16 ( mass number @xmath39 ) , can be written as @xmath40 where @xmath41 is the volume of the system and @xmath42 ( @xmath43 e@xmath44 ) and @xmath45 ( @xmath43 e@xmath46 ) are the primary ` free ' neutron and proton densities .
the exponents @xmath47 and @xmath48 are the neutron and the proton chemical potentials , and @xmath49 is the intrinsic partition function of the excited fragment .
the quantity @xmath50 , is the ground state binding energy of the fragment and @xmath6 is the temperature . in the above formula
, the effect of coulomb interaction on fragment yield is neglected by introducing @xmath42 and @xmath45 .
the actual isotope yields then reduce to an approximation appropriate for the thermodynamical limit at high excitation energy @xcite . as discussed in the introduction ,
taking the ratios of the fragment yields from two different systems which differ only in their isospin @xmath51 content reduces the uncertainties in the quantities shown in eq .
the isotopic yield distribution of the fragments in terms of relative reduced neutron density can then be written as , @xmath52 where @xmath53 , corresponds to various isotopes of an element that can be used to determine the double ratio .
the quantity @xmath54 is the yield for the @xmath0ca + @xmath1ni reaction with respect to which all the ratios are taken in this work .
a similar expression for the relative reduced proton density from the isotonic yield ratios can also be written as @xmath55 in the statistical limit for a dilute non - interacting gas , the relative nucleon densities are related to the isoscaling parameters @xmath2 and @xmath27 , through a relation , @xmath56 = @xmath57 and @xmath58 = @xmath59 , where @xmath60 and @xmath61 , with @xmath62 and @xmath63 being the difference in neutron and proton chemical potentials . in figure 8 , we show the experimentally obtained relative reduced neutron and proton density as a function of the difference in the @xmath13 of the systems for the ar + ni / ca + ni and ar + fe / ca + ni pairs of reactions . all densities shown are relative to those of ca + ni reaction .
the circle symbols correspond to the 25 mev / nucleon , squares to the 33 mev / nucleon and the triangles to the 45 mev / nucleon beam energies .
the @xmath2 and the @xmath27 values were taken from the fit to the isotopic and isotonic yield ratios of figures 4 and 5 .
the figure shows a steady decrease in the reduced neutron density and an increase in the proton density with increasing beam energies .
the effect is stronger for the @xmath0ar + @xmath1fe and @xmath0ca + @xmath1ni reaction pair which has the highest difference in @xmath13 .
an important feature of the data shown in figure 8 , is the decrease in the relative neutron - proton asymmetry ( @xmath64 ) with increasing beam energy .
the asymmetry is found to decrease from @xmath21 1.0 at 25 mev / nucleon to @xmath21 0.6 at 45 mev / nucleon for the ar + fe / ca + ni pair of reaction .
this is consistent with the prediction of phase coexistence of two conserved charges with isospin dependent interaction of lattice gas calculation @xcite .
the large asymmetry or the strong enrichment of the gas phase can be understood in terms of an isospin dependent interaction between the protons and the neutrons . in the absence of an isospin dependent interaction , the fragments
are populated close to the stability line and have similar @xmath13 ratio s as the initial source . with the inclusion of isospin dependence ,
the heavy fragments are strongly favored from the energetic point of view to be populated closer to the bottom of the stability valley .
only a few protons are left to be shared between the light fragments leading to the observed enhancement of neutrons in the gas phase . with the increase in temperature ,
the fragment distribution is centered closer to the @xmath65 line resulting in a decrease in the asymmetry of the gas phase .
the observed decrease in the relative neutron and proton densities with increasing beam energy may thus be attributed to the decrease in the sensitivity of the isospin effect with increasing temperature .
the scaling parameter is also sensitive to the breakup density , and the observed decrease in the neutron - proton density could result from a decrease in the break - up density , though it is expected to be less sensitive .
statistical models @xcite are widely used for describing multifragmentation reactions @xcite .
they are based on the assumption of statistical equilibrium at a low density freeze - out stage . in the statistical multifragmentation model ( smm )
@xcite , all break - up channels composed of nucleons and excited fragments are taken into account and considered as partitions . during each partition ,
the conservation of mass , charge , energy and angular momentum is taken into account , and the partitions are sampled uniformly in the phase space according to their statistical weights using the monte carlo sampling . in the present calculations the coulomb interaction between the fragments is treated in the wigner - seitz approximation .
light fragments with mass number @xmath14 @xmath17 4 are considered as elementary particles with only translational degrees of freedom ( nuclear gas " ) .
fragments with @xmath14 @xmath18 4 are treated as heated nuclear liquid drops , and their individual free energies @xmath66 are parametrized as a sum of the volume , surface , coulomb , and symmetry energy , @xmath67 where @xmath68 with parameter @xmath69 related to the level density and @xmath70 = 16 mev being the binding energy of infinite nuclear matter .
@xmath71^{5/4}$ ] , with @xmath72 = 18 mev being the surface co - efficient and @xmath73 = 18 mev being the critical temperature of infinite nuclear matter .
@xmath74 , where @xmath75 $ ] , is the coulomb parameter obtained in the wigner - seitz approximation with charge unit @xmath76 , and @xmath77 = 1.17 fm .
@xmath78 , where @xmath79 = 25 mev is the symmetry energy co - efficient .
these parameters are those adopted from the bethe - weizsacker mass formula and correspond to the assumption of isolated fragments with normal density in the freeze - out configuration . the value of the symmetry energy co - efficient @xmath79 is taken from the fit to the binding energies of isolated cold nuclei in their ground states . in a multifragmentation process
the primary fragments are not only excited but also expanded .
the fragments continue to interact in - medium with each other inside the freeze - out volume and modify their parameters . by comparing the experimentally determined fragment yield distribution with the smm calculation
, the parameters of hot nuclei under multifragmentation conditions , including the symmetry energy , can be extracted . in the following
, it will be shown how this information can be obtained from the isoscaling phenomena .
isotopic scaling or isoscaling arise naturally in statistical equilibrium models of multifragmentation . in these models
the difference in the chemical potential of systems with different @xmath13 is directly related to the scaling parameter @xmath2 .
it has been shown that the isoscaling parameter @xmath2 is proportional to the symmetry energy part of the fragment binding energy through a relation , @xmath80 where @xmath81 , @xmath82 and @xmath83 , @xmath84 are the charge and the mass numbers of the fragmenting systems , @xmath6 is the temperature of the system and @xmath79 , the symmetry energy co - efficient @xcite .
the above formula in the statistical model approach is valid at the freeze - out stage where the primary hot fragments are formed at reduced density . in order to extract information on symmetry energy @xmath79 , from the observed cold secondary fragments ,
one has to take into account the process of secondary de - excitation . in smm
, the secondary de - excitation of large fragments with @xmath14 @xmath18 16 is described by weisskopf - type evaporation and bohr - wheeler - type fission models @xcite .
the decay of smaller fragments is treated with the fermi - breakup model .
all ground and nucleon - stable excited states of light fragments are taken into account and the population probabilities of these states are calculated according to the available phase space @xcite .
ca + @xmath1ni ( stars and solid lines ) , @xmath0ar + @xmath1ni ( circles and dashed lines ) and @xmath0ar + @xmath1fe ( squares and dotted lines ) reactions at various excitation energies .
the calculations are for @xmath79 = 25 mev.,scaledwidth=47.0% ] ar + @xmath1ni and @xmath0ca + @xmath1ni pairs at various excitation energies .
the calculations are for @xmath79 = 25 mev .
the different symbols shown correspond to @xmath16 = 3 ( circles ) , @xmath16 = 4 ( open stars ) , @xmath16 = 5 ( triangles ) , @xmath16 = 6 ( square ) and @xmath16 = 7 ( solid stars ) elements .
the lines are the exponential fits to the ratios.,scaledwidth=47.0% ]
in order to compare the experimentally observed results to the theoretical predictions , we have calculated the primary and the secondary fragment isotopic yield distributions using the above described smm .
the calculations were carried out for the @xmath0ca + @xmath1ni , @xmath0ar + @xmath1ni and @xmath0ar + @xmath1fe reactions at various excitation energies . the excitation energy per nucleon of the initial system for the calculation depends strongly on the matching condition between the dynamical and statistical stage of the collision .
this quantity is presently difficult to calculate accurately .
a range of values for the excitation energy per nucleon from @xmath85 = 4 - 10 mev / nucleon were therefore assumed .
the excitation energy corresponding to each beam energy was also verified by an independent calculation using buu - gemini ( see table ii in ref.@xcite ) , and the systematic calorimetric measurements available in the literature for systems with mass a @xmath21 100 @xcite .
the mass and the charge of the initial systems were assumed to be those of the initial compound nucleus . to check for the possible uncertainty in the source size due to loss of nucleons during pre - equilibrium emission ,
the calculation was also performed for sources with 80@xmath24 of the total mass .
no significant change in the isospin characteristics under study was observed .
the freeze - out density in the calculation was assumed to be 1/3 of the normal nuclear density and the symmetry energy co - efficient @xmath79 was taken to be 25 mev . the calculated primary and secondary fragment yield distributions for the carbon isotopes in @xmath0ca + @xmath1ni , @xmath0ar + @xmath1ni and @xmath0ar + @xmath1fe reactions at various excitation energies are as shown in figure 9 . the characteristics of the hot primary fragment yield distribution , shown in the left column of the figure , change significantly after the secondary de - excitation .
the primary yield distribution for the three systems , shown by the dotted , dashed and solid curve , for each excitation energy clearly show isospin effect .
the most neutron rich system has the largest yield for the neutron rich isotopes and the lowest yield for the neutron deficient isotopes .
this effect appears to decrease with increasing excitation energy .
a similar feature is also observed in the secondary fragment distribution shown in the right column ( lines ) with the experimentally determined @xmath2 ( symbols ) as a function of excitation energy for different values of the symmetry energy co - efficient @xmath79 .
the dotted lines correspond to the primary fragments and the solid lines to the secondary fragments .
the left column shows the comparison for the @xmath0ar + @xmath1ni and @xmath0ca + @xmath1ni pair , and the right column shows the comparison for the @xmath0ar + @xmath1fe and @xmath0ca + @xmath1ni pair.,scaledwidth=47.0% ] as a function of symmetry energy co - efficient for various excitation energies .
the open circles joined by dotted lines correspond to the primary fragments and the open stars joined by solid lines to the secondary fragments .
the left column shows the calculation for @xmath0ar + @xmath1ni and @xmath0ca + @xmath1ni pair , and the right column for the @xmath0ar + @xmath1fe and @xmath0ca + @xmath1ni pair.,scaledwidth=47.0% ] of the figure , though the effect is observed to be weakened significantly .
furthermore , the mean of the distribution is also observed to decrease along with the width for the secondary fragments .
qualitatively , the smm simulates quite well the overall features of the experimentally observed isotopic yield distribution shown in figure 2 . the isotopic yield ratios using the primary and the secondary fragment distribution from the statistical multifragmentation model are shown in figure 10 for the @xmath0ar + @xmath1ni and @xmath0ca + @xmath1ni pair of reaction .
it is observed that the yield ratios for both the primary and the secondary distribution obey the isoscaling relation quite well .
very little difference in the scaling parameter @xmath2 , obtained from the primary and the secondary yield distribution is observed .
furthermore , the scaling parameter shows a gradual decrease in its value with increasing excitation energy similar to those observed experimentally and shown in figure 6 .
though the overall feature of the scaling parameter calculated from the statistical multifragmentation model is reproduced quite well , the absolute values do not quite agree with the experimentally determined @xmath2 .
this is shown in the top panel of figure 11 , where a comparison is made between the smm calculated and the experimentally observed values of @xmath2 .
in the figure , the left column correspond to the @xmath0ar + @xmath1ni and @xmath0ca + @xmath1ni pair of reaction and the right to the @xmath0ar + @xmath1fe and @xmath0ca + @xmath1ni pair .
the dotted lines correspond to @xmath2 calculated from the primary fragment distribution and the solid lines to those calculated from the secondary fragment distribution .
the symbols correspond to the experimentally determined @xmath2 s .
it is observed that the experimentally determined @xmath2 s are significantly lower than the calculated values of @xmath2 using the standard value of the symmetry energy co - efficient , @xmath79 = 25 mev , for the isolated cold nuclei in their ground states . in order to explain the observed dependence of the isoscaling parameter @xmath2 on excitation energy
, we varied the @xmath79 of the hot primary fragment in the smm input in the range 25 - 15 mev .
as shown in the center and the bottom panel of the figure , the isoscaling parameter decreases slowly with decreasing symmetry energy . the experimentally determined @xmath2 could be reproduced for both pairs of systems at all excitation energies using a symmetry energy value of @xmath79 = 15 mev .
this value of symmetry energy is significantly lower than the value of @xmath79 = 25 mev used for ground state nuclei .
figure 12 . shows the calculated @xmath79 dependence of the isoscaling parameter @xmath2 from the hot primary fragment distribution , and from the cold secondary fragment distribution , for the two pairs of systems at various excitation energies .
the @xmath2 as a function of symmetry energy for each excitation energy and system is observed to decrease with decreasing symmetry energy .
the difference between the primary fragment @xmath2 and the secondary fragment @xmath2 is negligible for the ar + ni and ca + ni reaction pair , which has the lowest difference in neutron - to - proton ratio .
the difference for the ar + fe and ca + ni pair , which has the highest difference in neutron - to - proton ratio , however is slightly larger at higher excitation energies . in the above described statistical multifragmentation model calculations the masses of the fragments used
were those of cold isolated nuclei .
the fragments in their primary stage are usually hot and the properties of hot nuclei ( i.e. , their binding energies and masses ) differ from those of cold nuclei .
if hot fragments in the freeze - out configuration have smaller @xmath79 , their masses at the beginning of the secondary de - excitation will be different , and this effect should be taken into account in the evaporation process . recently ,
buyukcizmeci _ et al . _
@xcite adopted a phenomenological approach to estimate the effect of the symmetry energy evolution during the sequential evaporation . in this approach , they assume liquid drop masses @xmath86 for the evaporation of the light particles ( @xmath87 , @xmath88 , @xmath89 , @xmath90 , @xmath91 , @xmath2 ) , if the internal excitation energy of the fragment is large ( @xmath92 ) . at lower excitation energies ( @xmath93 )
they assume a smooth transition to standard experimental masses with shell effects(@xmath94 ) using the following dependence , @xmath95 the excitation energy is determined from the energy balance taking into account the mass @xmath96 at the given excitation .
the above corrections were incorporated in the statistical model calculations described in the previous section to study the effect of @xmath79 during the sequential de - excitation of the hot primary fragments .
figure 13 shows the result of the statistical model calculation using the modified secondary de - excitation correction .
the isoscaling parameter @xmath2 is plotted as a function of excitation energy for the two pairs of systems ( left and right column ) .
the top panels show the calculations using symmetry energy value of 25 mev .
as noted in the previous section , the new calculations are not able to reproduce the experimentally determined alpha for both pairs of systems . with decreasing values of the symmetry energy , the calculated @xmath2 values for the ar + ni and ca + ni pair ( left column ) decrease and are in better agreement with the experimental values at @xmath79 = 13 mev . on the other hand ,
the calculated values for the ar + fe and ca + ni reaction pair ( right column ) are in good agreement with the experimental values at @xmath79 = 10 mev . in general
, one observes that the modified version of secondary de - excitation in smm leads to a symmetry energy value of 10 - 13 mev .
this is slightly lower than the value of 15 mev obtained from the standard version of the smm calculation shown in figure 11 .
the dependence of the isoscaling parameter as a function of the symmetry energy for primary and secondary fragments at various excitation energies are as shown in figure 14 .
once again , the difference between the primary fragment @xmath2 and the secondary fragment @xmath2 are extremely small for the ar + ni and ca + ni pair of reaction ( which has the lowest difference in neutron - to - proton ratio ) , and only slightly larger for the ar + fe and ca + ni pair ( which has the highest difference in neutron - to - proton ratio ) . however , the main difference between the dependence shown in figure 12 and 14 is the rate at which the isoscaling parameter @xmath2 decrease with decreasing symmetry energy .
the decrease is much slower in the calculation where the symmetry energy dependence of the mass is taken into account during the secondary de - excitation .
the slower decrease in the isoscaling parameter results in the calculation being able to reproduce the experimental value at a slightly lower value of symmetry energy .
one also observes that the primary fragment @xmath2 in the modified version of the secondary decay calculation are consistently smaller than the secondary fragment @xmath2 at lower symmetry energies .
ar + @xmath1fe reaction at @xmath85 = 6 mev / nucleon .
the left panels correspond to the old and the right to the new de - excitation prescription used in the smm calculation .
the solid and the dashed curves correspond to the calculations using two different values of the symmetry energy.,scaledwidth=47.0% ] in order to understand the difference in the symmetry energy , we show in figure 15 , the calculated isotopic yield distribution for the carbon element in @xmath0ar + @xmath1fe reaction at @xmath85 = 6 mev / nucleon .
the figure shows the primary and the secondary yield distribution using two different prescriptions for the secondary de - excitations in smm .
the left column in the figure corresponds to the smm calculations , where the fragment masses used are those of cold isolated nuclei , and the right column corresponds to the smm calculations , where the masses ( symmetry energy ) evolve with their excitation energy during secondary de - excitation .
the panels on the top correspond to the primary yield distribution and those in the bottom to the secondary yield distribution .
the dotted and the solid curves in each panel corresponds to the calculations assuming two different values of the symmetry energy , 15 and 25 mev respectively . from the figure , it is evident that there exist a subtle difference between the two final ( secondary ) distributions , shown by the dotted and the solid curves in the bottom panels .
one also observes that the distribution depends on whether the mass ( symmetry energy ) evolves during the evaporation or not .
the smm calculation with the standard de - excitation ( i.e the old de - excitation ) leads to a narrow final distribution and the isotopes are concentrated close to the @xmath27-stability line .
the difference in the final yield distributions for @xmath79 = 15 mev and @xmath79 = 25 mev is very small .
this difference is however much more pronounced in the new de - excitation calculation .
the final isotope distributions in this case are considerably wider , and shifted toward neutron - rich side . the smm calculation assuming the mass ( symmetry energy ) evolution during the evaporation therefore leads to larger yields for neutron rich fragments .
a similar observation was made by buyukcizmeci _
@xcite in their calculation of the primary and secondary fragment isotopic distributions in @xmath97au , @xmath98sn and @xmath98la systems . by using the experimental masses during the evaporation from the primary fragments ,
the emission of the charged particles are suppressed by the binding energy and the coulomb barrier . in the case of small @xmath79 ,
the binding energy in the beginning of evaporation process essentially favors emission of charged particles .
when the nucleus has cooled down sufficiently to restore the normal symmetry energy , the remaining excitation energy is rather low ( @xmath99 @xmath100 1 mev ) for the nucleus to evaporate many neutrons .
the above comparison of the experimentally observed isoscaling properties with the statistical multifragmentation model shows that , irrespective of the secondary de - excitation , the final fragment distribution depends strongly on the available free energy and the strength of the symmetry energy
. a significantly lower value of the symmetry energy than that assumed for cold isolated nuclei is required to explain the isotopic compositions of the fragments produced in multifragmentation reaction .
the difference between the two kinds of evaporation calculations gives a measure of the uncertainty expected in the present analysis .
the results above indicate that the properties of nuclei produced at high excitation energy , isospin and reduced density could be significantly different from those of the cold isolated nuclei .
such information can provide important inputs for the understanding of the nuclear composition of supernova matter where hot and neutron rich nuclei are routinely produced @xcite .
in summary , we have measured the isotopic yield distribution of the fragments produced in the multifragmentation of @xmath0ar , @xmath0ca + @xmath1fe , @xmath1ni reactions at 25 - 53 mev / nucleon .
the symmetry energy and the isoscaling properties of the fragments produced were studied within the framework of statistical multifragmentation model .
it is observed that the isoscaling parameter @xmath2 for the hot fragments decrease with increasing excitation energy and decreasing symmetry energy .
the @xmath2 values increase with increasing difference in the isospin of the fragmenting system .
similar behavior is also observed for the cold secondary fragments .
the sequential decay of the primary fragments to secondary fragments is observed to have very little influence on the isoscaling parameter as a function of excitation energies and isospin of the fragmenting system .
the symmetry energy however , strongly influences the isospin properties of the hot fragments .
the experimentally determined scaling parameters could be explained by a symmetry energy that is as low as 10 - 15 mev , and significantly lower than that for the normal ( cold ) nuclei at saturation density .
the present results indicate that the isospin properties of the fragments produced at high excitation energy and reduced density in multifragmentation reaction are sensitive to the symmetry energy .
the authors wish to thank the cyclotron institute staff for the excellent beam quality .
this work was supported in parts by the robert a. welch foundation through grant no .
a-1266 , and the department of energy through grant no .
de - fg03 - 93er40773 . d.v .
shetty , a.s .
botvina , s.j .
yennello , a. keksis , e. martin and g.a .
souliotis , phys .
c * 71 * , 024602 ( 2005 ) .
a. lefevre _ et al .
rev . lett . * 94 * , 162701 ( 2005 ) .
d. henzlova , a.s .
botvina , k.h .
schmidt , v. henzl , p. napolitani , m.v .
ricciardi , nucl - ex/0507003 ( 2005 ) .
a.s . botvina and i.n .
mishustin , phys .
b * 584 * , 233 ( 2004 ) .
botvina , o.v .
lozhkin , and w. trautmann , phys .
c * 65 * , 044610 ( 2002 ) .
j. lattimer , c. pethick , m. prakash and p. hansel , phys .
* 66 * , 2701 ( 1991 ) . c. lee , phys
. rep . * 275 * , 255 ( 1996 ) .
pethick and d.g .
ravenhall , annu .
nucl . part .
* 45 * , 429 ( 1995 ) .
lattimer and m. prakash , phys .
* 333 * , 121 ( 2000 ) .
hix _ et al . _ ,
lett . * 91 * , 201102 ( 2003 ) .
lattimer and m. prakash , astrophys .
j. * 550 * , 426 ( 2001 ) .
a.s . botvina and i.n .
mishustin , phys .
c * 72 * , 048801 ( 2005 ) .
brown , phys .
lett * 85 * , 5296 ( 2000 ) .
c.j . horowitz and j. piekarewicz , phys .
* 86 * , 5647 ( 2001 ) .
furnstahl , nucl .
a * 706 * , 85 ( 2002 ) .
k. oyamatsu , i. tanihata , y. sugahara , k. sumiyoshi and h. toki , nucl .
a * 634 * , 3 ( 1998 ) .
a. ono , p. danielewicz , w.a .
friedman , w.g .
lynch and m.b .
tsang , phys .
c * 68 * , 051601 ( 2003 ) .
chen , c.m .
ko and b.a .
li , phys .
lett . * 94 * , 032701 ( 2005 ) .
shetty _ et al .
c * 70 * , 011601 ( 2004 ) .
tsang _ et al .
lett . * 92 * , 062701 ( 2004 ) . b.a .
li , c.m .
ko and w. bauer , int .
e * 7 * , 147 ( 1998 ) . v. baran , m. colonna , v. greco and m. ditoro , phys .
rep . * 410 * , 335 ( 2005 ) .
v. baran , m. colonna , m. di toro , v. greco , m. zielinska - pfabe and h.h .
wolter , nucl .
a * 703 * , 603 ( 2002 ) .
liu _ et al .
c * 69 * , 014603 ( 2004 ) .
tsang _ et al .
_ , phys . rev .
c * 64 * , 054615 ( 2001 ) .
tsang _ et al . _ , phys .
lett . * 86 * , 5023 ( 2001 ) .
w.p . tan _
et al . _ ,
c * 64 * , 051901 ( 2001 ) .
d. lunney , j.m .
pearson and c. thibault , rev .
phys . * 75 * , 1021 ( 2003 ) .
dieperink , y. dewulf , d. van neck , m. waroquier and v. rodin , phys .
c * 68 * , 064307 ( 2003 ) .
w. zuo , i. bombaci and u. lombardo , phys .
c * 60 * , 024605 ( 1999 ) .
m. brack , c. guet and h.b .
hakansson , phys .
rep . * 123 * , 276 ( 1985 ) .
pearson and r.c .
nayak , nucl .
a * 668 * , 163 ( 2000 ) w.d .
myers and w.j .
swiatecki , nucl . phys . * 81 * , 1 ( 1966 ) . k. pomorski and j. dudek , phys .
c * 67 * , 044316 ( 2003 ) .
b. borderie _
lett . * 86 * , 3252 ( 2001 ) .
bowman _ et al .
* 67 * , 1527 ( 1991 ) .
m. dagostino _ et al .
b * 371 * , 175 ( 1996 ) .
l. beaulieu _ et al .
* 84 * , 5971 ( 2000 ) .
w.p . tan , s.r .
souza , r.j .
charity , r. donangelo , w.g .
lynch , and m.b .
tsang , phys .
c * 68 * , 034609 ( 2003 ) .
h. xi , w.g .
lynch , m.b .
tsang , w.a .
friedman , and d. durand , phys .
c * 59 * , 1567 ( 1999 ) .
h. xi , w.g .
lynch , m.b .
tsang , and w.a .
friedman , phys .
c * 54 * , 2163 ( 1996 ) .
h. johnston _ et al .
c * 56 * , 1972 ( 1997 ) .
schmitt _ et al .
methods phys .
a * 354 * , 487 ( 1995 ) .
raduta , eur . phys .
a*24 * , 85 ( 2005 )
tsang _ et al .
c * 64 * , 041603 ( 2001 ) .
s. albergo _ et al .
_ , nuovo cimento a*89 * , 1 ( 1985 ) .
bondorf , a.s .
botvina , a.s .
iljinov , i.n .
mishustin and k. sneppen , phys . rep . *
257 * , 133 ( 1995 ) .
j. randrup and s. koonin , nucl .
a * 356 * , 223 ( 1981 ) .
botvina , a.s .
iljinov , i.n .
mishustin , j.p .
bondorf , r. donangelo and k. sneppen , nucl .
a * 475 * , 663 ( 1987 ) .
d.h.e . gross _
. progr . phys . *
53 * , 605 ( 1990 ) .
d.h.e . gross _
et al . _ , z. phys .
a * 309 * , 41 ( 1982 ) .
x.z . zhang , d.h.e .
gross , s. xu , and y.m .
zheng , nucl .
a * 461 * , 641 ( 1987 ) . edited by b .- a .
li and w. schroder ( nova science , new york , 2001 ) .
p. chomaz and f. gulminelli , phys .
b * 447 * , 221 ( 1999 ) .
botvina _ et al .
a * 584 * , 737 ( 1995 ) .
scharenberg _ et al .
c * 64 * , 054602 ( 2001 ) .
n. bellaize _
et al . _ ,
nucl . phys .
a * 709 * , 367 ( 2002 ) .
avdeyev _ et al .
a * 709 * , 392 ( 2002 ) .
j. pochodzalla _ et al .
rev . lett . * 75 * , 1040 ( 1995 ) .
m. dagostino _ et al .
_ , nucl . phys .
a * 650 * , 329 ( 1999 ) .
d. cussol _ et al .
a * 561 * , 298 ( 1993 ) .
a.s . botvina and i.n .
mishustin , phys .
c * 63 * , 061601 ( 2001 ) .
n. buyukcizmeci , r. ogul , and a.s .
botvina , eur .
j. * 25 * , 57 ( 2005 ) . | the symmetry energy and the isoscaling properties of the fragments produced in the multifragmentation of @xmath0ar , @xmath0ca + @xmath1fe , @xmath1ni reactions at 25 - 53 mev / nucleon were investigated within the framework of statistical multifragmentation model .
the isoscaling parameters @xmath2 , from the primary ( hot ) and secondary ( cold ) fragment yield distributions , were studied as a function of excitation energy , isospin ( neutron - to - proton asymmetry ) and fragment symmetry energy .
it is observed that the isoscaling parameter @xmath2 decreases with increasing excitation energy and decreasing symmetry energy .
the parameter @xmath2 is also observed to increase with increasing difference in the isospin of the fragmenting system .
the sequential decay of the primary fragments into secondary fragments , when studied as a function of excitation energy and isospin of the fragmenting system , show very little influence on the isoscaling parameter .
the symmetry energy however , has a strong influence on the isospin properties of the hot fragments .
the experimentally observed scaling parameters can be explained by symmetry energy that is significantly lower than that for the ground state nuclei near saturation density .
the results indicate that the properties of hot nuclei at excitation energies , densities and isospin away from the normal ground state nuclei could be significantly different . |
the effect of atomic disorder on the electronic properties of correlated electron systems , particularly those close to a quantum critical point ( qcp ) @xcite has been a topic of active research . in the critical regime ,
the system is at the threshold of an instability and even weak perturbations , e.g. , disorder can cause significant effects by changing the nature of the quantum macro state . in these disordered systems , a rather large residual resistivity @xmath11
is often encountered , even for single crystals , which means that even weak disorder is influential . as was argued theoretically @xcite ,
such a drastic influence is possible because the band width of a few ev and an effective hubbard interaction @xmath12 of the same order of magnitude result in a much more subtle energy balance that atomic disorder can disturb more easily .
therefore , investigations of atomic scale disorder in the form of defects and vacancies , granularity , and the effective increase in disorder by doping have received renewed attention in recent times particularly because of observations of novel phenomena in strongly correlated materials .
the kondo insulators are an example of thermoelectric materials where the defects lead to a high value of figure - of - merit @xmath13 , where @xmath14 is the seebeck coefficient , @xmath15 is the electrical conductivity , @xmath16 is the electronic thermal conductivity , and @xmath17 is the lattice contribution to the thermal conductivity @xcite due to the reduction of the lattice contribution to the thermal conductivity .
the effect of disorder on superconducting properties has inspired a great deal of research , with the discovery of unconventional superconductivity in heavy fermion compounds @xcite and associated quantum critical behavior . in many superconductors , the critical temperature @xmath6 decreases with increasing disorder and sufficiently strong disorder can , in fact , destroy superconductivity . as this
disorder driven transition from a superconducting to a non - superconducting ground state occurs , the localization effects become so strong that often an insulating material results ( at @xmath18 this is known as a quantum phase transition ) .
this transition is referred to as a superconductor - insulator transition @xcite .
there are known strongly correlated superconductors that show evidence of nanoscale disorder , meaning that the sample exhibits electronic inhomogeneity over the length scale of the coherence length . such substantial nanoscale electronic disorder is characteristic of bi@xmath19sr@xmath19cacu@xmath19o@xmath20 high-@xmath6 materials , as well as pros@xmath1sb@xmath21 @xcite , cept@xmath0si @xcite , and ceirin@xmath22 @xcite .
our recent investigation of the filled cage superconductors la@xmath23sn@xmath2 with @xmath24 rh @xcite and ru @xcite have shown evidence of two superconducting phases : an inhomogeneous superconducting state below @xmath5 and the superconducting phase at @xmath25 .
this anomaly was interpreted in the context of the presence of an inhomogeneous superconducting phase between @xmath6 and @xmath5 . in this work ,
we present a comprehensive thermodynamic and high - pressure electrical resistivity study on la@xmath0co@xmath8ru@xmath9sn@xmath2 to explain the superconductivity in the presence of disorder .
la@xmath0co@xmath1sn@xmath2 clearly exhibits a homogeneous superconducting phase at @xmath6 , while in contrast la@xmath0ru@xmath1sn@xmath2 and its co - alloys show evidence of nanoscale inhomogeneity with the presence of @xmath5 .
the impact of disorder on the ground state of superconducting materials has played an important role in condensed matter physics over the years .
we believe that our results contribute towards developing a broader understanding of the complex behavior in novel superconducting strongly correlated electron systems .
polycrystalline la@xmath0co@xmath1sn@xmath2 and la@xmath0ru@xmath1sn@xmath2 samples were prepared by arc melting the constituent elements on a water cooled copper hearth in a high - purity argon atmosphere with an al getter .
the dilute la@xmath0co@xmath8ru@xmath9sn@xmath2 alloys were prepared by diluting nominal compositions of the parent compounds .
the samples were then annealed at 870 @xmath26c for 2 weeks .
all samples were carefully examined by x - ray diffraction analysis and found to have a cubic structure ( space group @xmath27 ) @xcite and for @xmath28 and 3.5 , the samples were single phase while for @xmath29 and 3 , the alloys were a mixture of two phases .
stoichiometry and homogeneity were verified by the microprobe technique ( scanning microscope jsm-5410 ) and by xps analysis . as an example , measurements of la@xmath0co@xmath1sn@xmath2 showed a composition close to the nominal ratio 3:4:13 ( i.e. , 14.87:19.93:65.20 for la : co : sn ) .
for the la@xmath0ru@xmath1sn@xmath2 and la@xmath0co@xmath8ru@xmath9sn@xmath2 alloys , the composition of the samples were also close to the nominal ratio 3:4:13 stochiometry , and in table [ tab : table1 ] , we present the results from measurements of la@xmath0ru@xmath30cosn@xmath2 noted at different points of the surface . [ cols="^,^,^,^,^,^ " , ] shown in fig .
[ fig : fig3 ] is the specific heat @xmath31 plotted as @xmath31 vs @xmath4 at various magnetic fields ( in panel @xmath32 ) , and ac and dc magnetic susceptibility ( panel @xmath33 ) for la@xmath30co@xmath34ru@xmath35sn@xmath2 .
the heat capacity data for la@xmath30co@xmath36sn@xmath2 ( not shown in fig . [
fig : fig3 ] , c.f . ref . ) indicates bulk superconductivity below @xmath37 k in agreement with resistivity data , while la@xmath30co@xmath34ru@xmath35sn@xmath2 shows a broad transition to the superconducting state with the same @xmath6 from the resistivity and susceptibility data .
( @xmath32 ) the temperature dependence of the specific heat @xmath38 of la@xmath30co@xmath34ru@xmath35sn@xmath2 at different magnetic fields @xmath39 . ( @xmath33 ) the ac magnetic susceptibility @xmath40 at @xmath41 gs divided by theoretical value of the full meissner state @xmath42 , and zero - field - cooled ( zfc ) and field - cooled ( fc ) dc magnetic susceptibility in an applied field of @xmath43 gs . , scaledwidth=47.0% ] for la@xmath30coru@xmath30sn@xmath2 , the superconductivity shown in @xmath44 data ( in fig .
[ fig : fig4]@xmath32 and fig .
[ fig : fig5 ] ) also shows broad transition with the maximum in @xmath45 at @xmath46 k spanning the maximum in @xmath47 in fig .
[ fig : fig4]@xmath33 . under certain conditions ,
the ac losses in superconducting transition can exceed those of a normal metal , leading to a peak in @xmath47 vs @xmath4 @xcite .
however , it was argued that a @xmath47 maximum can occur in surface superconductors at sufficiently low frequencies , this is not the case in the ac magnetic susceptibility shown in fig .
[ fig : fig3 ] @xcite .
the perfect diamagnetism of the full meissner state @xmath48 emu / g for mass density @xmath49 g/@xmath50 ( refs . )
is reached below the temperature of the maximum in @xmath47 , it should also be noted that @xmath47 depends on the frequency of the magnetic field , and that is characteristic of magnetically inhomogeneous materials .
we believe that the superconductivity in la@xmath30coru@xmath30sn@xmath2 is completely inhomogeneous superconductivity to explain the anomalies in the specific heat and @xmath47 .
( @xmath32 ) the temperature dependence of the specific heat @xmath51 with a gaussian gap distribution fit @xmath52 and resistivity @xmath53 , both at @xmath54 for la@xmath30coru@xmath30sn@xmath2 . for the sample under @xmath54 , @xmath55 ,
[ fig : fig5 ] .
( @xmath33 ) the ac magnetic susceptibility ( @xmath41 gs ) @xmath56 and @xmath47 at different frequencies , and zfc and fc dc magnetic susceptibility ( @xmath43 gs )
. , scaledwidth=48.0% ] the specific heat @xmath51 under various magnetic fields for la@xmath30coru@xmath30sn@xmath2 .
the arrow indicates the beginning of the superconductivity at @xmath54 , the transitions under magnetic fields @xmath57 are similarly broad .
the insets display the @xmath44 data near @xmath6 for la@xmath0ru@xmath1sn@xmath2 and la@xmath0co@xmath1sn@xmath2 . in case of la@xmath0ru@xmath1sn@xmath2 the bulk effect at @xmath6 and inhomogeneous superconducting phase between @xmath5 and @xmath6
are both shown .
the dotted line is the best fit to the data for the expression @xmath58 .
, scaledwidth=48.0% ] namely , we believe that the resistivity drop marks the onset of an inhomogeneous superconducting phase with spatial distribution of the magnitude of the superconducting gap , as a bulk property of the sample . since the drop of the resistivity at @xmath5 is not accompanied by a change of @xmath47 , the volume occupied by the inhomogeneous phase is too small to cancel out normal - state paramagnetic contributions .
on the other hand , the superconducting regions must be arranged as to form the necessary continuous paths reflected in the resistivity measurements . following ref .
we assume a simple gaussian gap distribution @xmath59,\ ] ] where @xmath60 and @xmath61 are treated as fitting parameters .
the best fit of @xmath52 to the experimental data @xmath51 gives the points @xmath62 in fig .
[ fig : fig2 ] , in good agreement with the points @xmath63 in fig .
[ fig : fig2 ] .
points @xmath61 represents the temperature of the maximum in @xmath47 . the behavior observed in this strongly disordered alloy is qualitatively different than that in la@xmath0ru@xmath36sn@xmath2 @xcite , or la@xmath0rh@xmath36sn@xmath2
@xcite with clear evidence for two superconducting phases : the _ high temperature _ inhomogeneous superconducting state below @xmath5 and the second ( bulk ) superconducting phase below @xmath6 , where @xmath64 .
we also note that the @xmath65 data for la@xmath30coru@xmath30sn@xmath2 is not well approximated by @xmath66 , while the bulk superconducting phases in both la@xmath30ru@xmath36sn@xmath2 and la@xmath30co@xmath36sn@xmath2 are well fit by this expression ( c.f . ,
[ fig : fig5 ] ) . excluding the case of la@xmath30coru@xmath30sn@xmath2
, we found @xmath38 follows the behavior described by the bcs theory in the weak - coupling limit , which indicates @xmath67-wave superconductivity .
the bcs theory for @xmath67-wave superconductors provides a relation @xmath68 between the jump of the specific heat @xmath69 at the critical temperature @xmath6 and the normal metallic state contribution @xmath70 ; the theoretical value @xmath68 is very close to the values presented in table [ tab : table2 ] . in fig .
[ fig : fig6 ] , we show that the hysteresis loop in the superconducting state of la@xmath30coru@xmath30sn@xmath2 is about 3 t , while in the case of the remaining compounds , it is nearly an order of magnitude smaller .
the broad hysteresis loop suggests strongly inhomogeneous material .
magnetization @xmath71 per formula unit vs magnetic field at various temperatures .
the inset shows a symmetric hysteresis loop at @xmath72 k for the superconducting state in la@xmath0co@xmath36sn@xmath2 .
panel ( @xmath32 ) shows the data for la@xmath0co@xmath34ru@xmath35sn@xmath2 , panel ( @xmath33 ) displays @xmath71 for la@xmath30coru@xmath30sn@xmath2 .
, scaledwidth=48.0% ] we expect that external pressure applied to strongly disordered materials should drive lattice instabilities from the compounds by varying the dominant parameters of the superconducting state , e.g. , electronic density of states at the fermi level .
most of the known superconductors show a decrease of @xmath6 with increasing applied pressure @xcite ; however , increasing pressure should also partially mitigate the inhomogeneity and stabilize the structural properties of the disordered system , and as a consequence , @xmath5 is also expected to decrease with pressure .
the evidence of this is shown in figs .
[ fig : fig7 ] - [ fig : fig9 ] and summarized in fig .
[ fig : fig10 ] . the observed increase of @xmath6 with pressure shown in fig .
[ fig : fig10 ] for la@xmath0co@xmath36sn@xmath2 was recently discussed as a possible result of a subtle structural distortion below @xmath73 k @xcite .
the pressure coefficients @xmath74 and @xmath75 obtained from the respective @xmath6 vs @xmath76 data shown in fig .
[ fig : fig10 ] are listed in table [ tab : table2 ] .
the pressure coefficients of @xmath5 are almost twice large as those of @xmath6 , while for la@xmath0co@xmath1sn@xmath2 , the @xmath77 k / gpa is positive .
the @xmath76-dependence of @xmath6 has been discussed according to the of eliashberg theory of strong - coupling superconductivity @xcite .
we employ the mcmillan expression @xcite @xmath78 which is a solution of the finite - temperature eliashberg equations , to connect the value of @xmath6 with the electron - phonon coupling parameter @xmath79 , debye temperature @xmath80 and the coulomb repulsion @xmath81 ( the value of @xmath82 was chosen to be 0.1 as is typical for @xmath67 and @xmath83 band superconductors ) .
this yields @xmath84 for @xmath6s and a larger @xmath79 value @xmath85 for @xmath5s .
however , in the both superconducting states , relatively small @xmath79 negates the strong coupling superconductivity .
the coupling @xmath79 is given by @xmath86 where @xmath87 is the square of the electronic matrix element of electron phonon interactions averaged over the fermi surface , @xmath88 is an average of the square of the phonon frequency , and @xmath71 is the atomic mass .
usually , @xmath82 and @xmath87 are very weakly pressure dependent , so that the main pressure effect on the transition temperature comes from @xmath89 and @xmath90 ( @xmath88 depends on @xmath89 ) .
the pressure dependence of @xmath80 is given by the grneisen parameter @xmath91 , which represents the lattice stiffening . using the mcmillan expression it was found @xcite
that @xmath92 strongly determines the magnitude and sign of @xmath74 .
our data suggest a larger @xmath92 for the inhomogeneous superconducting state with respect to the bulk effect observed below @xmath6 ; in case of la@xmath0co@xmath1sn@xmath2 , the grneisen parameter is expected to be smaller .
it is also possible that in the case of inhomogeneous superconductivity , the pressure dependence of the density of states at the fermi level is more pronounced than in bulk superconductors , and may lead to a larger value of @xmath93 than @xmath74 .
electrical resistivity @xmath94 of la@xmath30ru@xmath36sn@xmath2 at various applied pressures .
the inset displays the low temperature details , showing the smooth suppression of @xmath6 .
, scaledwidth=48.0% ] resistivity of la@xmath30coru@xmath30sn@xmath2 at different applied pressure .
the inset displays the details .
, scaledwidth=48.0% ] resistivity of la@xmath30co@xmath34ru@xmath35sn@xmath2 at different applied pressure . the inset displays the details .
, scaledwidth=48.0% ] critical temperatures @xmath6 and @xmath5 vs pressure @xmath76 .
the critical temperatures are obtained from resistivity under applied pressure at 50% of the normal state value . for la@xmath30ru@xmath36sn@xmath2
, @xmath6 is estimated as a very weak change in @xmath53 below @xmath5 .
, scaledwidth=48.0% ] since the co radius is smaller than that of ru , increasing the amount of co in the la@xmath0co@xmath8ru@xmath9sn@xmath2 system leads to an effectively negative internal pressure . with @xmath95 increasing from 0 to 1
, @xmath6 increases as well , but further , for @xmath95 going from 1 to 4 , @xmath6 decreases almost linearly from 5 k to 1.95 k ( see fig . [
fig : fig11 ] ) .
the dependence of @xmath5 on the chemical pressure is consistent with the effects of external pressure . with @xmath95 increasing from 0 to 1
, @xmath5 increases from about 3.76 k to 5.58 k , and can be interpreted as a continuation of the dependence on the external pressure for @xmath96 . in this case , however , @xmath97 is slightly less sensitive to the chemical pressure than @xmath6 .
ambient pressure @xmath6 and @xmath5 as a function of @xmath95 for la@xmath0co@xmath8ru@xmath9sn@xmath2 .
dashed lines are a guide to the eye .
the blue area represents the inhomogeneous superconducting phase ( is ) , whereas the red area represents bulk superconductivity ( bs).,scaledwidth=46.5% ] it is difficult to understand the abnormally large critical temperature for la@xmath30coru@xmath30sn@xmath2 , that is much larger than the @xmath5 characterizing the inhomogeneous superconducting phase in of la@xmath30ru@xmath36sn@xmath2 .
typically , disorder strongly reduces the critical temperature due to the elevated impurity scattering ; therefore , large value of @xmath5 is surprising .
it is , however , also possible to enhance superconductivity by local disorder @xcite . in this case
, the superconductor may be inhomogeneous with lower and higher @xmath6 regions . above @xmath97 ,
superconducting clusters appear , which at @xmath97 form a network of continuous paths trough the entire sample .
the random character of the co substitution leads to a statistical ( chaotic ) distribution of these clusters . despite the drop in the electrical resistivity ,
the fraction of the volume occupied by the superconducting state can still be small .
this is a typical percolation scenario @xcite . at a lower temperature @xmath6 ,
the previously normal regions becomes superconducting and a macroscopic ( bulk ) superconducting state is formed .
this is the transition that is seen in the specific heat and susceptibility measurements .
in most of the known superconductors , the transition temperature @xmath6 decreases as a consequence of increased disorder .
however , there are known examples of strongly correlated superconductors which show evidence of nanoscale disorder leading to an inhomogeneous superconducting state and , as a consequence , the critical temperature @xmath98 .
both superconducting phases : the @xmath6-bulk phase and the @xmath5 _ high - temperature _ inhomogeneous phase whose onset is observed between @xmath5 and @xmath6 are present in the skutterudite - related la@xmath0rh@xmath1sn@xmath2 and la@xmath0ru@xmath1sn@xmath2 compounds . in these compounds
, we observed a decrease of the critical temperature with the application of external pressure , however , the pressure coefficients @xmath75 are nearly twice as large as their respective @xmath74 values . in the case of la@xmath0co@xmath1sn@xmath2 , @xmath74 is positive .
the @xmath76-variations of @xmath6 were interpreted in the context of the eliasberg theory and discussed as a consequence of the lattice stiffening .
the results shown in this work should be of interest for understanding the @xmath95-dependent superconducting state of la@xmath0co@xmath8ru@xmath9sn@xmath2 where @xmath5 is larger than @xmath5 for la@xmath0ru@xmath1sn@xmath2 , ( e.g. , for la@xmath0coru@xmath30sn@xmath2 it is almost twice as large ) .
this unique observation is not predicted by the bcs theory and not observed in other chemically substituted superconductors .
we suggest that local disorder is responsible for the increase in @xmath6 in strongly inhomogeneous regions in the sample and/or the effect of chemical pressure when la@xmath0ru@xmath36sn@xmath2 is substituted with by co. this scenario should be verified theoretically .
the research was supported by national science centre ( ncn ) on the basis of decision no .
dec-2012/07/b / st3/03027 .
m.m.m . acknowledges support by ncn under grant dec-2013/11/b / st3/00824 .
high - pressure research at the university of california , san diego , was supported by the national nuclear security administration under the stewardship science academic alliance program through the u. s. department of energy under grant number de - na0001841 .
one of us ( a. . ) is grateful for the hospitality at the university of california , san diego ( ucsd ) .
j. p. remeika , g.p .
espinosa , a. s. cooper , h. barz , j.m .
rowel , d. b. mcwhan , j. m. vandenberg , d. e. moncton , z. fizk , l. d. woolf , h. c. hamaker , m. b. maple , g. shirane , and w. thomlinson , sol . state commun .
* 34 * , 923 ( 1980 ) ; j. l. hodeau , m. marezio , j. p. remeika , and c. h. chen , _ ibid . _ * 42 * , 97 ( 1982 ) .
a surface superconductivity effect can be observed above the bulk critical field @xmath3 for type ii superconductors .
saint - james and de gannes @xcite have shown theoretically that a surface superconducting layer with a critical field @xmath99 can exist above @xmath3 but only when the external magnetic field is parallel to the sample surface .
experimental evidence @xcite has confirmed the existence of the surface critical field @xmath100 . in case of
la@xmath30ru@xmath36sn@xmath2 the hypothetical @xmath101 ratio is closed to the calculated one , however , @xmath102 behavior observed for the surface and bulk superconductivity is not theoretically expected .
moreover , the _ high_-@xmath6 superconducting state is observed under the zero magnetic field .
the both arguments seem exclude the surface superconductivity effect in la@xmath30ru@xmath36sn@xmath2 and la@xmath30coru@xmath30sn@xmath2 .
i. martin , d. podolsky , and s. a. kivelson , phys .
b * 72 * , 060502(r ) ( 2005 ) ; m. m. maka , . ledz , k. czajka , and m. mierzejewski , phys . rev . lett . * 99 * , 147006 ( 2007 ) ; m. v. feigelman , l. b. ioffe , v. e. kravtsov , and e. a. yuzbashyan , phys . rev . lett .
98 , 027001 ( 2007 ) ; i. s. burmistrov , i. v. gornyi , and a. d. mirlin , phys . rev .
108 , 017002 ( 2012 ) . | la@xmath0co@xmath1sn@xmath2 and la@xmath0ru@xmath1sn@xmath2 were categorized as bcs superconductors . in a plot of the critical field @xmath3 vs @xmath4
, la@xmath0ru@xmath1sn@xmath2 displays a second superconducting phase at the higher critical temperature @xmath5 , characteristic of inhomogeneous superconductors , while la@xmath0co@xmath1sn@xmath2 shows bulk superconductivity below @xmath6 .
we observe a decrease in critical temperatures with external pressure and magnetic field for both compounds with @xmath7 .
the pressure dependences of @xmath6 are interpreted according to the mcmillan theory and understood to be a consequence of lattice stiffening .
the investigation of the superconducting state of la@xmath0co@xmath8ru@xmath9sn@xmath2 shows a @xmath5 that is larger then @xmath6 for @xmath10 .
this unique and unexpected observation is discussed as a result of the local disorder and/or the effect of chemical pressure when ru atoms are partially replaced by smaller co atoms . |
before beginning a discussion of shower monte carlo event generators , it is well to note that these programs are , for the most part , limited to leading order ( lo ) perturbative accuracy . on the other hand ,
there are programs available for a variety of important processes that perform purely perturbative ( non - shower ) calculations to next - to - leading order ( nlo ) accuracy .
although these programs have proved to be very useful , it is pertinent to ask why they can not tell the whole story . for this purpose , consider the cross section to produce three jets in electron - positron annihilation ( using a suitable definition of what one means by a jet ) .
the ratio of this cross section to the total cross section is the three - jet fraction , @xmath0 .
now , @xmath0 is an infrared safe observable that is amenable to calculation at nlo accuracy .
in such a calculation , the program produces simulated partonic events with three partons and others with four partons . in either case , if the parton momenta meet certain criteria , the event can be classified as a three jet event .
let us look at this calculation and ask for each jet in each three jet event what the mass of the jet is .
then we can plot the calculated probability of finding a jet in a given bin of jet mass @xmath1 , @xmath2 .
the result of this calculation is shown in figure [ fig : jetmass ] .
we see that there are two problems .
first , the cross section increases as @xmath3 . in figure
[ fig : jetmass ] , the apparent increase is limited only by the finite bin size .
second , the first bin reflects a large negative cross section .
in fact , the cross section for producing a jet with @xmath4 is negative and infinite .
clearly , this calculation is seriously at variance with the real world .
only after we integrate over @xmath1 do we get a sensible answer .
- 1.0 cm - 0.5 cm now let us look at what a parton shower monte carlo program does . in figure
[ fig : jetmass2 ] , we see that the standard parton shower monte carlo program pythia @xcite produces a sensible result in which the distribution @xmath2 peaks at about 8 gev .
this contrasts with the pure nlo prediction .
clearly it would be best to keep the nlo accuracy for @xmath0 while at the same time fixing the internal structure of the jets to be more like what one gets in pythia .
this can be done if one keeps track of what the parton shower algorithm does , expands the parton shower effects perturbatively , and subtracts the nlo contribution of the parton shower from the nlo term in the perturbative calculation .
the result labeled nlo+ps+had combines the nlo calculation with pythia @xcite .
we see that the result for @xmath2 closely follows the pure pythia result .
i do not display it as a graph , but the result of this program for just @xmath0 closely follows the pure nlo result .
the program just mentioned is for electron - positron annihilation to make three jets .
for hadron - hadron collisions , programs for several important processes are available in the package mc@nlo@xcite .
i will not say more about the technical methods involved in combining `` mc '' with `` nlo , '' but i will comment briefly on one further development later in this talk . in the rest of this talk ,
i will mainly concentrate on leading order aspects of monte carlo event generators .
- 1.0 cm - 0.5 cm
consider the parton shower picture of hadron - hadron scattering in which there is some sort of hard event , say jet production or squark - pair production .
( i leave out the description of the `` underlying event '' that accompanies the hard scattering and i omit entirely the parton shower description of the soft - scattering events that make up most of the hadron - hadron scattering cross section . )
the first thing to understand is that the parton shower description starts from the hard scattering and proceeds toward softer scatterings . for final state partons ,
one is thus working forwards in time , but for the initial state partons one is working backwards in time .
this is depicted in figure [ fig : insideout ] . in the right hand part of the figure
, @xmath5 is monte carlo time , for which @xmath6 represents the hardest interaction and @xmath7 as parton virtualities tend to zero .
thus some of the lines in the right hand picture are moving backward in physical time .
although the development of the parton shower description of hadron scattering dates from about 1980 @xcite , it was not until somewhat later that this backwards evolution scheme was developed @xcite . 1 cm i should mention that in the description that i present here , shower evolution is organized according to a hardness measure like virtuality or the transverse momentum in a splitting .
the important program herwig @xcite is organized differently , with splittings at the widest angles done first .
i do not discuss the herwig scheme here .
rather , the description here is more attuned to the program pythia @xcite .
now let us think about soft gluon radiation .
i consider three jet production in electron - positron annihilation as an example . at the born level ,
one has a @xmath8 final state .
the gluon is a color @xmath9 , but to leading order in an expansion in powers of @xmath10 , where @xmath11 is the number of colors , the gluon can be considered to be a @xmath12 state .
then the outgoing quark and the @xmath13 part of the gluon constitute a @xmath12 dipole , while the outgoing antiquark and the @xmath14 part of the gluon constitute another @xmath12 dipole , as depicted in figure [ fig : coherence ] .
( these are `` dipoles '' in the sense that there is a color charge and an opposite color charge moving away from each other .
however , the color charge separation is not small as in the traditional multipole approximation . )
1 cm the two dipoles will radiate soft gluons .
given the ( approximate ) color structure , the two dipoles radiate independently : there is no quantum interference between a gluon radiation from one dipole and radiation from the other dipole . the radiation pattern is depicted in figure [ fig : dipoleradiation ] . for each dipole , there is soft - collinear radiation that is concentrated in the directions of the two outgoing partons for that dipole .
there is also a wider angle component that is , approximately , spread over the angular region between the parton directions .
thus the wide angle dipole has soft radiation spread over a wide angular region while the narrow angle dipole has soft radiation spread over a narrow angular region .
1 cm in a parton shower monte carlo program , one can work to leading order in @xmath10 ( as parton shower programs generally do ) and make sure that the parton splitting formulas properly take into account the interference between gluons emitted from the two parts of a color dipole .
the program ariadne is based on this kind of picture @xcite . the latest version of pythia is also based on a dipole picture @xcite .
the present authors have found that the catani - seymour dipole formalism @xcite for generating the subtractions for perturbative nlo calculations is also quite useful as the basis for splittings in a parton shower @xcite .
there is another way to do this .
one can simply generate independent emissions from each parton and then impose a restriction on the angles of the emissions .
this is the method of herwig @xcite . in herwig ,
a wide angle soft gluon emission as depicted in the left - hand part of figure [ fig : dipoleradiation ] is generated first , before the splitting of the quark into a hard quark and a hard gluon .
the algorithm enforces that the angles between daughter partons in a splitting decrease for splittings generated later in the algorithm evolution .
the recognition of the importance of this ordering was important in the development of parton shower algorithms @xcite .
shower evolution can be represented using an evolution equation of the form @xmath15 here @xmath16 is a linear operator acting on a function @xmath17 giving the probability for having a given partonic state at evolution time @xmath18 .
then @xmath19 is the probability function for having a particular partonic state at a later evolution time @xmath20 .
the evolution equation expresses @xmath21 as a sum of two terms . in the first term
, the partonic state stays the same .
this is represented as a no splitting operator @xmath22 that inserts a sudakov factor giving the probability that no splitting occurred between times @xmath18 and @xmath20 .
this factor is an exponential of minus the integral from @xmath18 to @xmath20 of the differential splitting probability . in the second term
, there is a no splitting operator @xmath23 , followed by a splitting operator @xmath24 at time @xmath25 , followed by complete evolution @xmath26 for times after the splitting .
there is an integration over the intermediate time @xmath25 at which the splitting occurs .
the @xmath27 splitting operator @xmath24 uses an approximation to the qcd squared matrix element .
this shower evolution can be represented graphically as in figure [ fig : evolution ] .
the ovals represent the complete shower evolution operator @xmath21 .
the narrow rounded rectangles represent the sudakov operator @xmath28 .
the splitting is represented by a small circle with one parton coming in and two partons going out . to generate a cross section with a shower monte carlo event generator
, one can start with a hard squared matrix element for @xmath29 scattering , then apply the shower operator to the two incoming and two outgoing partons .
when the shower evolution equation is iterated , one obtains terms representing @xmath30 splittings with sudakov factors for the intervals with no splittings , as depicted in figure [ fig : iterated ] .
the standard shower depicted in figure [ fig : iterated ] has a deficiency , which is illustrated in figure [ fig : deficiency ] .
the left - hand picture depicts a term contributing to the standard shower . in this term
, there are sudakov factors and @xmath27 parton splitting functions .
if we omit the sudakov factors , we have the @xmath27 parton splittings as depicted in the middle picture . these splittings are approximations based on the splitting angles being small or one of the daughter partons having small momentum .
thus the shower splitting probability with two splittings approximates the exact squared matrix element for @xmath31 scattering .
the approximation is good in parts of the final state phase space , but not in all of it . thus one might want to replace the approximate squared matrix element of the middle picture with the exact squared matrix element of the right - hand picture .
however , if we use the exact squared matrix element , we lack the sudakov factors .
2 cm 2.5 cm one can improve the approximation as illustrated in figure [ fig : improved ] . we reweight the exact squared matrix element by the ratio of the shower approximation with sudakov factors to the shower approximation without sudakov factors .
the idea is to insert the sudakov factors into the exact squared matrix element .
this is the essential idea in the paper of catani , krauss , kuhn , and webber @xcite .
they use the @xmath32 jet algorithm to define the ratio needed to calculate the sudakov reweighting factor .
there is a further step in implementing this idea .
ckkw divide the shower evolution into two stages , @xmath33 and @xmath34 , where @xmath35 is a parameter that represents a moderate @xmath36 scale and @xmath37 represents the very small @xmath36 scale at which showers stop and hadronization is simulated . with this division
, the sudakov reweighting can be performed for the part of the shower at scale harder than @xmath35 , as depicted in figure [ fig : ckkw ] .
the first term has no splittings at scale harder than @xmath35 . in the second term
there is one splitting , generated via the exact matrix element with a sudakov correction as discussed above . in the next term
there are two splittings .
if we suppose that we do not have exact matrix elements for more than @xmath31 partons , states at scale @xmath35 with more partons are generated with the ordinary parton shower .
however , this contribution is suppressed by factors of @xmath38 .
evolution from @xmath35 to @xmath37 is done via the ordinary shower algorithm . to state the main idea of this jet number matching in a little different language
, we can consider the cross section for an observable @xmath39 . in the ckkw method
, we break @xmath40 $ ] into a sum of contributions @xmath41 $ ] from final states with @xmath42 jets at resolution scale @xmath35
. then @xmath41 $ ] is evaluated using the exact tree level matrix element for @xmath43 parton scattering , supplemented by sudakov reweighting and further supplemented by showering of the @xmath44 partons at scales softer than @xmath35 . if @xmath39 is an infrared safe observable , this method gets @xmath41 $ ] correct to the leading perturbative order , @xmath45 .
the method can be extended .
the present authors have shown ( at least for the case of electron - positron annihilation ) how to get @xmath41 $ ] for an infrared safe observable correct to next - to - leading order , @xmath46 @xcite .
the required nlo adjustments are a little complicated , so i do not discuss them here .
there is an alternative way to organize the shower improvement so as to include exact tree level matrix elements @xcite . one does not really need to split the evolution at a scale @xmath35 .
suppose that one has the exact tree level matrix elements for @xmath47 partons for @xmath48 .
then the partonic cross section at a final very soft scale @xmath37 before hadronization is the sum of the @xmath29 , @xmath49 ,
@xmath50 cross sections with sudakov factors plus one more term , which is the most important term . in the last term
, we have the sudakov improved @xmath50 squared matrix element in which the softest splitting has scale @xmath5 and we integrate over @xmath5 .
this is convoluted with the simple shower approximation for splittings softer than @xmath5 , down all the way to @xmath37 .
this is depicted in figure [ fig : ckkwalt ] .
the terms before the last one are included in the calculation but are not important because they contain the sudakov suppression for only a small number of splittings to occur down to a very soft scale @xmath37 . in the term that really matters ,
we use the sudakov improved @xmath50 squared matrix element with an arbitrary number of further splittings generated in the collinear / soft approximation , all of this with sudakov suppression factors .
parton shower event generators have proved to be an essential tool for particle physics .
these tools have been undergoing substantial improvement in recent years .
there are quite a number of very good people working on this subject and some of the most recent progress will be reported at this conference .
this work was supported in part the united states department of energy and by the swiss national science foundation ( snf ) through grant no.200020 - 109162 and by the hungarian scientific research fund grants otka k-60432 .
d. e. soper , phys .
lett . * 81 * , 2638 ( 1998 ) [ arxiv : hep - ph/9804454 ] .
t. sjostrand , s. mrenna and p. skands , jhep * 0605 * , 026 ( 2006 ) [ arxiv : hep - ph/0603175 ] .
m. kramer and d. e. soper , phys .
d * 69 * , 054019 ( 2004 ) [ arxiv : hep - ph/0306222 ] ; d. e. soper , phys . rev .
d * 69 * , 054020 ( 2004 ) [ arxiv : hep - ph/0306268 ] ; m. kramer , s. mrenna and d. e. soper , phys . rev .
d * 73 * , 014022 ( 2006 ) [ arxiv : hep - ph/0509127 ] .
s. frixione , p. nason and b. r. webber , jhep * 0308 * , 007 ( 2003 ) [ arxiv : hep - ph/0305252 ] ; s. frixione and b. r. webber , jhep * 0206 * , 029 ( 2002 ) [ arxiv : hep - ph/0204244 ] .
for example , r. odorico , nucl .
b * 172 * , 157 ( 1980 ) .
t. sjostrand , phys .
b * 157 * , 321 ( 1985 ) ; t. d. gottschalk , nucl .
b * 277 * , 700 ( 1986 ) .
g. corcella _ et al .
_ , jhep * 0101 * , 010 ( 2001 ) [ arxiv : hep - ph/0011363 ]
. l. lonnblad , comput .
commun .
* 71 * , 15 ( 1992 ) .
t. sjostrand and p. z. skands , eur .
j. c * 39 * , 129 ( 2005 ) [ arxiv : hep - ph/0408302 ] .
s. catani and m. h. seymour , nucl .
b * 485 * , 291 ( 1997 ) [ erratum - ibid .
b * 510 * , 503 ( 1997 ) ] [ arxiv : hep - ph/9605323 ] .
z. nagy and d. e. soper , jhep * 0510 * , 024 ( 2005 ) [ arxiv : hep - ph/0503053 ] ; z. nagy and d. e. soper , arxiv : hep - ph/0601021 .
g. marchesini and b. r. webber , nucl .
b * 238 * , 1 ( 1984 ) ; r. k. ellis , g. marchesini and b. r. webber , nucl .
b * 286 * , 643 ( 1987 ) [ erratum - ibid .
b * 294 * , 1180 ( 1987 ) ] .
s. catani , f. krauss , r. kuhn and b. r. webber , jhep * 0111 * , 063 ( 2001 ) [ arxiv : hep - ph/0109231 ] . | shower monte carlo event generators have played an important role in particle physics .
modern experiments would hardly be possible without them . in this talk
i discuss how qcd physics is incorporated into the mathematical structure of these programs and i outline recent developments including matching between events with different numbers of hard jets and the inclusion of next - to - leading order effects . |
the sg experiment is viewed as the standard evidence of the quantum nature of the spin of a particle .
when a beam of spin 1/2 particles , in the eigenstate @xmath0 of @xmath1 , goes through a variable magnetic field in the @xmath2 direction and the emerging particles are detected on a screen , one observes the presence of two distinct peaks corresponding to the spins in the positive and negative @xmath2 direction .
besides furnishing the evidence of the spin quantization , this experiment is considered the paradigm of the measurement process being the simplest example of coherence and entanglement ; two features of quantum mechanics of which one finds no analogue in the classical world . in the sg experiment
the particle enters the magnet in a pure state , for example , the eigenstate @xmath3 of @xmath1 , and for the effect of measurement of the @xmath4 spin component it is described as a coherent superposition of the eigenstates of @xmath4 . later on , as an effect of the interaction with the non- uniform part of the magnetic field , these spin degrees of freedom entangle with the spatial coordinates generating the state @xmath5 .
in the end , as a result of the measuring process , the quantum state collapses into one of these eigenstates . coherence can be observed by measuring the spin in the @xmath6 direction to obtain @xmath7 .
we know that spin coherence is lost as the state evolves in time and the two spatial parts become orthogonal .
since the measurement process naturally involves the partial trace over the spatial part of the initially pure global state the remaining spin part becomes a mixture .
but how is this spin coherence lost in the sg experiment ?
is it lost just when the two beams are far away ?
these are interesting questions that address the very origin of the entanglement between the states representing different degrees of freedom of the particles . here
we intend to study the evolution of the spin coherence in the sg experiment and how the entanglement between the spin and spatial coordinates is affected in the course of time to show the double role played by entanglement and coherence .
we are going to consider the usual sg model , where we just take into account one direction of the magnetic field . for discussions on these approximations
see @xcite . within this model
our hamiltonian is @xmath8 + where @xmath9 and we are not considering the uniform part of the magnetic field , that is just responsible for the spin precession .
this hamiltonian takes us to the following propagator@xcite @xmath10 \right\}\end{aligned}\ ] ] with @xmath11 and @xmath12 this propagator can be used to perform the temporal evolution of the physical state in the schrdinger prescription and we employ the superposition below as the initial state : @xmath13 here @xmath14 is the spatial part of the physical state which is not initially entangled with the spin . considering this initial spatial part as a wave packet with minimum uncertainty we have @xmath15 which is a gaussian packet with width @xmath16 and centered at the origin . performing the temporal evolution we got @xmath17 with @xmath18^{2}+\right .
\nonumber \\ & + & i\left[\frac{m}{2\hbar t}z^{2}\pm\frac{m}{\hbar t}\triangle z\left(t\right)z+ \frac{f^{2}t^{3}}{24m\hbar^{2}}+\right.\nonumber \\ & - & \left.\left.\frac{m}{2\hbar t}\left(\frac{\sigma}{\sigma\left(t\right)}\right)^{2 } \left(z\mp\overline{\triangle z}\left(t\right)\right)^{2}\right]\right\ } \label{eq:4}\end{aligned}\ ] ] and @xmath19 since the function @xmath20 appears only as a time dependent exponent of a complex phase it will not contribute to the evaluation of the probability amplitudes . as expected , after having interacted with the magnetic field the spatial and spin degrees of freedom of the particle are now entangled . within a time interval
@xmath21 the spatial part of the wave function is a gaussian whose center follows the classical trajectory with a time dependent width given by @xmath22 .
as we are interested in the spin coherence we should trace over the spatial coordinates and look at the off - diagonal elements of the density operator in the spin space which reads , for @xmath23 , @xmath24 @xmath25^{2}+\right.\\ & - & \left.\frac{1}{2}\left(\frac{\overline{\triangle z}\left(t\right ) } { \sigma\left(t\right)}\right)^{2}\right\}.\end{aligned}\ ] ] in this expression we can see that we have two terms contributing to the loss of coherence .
the first one is the distance between the centers of the packets in momentum space , as measured in unities of the spread of the packets in this space , namely @xmath26 .
the second term also measures the distance between the packets but in coordinate space instead . in this case
this measure is taken using the spread of the packet in coordinate space , @xmath22 , as the standard .
we are interested in knowing how long it takes for the spin coherence to be lost and what is the contribution of each of those two above - mentioned terms . defining a `` decoherence '' time , @xmath27 , as the time scale
within which the off - diagonal element decay to @xmath28 , we can show that@xmath29^{1/2}.\ ] ] as this expression does not tell us much if it is analyzed in its full extent we are going to do it for two particular limits , @xmath30 + in the first case we approximate the decoherence time by @xmath31 whereas in the second case we do it by @xmath32 .
+ now we want to investigate how the separation between the packets evolves during the loss of coherence .
this distance , in the coordinate representation , is given by the fraction @xmath33 which assumes the following values @xmath34 and @xmath35 we see that in both cases the packets are not well separated in space when the coherence is lost , and , therefore , the separation of the packets in momentum space must be responsible for the loss of spin coherence in the sg experiment .
this can be viewed in the figures 1 , 2 and 3 , where we plot the loss of coherence , as given by the off - diagonal elements of the density operator in the spin space , and the probability amplitudes for the two spatial parts .
these plots were made using the typical values of a sg experiment @xcite : @xmath36 ( cooper atom mass ) , @xmath37 and @xmath38 .
one should also notice that in the first case the spread of the packets in coordinate space is not relevant , @xmath39 , whereas in the second case it has to be taken in account since @xmath40 .
another way to understanding why the momentum separation is responsible for the loss of coherence is to note that in the beginning , when @xmath41 , @xmath42 while at long times , when @xmath40 , one has @xmath43 nevertheless , the momentum separation is given by @xmath44 being always linear in @xmath21 .
the only possibility the space separation between the packets exceeds their momentum separation is for long times when the space separation is no longer quadratic but linear instead .
another interesting thing to look at is the behaviour of the entanglement between the spin and spatial coordinates .
as we are dealing with a global pure state we can use either the von neumann @xcite or the linear entropy of one of the subsystems as a measure of entanglement .
we shall develop the latter in what follows .
usually we have a bipartite system with the global state given by @xmath45 + and as the coefficients @xmath46 and @xmath47 vary the degree of entanglement of the state of the system also changes .
two extremes cases are the state of maximum entanglement , the bell state , in which the coefficients are both @xmath48 and the separable state ( non - entangled ) in which one of the coefficients is zero .
in our case we have a different physical situation since what is varying is not the coefficients but the states themselves .
in other words , we are varying the states @xmath49 and @xmath50 that start `` parallel '' to one another and become orthogonal as time evolves .
the entanglement between the spatial and spin degrees of freedom can be given , in terms of the linear entropy of the spin subsystem , by @xmath51 .
in this expression we can see the close connection between entanglement and coherence ; as one of them increases the other is fated to decrease .
this behaviour can be observed in fig .
4 where we have ploted @xmath52 and @xmath53 in the typical sg experiment .
as expected , the entanglement vanishes at the beginning and increases as the states become orthogonal , having as a consequence the loss of spin coherence .
this example clearly shows the double role played by entanglement ; it allows for the creation of superpositions of macroscopic objects but at the same time makes subsystems lose their quantum mechanical coherence . for the sake of completeness
, we should mention that one could also think about tracing out the spin degrees of freedom and see the coherence in the coordinate representation as an interference pattern between the two beams in a double slit experiment .
the only `` problem '' here is that the spin degrees of freedom are always orthogonal which makes the reduced density matrix in coordinate space diagonal and destroy the possible coherence at any time . finally , since the present analysis makes the connection between entanglement and coherence clear only for pure states it would be desirable to extend it to mixed states as well .
we have given a simple example of the connection between entanglement and coherence showing how the spin coherence is lost in a sg experiment as the spatial and spin degrees of freedom become entangled .
we have also observed that , contrary to the expectations , the spin coherence is lost much faster than the two beams become clearly separated .
it is worth commenting that we have used the word decoherence with a somewhat different meaning from that used in the current literature .
here we do not have a real environment as the cause of decoherence , and therefore are referring to a possible case of reversible decoherence .
recovery of coherence should be achieved simply by recombining the two beams ( see ref.s @xcite for more details on this ) .
1 d. e. platt , am .
60 ( 4 ) , april ( 1992 ) .
p. alstrm , p. hjorth and r. mattuck , am .
50 ( 8) , aug .
( 1982 ) . l. s. schulman , techniques and applications of path integration , first edition ( wiley - vhc berlin ) .
b. friedrich and d. herschbach , physics today december , 53 ( 2003 )
. g. b. roston , m. casas , a. plastino , and a. r. plastino , eur .
* 26 * , 657 ( 2005 ) .
berthold - georg englert , et .
al . , found .
18 , 1045 ( 1998 ) .
thiago r. de oliveira , a. o. caldeira , phys
a 73 , 042502 ( 2006 ) .
m. dugic , eur .
j. d 29 , 173 ( 2004 ) . | we give a simple example of the tight connection between entanglement and coherence for pure bipartite systems showing the double role played by entanglement ; it allows for the creation of superpositions of macroscopic objects but at the same time makes subsystems lose their quantum mechanical coherence . for this
we study the time evolution of the spin coherence in the stern - gerlach ( sg ) experiment .
we also show that , contrary to the naive intuition , the spin coherence is lost before the two beams become separated in the spatial coordinates . |
consider a compact oriented surface @xmath0 of genus @xmath1 together with a finite subset @xmath2 , such that @xmath3 .
the moduli space @xmath4 is the set of all @xmath5-pointed riemann surfaces of genus @xmath1 up to isomorphism .
its universal cover can be identified with the teichmller space @xmath6 , which parametrizes complex structures on @xmath0 up to isotopy ( relative to @xmath5 ) ; equivalently , @xmath6 parametrizes isomorphism classes of @xmath7-marked riemann surfaces .
thus , @xmath4 is the quotient of @xmath6 under the action of the mapping class group @xmath8 .
as @xmath6 is contractible ( teichmller @xcite ) , we also have that @xmath9 . however , @xmath10 acts on @xmath6 discontinuously but with finite stabilizers .
thus , @xmath4 is naturally an orbifold and @xmath9 must be intended in the orbifold category . as compact riemann surfaces are complex algebraic curves
, @xmath4 has an algebraic structure and is in fact a deligne - mumford stack , which is the algebraic analogue of an orbifold .
the underlying space @xmath11 ( forgetting the isotropy groups ) is a quasi - projective variety .
the interest for enumerative geometry of algebraic curves naturally led to seeking for a suitable compactification of @xmath4 .
deligne and mumford @xcite understood that it was sufficient to consider algebraic curves with mild singularities to compactify @xmath4 .
in fact , their compactification @xmath12 is the moduli space of @xmath5-pointed stable ( algebraic ) curves of genus @xmath1 , where a complex projective curve @xmath13 is `` stable '' if its only singularities are nodes ( that is , in local analytic coordinates @xmath13 looks like @xmath14 ) and every irreducible component of the smooth locus of @xmath15 has negative euler characteristic . the main tool to prove the completeness of @xmath12 is the stable reduction theorem , which essentially says that a smooth holomorphic family @xmath16 of @xmath5-pointed riemann surfaces of genus @xmath1 over the pointed disc can be completed to a family over @xmath17 ( after a suitable change of base @xmath18 ) using a stable curve .
the beauty of @xmath12 is that it is smooth ( as an orbifold ) and that its coarse space @xmath19 is a projective variety ( mumford @xcite , gieseker @xcite , knudsen @xcite @xcite , kollr @xcite and cornalba @xcite ) .
the map @xmath20 that forgets the @xmath21-point can be identified to the universal curve over @xmath12 and is the first example of tautological map .
moreover , @xmath12 has a natural algebraic stratification , in which each stratum corresponds to a topological type of curve : for instance , smooth curves correspond to the open stratum @xmath4 . as another example : irreducible curves with one node correspond to an irreducible locally closed subvariety of ( complex ) codimension @xmath22 , which is the image of the ( generically @xmath23 ) tautological boundary map @xmath24 that glues @xmath25 to @xmath26 .
thus , every stratum is the image of a ( finite - to - one ) tautological boundary map , and thus is isomorphic to a finite quotient of a product of smaller moduli spaces .
teichmller theorists are more interested in compactifying @xmath6 rather than @xmath4 .
one of the most popular way to do it is due to thurston ( see @xcite ) : the boundary of @xmath6 is thus made of projective measured laminations and it is homeomorphic to a sphere .
clearly , there can not be any clear link between a compactification of @xmath6 and of @xmath4 , as the infinite discrete group @xmath10 would not act discontinuously on a compact boundary @xmath27 .
thus , the @xmath10-equivariant bordification of @xmath6 whose quotient is @xmath4 can not be compact . a way to understand it is to endow @xmath4 ( and @xmath6 ) with the weil - petersson metric @xcite and to show that its completion is exactly @xmath12 @xcite .
hence , the weil - petersson completion @xmath28 can be identified to the set of @xmath7-marked stable riemann surfaces .
similarly to @xmath12 , also @xmath28 has a stratification by topological type and each stratum is a ( finite quotient of a ) product of smaller teichmller spaces .
the moduli space @xmath12 comes equipped with natural vector bundles : for instance , @xmath29 is the holomorphic line bundle whose fiber at @xmath30 $ ] is @xmath31 .
chern classes of these line bundles and their push - forward through tautological maps generate the so - called tautological classes ( which can be seen in the chow ring or in cohomology ) .
the @xmath32 classes were first defined by mumford @xcite and morita @xcite and then modified ( to make them behave better under tautological maps ) by arbarello and cornalba @xcite .
the @xmath33 classes were defined by e.miller @xcite and their importance was successively rediscovered by witten @xcite .
the importance of the tautological classes is due to the following facts ( among others ) : * their geometric meaning appears quite clear * they behave very naturally under the tautological maps ( see , for instance , @xcite ) * they often occur in computations of enumerative geometry ; that is , poincar duals of interesting algebraic loci are often tautological ( see @xcite ) but not always ( see @xcite ) ! * they are defined on @xmath12 for every @xmath1 and @xmath5 ( provided @xmath34 ) , and they generate the stable cohomology ring over @xmath35 due to madsen - weiss s solution @xcite of mumford s conjecture ( see section [ ss : stability ] ) * there is a set of generators ( @xmath33 s and @xmath32 s ) which have non - negativity properties ( see @xcite and @xcite ) * they are strictly related to the weil - petersson geometry of @xmath12 ( see @xcite , @xcite , @xcite and @xcite ) .
one way to analyze the ( co)homology of @xmath4 , and so of @xmath10 , is to construct a highly connected simplicial complex on which @xmath10 acts .
this is usually achieved by considering complexes of disjoint , pairwise non - homotopic simple closed curves on @xmath36 with suitable properties ( for instance , harvey s complex of curves @xcite ) .
if @xmath5 is nonempty ( or if @xmath0 has boundary ) , then one can construct a complex using systems of homotopically nontrivial , disjoint arcs joining two ( not necessarily distinct ) points in @xmath5 ( or in @xmath37 ) , thus obtaining the arc complex @xmath38 ( see @xcite ) .
it has an `` interior '' @xmath39 made of systems of arcs that cut @xmath36 in discs ( or pointed discs ) and a complementary `` boundary '' @xmath40 .
an important result , which has many fathers ( harer - mumford - thurston @xcite , penner @xcite , bowditch - epstein @xcite ) , says that @xmath41 is @xmath10-equivariantly homeomorphic to @xmath42 ( where @xmath43 is the standard simplex in @xmath44 ) .
thus , we can transfer the cell structure of @xmath41 to an ( orbi)cell structure on @xmath45 .
the homeomorphism is realized by coherently associating a weighted system of arcs to every @xmath5-marked riemann surface , equipped with a decoration @xmath46 .
there are two traditional ways to do this : using the flat structure arising from a jenkins - strebel quadratic differential ( harer - mumford - thurston ) with prescribed residues at @xmath5 or using the hyperbolic metric coming from the uniformization theorem ( penner and bowditch - epstein ) .
quite recently , several other ways have been introduced ( see @xcite , @xcite , @xcite and @xcite ) . to better understand the homeomorphism between @xmath41 and @xmath47 , it is often convenient to adopt a dual point of view , that is to think of weighted systems of arcs as of metrized graphs @xmath48 , embedded in @xmath36 through a homotopy equivalence .
this can be done by picking a vertex in each disc cut by the system of arcs and joining these vertices by adding an edge transverse to each arc .
what we obtain is an @xmath7-marked metrized ribbon graph .
thus , points in @xmath49 correspond to metrized @xmath5-marked ribbon graphs of genus @xmath1 .
this point of view is particularly useful to understand singular surfaces ( see also @xcite , @xcite , @xcite , @xcite , @xcite , @xcite and @xcite ) .
the object dual to a weighted system of arcs in @xmath40 is a collection of data that we called an @xmath7-marked `` enriched '' ribbon graph .
notice that an @xmath5-marked `` enriched '' metrized ribbon graph does not carry all the information needed to construct a stable riemann surface .
hence , the map @xmath50 is not a injective on the locus of singular curves , but still it is a homeomorphism on a dense open subset .
the utility of the @xmath10-equivariant homotopy equivalence @xmath51 is the possibility of making topological computations on @xmath41 .
for instance , harer @xcite determined the virtual cohomological dimension of @xmath10 ( and so of @xmath4 ) using the high connectivity of @xmath52 and he has established that @xmath10 is a virtual duality group , by showing that @xmath52 is spherical .
an analysis of the singularities of @xmath53 is in @xcite .
successively , harer - zagier @xcite and penner @xcite have computed the orbifold euler characteristic of @xmath4 , where by `` orbifold '' we mean that a cell with stabilizer @xmath54 has euler characteristic @xmath55 .
because of the cellularization , the problem translates into enumerating @xmath5-marked ribbon graphs of genus @xmath1 and counting them with the correct sign .
techniques for enumerating graphs and ribbon graphs ( see , for instance , @xcite ) have been known to physicists for long time : they use asymptotic expansions of gaussian integrals over spaces of matrices .
the combinatorics of iterated integrations by parts is responsible for the appearance of ( ribbon ) graphs ( wick s lemma ) .
thus , the problem of computing @xmath56 can be reduced to evaluating a matrix integral ( a quick solution is also given by kontsevich in appendix d of @xcite ) . as @xmath45
is not just homotopy equivalent to @xmath57 but actually homeomorphic ( through a piecewise - linear real - analytic diffeomorphism ) , it is clear that one can try to rephrase integrals over @xmath4 as integrals over @xmath57 , that is as sums over maximal systems of arcs of integrals over a single simplex .
this approach looked promising in order to compute weil - petersson volumes ( see penner @xcite ) .
kontsevich @xcite used it to compute volumes coming from a `` symplectic form '' @xmath58 , thus solving witten s conjecture @xcite on the intersection numbers of the @xmath33 classes .
however , in witten s paper @xcite matrix integrals entered in a different way .
the idea was that , in order to integrate over the space of all conformal structures on @xmath0 , one can pick a random decomposition of @xmath0 into polygons , give each polygon a natural euclidean structure and extend it to a conformal structure on @xmath0 , thus obtaining a `` random '' point of @xmath4 .
refining the polygonalization of @xmath0 leads to a measure on @xmath4 .
matrix integrals are used to enumerate these polygonalizations .
witten also noticed that this refinement procedure may lead to different limits , depending on which polygons we allow .
for instance , we can consider decompositions into @xmath59 squares , or into @xmath59 squares and @xmath60 hexagons , and so on .
dualizing this last polygonalization , we obtain ribbon graphs embedded in @xmath0 with @xmath59 vertices of valence @xmath61 and @xmath60 vertices of valence @xmath62 . the corresponding locus in @xmath41
is called a witten subcomplex .
kontsevich @xcite and penner @xcite proved that witten subcomplexes obtained by requiring that the ribbon graphs have @xmath63 vertices of valence @xmath64 can be oriented ( see also @xcite ) and they give cycles in @xmath65 , which are denoted by @xmath66 .
the @xmath67-volumes of these @xmath66 are also computable using matrix integrals @xcite ( see also @xcite ) . in @xcite ,
kontsevich constructed similar cycles using structure constants of finite - dimensional cyclic @xmath68-algebras with positive - definite scalar product and he also claimed that the classes @xmath69 ( restriction of @xmath66 to @xmath4 ) are poincar dual to tautological classes . this last statement ( usually called witten - kontsevich s conjecture ) was settled independently by igusa @xcite @xcite and mondello @xcite , while very little is known about the nature of the ( non - homogeneous ) @xmath70-classes .
the key point of all constructions of a ribbon graph out of a surface is that @xmath5 must be nonempty , so that @xmath36 can be retracted by deformation onto a graph .
in fact , it is not difficult to see that the spine construction of penner and bowditch - epstein can be performed ( even in a more natural way ) on hyperbolic surfaces @xmath71 with geodesic boundary .
the associated cellularization of the corresponding moduli space is due to luo @xcite ( for smooth surfaces ) and by mondello @xcite ( also for singular surfaces , using luo s result ) .
the interesting fact ( see @xcite and @xcite ) is that gluing semi - infinite cylinders at @xmath72 produces ( conformally ) punctured surfaces that `` interpolate '' between hyperbolic surfaces with cusps and flat surfaces arising from jenkins - strebel differentials . in sections [ ss : arcs ] and [ ss : ribbon ]
, we carefully define systems of arcs and ribbon graphs , both in the singular and in the nonsingular case , and we explain how the duality between the two works . moreover , we recall harer s results on @xmath39 and @xmath40 and we state a simple criterion for compactness inside @xmath57 . in sections [ ss : deligne - mumford ] and
[ ss : system ] , we describe the deligne - mumford moduli space of curves and the structure of its boundary , the associated stratification and boundary maps . in [ ss : augmented ] , we explain how the analogous bordification of the teichmller space @xmath28 can be obtained as completion with respect to the weil - petersson metric . tautological classes and rings are introduced in [ ss : tautological ] and kontsevich s compactification of @xmath4 is described in [ ss : konts ] . in [ ss : hmt ] , we explain and sketch a proof of harer - mumford - thurston cellularization of the moduli space and we illustrate the analogous result of penner - bowditch - epstein in [ ss : pbe ] . in [ ss : boundary ] , we quickly discuss the relations between the two constructions using hyperbolic surfaces with geodesic boundary .
in [ ss : witten ] , we define witten subcomplexes and witten cycles and we prove ( after kontsevich ) that @xmath67 orients them .
we sketch the ideas involved in the proof the witten cycles are tautological in section [ ss : witten - tautological ] .
finally , in [ ss : stability ] , we recall harer s stability theorem and we exhibit a combinatorial construction that shows that witten cycles are stable .
the fact ( and probably also the construction ) is well - known and it is also a direct consequence of witten - kontsevich s conjecture and miller s work .
it is a pleasure to thank shigeyuki morita , athanase papadopoulos and robert c. penner for the stimulating workshop `` teichmller space ( classical and quantum ) ''
they organized in oberwolfach ( may 28th - june 3rd , 2006 ) and the mfo for the hospitality .
i would like to thank enrico arbarello for all i learnt from him about riemann surfaces and for his constant encouragement .
let @xmath0 be a compact oriented differentiable surface of genus @xmath1 with @xmath73 distinct marked points @xmath74 .
we will always assume that the euler characteristic of the punctured surface @xmath36 is negative , that is @xmath75 .
this restriction only rules out the cases in which @xmath36 is the sphere with less than @xmath76 punctures .
let @xmath77 be the group of orientation - preserving diffeomorphisms of @xmath0 that fix @xmath5 pointwise .
the _ mapping class group _
@xmath10 is the group of connected components of @xmath77 . in what follows
, we borrow some notation and some ideas from @xcite .
an oriented _ arc _ in @xmath0 is a smooth path @xmath78\rar s$ ] such that @xmath79)\cap x= \{\ora{\a}(0),\ora{\a}(1)\}$ ] , up to reparametrization .
let @xmath80 be the space of oriented arcs in @xmath0 , endowed with its natural topology .
define @xmath81 to be the orientation - reversing operator and we will write @xmath82 .
call @xmath83 the @xmath84-orbit of @xmath85 and denote by @xmath86 the ( quotient ) space of @xmath84-orbits in @xmath80 .
a _ system of @xmath87-arcs _ in @xmath0 is a collection @xmath88 of @xmath89 unoriented arcs such that : * if @xmath90 , then the intersection of @xmath91 and @xmath92 is contained in @xmath5 * no arc in @xmath93 is homotopically trivial * no pair of arcs in @xmath93 are homotopic to each other .
we will denote by @xmath94 the _ complementary subsurface _ of @xmath0 obtained by removing @xmath95 .
each connected component of the space of systems of @xmath87-arcs @xmath96 is clearly contractible , with the topology induced by the inclusion @xmath97 .
let @xmath98 be the set of homotopy classes of systems of @xmath89 arcs , that is @xmath99 .
the _ arc complex _ is the simplicial complex @xmath100 .
we will implicitly identify arc systems @xmath93 and @xmath101 that are homotopic to each other .
similarly , we will identify the isotopic subsurfaces @xmath94 and @xmath102 .
an arc system @xmath103 _ fills _
( resp . _ quasi - fills _ )
@xmath0 if @xmath94 is a disjoint union of subsurfaces homeomorphic to discs ( resp .
discs and pointed discs ) .
it is easy to check that the star of @xmath93 is finite if and only if @xmath93 quasi - fills @xmath0 . in this case
, we also say that @xmath93 is a _ proper _ simplex of @xmath38 denote by @xmath104 the subcomplex of non - proper simplices and let @xmath105 be the collection of proper ones .
we denote by @xmath52 and @xmath106 the topological realizations of @xmath40 and @xmath38 .
we will use the symbol @xmath41 to mean the complement of @xmath52 inside @xmath106 .
the realization @xmath106 of the arc complex can be endowed with two natural topologies ( as is remarked in @xcite , @xcite and @xcite ) .
the former ( which we call _ standard _ ) is the finest topology that makes the inclusions @xmath107 continuous for all @xmath103 ; in other words , a subset @xmath108 is declared to be open if and only if @xmath109 is open for every @xmath103 .
the latter topology is induced by the path _ metric _ @xmath110 , which is the largest metric that restricts to the euclidean one on each closed simplex .
the two topologies are the same where @xmath106 is locally finite , but the latter is coarser elsewhere .
we will always consider all realizations to be endowed with the metric topology .
for every system of arcs @xmath103 , define @xmath111 to be the largest isotopy class of open subsurfaces of @xmath0 such that * every arc in @xmath93 is contained in @xmath111 * @xmath93 quasi - fills @xmath111 .
the _ visible subsurface _
@xmath111 can be constructed by taking the union of a thickening a representative of @xmath93 inside @xmath0 and all those connected components of @xmath94 which are homeomorphic to discs or punctured discs ( this construction appears first in @xcite ) .
we will always consider @xmath111 as an open subsurface ( up to isotopy ) , homotopically equivalent to its closure @xmath112 , which is an embedded surface with boundary .
one can rephrase [ sss : proper ] by saying that @xmath93 is proper if and only if all @xmath0 is @xmath93-visible .
we call _ invisible subsurface _ @xmath113 associated to @xmath93 the union of the connected components of @xmath114 which are not unmarked cylinders .
we also say that a marked point @xmath115 is ( in)visible for @xmath93 if it belongs to the @xmath93-(in)visible subsurface . a maximal system of arcs @xmath103 is also called an _ ideal triangulation _ of @xmath0 .
in fact , it is easy to check that , in this case , each component of @xmath116 bounded by three arcs and so is a `` triangle '' .
( the term `` ideal '' comes from the fact that one often thinks of @xmath7 as a hyperbolic surface with cusps at @xmath5 and of @xmath93 as a collection of hyperbolic geodesics . )
it is also clear that such an @xmath93 is proper .
a simple calculation with the euler characteristic of @xmath0 shows that an ideal triangulation is made of exactly @xmath117 arcs .
consider the baricentric subdivision @xmath118 , whose @xmath119-simplices are chains @xmath120 . there is an obvious piecewise - affine homeomorphism @xmath121 , that sends a vertex @xmath122 to the baricenter of @xmath123 .
call @xmath124 the subcomplex of @xmath118 , whose simplices are chains of simplices that belong to @xmath39 . clearly , @xmath125 is contained in @xmath126 through the homeomorphism above .
it is a general fact that there is a deformation retraction of @xmath41 onto @xmath127 : on each simplex of @xmath128 this is given by projecting onto the face contained in @xmath127 .
it is also clear that the retraction is @xmath10-equivariant . in the special case of @xmath129 ,
a proper system contains at least @xmath130 arcs ; whereas a maximal system contains exactly @xmath131 arcs .
thus , the ( real ) dimension of @xmath127 is @xmath132 .
[ prop : spine ] if @xmath129 , the spine @xmath127 has dimension @xmath133 . for every arc system @xmath135 , denote by @xmath136 the subset @xmath137 of @xmath138 .
the action of @xmath84 clearly restricts to @xmath136 . for each @xmath139 , the orientation of @xmath0 induces a cyclic ordering of the oriented arcs in @xmath136 outgoing from @xmath115 .
if @xmath140 starts at @xmath115 , then define @xmath141 to be the oriented arc in @xmath136 outgoing from @xmath115 that comes just _ before _
moreover , @xmath142 is defined by @xmath143 .
if we call @xmath144 the orbits of @xmath136 under the action of @xmath145 , then * @xmath146 can be identified with @xmath93 * @xmath147 can be identified with the set of @xmath93-visible marked points * @xmath148 can be identified to the set of connected components of @xmath149 .
denote by @xmath150_t$ ] the @xmath145-orbit of @xmath140 , so that @xmath150_1=\a_j$ ] and @xmath150_\infty$ ] is the starting point of @xmath140 , whereas @xmath150_0 $ ] is the component of @xmath149 adjacent to @xmath92 and which induces the orientation @xmath140 on it .
there is a natural right action of the mapping class group @xmath151 & & \mathcal{a}(s , x ) \\
( \a , g ) \ar@{|->}[rr ] & & \a\circ g } \ ] ] the induced action on @xmath38 preserves @xmath40 and so @xmath39 .
it is easy to see that the stabilizer ( under @xmath10 ) of a simplex @xmath93 fits in the following exact sequence @xmath152 where @xmath153 is the group of permutations of @xmath93 and @xmath154 is the mapping class group of orientation - preserving diffeomorphisms of @xmath94 _ with compact support _ that fix @xmath5 .
define the image of @xmath155 to be the _ automorphism group of @xmath93_. we can immediately conclude that @xmath93 is proper if and only if @xmath156 is finite ( equivalently , if and only if @xmath154 is trivial ) .
a point @xmath157 consists of a map @xmath158 $ ] such that * the support of @xmath159 is a simplex @xmath160 * @xmath161 .
we will call @xmath159 the _ ( projective ) weight _ of @xmath93 .
a _ weight _ for @xmath93 is a point of @xmath162 , that is a map @xmath163 with support on @xmath93 .
call @xmath159 its associated projective weight .
we are going to prove a simple criterion for a subset of @xmath57 to be compact .
call @xmath164 the set of free homotopy classes of simple closed curves on @xmath36 , which are neither contractible nor homotopic to a puncture .
define the `` intersection product '' @xmath165 as @xmath166 , where @xmath167 is the _ geometric _ intersection number .
we will also refer to @xmath168 as to the _ length _ of @xmath169 at @xmath159 .
consequently , we will say that the _ systol _ at @xmath159 is @xmath170 clearly , the function @xmath171 descends to @xmath172 [ lemma : compact ] a closed subset @xmath173 is _ compact _ if and only if @xmath174 such that @xmath175)\geq \e$ ] for all @xmath176\in k$ ] .
in @xmath177 we easily have @xmath178 , where @xmath179 is the @xmath180-distance .
similarly , in @xmath106 we have @xmath181 where @xmath182 .
the same holds in @xmath53 .
thus , if @xmath183\in\ao(s , x)/\g(s , x)$ ] , then @xmath184 is compact for every @xmath185 .
as @xmath57 contains finitely many cells , we conclude that @xmath186 is compact .
vice versa , if @xmath187 is not bounded from below , then we can find a sequence @xmath188 \subset k$ ] such that @xmath189 .
thus , @xmath188 $ ] approaches @xmath53 and so is divergent in @xmath57 .
let @xmath43 be the standard simplex in @xmath44 .
the _ boundary weight map _
@xmath190 is the piecewise - linear map that sends @xmath191_{\infty}+[\ola{\a}]_{\infty}$ ] .
the projective boundary weight map @xmath192 instead sends @xmath193_{\infty}+\frac{1}{2 } [ \ola{\a}]_{\infty}$ ] .
a few things are known about the topology of @xmath106 . * the space of proper arc systems @xmath41
can be naturally given the structure of piecewise - affine topological manifold with boundary ( hubbard - masur @xcite , credited to whitney ) of ( real ) dimension @xmath194 .
* the space @xmath41 is @xmath10-equivariantly homeomorphic to @xmath47 , where @xmath6 is the teichmller space of @xmath7 ( see [ sss : teichmueller ] for definitions and section [ sec : triangulation ] for an extensive discussion on this result ) , and so is contractible .
this result could also be probably extracted from @xcite , but it is first more explicitly stated in harer @xcite ( who attributes it to mumford and thurston ) , penner @xcite and bowditch - epstein @xcite . as the moduli space of @xmath5-marked riemann surfaces of genus @xmath1 can be obtained as @xmath195 ( see [ sss : moduli ] ) , then @xmath9 in the orbifold category . *
the space @xmath52 is homotopy equivalent to an infinite wedge of spheres of dimension @xmath196 ( harer @xcite ) .
results ( b ) and ( c ) are the key step in the following . @xmath10 is a virtual duality group ( that is , it has a subgroup of finite index which is a duality group ) of dimension @xmath197 for @xmath73 ( and @xmath198 for @xmath199 ) . actually , it is sufficient to work with @xmath129 , in which case the upper bound is given by ( b ) and proposition [ prop : spine ] , and the duality by ( c ) .
a _ graph _ @xmath54 is a triple @xmath200 , where @xmath201 is a finite set , @xmath202 is a fixed - point - free involution and @xmath203 is an equivalence relation on @xmath201 . in ordinary language * @xmath201 is the set of _ oriented edges _ of the graph * @xmath84 is the orientation - reversing involution of @xmath201 , so that the set of unoriented edges is @xmath204 * two oriented edges are equivalent if and only if they come out from the same vertex , so that the set @xmath205 of vertices is @xmath206 and the valence of @xmath207 is exactly @xmath208 . a _ ribbon graph _
@xmath48 is a triple @xmath209 , where @xmath201 is a ( finite ) set , @xmath202 is a fixed - point - free involution and @xmath202 is a permutation .
define @xmath210 and call @xmath211 the set of orbits of @xmath145 and @xmath212_t : e\rar e_t$ ] the natural projection .
a disjoint union of two ribbon graphs is defined in the natural way .
given a ribbon graph @xmath48 , the underlying ordinary graph @xmath213 is obtained by declaring that oriented edges in the same @xmath142-orbit are equivalent and forgetting about the precise action of @xmath142 .
in ordinary language , a ribbon graph is an ordinary graph endowed with a cyclic ordering of the oriented edges outgoing from each vertex .
the @xmath214-orbits are sometimes called _
holes_. a _ connected component _ of @xmath48 is an orbit of @xmath215 under the action of @xmath216 .
the _ euler characteristic _ of a ribbon graph @xmath48 is @xmath217 and its _ genus _
is @xmath218 .
a _ ( ribbon ) tree _ is a connected ( ribbon ) graph of genus zero with one hole .
let @xmath219 be a ribbon graph and let @xmath220 be a nonempty subset of edges .
the _ subgraph @xmath221 _ is given by @xmath222 , where @xmath223 and @xmath224 are the induced operators ( that is , for every @xmath225 we define @xmath226 , where @xmath227 ) .
similarly , the _ quotient @xmath228 _ is @xmath229 , where @xmath230 and @xmath231 are the operators induced on @xmath232 and @xmath233 is defined accordingly .
a _ new vertex _ of @xmath228 is a @xmath233-orbit of @xmath234 , which is not a @xmath142-orbit .
a _ bicolored graph _
@xmath235 is a finite connected graph with a partition @xmath236 of its vertices .
we say that @xmath235 is _ reduced _ if no two vertices of @xmath237 are adjacent .
if not differently specified , we will always understand that bicolored graphs are reduced .
if @xmath235 contains an edge @xmath238 that joins @xmath239 , then we can obtain a new graph @xmath240 _ merging _ @xmath241 and @xmath242 along @xmath238 into a new vertex @xmath243 ( by simply forgetting @xmath244 and @xmath245 and by declaring that vertices outgoing from @xmath241 are equivalent to vertices outgoing from @xmath242 ) .
if @xmath235 comes equipped with a function @xmath246 , then @xmath247 is defined so that @xmath248 if @xmath249 , or @xmath250 if @xmath251 .
as merging reduces the number of edges , we can iterate the process only a finite number of times .
the result is independent of the choice of which edges to merge first and is a reduced graph @xmath252 ( possibly with a @xmath253 ) .
an _ enriched @xmath5-marked ribbon graph @xmath254 _ is the datum of * a connected bicolored graph @xmath255 * a ribbon graph @xmath48 plus a bijection @xmath256 * an ( invisible ) genus function @xmath257 * a map @xmath258 such that the restriction @xmath259 is bijective * an injection @xmath260 ( vertices of @xmath261 in the image are called _ special _ ) for every @xmath262 that satisfy the following properties : * for every @xmath262 and @xmath263 we have @xmath264 ( i.e. no more than one marking or one node at each vertex of @xmath261 ) * @xmath265 + @xmath266 for every @xmath267 ( stability condition ) * every non - special vertex of @xmath261 must be at least trivalent for all @xmath262 .
we say that @xmath254 is _ reduced _ if @xmath235 is .
if the graph @xmath235 is not reduced , then we can merge two vertices of @xmath235 along an edge of @xmath235 and obtain a new enriched @xmath5-marked ribbon graph .
@xmath268 and @xmath269 are considered equivalent if they are related by a sequence of merging operations .
it is clear that each equivalence class can be identified to its reduced representative . unless differently specified
, we will always refer to an enriched graph as the canonical reduced representative .
the total _ genus _ of @xmath254 is @xmath270 . in figure
[ fig : bicolored ] , the genus of each vertex is written inside , @xmath271 and @xmath272 are marking the two holes of @xmath48 ( sitting in different components ) , whereas @xmath273 is an invisible marked point .
moreover , @xmath274 ( resp .
@xmath275 ) are distinct ( special ) vertices of the visible component of genus @xmath276 ( resp . of genus @xmath76 ) .
the total genus of the associated @xmath254 is @xmath277 .
if an edge @xmath238 of @xmath235 joins @xmath262 and @xmath278 and this edge is marked by the special vertex @xmath263 , then we will say , for brevity , that @xmath238 joins @xmath279 to @xmath21 . an enriched @xmath5-marked ribbon graph is _ nonsingular _ if @xmath235 consists of a single visible vertex .
equivalently , an enriched nonsingular @xmath5-marked ribbon graph consists of a connected ribbon graph @xmath48 together with an injection @xmath280 , whose image is exactly @xmath281 , such that non - special vertices are at least trivalent and @xmath282 .
a _ morphism _ of nonsingular @xmath5-marked ribbon graphs @xmath283 is an injective map @xmath284 such that * @xmath285 commutes with @xmath84 , @xmath214 and respects the @xmath5-marking * @xmath286 is a disjoint union of trees , where @xmath287 .
notice that , as @xmath285 preserves the @xmath5-markings ( which are _ injections _
@xmath288 ) , then each component of @xmath289 may contain at most one special vertex .
vice versa , if @xmath48 is a nonsingular @xmath5-marked ribbon graph and @xmath290 such that @xmath221 is a disjoint union of trees ( each one containing at most a special vertex ) , then the inclusion @xmath291 induces a morphism of nonsingular ribbon graphs @xmath292 .
a morphism is an _ isomorphism _ if and only if @xmath285 is bijective .
@xmath293 is the small category whose objects are nonsingular @xmath5-marked ribbon graphs @xmath48 ( where we assume that @xmath215 is contained in a fixed countable set ) with the morphisms defined above .
we use the symbol @xmath294 to denote the full subcategory of ribbon graphs of genus @xmath1 .
a topological realization @xmath295 of the graph @xmath296 is the one - dimensional cw - complex obtained from @xmath297 ( where @xmath298 $ ] ) by identifying * @xmath299 for all @xmath300 and @xmath301 * @xmath302 whenever @xmath303 . a _ topological realization _
@xmath304 of the nonsingular @xmath5-marked ribbon graph @xmath219 is the oriented surface obtained from @xmath305 ( where @xmath306/i\times\{\infty\}$ ] ) by identifying * @xmath307 for all @xmath301 * @xmath308 for all @xmath301 and @xmath309 $ ] .
if @xmath54 is the ordinary graph underlying @xmath48 , then there is a natural embedding @xmath310 , which we call the _
spine_. the points at infinity in @xmath304 are called _ centers _ of the holes and can be identified to @xmath311 .
thus , @xmath304 is naturally an @xmath5-marked surface .
notice that a morphism of nonsingular @xmath5-marked ribbon graphs @xmath312 induces an isotopy class of orientation - preserving homeomorphisms @xmath313 that respect the @xmath5-marking .
an _ @xmath7-marking _ of the nonsingular @xmath5-marked ribbon graph @xmath48 is an orientation - preserving homeomorphism @xmath314 , compatible with @xmath315 .
define @xmath316 to be the category whose objects are @xmath7-marked nonsingular ribbon graphs @xmath317 and whose morphisms @xmath318 are morphisms @xmath312 such that @xmath319 is homotopic to @xmath320 . as usual , there is a right action of the mapping class group @xmath10 on @xmath316 and @xmath321 is equivalent to @xmath294 .
let @xmath322 be a proper arc system and let @xmath323 the corresponding operators on the set of oriented arcs @xmath136 .
the _ ribbon graph dual to @xmath93 _ is @xmath324 , which comes naturally equipped with an @xmath5-marking ( see [ sss : sigma - arc ] ) .
define the @xmath7-marking @xmath314 in the following way .
fix a point @xmath325 in each component @xmath326 of @xmath116 ( which must be exactly the marked point , if the component is a pointed disc ) and let @xmath285 send it to the corresponding vertex @xmath326 of @xmath304 . for each arc @xmath327 , consider a transverse path @xmath328 from @xmath329 to @xmath330 that joins the two components @xmath331 and @xmath332 separated by @xmath91 , intersecting @xmath91 exactly once , in such a way that @xmath333 if @xmath90 . define @xmath285 to be a homeomorphism of @xmath328 onto the oriented edge in @xmath304 corresponding to @xmath91 that runs from @xmath331 to @xmath332 . because all components of @xmath94 are discs ( or pointed discs ) , it is easy to see that there is a unique way of extending @xmath285 to a homeomorphism ( up to isotopy ) .
the association above defines a @xmath10-equivariant equivalence of categories @xmath334 where @xmath335 is the category of proper arc systems , whose morphisms are reversed inclusions .
in fact , an inclusion @xmath336 of proper systems induces a morphism @xmath337 of nonsingular @xmath7-marked ribbon graphs .
a pseudo - inverse is constructed as follows .
let @xmath314 be a nonsingular @xmath7-marked ribbon graph and let @xmath310 be the spine .
the graph @xmath338 decomposes @xmath0 into a disjoint union of one - pointed discs . for each edge @xmath339 of @xmath295 ,
let @xmath340 be the simple arc joining the points in the two discs separated by @xmath339 .
thus , we can associate the system of arcs @xmath341 to @xmath317 and this defines a pseudo - inverse @xmath342 .
a _ metric _ on a ribbon graph @xmath48 is a map @xmath343 .
given a simple closed curve @xmath344 and an @xmath7-marked nonsingular ribbon graph @xmath345 , there is a unique simple closed curve @xmath346 contained inside @xmath347 such that @xmath348 is freely homotopic to @xmath169 .
if @xmath48 is metrized , then we can define the _ length _
@xmath349 to be @xmath350 .
consequently , the _ systol _ is given by @xmath351 .
given a proper weighted arc system @xmath352 , supported on @xmath353 , we can endow the corresponding ribbon graph @xmath354 with a _
, by simply setting @xmath355 .
thus , one can extend the correspondence to proper weighted arc systems and metrized @xmath7-marked nonsingular ribbon graphs .
moreover , the notions of length and systol agree with those given in [ sss : compact ] .
notice the similarity between lemma [ lemma : compact ] and mumford - mahler criterion for compactness in @xmath356 .
an _ isomorphism _ of enriched @xmath5-marked ribbon graphs @xmath357 is the datum of compatible isomorphisms of their ( reduced ) graphs @xmath358 and of the ribbon graphs @xmath312 , such that @xmath359 and they respect the rest of the data .
let @xmath254 be an enriched @xmath5-marked ribbon graph and let @xmath360 , where @xmath262 .
assume that @xmath361 or that @xmath362 .
we define @xmath363 in the following way . *
if @xmath339 is the only edge of @xmath261 , then we just turn @xmath326 into an invisible component and we define @xmath364 and @xmath365 for all @xmath366 that marked a hole or a vertex of @xmath261 . in what follows , suppose that @xmath362 . * if @xmath367_0 $ ] and @xmath368_0 $ ] are distinct and not both special , then we obtain @xmath363 from @xmath254 by simply replacing @xmath261 by @xmath369 . *
if @xmath367_0=[\ola{e}]_0 $ ] not special , then replace @xmath261 by @xmath369 .
if @xmath370 was a hole marked by @xmath371 , then mark the new vertex of @xmath369 by @xmath371 . otherwise , add an edge to @xmath235 that joins the two new vertices of @xmath369 ( which may or may not split into two visible components ) . * in this last case , add a new invisible component @xmath279 of genus @xmath276 to @xmath235 , replace @xmath261 by @xmath369 ( if @xmath369 is disconnected , the vertex @xmath326 splits ) and join @xmath279 to the new vertices ( one or two ) of @xmath369 and to the old edges @xmath372_0)\cup s_v^{-1}([\ola{e}]_0)$ ] . moreover ,
if @xmath370 was a hole marked by @xmath371 , then mark @xmath279 by @xmath371 .
notice that @xmath363 can be not reduced , so we may want to consider the reduced enriched graph @xmath373 associated to it .
we define @xmath374 to be an _ elementary contraction_. @xmath5-marked enriched ribbon graphs form a ( small ) category @xmath375 , whose morphisms are compositions of isomorphisms and elementary contractions .
call @xmath376 the full subcategory of @xmath375 whose objects are ribbon graphs of genus @xmath1 . really , the automorphism group of an enriched ribbon graph must be defined as the product of the automorphism group as defined above by @xmath377 , where @xmath378 is the group of automorphisms of the generic riemann surface of type @xmath379 ( where @xmath380 is the number of oriented edges of @xmath235 outgoing from @xmath326 ) .
fortunately , @xmath378 is almost always trivial , except if @xmath381 , when @xmath382 .
the _ topological realization _ of the enriched @xmath5-marked ribbon graph @xmath254 is the nodal @xmath5-marked oriented surface @xmath383 obtained as a quotient of @xmath384 by a suitable equivalence relation , where @xmath385 is a compact oriented surface of genus @xmath386 with marked points given by @xmath387 and by the oriented edges of @xmath235 outgoing from @xmath279 .
the equivalence relation identifies couples of points ( two special vertices of @xmath48 or a special vertex on a visible component and a point on an invisible one ) corresponding to the same edge of @xmath235 . as in the nonsingular case , for each @xmath262 the positive component @xmath388 naturally contains an embedded _ spine _ @xmath389 .
notice that there is an obvious correspondence between edges of @xmath235 and nodes of @xmath383 .
moreover , the elementary contraction @xmath390 to the non - reduced @xmath363 defines a unique homotopy class of maps @xmath391 , which may shrink a circle inside a positive component of @xmath383 to a point ( only in cases ( c ) and ( d ) ) , and which are homeomorphisms elsewhere .
if @xmath373 is the reduced graph associated to @xmath363 , then we also have a map @xmath392 that shrinks some circles inside the invisible components to points and is a homeomorphism elsewhere . @xmath393 an _ @xmath7-marking _ of an enriched @xmath5-marked ribbon graph @xmath254 is a map @xmath394 compatible with @xmath395 such that @xmath396 is a disjoint union of circles and @xmath285 is an orientation - preserving homeomorphism elsewhere .
the subsurface @xmath397 is the _
visible subsurface_. an _ isomorphism _ of @xmath7-marked ( reduced ) enriched ribbon graphs is an isomorphism @xmath398 such that @xmath399 is homotopic to @xmath400 .
given @xmath7-markings @xmath394 and @xmath401 such that @xmath402 is homotopic to @xmath403 , then we define @xmath404 to be an _ elementary contraction _ of @xmath7-marked enriched ribbon graphs .
define @xmath405 to be the category whose objects are ( equivalence classes of ) @xmath7-marked enriched ribbon graphs @xmath406 and whose morphisms are compositions of isomorphisms and elementary contractions . again , the mapping class group @xmath10 acts on @xmath405 and the quotient @xmath407 is equivalent to @xmath376 .
let @xmath322 be an arc system and let @xmath323 the corresponding operators on the set of oriented arcs @xmath136 .
define @xmath408 to be the set of connected components of @xmath111 and @xmath237 the set of components of @xmath113 .
let @xmath235 be a graph whose vertices are @xmath236 and whose edges correspond to connected components of @xmath409 , where an edge connects @xmath326 and @xmath279 ( possibly @xmath410 ) if the associated component bounds @xmath326 and @xmath279 . define @xmath246 to be the genus function associated to the connected components of @xmath113 .
call @xmath411 the subsurface associated to @xmath262 and let @xmath412 be the quotient of @xmath413 obtained by identifying each component of @xmath414 to a point . as @xmath415 quasi - fills @xmath412 , we can construct a dual ribbon graph @xmath261 and a homeomorphism @xmath416 that sends @xmath414 to special vertices of @xmath388 and marked points on @xmath412 to centers of @xmath388 . these homeomorphisms glue to give a map @xmath417 that shrinks circles and cylinders in @xmath418 to nodes and is a homeomorphism elsewhere , which is thus homotopic to a marking of @xmath383 .
we have obtain an enriched @xmath7-marked ( reduced ) ribbon graph @xmath419 _ dual to @xmath93_. the construction above defines a @xmath10-equivariant equivalence of categories @xmath420 where @xmath421 is the category of proper arc systems , whose morphisms are reversed inclusions .
as before , an inclusion @xmath336 of systems of arcs induces a morphism @xmath422 of nonsingular @xmath7-marked enriched ribbon graphs . to construct a pseudo - inverse , start with @xmath406 and call @xmath412 the surface obtained from @xmath423 by shrinking each boundary circle to a point . by nonsingular duality , we can construct a system of arcs @xmath424 inside @xmath412 dual to @xmath425 .
as the arcs miss the vertices of @xmath426 by construction , @xmath424 can be lifted to @xmath0 .
the wanted arc system on @xmath0 is @xmath427 .
a metric on @xmath254 is a map @xmath428 .
given @xmath344 and an @xmath7-marking @xmath394 , we can define @xmath429 . as in the nonsingular case
, there is a unique @xmath430 inside @xmath347 such that @xmath431 .
hence , we can define @xmath432 .
clearly , @xmath433 , where @xmath279 is the weight function supported on the arc system dual to @xmath406 .
thus , the arc - graph duality also establishes a correspondence between weighted arc systems on @xmath7 and metrized @xmath7-marked enriched ribbon graphs .
fix a compact oriented surface @xmath0 of genus @xmath1 and a subset @xmath74 such that @xmath3 .
a _ smooth family _ of @xmath7-marked riemann surfaces is a commutative diagram @xmath434^f \ar[rrd ] & & \mathcal{c } \ar[d]^\pi \\ & & b } \ ] ] where @xmath285 is an relatively ( over @xmath60 ) oriented diffeomorphism , @xmath435 is the projection on the first factor and the fibers @xmath436 of @xmath437 are riemann surfaces , whose complex structure varies smoothly with @xmath438 .
two families @xmath439 and @xmath440 over @xmath60 are _ isomorphic _ if there exists a continuous map @xmath441 such that * @xmath442 is homotopic to @xmath443 for every @xmath438 * @xmath444 is biholomorphic for every @xmath438 .
the functor @xmath445 defined by @xmath446 is represented by the _ teichmller space _ @xmath6 .
it is a classical result that @xmath6 is a complex - analytic manifold of ( complex ) dimension @xmath447 ( ahlfors @xcite , bers @xcite and ahlfors - bers @xcite ) and is diffeomorphic to a ball ( teichmller @xcite ) .
a _ smooth family _ of @xmath5-marked riemann surfaces of genus @xmath1 is * a submersion @xmath448 * a smooth embedding @xmath449 such that the fibers @xmath436 are riemann surfaces of genus @xmath1 , whose complex structure varies smoothly in @xmath438 , and @xmath450 is a section for every @xmath366 .
two families @xmath451 and @xmath452 over @xmath60 are isomorphic if there exists a diffeomorphism @xmath453 such that @xmath454 , the restriction of @xmath455 to each fiber @xmath444 is a biholomorphism and @xmath456 .
the existence of riemann surfaces with nontrivial automorphisms ( for @xmath457 ) prevents the functor @xmath458 & ( \text{sets } ) \\ b \ar@{|->}[r ] & \left\{\substack{\dis \text{smooth families of $ x$-marked}\\ \dis\text{riemann surfaces over $ b$}}\right\}/\text{iso } } \ ] ] from being representable .
however , riemann surfaces with @xmath3 have finitely many automorphisms and so @xmath4 is actually represented by an orbifold , which is in fact @xmath459 ( in the orbifold sense ) . in the algebraic category , we would rather say that @xmath4 is a deligne - mumford stack with quasi - projective coarse space .
enumerative geometry is traditionally reduced to intersection theory on suitable moduli spaces . in our case
, @xmath4 is not a compact orbifold . to compactify it in an algebraically meaningful way
, we need to look at how algebraic families of complex projective curves can degenerate . in particular , given a holomorphic family @xmath16 of algebraic curves over the punctured disc , we must understand how to complete the family over @xmath17 .
consider the family @xmath460)\in\delta^*\times{\mathbb{c}}\mathbb{p}^2\,|\ , y^2 z = x(x - bz)(x-2z)\}$ ] of curves of genus @xmath22 with the marked point @xmath461\in { \mathbb{c}}\mathbb{p}^2 $ ] , parametrized by @xmath462 .
notice that the projection @xmath463 given by @xmath464\mapsto [ x : z]$ ] ( where @xmath465\mapsto [ 1:0]$ ] ) is a @xmath23 cover , branched over @xmath466 . fix a @xmath467 and consider a closed curve @xmath468 that separates @xmath469 from @xmath470 and pick one of the two ( simple closed ) lifts @xmath471 .
this @xmath472 determines a nontrivial element of @xmath473 .
a quick analysis tells us that the endomorphism @xmath474 induced by the monodromy around a generator of @xmath475 is nontrivial .
thus , the family @xmath476 can not be completed over @xmath17 as smooth family ( because it would have trivial monodromy ) .
if we want to compactify our moduli space , we must allow our curves to acquire some singularities .
thus , it makes no longer sense to ask them to be submersions .
instead , we will require them to be _
flat_. given an open subset @xmath477 , a flat family of connected projective curves @xmath478 may typically look like ( up to shrinking @xmath60 ) * @xmath479 around a smooth point of @xmath480 * @xmath481 around a node of @xmath480 that persists on each @xmath436 * @xmath482 around a node of @xmath480 that does not persist on the other curves @xmath436 with @xmath483 in local analytic coordinates .
notice that the ( arithmetic ) genus of each fiber @xmath484 $ ] is constant in @xmath485 . to prove that allowing nodal curves is enough to compactify @xmath4
, one must show that it is always possible to complete any family @xmath486 to a family over @xmath17 .
however , because nodal curves may have nontrivial automorphisms , we shall consider also the case in which @xmath487 is an orbifold point .
thus , it is sufficient to be able to complete not exactly the family @xmath476 but its pull - back under a suitable map @xmath488 given by @xmath18 .
this is exactly the _ semi - stable reduction theorem_. one can observe that it is always possible to avoid producing genus @xmath276 components with @xmath22 or @xmath489 nodes .
thus , we can consider only _ stable curves _
, that is nodal projective ( connected ) curves such that all irreducible components have finitely many automorphisms ( equivalently , no irreducible component is a sphere with less than three nodes / marked points ) .
the _ deligne - mumford compactification _
@xmath12 of @xmath4 is the moduli space of @xmath5-marked stable curves of genus @xmath1 , which is a compact orbifold ( algebraically , a deligne - mumford stack with projective coarse moduli space ) .
its underlying topological space is a projective variety of complex dimension @xmath447 .
many facts suggest that one should not look at the moduli spaces of @xmath5-pointed genus @xmath1 curves @xmath12 each one separately , but one must consider the whole system @xmath490 .
an evidence is given by the existence of three families of maps that relate different moduli spaces . 1 .
the _ forgetful map _ is a projective flat morphism @xmath491 that forgets the point @xmath492 and stabilizes the curve ( i.e. contracts a possible two - pointed sphere ) .
this map can be identified to the universal family and so is endowed with natural sections @xmath493 for all @xmath366 .
the _ boundary map _ corresponding to irreducible curves is the finite map @xmath494 ( defined for @xmath495 ) that glues @xmath496 and @xmath497 together .
it is generically @xmath23 and its image sits in the boundary of @xmath12 .
3 . the _ boundary maps _ corresponding to reducible curves are the finite maps @xmath498 ( defined for every @xmath499 and @xmath500 such that the spaces involved are nonempty ) that take two curves and glue them together identifying @xmath496 and @xmath497 .
they are generically @xmath501 ( except in the case @xmath502 and @xmath503 , when the map is generically @xmath23 ) and their images sit in the boundary of @xmath12 too .
let @xmath504 be the cartier divisor in @xmath505 corresponding to the image of the tautological section @xmath506 and call @xmath507 .
we observe that @xmath12 has a natural _ stratification _ by topological type of the complex curve .
in fact , we can attach to every stable curve @xmath71 its _ dual graph _
@xmath508 , whose vertices @xmath205 correspond to irreducible components and whose edges correspond to nodes of @xmath71 . moreover , we can define a genus function @xmath509 such that @xmath510 is the genus of the normalization of the irreducible component corresponding to @xmath326 and a marking function @xmath511 ( determined by requiring that @xmath115 is marking a point on the irreducible component corresponding to @xmath512 ) .
equivalently , we will also say that the vertex @xmath267 is _ labeled _ by @xmath513 . call
@xmath514 the singular points of @xmath515 . for
every such labeled graph @xmath235 , we can construct a boundary map @xmath516 which is a finite morphism .
fix @xmath0 a compact oriented surface of genus @xmath1 and let @xmath74 such that @xmath3 .
it is natural to look for natural _ bordifications _ of @xmath6 : that is , we look for a space @xmath517 that contains @xmath6 as a dense subspace and such that the action of the mapping class group @xmath10 extends to @xmath28 .
a remarkable example is given by thurston s compactification @xmath518 , in which points at infinity are projective measured lamination with compact support in @xmath36 .
thurston showed that @xmath519 is compact and homeomorphic to a sphere .
as @xmath10 is infinite and discrete , this means that the quotient @xmath520 can not be too good and so this does not sound like a convenient way to compactify @xmath4 .
we will see in section [ sec : triangulation ] that @xmath6 can be identified to @xmath41 .
thus , another remarkable example will be given by @xmath106 .
a natural question is how to define a bordification @xmath28 such that @xmath521 .
a _ ( continuous ) family of stable @xmath7-marked curves _ is a diagram @xmath434^f \ar[rrd ] & & \mathcal{c } \ar[d]^\pi \\ & & b } \ ] ] where @xmath435 is the projection on the first factor and * the family @xmath437 is obtained as a pull - back of a flat stable family of @xmath5-marked curves @xmath522 through a continuous map @xmath523 * if @xmath524 is the subset of nodes , then @xmath525 is a smooth loop in @xmath526 for every @xmath527 * for every @xmath438 the restriction @xmath528 is an orientation - preserving homeomorphism , compatible with the @xmath5-marking .
isomorphisms of such families are defined in the obvious way .
a way to construct such families is to start with a flat family @xmath529 such that @xmath530 are all homeomorphic for @xmath483 .
then consider the path @xmath531 and call @xmath532 . over @xmath533 ,
the family @xmath534 is topologically trivial , whereas @xmath480 may contain some new nodes .
consider a marking @xmath535 that pinches circles to nodes , is an oriented homeomorphism elsewhere and is compatible with @xmath5 .
the map @xmath536 extends @xmath537 , which is the wanted @xmath7-marking .
the _ deligne - mumford augmentation _ of @xmath6 is the topological space @xmath538 that classifies families of stable @xmath7-marked curves .
it follows easily that @xmath539 as topological spaces .
however , @xmath540 has infinite ramification at @xmath541 , due to the dehn twists around the pinched loops .
let @xmath542 $ ] be a point of @xmath6 .
as @xmath543 , the uniformization theorem provides a universal cover @xmath544 , which endows @xmath545 with a hyperbolic metric of finite volume , with cusps at @xmath546 .
in fact , we can interpret @xmath6 as the classifying space of @xmath7-marked families of hyperbolic surfaces .
it is clear that continuous variation of the complex structure corresponds to continuous variation of the hyperbolic metric ( uniformly on the compact subsets , for instance ) , and so to continuity of the holonomy map @xmath547 .
in particular , for every @xmath548 the function @xmath549 that associates to @xmath542 $ ] the length of the unique geodesic in the free homotopy class @xmath550 is continuous . as @xmath551
, one can check that @xmath552 can be reconstructed from sufficiently ( but finitely ) many length functions . so that the continuity of these is equivalent to the continuity of the family .
let @xmath553 be a maximal system of disjoint simple closed curves of @xmath36 ( and so @xmath554 ) such that no @xmath555 is contractible in @xmath36 or homotopic to a puncture and no couple @xmath556 bounds a cylinder contained in @xmath36 .
the system @xmath557 induces a _ pair of pants decomposition _ of @xmath0 , that is @xmath558 , and each @xmath559 is a pair of pants ( i.e. a surface of genus @xmath276 with @xmath560 ) .
given @xmath561\in\teich(s , x)$ ] , we have _ lengths _
@xmath562 for @xmath563 , which determine the hyperbolic type of all pants @xmath564 .
the information about how the pants are glued together is encoded in the _ twist parameters _
@xmath565 , which are well - defined up to some choices .
what is important is that , whatever choices we make , the difference @xmath566 is the same and it is well - defined . the _ fenchel - nielsen coordinates _ @xmath567 exhibit a real - analytic diffeomorphism @xmath568 ( which clearly depends on the choice of @xmath557 ) .
points of @xmath541 are @xmath7-marked stable curves or , equivalently ( using the uniformization theorem componentwise ) , @xmath7-marked hyperbolic surfaces with nodes , i.e. homotopy classes of maps @xmath569 , where @xmath71 is a hyperbolic surface with nodes @xmath570 , the fiber @xmath571 is a simple closed curve @xmath572 and @xmath285 is an oriented diffeomorphism outside the nodes .
complete @xmath573 to a maximal set @xmath557 of simple closed curves in @xmath7 and consider the associated fenchel - nielsen coordinates @xmath574 on @xmath6 .
as we approach the point @xmath575 $ ] , the holonomies @xmath576 tend to parabolics and so the lengths @xmath577 tend to zero . in fact , the hyperbolic metric on surface @xmath71 has a pair of cusps at each node @xmath578 .
this shows that the lengths functions @xmath577 extend to zero at @xmath575 $ ] with continuity . on the other hand ,
the twist parameters @xmath579 make no longer sense .
if we look at what happens on @xmath12 , we may notice that the couples @xmath580 behave like polar coordinate around @xmath581 $ ] , so that is seems natural to set @xmath582 for all @xmath583 and define consequently a map @xmath584 , that associates to @xmath585 the surface with fenchel - nielsen coordinates @xmath586 .
notice that the map is well - defined , because a twist along @xmath572 by @xmath587 is a diffeomorphism of the surface ( a dehn twist ) .
the map @xmath588 is an orbifold cover @xmath589 and its image contains @xmath581 $ ] .
varying @xmath557 , we can cover the whole @xmath12 and thus give it a _ fenchel - nielsen smooth structure_. the bad news , analyzed by wolpert @xcite , is that the fenchel - nielsen smooth structure is different ( at @xmath590 ) from the deligne - mumford one . in fact , if a boundary divisor is locally described by @xmath591 , then the length @xmath592 of the corresponding vanishing geodesic is related to @xmath593 by @xmath594 , which shows that the identity map @xmath595 is lipschitz , but its inverse it not hlder - continuous .
let @xmath71 be a riemann surface of genus @xmath1 with marked points @xmath596 such that @xmath3 .
first - order deformations of the complex structure can be rephrased in terms of @xmath597 operator as @xmath598 , where the _ beltrami differential _
@xmath599 can be locally written as @xmath600 with respect to some holomorphic coordinate @xmath238 on @xmath71 and @xmath601 vanishes at @xmath5 .
given a smooth vector field @xmath602 on @xmath71 that vanishes at @xmath5 , the deformations induced by @xmath603 and @xmath604 differ only by an isotopy of @xmath71 generated by @xmath205 ( which fixes @xmath5 ) .
thus , the _ tangent space _ @xmath605}\m_{g , x}$ ] can be identified to @xmath606 . as a consequence ,
the _ cotangent space _ @xmath607}\m_{g , x}$ ] identifies to the space @xmath608 of integrable holomorphic quadratic differentials on @xmath609 , that is , which are allowed to have a simple pole at each @xmath366 .
the duality between @xmath605}\m_{g , x}$ ] and @xmath607}\m_{g , x}$ ] is given by @xmath610 & & { \mathbb{c}}\\ ( \mu,\varphi ) \ar@{|->}[rr ] & & \dis\int_{\si } \mu\varphi } \ ] ] if @xmath609 is given the hyperbolic metric @xmath611 , then elements in @xmath606 can be identified to the space of _ harmonic beltrami differentials _ @xmath612 .
the _ weil - petersson hermitean metric _
@xmath613 ( defined by weil @xcite using petersson s pairing of modular forms ) is @xmath614 for @xmath615 .
this metric has a lot of properties : it is khler ( weil @xcite and ahlfors @xcite ) and it is mildly divergent at @xmath590 , so that the weil - petersson distance extends to a non - degenerate distance on @xmath12 and all points of @xmath590 are at finite distance ( masur @xcite , wolpert @xcite ) . because @xmath12 is compact and so wp - complete , the lifting of the weil - petersson metric on to @xmath28 is also complete .
thus , @xmath28 can be seen as the _ weil - petersson completion _ of @xmath6 .
we should emphasize that the weil - petersson symplectic form @xmath616 depends more directly on the hyperbolic metric on the surface than on its holomorphic structure .
in particular , wolpert @xcite has shown that @xmath617 on @xmath6 , where @xmath618 are fenchel - nielsen coordinates associated to any pair of pants decomposition of @xmath7 .
on the other hand , if we identify @xmath6 with an open subset of @xmath619 , then points of @xmath6 are associated @xmath620-local systems @xmath621 on @xmath36 ( with parabolic holonomies at @xmath5 and hyperbolic holonomies otherwise ) , where @xmath622 is endowed with the symmetric bilinear form @xmath623 .
goldman @xcite has proved that , in this description , the tangent space to @xmath6 at @xmath621 is naturally @xmath624 and that @xmath616 is given by @xmath625 $ ] .
another description of @xmath626 in terms of shear coordinates and thurston s symplectic form on measured laminations is given by bonahon - szen @xcite .
one can feel that the complex structure @xmath627 on @xmath6 inevitably shows up whenever we deal with the weil - petersson metric , as @xmath628 . on the other hand ,
the knowledge of @xmath626 is sufficient to compute volumes and characteristic classes .
all the maps between moduli spaces we have defined are in some sense tautological as they are very naturally constructed and they reflect intrinsic relations among the various moduli spaces .
it is evident that one can look at these as classifying maps to the deligne - mumford stack @xmath12 ( which obviously descend to maps between coarse moduli spaces ) .
hence , we can consider all the cycles obtained by pushing forward or pulling back via these maps as being `` tautologically '' defined .
moreover , there is an ingredient we have not considered yet : it is the relative dualizing sheaf of the universal curve @xmath629 .
one expects that it carries many informations and that it can produce many classes of interest .
the relative dualizing sheaf @xmath630 is the sheaf on @xmath505 , whose local sections are ( algebraically varying ) abelian differentials that are allowed to have simple poles at the nodes , provided the two residues at each node are opposite .
the local sections of @xmath631 ( the logarithmic variant of @xmath630 ) are sections of @xmath630 that may have simple poles at the @xmath5-marked points .
the _ miller classes _ are @xmath632 where @xmath633 and the modified ( by arbarello - cornalba ) _ mumford - morita classes _ as @xmath634 one could moreover define the @xmath635-th _ hodge bundle _ as and consider the chern classes of these bundles ( for example , the @xmath611 classes ) .
however , using grothendieck - riemann - roch , mumford @xcite and bini @xcite proved that @xmath636 can be expressed as a linear combination of mumford - morita classes up to elements in the boundary , so that they do not introduce anything really new .
when there is no risk of ambiguity , we will denote in the same way the classes @xmath33 and @xmath32 belonging to different @xmath12 s as it is now traditional .
wolpert has proven @xcite that , on @xmath637 , we have @xmath638/\pi^2 $ ] and that the amplitude of @xmath639 ( and so the projectivity of @xmath637 ) can be recovered from the fact that @xmath640 $ ] is an integral khler class @xcite .
he also showed that the cohomological identity @xmath640=\k_1=(\pi_q)_*\psi_q^2 $ ] admits a beautiful pointwise interpretation @xcite .
because of the natural definition of @xmath32 and @xmath33 classes , as explained before , the subring @xmath641 of @xmath642 they generate is called the _ tautological ring _ of @xmath4 .
its image @xmath643 through the cycle class map is called cohomology tautological ring . from an axiomatic point of view , the _ system of tautological rings _
@xmath644 is the minimal system of subrings of @xmath645 is the minimal system of subrings such that * every @xmath646 contains the fundamental class @xmath647 $ ] * the system is closed under push - forward maps @xmath648 , @xmath649 and @xmath650 .
@xmath641 is defined to be the image of the restriction map @xmath651 .
the definition for the rational cohomology is analogous ( where the role of @xmath647 $ ] is here played by its poincar dual @xmath652 ) .
it is a simple fact to remark that all tautological rings contain @xmath33 and @xmath653 classes and in fact that @xmath641 is generated by them . really , this was the original definition of @xmath641 .
the @xmath33 classes interact reasonably well with the forgetful maps .
in fact @xmath654 where the first one is the so - called _ string equation _ and the second one for @xmath655 is the _ dilaton equation _ ( see @xcite ) .
they have been generalized by faber for maps that forget more than one point : faber s formula ( which we are going to describe below ) can be proven using the second equation above and the relation @xmath656 ( proven in @xcite ) .
let @xmath657 and let @xmath658 be the forgetful map .
then @xmath659 where @xmath660 and @xmath661 is defined in the following way .
if @xmath662 is a cycle , then set @xmath663 .
if @xmath664 is the decomposition in disjoint cycles ( including 1-cycles ) , then we let @xmath665 .
we refer to @xcite for more details on faber s formula , to @xcite and @xcite for more properties of tautological classes and to @xcite for a conjectural description ( which is now partially proven ) of the tautological rings .
it has been observed by witten @xcite that the intersection theory of @xmath653 and @xmath33 classes can be reduced to that of @xmath33 classes only by using the push - pull formula with respect to the forgetful morphisms .
moreover recall that @xmath667 on @xmath12 , where @xmath668 ( as shown in @xcite ) .
so , in order to find a `` minimal '' projective compactification of @xmath4 where to compute the intersection numbers of the @xmath33 classes , it is natural to look at the maps induced by the linear system @xmath669 .
it is well - known that @xmath666 is nef and big ( arakelov @xcite and mumford @xcite ) , so that the problem is to decide whether @xmath666 is semi - ample and to determine its exceptional locus @xmath670 for @xmath671 .
it is easy to see that @xmath672 pulls back to the trivial line bundle via the boundary map , where @xmath13 is a fixed curve of genus @xmath673 with a @xmath674-marking and the map glues @xmath496 with @xmath497 .
hence the map induced by the linear system @xmath672 ( if base - point - free ) should restrict to the projection on these boundary components .
whereas @xmath666 is semi - ample in characteristic @xmath675 , it is not so in characteristic @xmath276 ( keel @xcite ) .
however , one can still topologically contract the exceptional ( with respect to @xmath666 ) curves to obtain kontsevich s map @xmath676 which is a proper continuous surjection of orbispaces .
a consequence of keel s result is that the coarse @xmath677 can not be given a scheme structure such that the contraction map is a morphism .
this is in some sense unexpected , because the morphism behaves as if it were algebraic : in particular , the fiber product @xmath678 is projective .
@xmath679 can be given the structure of a stratified orbispace , where the stratification is again by topological type of the generic curve in the fiber of @xmath680 .
also , the stabilizer of a point @xmath681 in @xmath679 will be the same as the stabilizer of the generic point in @xmath682 .
so now we leave the realm of algebraic geometry and proceed topologically to construct and describe this different compactification .
in fact we introduce a slight modification of kontsevich s construction ( see @xcite ) .
we realize it as a quotient of @xmath683 by an equivalence relation , where @xmath684 is the standard simplex in @xmath44 if @xmath685 is an element of @xmath683 , then we say that an irreducible component of @xmath71 ( and so the associated vertex of the dual graph @xmath508 ) is _ visible _ with respect to @xmath686 if it contains a point @xmath687 such that @xmath688 . next , we declare that @xmath685 is equivalent to @xmath689 if @xmath690 and there is a homeomorphism of pointed surfaces @xmath691 , which is biholomorphic on the visible components of @xmath71 . as this relation would not give back a hausdorff space we consider its closure , which we are now going to describe .
consider the following two moves on the dual graph @xmath692 : 1 .
if two non - positive vertices @xmath279 and @xmath693 are joined by an edge @xmath339 , then we can build a new graph discarding @xmath339 , merging @xmath279 and @xmath693 along @xmath339 , thus obtaining a new vertex @xmath694 , which we label with @xmath695 2 .
if a non - positive vertex @xmath279 has a loop @xmath339 , we can make a new graph discarding @xmath339 and relabeling @xmath279 with @xmath696 .
applying these moves to @xmath692 iteratively until the process ends , we end up with a _ reduced dual graph _ @xmath697 .
call @xmath698 the subset of invisible vertices and @xmath699 the subset of visible vertices of @xmath697 .
for every couple @xmath685 denote by @xmath700 the quotient of @xmath71 obtained collapsing every non - positive component to a point .
we say that @xmath685 and @xmath689 are _ visibly equivalent _ if @xmath690 and there exist a homeomorphism @xmath701 , whose restriction to each component is analytic , and a compatible isomorphism @xmath702 of reduced dual graphs . in other words , @xmath703 are visibly equivalent if and only if @xmath690 there exists a stable @xmath704 and maps @xmath705 and @xmath706 such that @xmath707 are biholomorphic on the visible components and are a stable marking on the invisible components of @xmath708 ( that is , they may shrink some disjoint simple closed curves to nodes and are homeomorphisms elsewhere ) . finally call @xmath709 the quotient map and remark that @xmath710 is compact and that @xmath711 commutes with the projection onto @xmath684 .
similarly , one can say that two @xmath7-marked stable surfaces @xmath712,\up)$ ] and @xmath713,\up')$ ] are visibly equivalent if there exists a stable @xmath7 marked @xmath714 $ ] and maps @xmath705 and @xmath706 such that @xmath715 , @xmath716 and @xmath703 are visibly equivalent through @xmath707 ( see the remark above ) .
consequently , we can define @xmath717 as the quotient of @xmath718 obtained by identifying visibly equivalent @xmath7-marked surfaces . for every @xmath686 in @xmath684
, we will denote by @xmath719 the subset of points of the type @xmath720 $ ] .
then it is clear that @xmath721 is in fact homeomorphic to a product @xmath722 for any given @xmath723 . observe that @xmath719 is isomorphic to @xmath679 for all in such a way that @xmath724 identifies to @xmath680 .
notice , by the way , that the fibers of @xmath711 are isomorphic to moduli spaces .
more precisely consider a point @xmath720 $ ] of @xmath710 .
for every @xmath725 , call @xmath514 the subset of edges of @xmath697 outgoing from @xmath279 .
then we have the natural isomorphism @xmath726)\cong \prod_{w\in v_-(\si,\up ) } \mbar_{g_w , x_w\cup q_w}\ ] ] according to the fact that is projective .
one traditional way to associate a weighted arc system to a riemann surface endowed with weights at its marked points is to look at critical trajectories of jenkins - strebel quadratic differentials .
equivalently , to decompose the punctured surface into a union of semi - infinite flat cylinders with assigned lengths of their circumference .
let @xmath71 be a compact riemann surface and let @xmath727 be a _
meromorphic quadratic differential _ , that is @xmath728 where @xmath238 is a local holomorphic coordinate and @xmath729 is a meromorphic function . being a quadratic differential means that , if @xmath730 is another local coordinate , then @xmath731 .
_ regular points _ of @xmath71 for @xmath727 are points where @xmath727 has neither a zero nor a pole ; _ critical points _ are zeroes or poles of @xmath727 .
we can attach a metric to @xmath727 , by simply setting @xmath732 . in coordinates , @xmath733 .
the metric is well - defined and flat at the regular points and it has conical singularities ( with angle @xmath734 ) at simple poles ( @xmath735 ) and at zeroes of order @xmath119 .
poles of order @xmath489 or higher are at infinite distance .
if @xmath736 is a regular point , we can pick a local holomorphic coordinate @xmath238 at @xmath737 such that @xmath738 and @xmath739 on @xmath740 .
the choice of @xmath238 is unique up sign .
thus , @xmath741 defines a real - analytic curve through @xmath736 on @xmath71 , which is called _ horizontal trajectory _ of @xmath727 .
similarly , @xmath742 defines the _ vertical trajectory _ of @xmath727 through @xmath736 .
horizontal ( resp .
vertical ) trajectories @xmath743 are intrinsically defined by asking that the restriction of @xmath727 to @xmath743 is a positive - definite ( resp .
negative - definite ) symmetric bilinear form on the tangent bundle of @xmath743 .
if @xmath727 has at worst double poles , then the local aspect of horizontal trajectories is as in figure [ fig : trajectories ] ( horizontal trajectories through @xmath492 are drawn thicker ) .
trajectories are called _ critical _ if they meet a critical point .
it follows from the general classification ( see @xcite ) that * a trajectory is _ closed _ if and only if it is either periodic or it starts and ends at a critical point ; * if a horizontal trajectory @xmath743 is periodic , then there exists a maximal open annular domain @xmath744 and a number @xmath745 such that @xmath746 and , under this identification , @xmath747 for some @xmath748 ; * if all horizontal trajectories are closed of finite length , then @xmath727 has at worst double poles , where it has negative quadratic residue ( i.e. at a double pole , it looks like @xmath749 , with @xmath750 ) .
there are many theorems about existence and uniqueness of quadratic differentials @xmath727 with specific behaviors of their trajectories and about their characterization using extremal properties of the associated metric @xmath751 ( see jenkins @xcite ) .
the following result is the one we are interested in .
[ thm : strebel ] let @xmath71 be a compact riemann surface of genus @xmath1 and @xmath752 such that @xmath3 . for every @xmath753 there exists a unique quadratic differential @xmath727 such that * @xmath727 is holomorphic on @xmath609 * all horizontal trajectories of @xmath727 are closed * it has a double pole at @xmath115 with quadratic residue @xmath754 *
the only annular domains of @xmath727 are pointed discs at the @xmath115 s .
moreover , @xmath727 depends continuously on @xmath71 and on @xmath755 .
notice that the previous result establishes the existence of a continuous map @xmath756 where @xmath757 is the vector bundle , whose fiber over @xmath561 $ ] is the space of quadratic differentials on @xmath71 , which can have double poles at @xmath5 and are holomorphic elsewhere .
hubbard and masur @xcite proved ( in a slightly different case , though ) that the sections of @xmath757 are piecewise real - analytic and gave precise equations for their image .
quadratic differentials that satisfy ( a ) and ( b ) are called _ jenkins - strebel differentials_. they are particularly easy to understand because their critical trajectories form a graph @xmath758 embedded inside the surface @xmath71 and @xmath54 decomposes @xmath71 into a union of cylinders ( with respect to the flat metric @xmath751 ) , of which horizontal trajectories are the circumferences .
property ( d ) is telling us that @xmath609 retracts by deformation onto @xmath54 , flowing along the vertical trajectories out of @xmath5 .
it can be easily seen that theorem [ thm : strebel ] still holds for @xmath759 but @xmath760 .
condition ( d ) can be rephrased by saying that every annular domain corresponds to some @xmath115 for which @xmath761 , and that @xmath762 if @xmath763 .
it is still true that @xmath609 retracts by deformation onto @xmath54 .
we sketch the traditional existence proof of theorem [ thm : strebel ] .
the _ modulus _ of a standard annulus latexmath:[$a(r , r)=\{z\in{\mathbb{c}}\ , and the modulus of an annulus @xmath59 is defined to be that of a standard annulus biholomorphic to @xmath59 .
given a simply connected domain @xmath765 and let @xmath238 be a holomorphic coordinate at @xmath276 .
the _ reduced modulus _ of the annulus @xmath766 is @xmath767 , which is independent of the choice of a sufficiently small @xmath185 .
notice that the _ extremal length _ @xmath768 of a circumference @xmath169 inside @xmath769 is exactly @xmath770 .
fix holomorphic coordinates @xmath771 at @xmath772 .
a _ system of annuli _ is a holomorphic injection @xmath773 such that @xmath774 , where @xmath17 is the unit disc in @xmath775 .
call @xmath776 the reduced modulus @xmath777 and define the functional @xmath778 & & { \mathbb{r}}\\ s \ar@{|->}[rr ] & & \dis \sum_{i=1}^n p_i^2 m_i(s ) } \ ] ] which is bounded above , because @xmath609 is hyperbolic .
a maximizing sequence @xmath779 converges ( up to extracting a subsequence ) to a system of annuli @xmath780 and let @xmath781 .
notice that the restriction of @xmath782 to @xmath783 is injective if @xmath761 and is constantly @xmath115 if @xmath784 .
clearly , @xmath780 is maximizing for every choice of @xmath771 and so we can assume that , whenever @xmath761 , @xmath785 is the coordinate induced by @xmath782 .
define the @xmath786-quadratic differential @xmath727 on @xmath71 as @xmath787 on @xmath788 ( if @xmath761 ) and @xmath789 elsewhere .
notice that @xmath790 , where the _ reduced norm _ is given by @xmath791 + \sum_{i=1}^n p_i\log(\e_i)\ ] ] which is independent of the choice of sufficiently small @xmath792 . as @xmath780 is a stationary point for @xmath793
, so is for @xmath794 .
thus , for every smooth vector field @xmath795 on @xmath71 , compactly supported on @xmath609 , the first order variation of @xmath796 must vanish , where @xmath797 .
thus , @xmath727 is holomorphic on @xmath609 by weyl s lemma and it satisfies all the requirements . using the construction described above
, we can attach to every @xmath798 a graph @xmath799 ( and thus an @xmath7-marked ribbon graph @xmath800 ) which is naturally metrized by @xmath751 . by arc - graph duality ( in the nonsingular case ,
see [ sss : ns - duality ] ) , we also have a weighted proper system of arcs in @xmath71 .
notice that , because of ( c ) , the boundary weights are exactly @xmath801 .
if @xmath542 $ ] is a point in @xmath6 and @xmath802 , then the previous construction ( which is first explicitly mentioned by harer in @xcite , where he attributes it to mumford and thurston ) provides a point in @xmath803 .
it is however clear that , if @xmath750 , then the strebel differential associated to @xmath804 is @xmath805 .
thus , we can just consider @xmath806 , so that the corresponding weighted arc system belongs to @xmath41 ( after multiplying by a factor @xmath489 ) .
because of the continuous dependence of @xmath727 on @xmath71 and @xmath686 , the map @xmath807 is _
continuous_. we now show that a point @xmath808 determines exactly one @xmath7-marked surface , which proves that @xmath809 is _
bijective_. by [ sss : metrized ] , we can associate a metrized @xmath7-marked nonsingular ribbon graph @xmath354 to each @xmath810 supported on @xmath93 . however , if we realize @xmath811 by gluing semi - infinite tiles @xmath812 of the type @xmath813_x\times[0,\infty)_y\subset \hat{{\mathbb{c}}}_z$ ] , which naturally come together with a complex structure and a quadratic differential @xmath814 , then @xmath811 becomes a riemann surface endowed with the ( unique ) jenkins - strebel quadratic differential @xmath727 determined by theorem [ thm : strebel ] .
thus , @xmath815,\up)$ ] , where @xmath816 is obtained from the quadratic residue of @xmath727 at @xmath115 .
moreover , the length function defined on @xmath817 exactly corresponds to the @xmath751-length function on @xmath4 .
notice that @xmath809 is @xmath10-equivariant by construction and so induces a continuous bijection @xmath818 on the quotient . if we prove that @xmath819 is _ proper _ , then @xmath819 is a homeomorphism . to conclude that @xmath809 is a homeomorphism too
, we will use the following .
[ lemma : isom ] let @xmath820 and @xmath289 be metric spaces acted on discontinuously by a discrete group of isometries @xmath54 and let @xmath821 be a @xmath54-equivariant continuous injection such that the induced map @xmath822 is a homeomorphism . then @xmath455 is a homeomorphism . to show that @xmath455 is surjective , let @xmath823 . because @xmath824 is bijective , @xmath825\in y / g$ ] such that @xmath826)=[z]$ ] .
hence , @xmath827 for some @xmath828 and so @xmath829 . to prove that @xmath830 is continuous ,
let @xmath831 be a sequence such that @xmath832 as @xmath833 for some @xmath834 .
clearly , @xmath835\rar[h(y)]$ ] in @xmath836 and so @xmath837\rar [ y]$ ] in @xmath838 , because @xmath824 is a homeomorphism .
let @xmath839 be a sequence such that @xmath840=[y_m]$ ] and @xmath841 and call @xmath842 the element such that @xmath843 . by continuity of @xmath455 , we have @xmath844 and by hypothesis @xmath845 .
hence , @xmath846 and so @xmath847 for large @xmath848 , because @xmath54 acts discontinuously on @xmath289 . as a consequence , @xmath849 and so @xmath830 is continuous .
the final step is the following .
@xmath818 is proper .
let @xmath850,\up_m)$ ] be a diverging sequence in @xmath45 and call @xmath851 the hyperbolic metric on @xmath609 . by mumford - mahler criterion , there exist simple closed hyperbolic geodesics @xmath852 such that @xmath853 . because the hyperbolic length and the extremal length are approximately proportional for short curves ,
we conclude that extremal length @xmath854 .
consider now the metric @xmath855 induced by the jenkins - strebel differential @xmath856 uniquely determined by @xmath857 .
call @xmath858 the length of the unique geodesic @xmath859 with respect to the metric @xmath855 , freely homotopic to @xmath852 .
notice that @xmath859 is a union of critical horizontal trajectories .
because @xmath855 has infinite area , define a modified metric @xmath860 on @xmath861 in the same conformal class as @xmath855 as follows .
* @xmath860 agrees with @xmath855 on the critical horizontal trajectories of @xmath856 * whenever @xmath862 , consider a coordinate @xmath238 at @xmath115 such that the annular domain of @xmath856 at @xmath115 is exactly @xmath863 and @xmath864 . then define @xmath860 to agree with @xmath855 on @xmath865 ( which becomes isometric to a cylinder of circumference @xmath866 and height @xmath22 , so with area @xmath866 ) and to be the metric of a flat euclidean disc of circumference @xmath866 centered at @xmath867 ( so with area @xmath868 ) on @xmath869 .
notice that the total area @xmath870 is @xmath871 .
call @xmath872 the length of the shortest @xmath860-geodesic @xmath873 in the class of @xmath874 . by definition , @xmath875 and so @xmath876 . as a @xmath860-geodesic
is either longer than @xmath22 or contained in the critical graph of @xmath727 , then @xmath873 coincides with @xmath859 for @xmath877 .
hence , @xmath878 and so @xmath189 . by lemma [ lemma : compact ] , we conclude that @xmath879 is diverging in @xmath57 .
[ rmk : zero ] suppose that @xmath880,\up_m)$ ] is converging to @xmath712,\up ) \in \ol{\teich}_{g , x}\times\delta_x$ ] and let @xmath881 be an invisible component .
then @xmath882 is bounded by simple closed curves @xmath883 and @xmath884 for @xmath885 . just analyzing the shape of the critical graph of @xmath856
, one can check that @xmath886 for all @xmath887 .
hence , @xmath888 and so @xmath889 tends to zero uniformly on the compact subsets of @xmath890 .
we want to extend the map @xmath809 to deligne - mumford s augmentation : will call still @xmath891 this extension . given @xmath712,\up)$ ]
, we can construct a jenkins - strebel differential @xmath727 on each visible component of @xmath71 , by considering nodes as marked points with zero weight .
extend @xmath727 to zero over the invisible components .
clearly , @xmath727 is a holomorphic section of @xmath892 ( the square of the logarithmic dualizing sheaf on @xmath71 ) : call it the jenkins - strebel differential associated to @xmath685 .
notice that it clearly maximizes the functional @xmath793 , used in the proof of theorem [ thm : strebel ] .
as @xmath727 defines a metrized ribbon graph for each visible component of @xmath71 , one can easily see that thus we have an @xmath7-marked enriched ribbon graph @xmath254 ( see [ sss : enriched ] ) , where @xmath235 is the dual graph of @xmath71 and @xmath408 is the set of visible components of @xmath685 , @xmath848 is determined by the @xmath5-marking and @xmath681 by the position of the nodes . by arc - graph duality ( see [ sss : arc - graph ] ) , we obtain a system of arcs @xmath93 in @xmath7 and the metrics provide a system of weights @xmath159 with support on @xmath93 .
this defines the set - theoretic extension of @xmath809 . clearly , it is still @xmath10-equivariant and it identifies visibly equivalent @xmath7-marked surfaces .
thus , it descends to a bijection @xmath893 and we also have @xmath894 where @xmath53 can be naturally given the structure of an orbispace ( essentially , forgetting the dehn twists along curves of @xmath0 that are shrunk to points , so that the stabilizer of an arc system just becomes the automorphism group of the corresponding enriched @xmath5-marked ribbon graph ) .
the only thing left to prove is that @xmath809 is continuous .
in fact , @xmath710 is compact and @xmath53 is hausdorff : hence , @xmath819 would be ( continuous and ) automatically proper , and so a homeomorphism .
using lemma [ lemma : isom ] again ( using a metric pulled back from @xmath710 ) , we could conclude that @xmath809 is a homeomorphism too . consider a differentiable stable family @xmath895 \ar[r]^{\qquad f } \ar[rd ] & \mathcal{c } \ar[d]\\ & [ 0,\e ] } \ ] ] of @xmath7-marked curves ( that is , obtained restricting to @xmath896 $ ] a holomorphic family over the unit disc @xmath17 ) , such that @xmath1 is topologically trivial over @xmath897 $ ] with fiber a curve with @xmath119 nodes .
let also @xmath898\rar \delta_x$ ] be a differentiable family of weights .
we can assume that there are disjoint simple closed curves @xmath899 such that @xmath900 is a node for all @xmath901 , that @xmath902 is a node for @xmath903 and that @xmath904 is smooth away from these nodes .
fix @xmath905 a nonempty open relatively compact subset of @xmath906 that intersects every connected component .
define a reduced @xmath907 norm of a section @xmath908 of @xmath909 to be @xmath910 .
notice that @xmath907 convergence of holomorphic sections @xmath908 as @xmath911 implies uniform convergence of @xmath912 on the compact subsets of @xmath906 .
call @xmath913 the jenkins - strebel differential associated to @xmath914 with annular domains @xmath915 . as all the components of @xmath904 are hyperbolic , @xmath916 is uniformly bounded and we can assume ( up to extracting a subsequence ) that @xmath913 converges to a holomorphic section @xmath917 of @xmath918 in the reduced norm .
clearly , @xmath917 will have double poles at @xmath115 with the prescribed residue .
remark [ rmk : zero ] implies that @xmath917 vanishes on the invisible components of @xmath480 , whereas it certainly does not on the visible ones .
for all those @xmath919 $ ] such that @xmath920 , let @xmath921 be the coordinate at @xmath115 ( uniquely defined up to phase ) given by @xmath922 and @xmath923 is continuous on @xmath924 and biholomorphic in the interior for all @xmath925 and @xmath926 for @xmath927 .
whenever @xmath928 , choose @xmath921 such that @xmath929 , with @xmath930 .
when @xmath920 , we can choose the phases of @xmath931 in such a way that @xmath931 vary continuously with @xmath927 . if @xmath932 , then set @xmath933 .
otherwise , @xmath934 and so @xmath935 can not shrink to @xmath936 ( because @xmath937 would go to @xmath938 as @xmath911 ) . in this case , call @xmath935 the region @xmath939 . notice that @xmath917 has a double pole at @xmath115 with residue @xmath934 and clearly @xmath940 .
we want to prove that the visible subsurface of @xmath480 is covered by @xmath941 and so @xmath917 is a jenkins - strebel differential on each visible component of @xmath480 . by uniqueness
, it must coincide with @xmath942 .
consider a point @xmath21 in the interior of @xmath943 .
for every @xmath925 there exists an @xmath944 such that @xmath945 does not belong to the critical graph of @xmath913 and the @xmath946-distance @xmath947 . as @xmath948 in reduced norm and @xmath949 ,
then @xmath950 as @xmath911 .
we can assume ( up to discarding some @xmath901 s ) that @xmath951 belongs to @xmath952 for a fixed @xmath953 and in particular that @xmath954 for some @xmath955 .
up to discarding some @xmath901 s , we can also assume that @xmath956 .
call @xmath957 the point given by @xmath958 .
@xmath959 and all terms go to zero as @xmath911 .
thus , every point in the smooth locus @xmath960 is at @xmath961-distance zero from some @xmath935 .
hence , @xmath942 is a jenkins - strebel differential on the visible components . with a few simple considerations ,
one can easily conclude that * the zeroes of @xmath913 move with continuity for @xmath962 $ ] * if @xmath963 is an edge of the critical graph of @xmath913 which starts at @xmath964 and ends at @xmath965 , and if @xmath966 for @xmath967 , then @xmath968 the corresponding edge of the critical graph of @xmath942 starting at @xmath969 and ending at @xmath970 ; moreover , @xmath971 * the critical graph of @xmath913 converges to that of @xmath942 for the gromov - hausdorff distance .
thus , the associated weighted arc systems @xmath972 converge to @xmath973 for @xmath911 .
thus , we have proved the following result , claimed first by kontsevich in @xcite ( see looijenga s @xcite and zvonkine s @xcite ) .
the map defined above @xmath974 is a @xmath10-equivariant homeomorphism , which commutes with the projection onto @xmath43 .
hence , @xmath975 is a homeomorphism of orbispaces too .
a consequence of the previous proposition and of [ sss : arc - graph ] is that the realization @xmath976 is the classifying space of @xmath10 and that @xmath977 is a homotopy equivalence ( in the orbifold category ) .
the other traditional way to obtain a weighted arc system out of a riemann surface with weighted marked points is to look at the spine of the truncated surface obtained by removing horoballs of prescribed circumference .
equivalently , to decompose the surface into a union of hyperbolic cusps .
let @xmath542 $ ] be an @xmath7-marked hyperbolic surface and let @xmath46 .
call @xmath978 the horoball at @xmath115 with circumference @xmath816 ( as @xmath979 , the horoball is embedded in @xmath71 ) and let @xmath980 be the _
truncated surface_. the datum @xmath981 is also called a _ decorated surface_. for every @xmath982 at finite distance from @xmath983 ,
let the _ valence _ @xmath984 be the number of paths that realize @xmath985 , which is generically @xmath22 . we will call a _ projection _ of @xmath21 a point on @xmath983 which is at shortest distance from @xmath21 : clearly , there are @xmath984 of them .
let the _ spine _ @xmath986 be the locus of points of @xmath71 which are at finite distance from @xmath983 and such that @xmath987 ( see figure [ fig : hcoord ] ) . in particular , @xmath988 is a disjoint union of finitely many geodesic arcs ( the _ edges _ ) and @xmath989 is a finite collection of points ( the _ vertices _ ) .
if @xmath784 , then we include @xmath115 in @xmath986 and we consider it a vertex .
its valence is defined to be the number of half - edges of the spine incident at @xmath115 .
there is a deformation retraction of @xmath990 ( where @xmath991 is the visible subsurface ) onto @xmath986 , defined on @xmath992 simply flowing away from @xmath983 along the unique geodesic that realizes the distance from @xmath983 .
this shows that @xmath986 defines an @xmath7-marked enriched ribbon graph @xmath993 . by arc - graph duality
, we also have an associated _ spinal arc system _ @xmath994 . as @xmath71 is a hyperbolic surface , we could metrize @xmath986 by inducing a length on each edge .
however , the relation between this metric and @xmath686 would be a little involved . instead ,
for every edge @xmath339 of @xmath993 ( that is , of @xmath986 ) , consider one of its two projections @xmath995 to @xmath983 and define @xmath996 to the be hyperbolic length of @xmath995 , which clearly does not depend on the chosen projection .
thus , the boundary weights vector @xmath997 is exactly @xmath686
. this endows @xmath993 with a metric and so @xmath998 with a projective weight @xmath999 .
notice that visibly equivalent surfaces are associated the same point of @xmath106 .
this defines a @xmath10-equivariant map @xmath1000 that commutes with the projection onto @xmath43 .
penner @xcite proved that the restriction of @xmath1001 to @xmath1002 is a homeomorphism ; the proof that @xmath1001 is a homeomorphism first appears in bowditch - epstein s @xcite ( and a very detailed treatment will appear in @xcite ) .
we refer to these papers for a proof of this result .
the purpose of this informal subsection is to briefly illustrate the bridge between the cellular decomposition of the teichmller space obtained using jenkins - strebel differentials and that obtained using spines of decorated surfaces .
fix @xmath0 a compact oriented surface as before and @xmath1003 a nonempty subset .
a ( stable ) hyperbolic surface @xmath71 is a nodal surface such that @xmath1004 is hyperbolic with geodesic boundary and/or cusps .
notice that , by convention , @xmath72 does not include the possible nodes of @xmath71 .
an @xmath5-marking of a ( stable ) hyperbolic surface @xmath71 is a bijection @xmath1005 .
an @xmath7-marking of the ( stable ) hyperbolic surface @xmath71 is an isotopy class of maps @xmath1006 , that may shrink disjoint simple closed curves to nodes and are homeomorphisms onto @xmath1007 elsewhere .
let @xmath1008 be the teichmller space of @xmath7-marked stable hyperbolic surfaces .
there is a natural map @xmath1009 that associates to @xmath542 $ ] the boundary lengths of @xmath71 , which thus descends to @xmath1010 . call @xmath1011 ( resp .
@xmath1012 ) the leaf @xmath1013 ( resp .
@xmath1014 ) . there is an obvious identification between @xmath1015 ( resp .
@xmath1016 ) and @xmath28 ( resp .
@xmath12 ) .
call @xmath1017 the blow - up of @xmath1018 along @xmath1016 : the exceptional locus can be naturally identified to the space of decorated surfaces with cusps ( which is homeomorphic to @xmath1019 ) .
define similarly , @xmath1020 .
the conformal analogue of a hyperbolic surface with geodesic boundary @xmath71 is a riemann surface with real boundary .
in fact , the double of @xmath71 is a hyperbolic surface with no boundary and an orientation - reversing involution , that is a riemann surface with an anti - holomorphic involution . as a consequence ,
@xmath72 is a real - analytic submanifold .
this means that first - order deformations are determined by beltrami differentials on @xmath71 which are real on @xmath1021 , and so @xmath605}\mbar^{\pa}_{g , x}\cong h^{0,1}(\si , t_{\si})$ ] , where @xmath1022 is the sheaf of tangent vector fields @xmath1023 , which are real on @xmath72 .
dually , the cotangent space @xmath607}\mbar^{\pa}_{g , x}$ ] is given by the space @xmath1024 of holomorphic quadratic differentials that are real on @xmath72 .
if we call @xmath1025 , where @xmath611 is the hyperbolic metric on @xmath71 , then @xmath1026 identifies to the space of harmonic beltrami differentials @xmath1027 .
as usual , if @xmath71 has a node , then quadratic differentials are allowed to have a double pole at the node , with the same quadratic residue on both branches . if a boundary component of @xmath71 collapses to a cusp @xmath115 , then the cotangent cone to @xmath1018 at @xmath581 $ ]
is given by quadratic differentials that may have at worst a double pole at @xmath115 with positive residue .
the phase of the residue being zero corresponds to the fact that , if we take fenchel - nielsen coordinates on the double of @xmath71 which are symmetric under the real involution , then the twists along @xmath72 are zero . mimicking what done for surfaces with cusps
, we can define hermitean pairings on @xmath1024 and @xmath1027 , where @xmath71 is a hyperbolic surface with boundary .
in particular , @xmath1028 where @xmath1029 and @xmath1030 . thus , if @xmath613 , then @xmath1 is the weil - petersson riemannian metric and @xmath626 is the weil - petersson form .
write similarly @xmath1031 , where @xmath1032 is the cometric dual to @xmath1 and @xmath1033 is the weil - petersson bivector field .
notice that @xmath626 and @xmath1033 are degenerate .
this can be easily seen , because wolpert s formula @xmath1034 still holds .
we can also conclude that the symplectic leaves of @xmath1033 are exactly the fibers of the boundary length map @xmath997 .
the spine construction can be carried on , even in a more natural way , on hyperbolic surfaces with geodesic boundary .
in fact , given such a @xmath71 whose boundary components are called @xmath772 , we can define the distance from @xmath72 and so the valence of a point in @xmath71 and consequently the spine @xmath1035 , with no need of further information .
similarly , if @xmath71 has also nodes ( that is , some holonomy degenerates to a parabolic element ) , then @xmath1035 is embedded inside the _ visible components of @xmath71 _ ,
i.e. those components of @xmath71 that contain a boundary circle of positive length
. the weight of an arc @xmath1036 dual to the edge @xmath1037 of @xmath1035 is still defined as the hyperbolic length of one of the two projections of @xmath1037 to @xmath72 .
thus , the construction above gives a point @xmath1038 .
it is easy to check ( see @xcite or @xcite ) that @xmath1039 converges to the @xmath1040 defined above when the hyperbolic surface with boundary converges to a decorated surface with cusps in @xmath1020 .
thus , the @xmath10-equivariant map @xmath1041 reduces to @xmath1042 for decorated surfaces with cusps .
the restriction of @xmath1043 to smooth surfaces with no boundary cusps gives a homeomorphism onto its image .
the continuity of the whole @xmath1043 is proven in @xcite , using luo s result .
+ the key point of luo s proof is the following . pick a generic hyperbolic surface with geodesic boundary @xmath71 and
suppose that the spinal arc system is the ideal triangulation @xmath1044 with weight @xmath1039 .
we can define the length @xmath1045 as the hyperbolic length of the shortest geodesic @xmath1046 in the free homotopy class of @xmath91 .
the curves @xmath1047 cut @xmath71 into hyperbolic hexagons , which are completely determined by @xmath1048 , where the @xmath1049 s are the sides of the hexagons lying on @xmath72 .
unfortunately , going from the @xmath1050 s to @xmath1039 is much easier than the converse .
in fact , @xmath1051 can be written as explicit linear combinations of the @xmath1050 s : in matrix notation , @xmath1052 is a solution of the system @xmath1053 , where @xmath1054 is a fixed @xmath1055-matrix ( that encodes the combinatorics is @xmath998 ) and @xmath1056 .
clearly , there is a whole affine space @xmath1057 of dimension @xmath1058 of solutions of @xmath1053 .
the problem is that a random point in @xmath1057 would determine hyperbolic structures on the hexagons of @xmath1059 that do not glue , because we are not requiring the two sides of each @xmath91 to have the same length .
starting from very natural quantities associated to hyperbolic hexagons with right angles , luo defines a functional on the space @xmath1060 . for every @xmath1061 ,
the space @xmath1057 is not empty ( which proves the surjectivity of @xmath1043 ) and the restriction of luo s functional to @xmath1057 is strictly concave and achieves its ( unique ) maximum exactly when @xmath1052 ( which proves the injectivity of @xmath1043 ) .
the geometric meaning of this functional is still not entirely clear , but it seems related to some volume of a three - dimensional hyperbolic manifold associated to @xmath71 .
quite recently , luo @xcite ( see also @xcite ) has introduced a modified functional @xmath1062 , which depends on a parameter @xmath1063 , and he has produced other realizations of the teichmller space as a polytope , and so different systems of `` simplicial '' coordinates . to close the circle , we must relate the limit of @xmath1043 for surfaces whose boundary lengths diverge to @xmath809 .
this is the topic of @xcite . here
, we only sketch the main ideas . to simplify the exposition , we will only deal with smooth surfaces .
consider an @xmath5-marked hyperbolic surface with geodesic boundary @xmath71 .
define @xmath1064 to be the surface obtained by gluing semi - infinite flat cylinders at @xmath72 of lengths @xmath1065 .
thus , @xmath1066 has a hyperbolic core and flat ends and the underlying conformal structure is that of an @xmath5-punctured riemann surface .
this _ grafting _ procedure defines a map @xmath1067 the map @xmath1068 is a @xmath10-equivariant homeomorphism .
the proof is a variation of scannell - wolf s @xcite that finite grafting is a self - homeomorphism of the teichmller space .
thus , the composition of @xmath1069 and @xmath1043 gives ( after blowing up the locus @xmath1070 ) the homeomorphism @xmath1071 the map @xmath1072 extends to a @xmath10-equivariant homeomorphism @xmath1073\lra |\ao(s , x)|\times[0,\infty]\ ] ] and @xmath1074 coincides with harer - mumford - thurston s @xmath809 . the main point is to show that a surface @xmath71 with large boundaries and with spine @xmath1035 is very close in @xmath6 to the flat surface whose jenkins - strebel differential has critical graph isomorphic to @xmath1035 ( as metrized ribbon graphs ) . to understand why this is reasonable ,
consider a sequence of hyperbolic surfaces @xmath861 whose spine has fixed isomorphism type @xmath48 and fixed projective metric and such that @xmath1075 , where @xmath1076 diverges as @xmath833 .
consider the grafted surfaces @xmath1077 and rescale them so that @xmath1078 .
the flat metric on the cylinders is naturally induced by a holomorphic quadratic differential , which has negative quadratic residue at @xmath5 .
extend this differential to zero on the hyperbolic core .
because of the rescaling , the distance between the flat cylinders and the spine goes to zero and the differential converges in @xmath1079 to a jenkins - strebel differential .
dumas @xcite has shown that an analogous phenomenon occurs for closed surfaces grafted along a measured lamination @xmath1080 as @xmath1081 . using wolpert s result and hyperbolic geometry
, penner @xcite has proved that pull - back of the weil - petersson form on the space of decorated hyperbolic surfaces with cusps , which can be identified to @xmath47 , can be neatly written in the following way .
fix a triangulation @xmath1082 . for every @xmath712,\up)\in\teich(s , x)\times\delta_x$ ] ,
let @xmath1046 be the geodesic representative in the class of @xmath1083 and call @xmath1084 , where @xmath1085 be the truncated hyperbolic surface .
then @xmath1086 where @xmath1087 is the projection , @xmath1088 is the set of ideal triangles in which the @xmath1046 s decompose @xmath71 , and the sides of @xmath901 are @xmath1089 in the cyclic order induced by the orientation of @xmath901 ( see figure [ fig : triang ] ) . to work on @xmath45 ( for instance , to compute weil - petersson volumes ) , one can restrict to the interior of the cells @xmath1090 whose associated system of arcs @xmath93 is triangulation and write the pull - back of @xmath616 with respect to @xmath93 . still using methods of wolpert @xcite
, one can generalize penner s formula to hyperbolic surfaces with boundary .
the result is better expressed using the weil - petersson bivector field than the @xmath489-form .
let @xmath71 be a hyperbolic surface with boundary components @xmath1091 and let @xmath1092 be a triangulation
. then the weil - petersson bivector field can be written as @xmath1093 where @xmath1094 and @xmath1095 is the length of the geodesic arc running from @xmath1096 to @xmath1097 along @xmath1098 in the positive direction ( according to the orientation induced by @xmath71 on @xmath1098 ) .
the idea is to use wolpert s formula @xmath1099 on the double @xmath1100 of @xmath71 with the pair of pants decomposition induced by doubling the arcs @xmath1101
. then one must compute the ( first - order ) effect on the @xmath1102 s of twisting @xmath1100 along @xmath92 .
though not immediate , the formula above can be shown to reduce to penner s , when the boundary lengths go to zero , as we approximate @xmath1103 for small @xmath1104 .
notice that penner s formula shows that @xmath626 linearizes ( with constant coefficients ! ) in the coordinates given by the @xmath1102 s .
more interesting is to analyze what happens for @xmath1105 with @xmath46 , as @xmath1081 .
assume the situation is generic and so @xmath1106 is supported on a triangulation , whose dual graph is @xmath48 .
once again , the formula dramatically simplifies as we approximate @xmath1107 for @xmath1108 . under the rescalings @xmath1109 and @xmath1110 with @xmath1111 , we obtain that @xmath1112 where @xmath1113 and @xmath1114 ( and @xmath1115 ) .
thus , the weil - petersson symplectic structure is again linearized ( and with constant coefficients ! ) , but in the system of coordinates given by the @xmath1116 s , which are in some sense dual to the @xmath1102 s . it would be nice to exhibit a clear geometric argument for the perfect symmetry of these two formulae .
we introduce some remarkable @xmath10-equivariant subcomplexes of @xmath38 , which define interesting cycles in the homology of @xmath679 as well as in the borel - moore homology of @xmath4 and so , by poincar duality , in the cohomology of @xmath4 ( that is , of @xmath10 ) .
these subcomplexes are informally defined as the locus of points of @xmath41 , whose associated ribbon graphs have prescribed odd valences of their vertices .
it can be easily shown that , if we assign even valence to some vertex , the subcomplex we obtain is not a cycle ( even with @xmath1117 coefficients ! ) .
notice that we are going to use the cellularization of the moduli space of curves given by @xmath809 , and so we will identify @xmath710 with the orbispace @xmath53 . as the arguments will be essentially combinatorial / topological , any of the decompositions described before would work .
the _ combinatorial subcomplex _
@xmath1122 is the smallest simplicial subcomplex that contains all proper simplices @xmath1123 such that @xmath94 is the disjoint union of exactly @xmath63 polygons with @xmath1124 sides .
it is convenient to set @xmath1125 . clearly , this subcomplex is @xmath10-equivariant .
hence , if we call @xmath1126 , then we can define @xmath1127 to be the subcomplex of @xmath1128 induced by @xmath1129 .
we can introduce also univalent vertices by allowing @xmath1130 .
it is still possible to define the complexes @xmath1129 and @xmath1131 , just allowing ( finitely many ) contractible loops ( i.e. unmarked tails in the corresponding ribbon graph picture ) .
however , @xmath1129 would no longer be a subcomplex of @xmath38 .
thus , we should construct an associated family of riemann surfaces over @xmath1127 ( which can be easily done ) and consider the classifying map @xmath1132 , whose existence is granted by the universal property of @xmath12 , but which would no longer be cellular .
define @xmath1137 as the space of couples @xmath1138 , where @xmath48 is a @xmath5-marked metrized ribbon graph in @xmath1139 and @xmath21 is a point of @xmath347 belonging to an edge that borders the @xmath115-th hole .
clearly @xmath1140 is a topological bundle with fiber homeomorphic to @xmath1141 .
it is easy to see that , for a fixed @xmath1142 such that @xmath761 , the pull - back of @xmath1137 via @xmath1143 is isomorphic ( as a topological bundle ) to the sphere bundle associated to @xmath1144 .
[ lemma : eta ] fix @xmath115 in @xmath5 and @xmath1145 such that @xmath761 . then on every simplex @xmath1146
define @xmath1147 where @xmath1148 and @xmath115 marks a hole with cyclically ordered sides @xmath1149 . these @xmath489-forms glue to give a piecewise - linear @xmath489-form @xmath1150 on @xmath1151 , that represents @xmath1152 .
hence , the pull - back class @xmath1153 $ ] is exactly @xmath1154 in @xmath1155 .
[ lemma : orient ] for every @xmath1157 the restriction of @xmath1158 to the maximal simplices of @xmath1159 is a non - degenerate symplectic form .
hence , @xmath1160 defines an orientation on @xmath1159 . also , @xmath1161 is a volume form on @xmath1127 . on @xmath1162 , the differentials @xmath1163 span the cotangent space .
as the @xmath816 s are fixed , we have the relation @xmath1164 for all @xmath139 . hence @xmath1165_0=x_i } de\,\big|\,i=1,\dots , n\right)\ ] ] on the other hand the tangent bundle is @xmath1166_0\in x_i}b_e=0 \quad\text{for all $ i=1,\dots , n$}\ , \right\}.\ ] ] in order to prove that @xmath1167 is non - degenerate , we construct its right - inverse .
define @xmath1168 as @xmath1169_1 } + \sum_{j=1}^{2 t } ( -1)^j \frac{\pa}{\pa [ \s_0^j(\ola{e})]_1}\ ] ] where @xmath1170 is any orientation of @xmath339 , while @xmath1171 and @xmath1172 are the cardinalities of @xmath367_0 $ ] and @xmath368_0 $ ] respectively .
we want to prove that @xmath1173 for every @xmath1174 . to shorten the notation , set @xmath1175_1 $ ] and @xmath1176_1 $ ] and call @xmath1177_{\infty}$ ] for @xmath1178 and @xmath1179_{\infty}$ ] for @xmath1180 the holes bordered respectively by @xmath1181 and @xmath1182
. finally call @xmath1183 and @xmath1184 the holes adjacent to @xmath339 as in figure [ fig : orient ] .
remark that neither the edges @xmath285 and @xmath455 nor the holes @xmath793 and @xmath552 are necessarily distinct .
this however has no importance in the following computation .
@xmath1185 it is easy to see ( using that the perimeters are constant ) that @xmath1186 and analogously for the @xmath455 s .
moreover @xmath1187 and similarly for @xmath1184 .
finally , we obtain @xmath1173 . given a top - dimensional cell @xmath1162 in @xmath1159 , each face in the boundary @xmath1193 is obtained shrinking one edge of @xmath354 .
this contraction may merge two vertices as in fig .
[ fig : contraction ] .
let @xmath1194 be the face of @xmath1162 obtained by shrinking the edge @xmath339 .
then @xmath1195 and so the dual of the orientation form induced by @xmath1162 on @xmath1196 is @xmath1197 , where @xmath1198 is the bivector field on @xmath1162 defined in lemma [ lemma : orient ] .
consider the graph @xmath1199 that occurs in the boundary of a top - dimensional cell of @xmath1159 .
suppose it is obtained merging two vertices of valences @xmath1200 and @xmath1201 in a vertex @xmath326 of valence @xmath1202 .
then @xmath1196 is in the boundary of exactly @xmath1202 cells of @xmath1159 or @xmath1203 ones in the case @xmath1204 . in any case , the number of cells @xmath1196 is bordered by are even : we need to prove that half of them induces on @xmath1196 an orientation and the other half induces the opposite one . if @xmath1199 is obtained from some @xmath354 contracting an edge @xmath339 , then we just have to compute the vector field @xmath1205 , which turns to be @xmath1206 where @xmath1207 are the edges of @xmath1199 outgoing from @xmath326 .
it is a straightforward computation to check that one obtains in half the cases a plus and in half the cases a minus .
when @xmath1208 has a node with @xmath1209 edges on one side ( which we will denote by @xmath1210 ) and @xmath1201 edges on the other side , the computation is similar .
the cell occurs as boundary of exactly @xmath1211 top - dimensional cells and , if @xmath1199 is obtained by @xmath354 contracting the edge @xmath339 , then @xmath1212 a quick check ensures that the signs cancel .
these _ generalized witten classes _ are related to the previous @xmath69 in an intuitively obvious way , because forgetting the markings of some vertices will map them onto one another .
we will omit the details and refer to @xcite .
* witten generalized cycles on @xmath4 are poincar dual to polynomials in the @xmath33 and the @xmath653 classes * ordinary and generalized witten cycles on @xmath1151 are push - forward of ( the poincar dual of ) tautological classes from @xmath12 ; an explicit recipe to produce such tautological classes is given .
we want to consider a combinatorial cycle on @xmath4 supported on ribbon graphs , whose vertices are generically all trivalent except one , which is @xmath1217-valent ( and @xmath1218 ) . to shorten the notation , call this witten cycle @xmath1219 .
we also define a generalized witten cycle on the universal curve @xmath1220 supported on the locus of ribbon graphs , which have a @xmath1217-valent vertex marked by @xmath21 and all the other vertices are trivalent and unmarked .
call @xmath1221 this cycle .
lemma [ lemma : eta ] gives us the nice piecewise - linear @xmath489-form @xmath1226 , that is pulled back to @xmath1227 through @xmath711 .
the only problem is that @xmath1226 is defined only for @xmath1228 , whereas @xmath1221 is exactly contained in the locus @xmath1229 . to compare the two
, one can look at the blow - up @xmath1230 of @xmath1231 along the locus @xmath1229 .
points in the exceptional locus @xmath201 can be identified with metrized ( nonsingular ) ribbon graphs @xmath48 , in which @xmath21 marks a vertex , plus _ angles _ @xmath1232 between consecutive oriented edges outgoing from @xmath21 .
one must think of these angles as of _ infinitesimal edges_. it is clear now that @xmath1226 extends to @xmath201 by @xmath1233 where @xmath1234 , @xmath21 marks a vertex with cyclically ordered outgoing edges @xmath1235 and @xmath1236 is the angle between @xmath1237 and @xmath1238 ( with @xmath1239 ) .
thus , pushing forward @xmath1240 through @xmath1241 , we obtain @xmath1242 plus other terms contained in the boundary , and the coefficient @xmath1223 is exactly the integral of @xmath1240 on a fiber ( that is , a simplex ) , which turns out to be @xmath1243 .
thus , @xmath1221 is poincar dual to @xmath1244 . as before
, we can identify @xmath1254 with the set of metrized ribbon graphs @xmath48 , with angles at the vertices @xmath21 and @xmath238 .
thus , pushing @xmath1255 forward through the blow - up map ( which forgets the angles at @xmath21 and @xmath238 ) , we obtain a multiple of the generalized combinatorial cycles given by @xmath21 marking a @xmath1217-valent vertex and @xmath238 marking a @xmath1256-valent ( distinct ) vertex .
the coefficient @xmath1257 will just be @xmath1258 .
points in @xmath1259 can be thought of as metrized ribbon graphs @xmath48 with two infinitesimal holes ( respectively marked by @xmath21 and @xmath238 ) adjacent to each other .
if we perform the push - forward of @xmath1255 forgetting first the angles at @xmath238 and then the angles at @xmath21 , then we obtain some contribution only from the loci in which the infinitesimal @xmath238-hole has @xmath1256 edges and the infinitesimal @xmath21-hole has @xmath1260 edges ( included the common one ) .
thus , we obtain the same contribution for each of the @xmath1261 configurations of two adjacent holes of valences @xmath1256 and @xmath1260 .
hence , @xmath1245 is poincar dual to a linear combination of generalized combinatorial cycles . as before , using the forgetful map , the same holds for the witten cycles obtained by deleting the @xmath21 and the @xmath238 markings .
the ( co)homologies of the mapping class groups have the remarkable property that they stabilize when the genus of the surface increases .
this was first proven by harer @xcite , and the stability bound was then improved by ivanov @xcite ( and successively again by harer for homology with rational coefficients , in an unpublished paper ) .
we now want to recall some of harer s results .
let @xmath1265 be a compact oriented surface of genus @xmath1 with @xmath1136 marked points and @xmath485 boundary components @xmath1266 .
call @xmath1267 the group of isotopy classes of diffeomorphisms of @xmath0 that fix the marked points and @xmath37 pointwise .
* gluing @xmath1265 and @xmath736 by identifying @xmath1098 with @xmath1270 , thus producing an oriented surface of genus @xmath1 with @xmath1136 marked points and @xmath1271 boundary components * identify @xmath1272 with @xmath1098 of @xmath1265 , thus producing an oriented surface of genus @xmath1273 with @xmath1136 marked points and @xmath1274 boundary components .
the exact bound is not important for our purposes .
we only want to stress that the theorem implies that @xmath1281 stabilizes for large @xmath1 . in particular , fixed @xmath1282 , the rational homology of @xmath356 stabilizes for large @xmath1 .
we have @xmath1283 , where @xmath1284 is the moduli space of riemann surfaces of genus @xmath1 with @xmath1285 marked points ( @xmath2 and @xmath1286 ) and a nonzero tangent vector at each point of @xmath1088 .
if @xmath1276 , then @xmath1284 is a smooth variety : in fact , an automorphism of a riemann surface that fixes a point and a tangent direction at that point is the identity ( it follows from uniformization and schwarz lemma ) .
recently , after works of tillmann ( for instance , @xcite ) and madsen - tillmann @xcite , madsen and weiss @xcite proved a much stronger statement of homotopy theory , which in particular implies mumford s conjecture .
for every @xmath1303 , move from the center @xmath1304 along a vertical trajectory @xmath572 of @xmath727 determined by the tangent vector @xmath1305 , until we hit the critical graph .
parametrize the opposite path @xmath1306 by arc - length , so that @xmath1307\rar \si$ ] , @xmath1308 lies on the critical graph and @xmath1309 . then , construct a new ribbon graph out of @xmath48 by `` adding '' a new vertex ( which we will call @xmath1310 ) and a new edge @xmath1311 of length @xmath1312 ( a _ tail _ ) , whose realization is @xmath1313)$ ] ( see figure [ fig : tail ] ) .
we set @xmath1329 , where we declare that @xmath1330 and @xmath1331 .
thus , we have a natural @xmath84 induced on @xmath215 .
moreover , we define @xmath142 acting on @xmath215 as @xmath1332)= \begin{cases } [ \s'_0(\ora{e } ) ] & \text{if $ \ora{e } \in e(\gg')$ and $ \ora{e}\neq\ora{e'}$}\\ [ \s''_0(\ora{e } ) ] & \text{if $ \ora{e } \in e(\gg'')$ and $ \ora{e}\neq\ora{e''}$ } \end{cases}\ ] ] if @xmath1321 and @xmath1322 are metrized , then we induce a metric on @xmath48 in a canonical way , declaring the length of the new edge of @xmath48 to be @xmath1333 .
suppose that @xmath1321 is marked by @xmath1334 and @xmath1335 is a tail contained in the hole @xmath1336 and that @xmath1322 is marked by @xmath1337 and if @xmath1338 is a tail contained in the hole @xmath1339 , then @xmath48 is marked by @xmath1340 , where @xmath901 is a new hole obtained _ merging _ the holes centered at @xmath1336 and @xmath1339 .
consider the combinatorial gluing maps @xmath1347 where @xmath1348 is obtained by simply gluing a graph @xmath48 in @xmath1349 with the fixed graph @xmath1343 , identifying the unique tail of @xmath1349 with the @xmath326-tail of @xmath1343 and renaming the new hole by @xmath901 .
on the other hand , consider the map @xmath1353 , that glues a torus @xmath1354 with one puncture and one boundary component to the unique boundary component of @xmath1355 , by identifying @xmath901 and @xmath1336 , and relabels the @xmath21-puncture by @xmath901 .
the composition of @xmath1356 followed by the map @xmath1357 that forgets the @xmath901-marking @xmath1358 induces an isomorphism on @xmath1359 for @xmath1360 , because it can be also obtained composing @xmath1351 and @xmath1352 .
let @xmath48 be a metrized ribbon graph in @xmath1363 . if @xmath901 is marking a vertex of valence @xmath76 or more , than just forget the @xmath901-marking .
if @xmath901 is marking a vertex of valence @xmath489 , then forget the @xmath901 marking and merge the two edges outgoing from @xmath901 in one new edge .
finally , if @xmath901 is marking a univalent vertex of @xmath48 lying on an edge @xmath339 , then replace @xmath48 by @xmath1364 and forget the @xmath901-marking .
the induced homomorphism on borel - moore homology @xmath1365 pulls @xmath1366 back to the combinatorial class @xmath1367 , corresponding to ( the closure of the locus of ) ribbon graphs with one univalent vertex marked by @xmath901 and @xmath1368 vertices of valence @xmath1369 for all @xmath1370 .
this last phenomenon can be understood by simply observing that @xmath1373 corresponds to opening the ( generically univalent ) @xmath901-marked vertex to a small @xmath901-marked hole , thus producing an extra trivalent vertex . as a consequence
, we have obtained that @xmath1376 is an isomorphism for @xmath1377 and pulls @xmath1366 back to @xmath1378 . + the other gluing map is much simpler : the induced @xmath1379 carries @xmath1380 to @xmath1381 , because @xmath1382 has @xmath61 trivalent vertices . + we recall that a class in @xmath1383 ( i.e. a stable class ) is a sequence of classes @xmath1384 , which are compatible with the stabilization maps , and that two sequences are equivalent ( i.e. they represent the same stable class ) if they are equal for large @xmath1 .
[ prop : witten - stable ] let @xmath1118 be a sequence of nonnegative integers such that @xmath1385 for large @xmath1386 and let @xmath1387 . define @xmath1388 and call @xmath1389 .
then , the collection + is a stable class , where @xmath1390 .
it is clear that an analogous statement can be proven for generalized witten cycles .
notice that proposition [ prop : witten - stable ] implies miller s result @xcite that @xmath33 and @xmath32 classes are stable .
carl - friedrich bdigheimer and ulrike tillmann , _ stripping and splitting decorated mapping class groups _ , cohomological methods in homotopy theory ( bellaterra , 1998 ) ,
196 , birkhuser , basel , 2001 , pp .
4757 .
david gieseker , _ lectures on moduli of curves _ , tata institute of fundamental research lectures on mathematics and physics , vol .
69 , published for the tata institute of fundamental research , bombay , 1982 .
william j. harvey , _ geometric structure of surface mapping class groups _ , homological group theory ( proc .
sympos . , durham , 1977 ) , london math .
lecture note ser .
36 , cambridge univ .
press , cambridge , 1979 , pp .
255269 .
nikolai v. ivanov , _ on the homology stability for teichmller modular groups : closed surfaces and twisted coefficients _
, mapping class groups and moduli spaces of riemann surfaces ( gttingen , 1991/seattle , wa , 1991 ) , contemp . math .
150 , amer .
soc . , providence , ri , 1993 , pp .
149194 .
finn f. knudsen , _ the projectivity of the moduli space of stable curves .
iii . the line bundles on @xmath1391 , and a proof of the projectivity of @xmath1392 in characteristic @xmath276 _ , math . scand .
* 52 * ( 1983 ) , no . 2 , 200212 . | we begin by describing the duality between arc systems and ribbon graphs embedded in a punctured surface and explaining how to cellularize the moduli space of curves in two different ways : using jenkins - strebel differentials and using hyperbolic geometry .
we also briefly discuss how these two methods are related .
next , we recall the definition of witten cycles and we illustrate their connection with tautological classes and weil - petersson geometry .
finally , we exhibit a simple argument to prove that witten classes are stable . 32g15 , 30f30 , 30f45 .
moduli of riemann surfaces , ribbon graphs , witten cycles . |
heavy baryons such as @xmath2 , @xmath3 , @xmath4 or @xmath5 containing one heavy quark can be successfully described by the heavy quark effective theory ( hqet ) , especially in leading order of the @xmath0-expansion , i.e. in the limit of heavy quark symmetry ( hqs ) ( for a review on hqet see @xcite , for details on hqs see @xcite ) . for each of the ground - state baryons
there are two independent current components , e.g. @xmath6q^k{\varepsilon}_{ijk}\quad\mbox{and}\quad j_{\lambda2}=[q^{it}c\tau\gamma_5\gamma_0q^j]q^k{\varepsilon}_{ijk},\ ] ] while the general structure of heavy baryon currents has the form @xmath7\gamma'q^k{\varepsilon}_{ijk},\ ] ] where the index @xmath8 means transposition , @xmath9 is the charge conjugation matrix , @xmath10 is a matrix in flavour space , and @xmath11 , @xmath12 and @xmath13 are colour indices .
the effective static field of the heavy quark is denoted by @xmath14 , and @xmath15 and @xmath16 are the dirac structures of the different current parts . in fig .
1 are shown some examples of diagrams up to two - loop order which we calculated .
the one- and two - loop diagrams can be grouped in three different classes , namely the heavy diquark system ( diagrams ( b ) , ( c ) , ( e ) , ( f ) and ( g ) ) , the light diquark system ( diagrams ( d ) and ( h ) ) and the irreducible vertex contribution ( diagrams ( i ) and ( j ) ) .
the first two can be handled in a manner similar to the meson case , and actually it turns out that the replacement @xmath17 leads from the diquark to the meson quark - antiquark case ( and not vice - versa ) .
so the mesonic case can be seen as a byproduct of our calculations .
the calculation of all these diagrams can be executed by nearly the same procedure .
the procedure is strictly algorithmic , so that the calculations could be automatized and programmed in mathematica . in the following
i want to emphasize the main steps of this calculation , and i want to do this for the first two - loop heavy diquark diagram which is displayed in fig . 2 ( the arrows on the gluon lines only indicate the momentum flow ) . before starting the main calculation , the colour structure can be treated .
in general it contains the gell - mann matrices , the epsilon tensor for resons of colourless , and the colour couplings of the gluons , which consists of a @xmath18 for a single gluon and a factor @xmath19 for a three - gluon vertex . in our case
we get @xmath20 @xmath21 is the number of colours , and the results can be expressed by the constants @xmath22 , @xmath23 , @xmath24 , @xmath25 and @xmath26 , which denotes the number of flavours . without the colour structure
we get @xmath27 the four - momentum of the heavy quark is given by @xmath28 , and we have @xmath29 .
this vertex correction can be split up into a _
dirac structure @xmath30 _ , a _ momentum structure @xmath31 _ and a _
scalar integral _ ( also called _
`` race integral '' _ ) @xmath32 with the help of the recursion formulas developed in @xcite it is possible to reduce each two - loop function @xmath33 of the hqet and each two - loop function @xmath34 of the massless qcd to corresponding one - loop functions @xmath35 and @xmath36 , which dissolve into euler s gamma functions .
the recursion formulas are one of the main parts of the developed program package , and we would like to thank david broadhurst for providing us with a reduce implementation of these formulas . by extracting the dirac structure ,
the remaining integral receives corresponding lorentz indices .
it can be evaluated covariantly , in this case into @xmath37 the parts @xmath38 and @xmath39 can be obtained by contracting the integral with the corresponding dual tensors .
this contraction pairs the momentum structure to the scalar products @xmath40 the scalar products `` operate '' on the entries of the two - loop function ( e.g. , @xmath41 decreases the first entry by one unit ) , and can therefore be identified with operators on the different entries of the two - loop functions .
the covariant evaluation in combination with the dirac structure gives @xmath42 for the heavy diquark system the matrix @xmath15 will always remain on the edge of the structure .
in contrast to this , for the light diquark system the dirac structure can be reduced to one of the main structures @xmath43
the result of the perturbation series to second order for the different ground - state hadron currents can be written in quite a compact form , if we make some slight assumption for @xmath15 . in this case , we can split off a scalar valued _ vertex function @xmath44_. if @xmath15 is totally antisymmetrized , @xmath47},\ldots\}$ ] , we get @xmath48 where @xmath49 is the space - time dimension , @xmath45 the number of dirac matrices and @xmath46 denotes the number of @xmath50 s in @xmath15 , i.e. being @xmath51 for an even and @xmath52 for an odd number of @xmath50 s
. problems will emerge from the appearence of @xmath1 , but i will postpone the discussion of this to the end of the talk . without looking at the parts from different classes of diagrams , the vertex function for the heavy baryon current in its generalized form
is given by @xmath53 this expression can be evaluated in powers of @xmath54 with @xmath55 and shows singularities which can be canceled by means of the minimal subtraction scheme ( ms ) .
indeed , the evaluation in @xmath56 is rather more simple .
this evaluation allows for the application of the modified minimal subtraction scheme ( @xmath57 ) . by using a general gauge parameter @xmath58
( @xmath59 means feynman gauge ) for the one - loop contributions , we get @xmath60 @xmath61 @xmath62 is omitted , because it has no significance for the following calculations .
the vertex function still contains uv - singularities . it is a bare quantity , from now on denoted by @xmath63 ,
which has to be renormalized .
this renormalization is done with the help of a renormalization factor .
dimensional regularization is used from the beginning , what remains is the subtraction with the aid of the modified minimal subtraction scheme , as indicated above . in the following
we simplify the notation by using the term @xmath64 instead of @xmath65 , but we keep in mind that we have used the @xmath57-scheme
. the divergencies of @xmath63 will be absorbed by the renormalization factor @xmath66 .
we have to state therefore that @xmath67 is a finite quantity . by expanding the renormalization factor @xmath66 in a double series @xmath68 a comparison of the coefficients
up to second order gives @xmath69 up to now we have left out the renormalization of the coupling constant @xmath70 and the gauge parameter @xmath58 , @xmath71 looking at terms up to second order in perturbation theory , we only need to correct the terms of first order . in this way we get corrections for the coefficient @xmath72 of the renormalization factor @xmath66 , namely @xmath73 where @xmath74 is the gauge dependent part of @xmath75 .
the renormalized coupling is an object in @xmath49 space - time dimensions .
therefore , it gets a double dependence in the renormalization scale @xmath76 , @xmath77 the beta - function for the coupling and the gauge , two essential tools for the renormalization group equation , can then be calculated as @xmath78 this can be used for the calculation of the anomalous dimension @xmath79 , @xmath80 by evaluating the finite quantities @xmath81 , @xmath82 and @xmath79 in a perturbation series , starting at first order , the comparison of the coefficients leads to a determining equation for the coefficients of the anomalous dimension , @xmath83 and on the other hand to a class of consistency checks , which can be revolved to the coefficients of the perturbation series , e.g. @xmath84 where @xmath85 again denotes the gauge dependent part of @xmath86 . for the derivation we should note that @xmath87
to renormalize the current , we also have to include the renormalization of the outer legs , because we have @xmath88\gamma'q^0=z_qz_q^{1/2}z_vj = z_jj\quad \rightarrow\quad\gamma_j=2\gamma_q+\gamma_q+\gamma_v.\ ] ] by taking this into account @xcite , to first order we get @xmath89 for the values @xmath90 and @xmath91 of @xmath92 , this expression is gauge independent .
so we can use these values , @xmath93 and @xmath25 and restrict ourselves to the feynman gauge . for the anomalous dimension of different baryon currents and different schemes we get @xmath94 the `` naive non - abelianization '' recipe ,
the estimate of the total result by calculating the quark - loop diagram and replacing @xmath26 by @xmath95 , does work for most of the cases with an accuracy of 15 to 30% . only in the case of the current @xmath96
it seems to fail .
note , however , that in this case the anomalous dimension is one order of magnitude smaller than in all other cases .
the difference of the anomalous dimension for the two different @xmath1-schemes turns out to be a correction of second order which is proportional to the coefficient @xmath97 of the beta - function of the coupling .
so we might think about connecting the already renormalized currents by some finite renormalization factor @xcite , namely @xmath98 so we get @xmath99 indeed , the ansatz @xmath100 leads to @xmath101 with @xmath102 o_{\gamma_5\gamma_i\gamma_j}\ = \
o_{\gamma_5\gamma_i\gamma_0}&=&0,\nonumber\end{aligned}\ ] ] and therefore @xmath103 the vertex structures @xmath104 , @xmath105 and @xmath106 are included for the sake of completeness , although they are not needed for the baryonic @xmath107-wave states discussed in this talk .
the meson case , already treated by trueman and larin @xcite , can again be obtained by the replacement @xmath17 .
+ * acknowledgments : * + this work was partially supported by the bmbf , frg , under contract 06mz566 .
moreover , i am grateful to the dfg , frg , for financial support .
i would like to thank j.g .
krner and o.i .
yakovlev for joining me in doing the presented calculations .
99 see for example : m.b .
wise , in `` particle physics the factory era '' , proceedings of the 6th lake louise winter institute , lake louise , canada 1991 , edited by b.a .
campbell , a.n .
kitching and f.g .
khanna ( world scientific , singapore , 1991 ) p. 222
m. neubert , `` heavy quark symmetry '' , phys .
* 245 * ( 1994 ) 259 s. groote , j.g .
krner and o.i .
yakovlev , phys .
* d54 * ( 1996 ) 3447 k.g .
chetyrkin and f.v .
tkachov , nucl .
* b192 * ( 1981 ) 159 ; + f.v .
tkachov , phys .
* b100 * ( 1981 ) 65 l.r .
surguladze and f.v .
tkachov , comput .
* 55 * ( 1989 ) 205 d.j . broadhurst and a.g .
grozin , phys .
* b267 * ( 1991 ) 105 d.r .
jones , nucl .
* b75 * ( 1974 ) 531 ; + e.s .
egorian and o.v .
tarasov , theor .
* 41 * ( 1979 ) 26 ; + x. ji and m.j .
musolf , phys .
lett . * b257 * ( 1991 ) 409 ; + v. gimenez , nucl . phys .
* b375 * ( 1992 ) 582 g. t hooft and m. veltman , nucl .
b44 * ( 1972 ) 189 ; * 50 * ( 1972 ) 318 ; + g. t hooft , nucl
. phys . * b61 * ( 1973 ) 455 p. breitenlohner and d. maison , commun . math .
* 52 * ( 1977 ) 11 ; * 52 * ( 1977 ) 39 d. kreimer , phys . lett .
* b237 * ( 1990 ) 59 ; + j.g .
krner , d. kreimer and k. schilcher , z. phys . * c54 * ( 1992 ) 503 ; + d. kreimer , utas - phys/94 - 01 ; hep - ph/9401354 ( unpublished ) t.l .
trueman , phys .
* b88 * ( 1979 ) 331 ; + s.a .
larin , phys .
* b303 * ( 1993 ) 113 | this talk presents details of the calculation of two - loop corrections to heavy baryon currents containing one heavy quark in leading order of the @xmath0-expansion , i.e. in the limit of heavy quark symmetry ( hqs ) .
the calculations lead to the determination of the anomalous dimension of these currents .
i also discuss problems with different @xmath1-schemes and their solution by the finite renormalization procedure .
0.75 cm 0.75 cm -1truecm 0truecm 0truecm 16truecm 24truecm mz - th/97 - 05 + hep - ph/9702374 + february 1997 + * two - loop anomalous dimensions * + * of heavy baryon currents * + * in heavy quark symmetry * + stefan groote + institut fr physik , johannes - gutenberg - universitt , + staudinger weg 7 , d-55099 mainz , germany + |
the momentum exchange between different species plays an exceptionally important role in complex ( dusty ) plasmas multi - component systems consisting of charged micron - sized ( `` dust '' ) grains embedded in a plasma @xcite . in a weakly ionized plasma , the momentum transfer in collisions with the neutral gas `` cool down '' the system , in particular grains and ions , and introduce some damping .
relative motion between the grains and other plasma components results in the so - called drag forces .
the neutral , ion , and ( to a lesser extent ) electron drag forces are known to affect and often dominate the grain component statical and dynamical properties @xcite .
relative motion between the grain and other plasma species is not the only mechanism which can be responsible for the momentum transfer .
a relevant example is the thermophoresis , which describes the phenomenon wherein small particles , suspended in a gas where a temperature gradient exists ( but macroscopic flows are absent ) , experience a force in the direction opposite to that of the gradient .
elementary consideration of this phenomenon has been given by einstein @xcite and then it was investigated in detail ( see e.g. refs .
@xcite and references therein ) . in the context of complex plasmas , jellum _
et al_. @xcite were apparently the first who recognized the possibility to manipulate the particles in gas discharges using the thermophoretic force . since then
, applying the vertical temperature gradients to compensate for the particle gravity has become a standard technique for controlled particle manipulation in laboratory experiments @xcite .
in complex plasmas , the thermophoretic force has its counterparts associated with the charged electron and ion components , provided the corresponding temperature gradients are present .
these electron and ion _ thermal _ forces appearing in plasmas have received much less attention than the conventional thermophoresis , or the corresponding drag forces .
brook @xcite calculated the force acting on a small _ uncharged _ particle suspended in a plasma in which temperature gradients and a magnetic field are present .
gnedovets @xcite took into account the particle charge and demonstrated that the electron and ion thermal forces can have significant effect on the particles which are smaller than the plasma screening length .
nevertheless , these forces have not yet been discussed in the context of complex ( dusty ) plasmas in any reasonable detail .
it seems of value , therefore , to revisit this issue with proper account of the progresses in the understanding of basic plasma - dust interactions achieved in the last two decades . in this paper
the expressions for the ion and electron thermal forces acting on a highly charged grain immersed in a weakly ionized plasma subject to temperature gradients are derived .
the focus is on spherical grains , which are smaller than the plasma screening length .
low collisionality is assumed , so that the mean free paths of all plasma species ( neutrals , ions , and electrons ) are long compared to the grain size , which corresponds to the limit of large knudsen number .
these conditions are typical for complex plasma experiments in laboratory gas discharges .
the obtained in this way expressions are simple enough and can be easily implemented for practical applications . the directions and the magnitudes of the forces
are analyzed and an interesting and non - trivial behavior is reported .
comparison with other forces indicates that the electron thermal force , in particular , can provide important contributions to the net force balance in typical complex plasma experiments .
the general expression for the force associated with the momentum transfer from a light species @xmath1 to the massive grain at rest is @xmath2 where @xmath3 , @xmath4 and @xmath5 are the corresponding mass , velocity distribution function , and ( velocity dependent ) momentum transfer cross section ( @xmath6 for neutrals , ions , and electrons , respectively ) .
the net momentum transfer occurs when the velocity distribution has some asymmetry ( e.g. relative motion ) . assuming weak asymmetry , we write @xmath7 , where the symmetric component @xmath8 is taken to be maxwellian @xmath9 here @xmath10 and @xmath11 are the density and temperature ( in energy units ) of the species @xmath1 .
the small asymmetric component @xmath12 , which gives contribution to the integral in ( [ mt ] ) , depends on the nature of the anisotropy . in the case of subthermal
drifts with relative velocity @xmath13 ( @xmath14 ) it reduces to @xmath15 , where @xmath16 is the thermal velocity .
this ansatz corresponds to the conventional calculation of the neutral , ion and electron drag forces for subthermal flows and has been thoroughly investigated earlier @xcite .
the focus of the present paper is on a complementary situation when relative drifts are absent , but the momentum transfer do occur due to the net momentum flux caused by the temperature gradients ( if both temperature gradients and subthermal drifts are present , the corresponding thermal and drag forces are added in a simple superposition . ) in this case , the asymmetric part of the velocity distribution function of the component @xmath1 can be approximated as @xmath17{\bf v}\nabla t_{\alpha},\ ] ] where @xmath18 is the thermal conductivity of the species @xmath1 .
this expression can be for instance derived by linearizing the kinetic equation with the bgk - like collision operator and expressing the effective collision frequency via the thermal conductivity .
it is easy to check that this form ensures that the self - consistent density gradient and/or electric field ( for charged species ) , build up in response to the temperature gradient , result in no net flux : @xmath19 ( i.e. , @xmath20 ) .
the fourier s law for heat transfer is also satisfied : @xmath21 .
equation ( [ f1 ] ) is an approximation , which is exact only for the special case of @xmath22 interactions ( for the rigorous mathematical treatment of non - uniform gases see ref .
its accuracy is , however , more than acceptable for our present purposes , especially in view of further simplifications involved in treating electron - grain and ion - grain collisions . substituting eq .
( [ f1 ] ) into ( [ mt ] ) we get after integration over the angles in spherical coordinates @xmath23 where the remaining integration is over the reduced velocity @xmath24 . for grain - neutral collisions
the momentum transfer cross section is velocity - independent , @xmath25 , and the integration yields @xmath26 , where @xmath27 is the particle radius .
this coincides with the celebrated expression by waldmann @xcite for the conventional thermophoretic force .
the force is directed towards lower gas temperature , because hotter atoms transfer more momentum to the grain than the colder ones .
it is convenient to write the generic expression for the thermal forces in complex plasmas as @xmath28 for the neutral component ( thermophoresis ) @xmath29 .
the factors @xmath30 account for the electrical interactions between the ions ( electrons ) and the charged grain .
they depend on the shape of the electrical potential around the grain .
we use the conventional debye - hckel ( yukawa ) form , @xmath31 , where @xmath32 is the grain charge and @xmath33 is the plasma screening length . in low - temperature gas discharges
the dominant charging process is the continuous absorption of electrons and ions on the grain surface . in this case
the charge is negative and proportional to the product of the grain radius and the electron temperature , viz .
@xmath34 , where @xmath35 is the coefficient of order unity , which depends on the plasma parameter regime @xcite . in the absence of substantial ion flows and strong nonlinearities in ion - grain interaction ,
the screening length is approximately given by the ion debye radius @xmath36 , provided @xmath37 .
when ion - grain interaction is strongly non - linear , the effective screening length exceeds the ion debye radius @xcite , approximate fits are available in the literature @xcite .
although it is well recognized that the long range asymptote of the electrical potential can be modified ( by e.g. continuous plasma absorption on the grain surface @xcite or plasma ionization / recombination effects @xcite ) , this is expected to affect merely grain - grain interactions , but not the momentum transfer from the ions and electrons .
normally , the grain radius in complex plasmas is much smaller than the plasma screening length . in this case
, the electron - grain interaction can be called `` weak '' , in the sense that its range the coulomb radius @xmath38 is much smaller than the screening length @xmath33 @xcite .
this implies that the coulomb scattering theory is appropriate to describe electron - grain elastic collisions .
the general momentum transfer cross section for scattering in the coulomb potential @xmath39 is @xmath40^{1/2},\ ] ] where @xmath41 and @xmath42 .
the maximum impact parameter @xmath43 is necessary to avoid the logarithmic divergence of the cross section . in the standard coulomb scattering theory @xmath44 , and
this choice is appropriate for the weak electron - grain interaction @xcite .
the minimum impact parameter @xmath45 is zero in the standard coulomb scattering theory . in the considered case , however
, some electrons are falling onto the grain instead of being elastically scattered in its electrical potential .
consequently , @xmath45 should be set equal to the maximum impact parameter corresponding to the electron collection by the grain , @xmath46 .
the orbital motion limited ( oml ) theory yields @xcite @xmath47 for @xmath48 and @xmath49 otherwise ( sufficiently slow electrons can not be collected even in head - on collisions with the grain due to electrical repulsion ) . the momentum transfer cross section for electron collection
is simply @xmath50 .
expressing now velocity in terms of @xmath51 , substituting @xmath52 into eq .
( [ force ] ) and performing the integration with the appropriate limits yields : @xmath53 where @xmath54 is the electron - grain coulomb logarithm @xmath55 the function @xmath56 is defined as @xmath57 .
equations ( [ generic ] ) , ( [ etf ] ) and ( [ ecl ] ) constitute the expression for the electron thermal force .
it is similar to that obtained earlier by gnedovets @xcite . to get
identical expressions one needs to neglect unity compared to @xmath58 in the equations of ref .
@xcite . and
charge @xmath35.,width=245 ] for an uncharged particle @xmath59 and the electron thermal force pushes the grains in the direction of lower electron temperature , similarly to the thermophoretic force . according to eq .
( [ etf ] ) the contributions from collection and scattering are directed oppositely to each other . for a sufficiently high charge ,
the scattering part becomes dominant and the force reverses direction .
the physical reason is that the ( coulomb ) scattering momentum transfer cross section quickly decreases with velocity [ @xmath60 , see eq .
( [ sigma ] ) ] , so that the cold electrons are more effective in transferring their momentum upon scattering . in this regime ( @xmath61 ) , the thermal force acts in the direction of higher electron temperature . in figure [ fig1 ] the curve separating the positive and negative values of @xmath62
is plotted in the plane ( @xmath63 , @xmath35 ) . for most experimental conditions @xmath61
should be expected , although the transition line does not seem unreachable @xcite .
the coulomb radius of ion - grain interaction @xmath64 is not necessary small compared to the plasma screening length due to the presence of a ( normally ) large factor @xmath65 electron - to - ion temperature ratio @xcite .
the interaction range can exceed @xmath33 and considerable amount of momentum transfer can occur for impact parameters beyond @xmath33 .
this implies that the standard coulomb scattering approach is inappropriate @xcite .
it makes sense to consider two regimes of ion scattering separately . in the regime of
_ moderate ion - grain interaction _ , an extension of the standard coulomb scattering theory is possible by taking into account the momentum transfer from the ions that _ approach _ the grain closer than @xmath33 @xcite .
this results in @xmath66 .
this approximation demonstrates good accuracy for @xmath67 and reduces to @xmath44 in the limit of weak ion - grain interaction [ @xmath68 .
the impact parameter corresponding to the ion collection is @xmath69 and the corresponding collection cross section is @xmath70 . combining the contributions from collection and scattering yields : @xmath71 where @xmath72 is the ( modified ) ion - grain coulomb logarithm @xmath73dx.\ ] ] in typical complex plasmas with @xmath74 , @xmath75 , and @xmath76 , the coulomb logarithm can be roughly estimated as @xmath77 , where @xmath78 and @xmath79 is the ion scattering parameter @xcite .
the approach is reliable up to @xmath80 .
since the product @xmath81 is normally quite large , @xmath82 , it follows that ( i ) the scattering provides dominant contribution to the momentum transfer ; ( ii ) @xmath83 , i.e. the grains are pushed into the region with higher ion temperature .
the physical reason is again fast decrease of the scattering momentum transfer cross section with the ion velocity , so that cold ions transfer more momentum to the grain .
this is not a unique example when the force acting on a charged grain is directed oppositely to the net ion momentum flux .
another example is related to the sign reversal of the ion drag force acting on an absorbing particle in the highly collisional ( continuum ) limit @xcite , although the detailed physics is different . in the regime of
very _ strong ion - grain interaction _ , the scattering is characterized by the formation of a potential barrier for ions with impact parameters above the critical one @xmath84 , which considerably exceeds @xmath33 . for @xmath85 scattering with large angles
occurs , which gives the major contribution to the momentum transfer .
relative importance of momentum transfer from distant collisions with @xmath86 decreases rapidly with the increase in ion - grain interaction strength .
the detailed consideration of the momentum transfer in this regime can be found in refs .
@xcite . in the limit @xmath87 ( i.e. @xmath88 @xcite )
the total momentum transfer cross section ( collection and scattering ) can be roughly approximated as @xmath89 , with @xmath90 + 2\ln[\beta(v)]\right\}$ ] .
the integration yields @xmath91dx.\ ] ] the upper limit of integration can be chosen as @xmath92 to avoid unphysical regime of negative cross section in this approximation .
however , due to the presence of the exponentially small term in @xmath56 , taking @xmath93 will not produce big errors for @xmath94 .
note that the sign of @xmath95 is changed from negative to positive upon increasing @xmath96 ( in the considered approximation for @xmath97 this happens at @xmath98 ; the exact value is rather sensitive to the functional dependence of the cross section on the ion velocity and , thus , is subject to significant uncertainty ) .
physically , the sign reversal occurs because scattering in the yukawa potential in the limit of strong interaction tends to that on a hard sphere with the radius @xmath84 , which only weakly ( logarithmically ) depends on the ion velocity .
having derived the expressions for the ion and electron thermal forces , let us discuss their relative importance .
the relationship between @xmath99 and @xmath100 depends on many factors and is , in principle , arbitrary .
in weakly ionized plasmas @xmath101 ( where @xmath102 is the transport cross sections for collisions with neutrals ) and the force ratio becomes @xmath103 , where the quasineutrality condition @xmath104 has been used .
assuming further that most of the contribution to the thermal forces comes from elastic scattering , and setting @xmath105 for simplicity , we end up with @xmath106 for comparable temperature gradients .
the first factor is small , whilst the second can be quite large , and thus various situations are possible .
another situation deserving attention is when a gradient of the neutral gas temperature is created ( e.g. by heating the lower electrode of a parallel plate rf discharge ) to compensate for grain gravity ( thermophoretic levitation ) @xcite . in this case , the ion temperature is likely coupled to the neutral gas temperature , @xmath107 .
the neutral and ion thermal forces are directed in the opposite directions ( in the regime of weak and moderate ion - grain coupling ) .
their ratio is @xmath108 . assuming @xmath109
@xcite and @xmath110 we find that @xmath100 dominates over @xmath111 for @xmath112 . for a more typical ( in laboratory gas discharges ) ionization fraction @xmath113
, the ion thermal force is only slightly reducing the effect of thermophoretic force .
it is worth to remind that in laboratory gas discharges , ions produced in the plasma bulk are drifting towards the walls and electrodes of the chamber .
so even if some gradients of the ion temperature are present , the related effects are likely masked by the more pronounced ion drag force associated with these drifts .
for example , consider the expression for the ion drag force derived in ref .
@xcite for the subthermal ion drift regime and weak - to - moderate ion - grain coupling , @xmath114 ( @xmath115 is the coulomb logarithm for ion scattering , which gives the dominant contribution to the ion drag force ) , along with the relation between the ion drift velocity and the electric field @xmath116 , where @xmath117 is the characteristic momentum transfer frequency .
the coulomb logarithms encountered in the calculations of the ion drag and ion thermal forces are not exactly the same , but are comparable @xmath118 .
this results in a particularly simple estimate @xmath119 .
the electric fields of the order of 1 v / cm are quite natural even for the bulk regions of gas discharges ( ambipolar fields ) .
the ion temperature is normally close to the neutral gas ( room ) temperature , and thus for most situations @xmath120 is orders of magnitude less than 1 ev / cm . this implies that in the considered case @xmath121 .
electrons are also drifting towards the discharge walls and electrodes .
in the regime of ambipolar diffusion electron and ion fluxes are equal to each other .
this implies that the electron drag force is negligible as compared to the ion drag force in this regime , due to small electron mass .
however , electron temperature gradients can result in the thermal force acting on the particles .
the existence of significant gradients of @xmath122 is known from earlier spatially resolved probe measurements of the electron energy distribution function in various gas discharges @xcite . for devices used in complex plasma research , the more recent results from probe measurements @xcite , optical emission spectroscopy @xcite , as well as from numerical modeling @xcite all revealed gradients in @xmath122 of the order of @xmath123(ev / cm ) .
the spatial distribution of @xmath122 generally depends on discharge geometry , plasma parameters , and can be affected by the presence of grains @xcite .
let us therefore assume @xmath124 ev / cm and restrict ourselves to the comparison of the absolute magnitudes of the electron thermal force and other forces acting on a small grain in the bulk of a gas discharge .
we take the plasma parameter set from ref .
@xcite used to estimate relative importance of the electrical and ion drag forces : argon gas at a pressure @xmath125 pa ( @xmath126 @xmath127 ) , @xmath128 @xmath127 , @xmath129 ev , @xmath130 ev , @xmath131 @xmath132 m ( @xmath133 ) , and @xmath134 ( estimated from the collisionless oml theory ) . from eqs .
( [ etf ] ) and ( [ ecl ] ) we get @xmath135 . the electron thermal conductivity in a weakly ionized plasma is @xmath136 with @xmath137 ( @xmath138 @xmath139 for @xmath140 ev in argon ) .
this results in @xmath141 dyne .
this force is more than three times larger than the force of gravity , experienced by the grain of this size ( and material density of @xmath142 g/@xmath143 ) in ground - based experiments .
the equivalent electric field @xmath144 , for which @xmath145 , is @xmath146 v / cm .
such a field would produce a significant plasma anisotropy , characterized by superthermal ion flows ( fig .
4b from ref .
@xcite ) . finally , this magnitude is comparable to the _ maximum _ value of the ion drag force the grain can experience in subsonic ion flows for this set of parameters ( @xmath147 dyne according to fig .
4a from ref .
@xcite ) .
observation of big grains , trapped in standing striations of a stratified dc glow discharge @xcite , provides another example where the electron thermal force can play a significant role .
the electron temperature is known to increase considerably in the head of the striation ( where their energy is of the order of the first excitation potential ) , which can explain that even very massive grains can be confined there @xcite .
other situations wherein plasma thermal forces may play significant role include charged aerosols , dust in planetary ( e.g. , earth ) atmospheres , fusion devices , and even quark - gluon plasma @xcite .
present results may not , however , be directly applicable to some of these cases . for example , the regime of fully ionized magnetized plasma , considered recently @xcite , is apparently more relevant in the context of dust in fusion devices . as a final remark
, we point out that complex plasmas represents a natural example where negative thermophoresis
( force pointing towards higher temperatures ) can exists .
a theoretical criterion for negative thermophoresis in dilute non - ionized gases has been discussed in a recent publication @xcite .
to conclude , the forces acting on small particles in a weakly ionized plasma subject to ion and electron temperature gradients have been analyzed with the main emphasis on applications to complex ( dusty ) plasmas .
the presented estimates demonstrate that the ion thermal force is usually of minor importance in gas discharges presently employed in complex plasma research .
in contrast , the electron thermal force can well be comparable to other forces acting on micron - size particles .
this finding should be properly addressed when developing new ( and updating existing ) numerical codes to model particle transport in gas discharges .
the expressions derived in this paper are easy for practical implementation and can serve as a theoretical basis for further detailed studies .
99 v. e. fortov and g. e. morfill , _ complex and dusty plasmas : from laboratory to space _ , ( crc press 2010 ) .
v. e. fortov , a. v. ivlev , s. a. khrapak , a. g. khrapak , and g. e. morfill , phys . rep . * 421 * , 1 ( 2005 ) .
s. a. khrapak , a. v. ivlev , and g. e. morfill , phys .
e * 70 * , 056405 ( 2004 ) .
a. einstein , z. phys .
* 27 * , 1 ( 1924 ) .
l. waldmann , z. naturforsch . *
14a * , 589 ( 1959 ) .
l. talbot , r. k. cheng , r. w. schefer , and d. r. willis , j. fluid mech . * 101 * , 737 ( 1980 ) .
z. li and h. wang , phys .
e * 70 * , 021205 ( 2004 ) .
g. m. jellum , j. e. daugherty , and d. b. graves , j. appl . phys . * 69 * , 6923 ( 1991 ) .
h. rothermel , t. hagl , g. e. morfill , m. h. thoma , and h. m. thomas , phys .
lett . * 89 * , 175001 ( 2002 ) .
m. schwabe , m. rubin - zuzic , s. zhdanov , h. m. thomas , and g. e. morfill .
lett . * 99 * , 095002 ( 2007 ) .
m. schwabe , m. rubin - zuzic , s. zhdanov , a. v. ivlev , h. m. thomas , and g. e. morfill , phys .
lett . * 102 * , 255005 ( 2009 ) . c. schmidt , o. arp , and a. piel , phys .
plasmas * 18 * , 013704 ( 2011 ) . v. n. tsytovich , phys . usp . * 167 * , 57 ( 1997 ) .
j. e. allen , b. m. annaratone , u. de angelis , j. plasma phys . * 63 * , 299 ( 2000 ) .
a. v. filippov , a. g. zagorodny , a. f. pal , a. n. starostin , and a. i. momot , jetp lett . * 86 * , 761 ( 2007 ) .
s. a. khrapak , b. a. klumov , and g. e. morfill , phys .
lett . * 100 * , 225003 ( 2008 ) .
s. a. khrapak , a. v. ivlev , and g. e. morfill , phys .
plasmas * 17 * , 042107 ( 2010 ) . in a recent experimental study of particle charging in a dc glow discharge at high pressures [ s. a. khrapak _
et al_. , epl * 97 * , 35001 ( 2012 ) ] the values of @xmath148 and @xmath149 have been reported , which correspond to the regime @xmath150 .
s. a. khrapak , s. k. zhdanov , a. v. ivlev , and g. e. morfill , j. appl . phys . * 101 * , 033307 ( 2007 ) . s. v. vladimirov , s. a. khrapak , m. chaudhuri , and g. e. morfill , phys .
* 100 * , 055002 ( 2008 ) .
a. v. filippov , a. g. zagorodny , and a. i. momot , jetp lett .
* 88 * , 24 ( 2008 ) .
m. wolter , a. melzer , o. arp , m. klindworth , and a. piel , phys .
plasmas * 14 * , 123707 ( 2007 ) .
s. mitic , b. a. klumov , m. y. pustilnik , and g. e. morfill , jetp lett . *
91 * , 231 ( 2010 ) .
v. land and w. j. goedheer , new
. j. phys .
* 9 * , 246 ( 2007 ) .
o. arp , j. goree , and a. piel , phys .
e * 85 * , 046409 ( 2012 ) . | expressions for the ion and electron thermal forces acting on a charged grain , suspended in a weakly ionized plasma subject to temperature gradients , are derived .
the main emphasize is on the conditions pertinent to the investigations of complex ( dusty ) plasmas in gas discharges .
estimates show that for the electron temperature gradients @xmath0(ev / cm ) typically encountered in laboratory gas discharges , the electron thermal force can become an important player among other forces acting on micron - size grains . |
stripe ordering phenomena on the nanoscale seem to be an inherent consequence of electronic correlations in high - superconducting materials .
they are most prominent close to @xmath3 hole doping @xcite but they were also identified within a broader doping range in the pseudo - gap regime @xcite .
however , the nature of the stripe order varies significantly for different cuprate materials .
unidirectional charge- ( cdw ) and spin - density waves ( sdw ) have been detected in many cuprates by neutron scattering and x - ray experiments @xcite .
however , the details are strongly material dependent .
neutron scattering experiments on la@xmath4nd@xmath5sr@xmath6cuo@xmath7 ( lnsco ) at @xmath8 @xcite revealed static antiferromagnetic ( af ) spin - density wave order with a period of eight lattice constants and a concomitant charge - density wave with half this period .
a similar spin structure was found in la@xmath9ba@xmath6cuo@xmath7 ( lbco ) @xcite , in la@xmath9ba@xmath10sr@xmath11cuo@xmath7 ( lbsco ) ( with @xmath12 ) @xcite , and in la@xmath4eu@xmath10sr@xmath6cuo@xmath7 @xcite , where sdw and cdw coexist at and near @xmath8 .
a common feature of these cuprates is an anisotropic low - temperature tetragonal ( ltt ) phase and , in addition , a strong dopant disorder ; both , the ltt structure and the dopant disorder are supposed to pin stripes .
however , not all cuprates exhibit a ltt phase or dopant disorder .
while every chemical doping of the cuprate parent compounds introduces disorder , its impact on the superconducting properties depend decisively on the distance between the cuo@xmath13 planes and the dopants . in la@xmath9sr@xmath6cuo@xmath7 ( lsco )
dopants are randomly positioned close to the cuo@xmath13 planes generating effective disordered impurity potentials to the in - plane electrons .
in contrast , oxygen dopants order in cuo chains in yba@xmath13cu@xmath14o@xmath15 ( ybco ) that are separated from the cuo@xmath13 planes by a bao plane and they are roughly at twice the distance from the cuo@xmath13 planes than sr is in lsco .
hence ybco is minimally affected by the dopants impurity potentials and is therefore considered as the cleanest material in the cuprate family . in pure lsco , which exhibits no ltt phase , spin stripes
have been detected below @xmath16 , but no cdw order @xcite .
although no static stripes have so far been reported for ybco , electron - nematic order is inferred from anisotropies @xcite found in neutron scattering @xcite and thermoelectric transport @xcite measurements .
in addition , incommensurate spin fluctuations are detected in superconducting ybco and a static cdw appears in the presence of an external magnetic field @xcite .
the different behavior of ybco may arise from both the absence of dopant disorder and the ltt structure .
stripe phenomena in cuprates must therefore be considered strongly material specific . here
we present a comprehensive analysis of their delicate response to the presence of disorder . in the past few years
the theoretical understanding of stripe formation in cuprates has advanced . early on neutron - scattering data on lnsco at
hole doping @xmath8 @xcite suggested the formation of a spin - ladder structure in the cuo@xmath13 planes .
this structure is built from half - filled , three - legged spin ladders , separated by quarter filled chains . the af spin structure changes sign from one ladder to the next , resulting in a wavelength of eight lattice constants .
this same spin structure was later detected also in lbco @xcite .
indeed , theoretical models based on coupled spin - ladders describe @xcite with some success inelastic neutron scattering data on lbco @xcite . already before the experimental discovery of stripes , zaanen and gunnarsson @xcite and
machida @xcite predicted the formation of spin stripes in doped antiferromagnets from mean - field analyses of the hubbard model .
the spin stripes suggested by tranquada and coworkers were found also in various numerical calculations using the hubbard model @xcite , the @xmath17-@xmath18 model @xcite or the spin - fermion model @xcite , and their existence is by now well established .
a more delicate problem is the understanding of the coexistence of or competition between spin- and charge - stripe order and superconductivity . in disordered systems ,
@xmath19-wave superconductivity and antiferromagnetism can coexist @xcite . on the other hand , within a mean - field treatment of the @xmath17-@xmath18 model and variants of it ,
the above described spin - ladder state was found to coexist with a striped form of @xmath19-wave superconductivity @xcite .
this superconducting ( sc ) state is modulated in space with the same period as the spin structure ; its sc order parameter is minimal in the center of the af spin ladder and maximal in between the spin stripes .
striped superconductivity in this context corresponds to a unidirectional pair - density wave ( pdw ) state @xcite .
notably this pdw oscillates with twice the wavelength of the accompanying cdw , which is caused by periodic zero - crossings . below we discuss and contrast the pdw with the periodically modulated @xmath19-wave superconductivity ( mdsc ) , which lacks a sign change , thus oscillating with the same wavelength as the cdw . while the emerging spin order in these models is typically the same in a wide range of parameters for constant hole density , the stability of coexisting spin order and superconductivity varies strongly .
the @xmath17-@xmath18 model displays two distinct limits which have been assessed within a mean - field approach in ref .
@xcite : ( @xmath20 ) the strong @xmath18 limit , referred to as the so - called `` @xmath21-model '' , where @xmath21 parametrizes the attractive nearest - neighbor interaction and represents the dominating contribution to the interaction ( see sec . [ subsec : vmodel ] ) and ( @xmath22 ) the weak @xmath18 limit .
the latter is well described by a bcs model which includes a repulsive on - site interaction and is appropriately named the `` @xmath23-model '' ( see sec .
[ subsec : umodel ] ) . in both limits ,
spin stripes form .
while in the @xmath21-model the spin ladders are separated by quasi - one dimensional ( 1d ) @xcite , mutually uncorrelated sc stripes , antiferromagnetism is weak in the @xmath23-model and coexists with two - dimensional ( 2d ) superconductivity @xcite .
the cuprates are estimated to be placed in between the two limits .
some cuprate materials show characteristics which are qualitatively explained by the @xmath21-model , whereas others come closer to the physics of the @xmath23-model . as pointed out in ref .
@xcite , for clean systems there is a balance between af and sc correlations in a certain hole - doping range within which specific details matter .
this regime reacts sensitively to dopant disorder and is likely to describe the physics of 214-compounds . in this paper
we discuss the impact of inhomogeneities on striped superconductors and compare the theoretical results with the experimental observations . in sec .
[ sec : ham ] we introduce the @xmath23- and the @xmath21-models as derived in ref
. @xcite and recollect their respective mean - field theories . in sec .
[ sec : res ] we discuss the solutions of the @xmath23-model in the presence of disorder and vortices , and the effects of disorder on the solutions of the @xmath21-model . in sec .
[ sec : sum ] we summarize and discuss our results .
we follow the general idea that the one - band hubbard model describes well the low - energy physics of the cuo@xmath13 planes of the cuprates , including antiferromagnetism and superconductivity @xcite . at strong coupling , a unitary transformation maps the hubbard model onto the @xmath17-@xmath18 model with an af exchange coupling @xmath24 @xcite .
the non - local interaction in the @xmath17-@xmath18 model accounts for both superconductivity and antiferromagnetism already on the mean - field level @xcite . here
we use an ansatz introduced by kagan and rice , in which the projection to exclude doubly occupied sites is replaced by an on - site repulsion term @xmath25 , leading back to the original @xmath17-@xmath18 model in the limit @xmath26 @xcite . on mean - field level , this ansatz is equivalent to the fully decoupled bcs hamiltonian @xmath27 with attractive nearest - neighbor interaction of strength @xmath21 and an on - site repulsion of strength @xmath23 , if @xmath21 is identified with @xmath18 @xcite . we employ this model on a square lattice including randomly positioned on - site impurity potentials @xmath28 and a perpendicular orbital magnetic field .
specifically the model hamiltonian reads @xmath29 where @xmath30 creates an electron on site @xmath20 with spin @xmath31 and @xmath32 .
the hopping matrix elements between nearest and next - nearest neighbor sites are denoted by @xmath33 and @xmath34 , respectively .
an electron moving in the external magnetic field from site @xmath35 to @xmath20 acquires the peierls phase @xmath36 , where @xmath37 and @xmath38 is the vector potential in the landau gauge .
the attractive nearest - neighbor interaction is parametrized by @xmath39 and the chemical potential @xmath40 is adjusted to fix the electron density @xmath41 , where @xmath42 is the hole concentration a mean - field decoupling of the interaction term @xmath43 , leads to the standard term for bcs type superconductivity , plus an af interaction term of the form @xmath44 .
the latter results in an energy gain @xmath21 for two electrons on nearest - neighbor sites with antiparallel aligned spins .
consequently , the system exhibits an af phase above a critical @xmath21 ( @xmath45 ) @xcite . in this regime
double occupancies become very rare .
hence the term @xmath46 is small compared to the @xmath21-term and has no qualitative consequences in this scenario .
this case is represented by the @xmath21-model . on the other hand ,
if @xmath47 , no af order originates from @xmath21 and antiferromagnetism is controlled by @xmath23 .
this case is adequately depicted by the @xmath23-model @xcite . as discussed above
, the interaction @xmath49 is not important for magnetism in the limit @xmath47 , but leads to the standard bcs expression for the sc order parameter .
thus the hartree - fock decoupling of @xmath50 can be restricted to the @xmath23-model : @xmath51 where @xmath52 . in the following
we will focus mainly on hole densities at and near @xmath8 .
the non - local pairing amplitude is defined as @xmath53 the @xmath23-model has been the starting point for numerous investigations on disorder- @xcite and field - induced antiferromagnetism @xcite . in the following we fix the pairing interaction at @xmath54 and set the strength of the non - magnetic impurity potential @xmath28 to @xmath55 .
all fields , i.e. @xmath56 , the local electron density @xmath57 , and the local magnetization @xmath58 are calculated self - consistently from the solutions of the associated bogoliubov - de gennes ( bdg ) equations . a detailed derivation of the bdg equations is presented in refs .
the @xmath19-wave order parameter on a lattice site @xmath20 is defined as @xmath59 where @xmath60 . in order to describe a possibly appearing modulation of the superconducting order parameter , as in the pair - density wave state @xcite
, we subdivide the pairing order parameter into three contributions @xcite @xmath61 where the wavevector @xmath62 parametrizes the modulation of the pairing amplitude and reflects a finite center - of - mass momentum @xmath63 of the electron pairs .
@xmath64 accounts for a homogeneous component of the order parameter
. two distinct types of sc states are possible solutions for @xmath65 : if @xmath66 , the sc order parameter shows a stripe pattern modulated with wave vector @xmath62 , where the superconducting stripes are separated by channels of zero pairing amplitude across which the sc order parameter changes sign .
if the absolute values of @xmath67 and @xmath68 are smaller than the homogeneous component @xmath69 , @xmath70 is always finite and modulated with wave vector @xmath71 .
although the `` pure '' pdw with @xmath66 was phenomenologically suggested to be the ground state of striped high-@xmath72 superconductors @xcite , all calculations of mean - field type so far led to ground states with dominating uniform component @xmath64 , c.f .
@xcite .
for large attractive interactions @xmath45 the hartree - fock decoupled hamiltonian of eq .
is effectively described by the @xmath21-model hamiltonian @xmath74 where @xmath56 is defined as in eq . .
the term @xmath75 is responsible for antiferromagnetism ; it is controlled by the same interaction parameter @xmath21 as the sc order parameter . since in this limit
af correlations are strong enough to separate the system locally into af regions close to half filling and into hole - rich regions far from half filling , the @xmath23-term contributes little and is therefore omitted from @xmath76 . for hole doping around @xmath8 ,
the solutions of the bdg equations for @xmath77 in the impurity - free case are very similar to those of the @xmath23-model @xcite .
the main difference is the stronger magnetization of the magnetic stripes in the @xmath21-model . within the af stripes ,
superconductivity vanishes , and the regions with dominant superconductivity are spatially separated into uncorrelated , quasi 1d filaments .
therefore the phase relation between them is not fixed and the `` pure '' pdw state is degenerate to a solution with finite @xmath64 . because of the one - dimensional character of the solutions of the @xmath21-model , they barely adjust to impurities in comparison to the solutions of the @xmath23-model . this characteristic difference is discussed in sec .
[ sec : v ] . for a thorough analysis of our results
we also discuss the following momentum space quantities . with the fourier transform of the factorized square of the magnetization @xmath78 we approximate the magnetic structure factor .
in addition we discuss the absolute squares of the fourier transformed @xmath19-wave order parameter @xmath79 and the deviation from the average electron density @xmath80 @xmath81 in the strong disorder limit , the suppression of the electron density at the random impurity sites dominates the charge distribution even for small impurity potentials .
coherent charge oscillations with small amplitudes are therefore barely visible in the fourier transform @xmath82 .
since we are especially interested in these coherent charge modulations we exclude the impurity sites in the presence of strong disorder from the sum over sites in the fourier transformation .
the momentum distribution is given by @xmath83 where the fermionic operators @xmath84 are the fourier transform of the real - space operators @xmath85 .
the spectral density at the fermi energy @xmath86 is determined from the imaginary part of the retarded green s function @xmath87 @xmath88 allows for the determination of partially reconstructed fermi surfaces which are characteristic for inhomogeneous superconductors .
homogeneous @xmath19-wave superconductors exhibit a finite gap away from the nodes on the brillouin zone diagonals .
@xmath89 is therefore continuous everywhere except across the nodes .
however , stripe formation leads to a reconstruction of the fermi surface attributed to finite pairing - momentum . in this case
the pair density @xmath90 ( eq . ) is readjusted and , as a result , the superconducting energy gap no longer covers all sectors of the underlying fermi surface .
impurities , on the other hand , induce bound states within the superconducting gap , leading to a redistribution of spectral weight into the gap which can be observed in @xmath88 . in this process the transition from partially filled to empty states in @xmath89 narrows , implicating a partial reconstruction of the fermi surface .
the momentum distribution of the pair density @xmath90 is defined as @xcite @xmath91 it measures the average correlation of the two occupied electron states @xmath92 and @xmath93 and thereby the parts of the brillouin zone where electron pairing occurs .
eventually , we discuss the information contained in @xmath94 on the spin state .
this quantity is maximal in those parts of the brillouin zone where antiferromagnetism dominates .
thus the competition between superconductivity and antiferromagnetism can be analyzed by comparing @xmath95 and @xmath96 .
the hamiltonian gives rise to unidirectional stripes which have been identified in the real - space quantities of clean @xmath19-wave superconductors above a critical on - site repulsion @xmath98 @xcite .
for a clean ( @xmath99 ) @xmath19-wave superconductor ( @xmath19sc ) the free energy of stripe solutions is typically close to the homogeneous @xmath19sc solutions .
depending sensitively on the initial conditions of the self - consistent bdg calculations we find either stripes or homogeneous af order , both coexisting with superconductivity .
in clean systems , regular stripe solutions emerge only if the self - consistency loop is started from a striped initial state .
however , in the presence of perturbations , such as impurities or vortices , or by a rectangular lattice geometry , stripes are pinned and the stripe solutions become robust against changes of the initial conditions .
this relates to the notion that in the unperturbed superconductor fluctuating stripes are present which become static by the pinning to defects @xcite . in the @xmath23-model below a critical pairing interaction an emerging stripe state
exhibits a dominant uniform component of the superconducting order parameter , i. e. @xmath100 in eq .
, in addition to the finite - momentum pairing amplitudes . in this case the superconducting order parameter @xmath70 does not feature a sign change and we call this state a modulated @xmath19-wave supercondcutor ( mdsc ) . in ref .
@xcite it was shown within a momentum space formulation that a translation - invariant hamiltonian in zero magnetic field can nevertheless support a `` pure '' pdw groundstate with @xmath101 for a similar set of parameters but with a stronger pairing interaction @xmath102 . there , an analytic approximation to gorkov s equations was considered .
solutions of this kind are also found in the impurity - free real - space model used here , although the pdw solution converges into a _ local _ energy minimum and is slightly higher in energy then the mdsc solution .
stripe solutions are indeed observed in a broad hole - doping range . in the strong disorder limit , however , we find stripe states only close to @xmath8 .
fig1a.pdf ( -3,59)*a * fig1b.pdf ( -3,59)*b * fig1c.pdf ( 3,157)*c * choosing sinus - shaped initial values with a wavelength of @xmath103 for the self - consistent fields @xmath56 and @xmath104 , the impurity - free system ( @xmath99 ) exhibits horizontal ( or vertical ) stripes in the real - space quantities over a wide hole - doping range . in fig .
[ fig : impfree ] the order parameter , the electron density , and the staggered magnetization are shown for @xmath8 .
obviously , a cdw , a sdw , and a modulated pair - density ( mdsc ) ( @xmath105 ) coexist in such a striped superconductor .
the superconducting order parameter is _ finite _ everywhere , modulated only with an amplitude of a few percent of the average order parameter @xmath106 .
the cdw oscillates around the average density @xmath80 with an amplitude of only 1% .
the staggered magnetization exhibits the strongest modulation with an amplitude of about @xmath107 including a periodic sign change in the vertical direction corresponding to anti - phase domain walls between the antiferromagnetically ordered stripes .
this solution is the groundstate of the @xmath23-model for the chosen parameters and initial conditions .
the periodic modulation of the magnetization in the vertical direction translates into two distinct peaks in the magnetic structure factor at the wavevectors @xmath108 with @xmath109 .
these magnetic ordering wavevectors @xmath110 correspond to a sdw with period @xmath103 perpendicular to the stripes as is also obvious from the real space pattern of the staggered magnetization in fig .
[ fig : impfree ] ( c ) .
the concomitantly emerging cdw ( fig .
[ fig : impfree ] ( c ) ) and mdsc ( fig .
[ fig : impfree ] ( a ) ) modulate with half the wavelength compared to the fluctuating sdw .
their fourier transforms peak at @xmath111 with @xmath112 indicating a vertical oscillation with wavelength @xmath113 . the period doubling of the sdw results from the @xmath114-phase shift that the spin order experiences between neighboring charge stripes .
in contrast to the pdw which contains _ no _ uniform @xmath115 pairing component , the sc order parameter is here always of equal sign . these wavelengths stay fixed in a hole doping range from @xmath116 up to @xmath117 . for @xmath118 , the additional holes collect in the already present stripes which deepens the hole rich channels thus increasing the oscillation amplitude of the cdw .
fig2a.pdf ( -3,59)*a * fig2b.pdf ( -3,59)*b * fig2c.pdf ( -3,59)*c * fig2d.pdf ( -3,59)*d * for a further characterization of the impurity - free case we present in fig .
[ fig : impfree_qspace ] momentum space quantities .
the momentum distribution @xmath89 is shown in fig . [
fig : impfree_qspace ] ( a ) and exhibits an anisotropy in @xmath119 and @xmath120 . in the presence of horizontal spin stripes , states with wavevectors near @xmath121 are less probably occupied than states near @xmath122 . in a homogeneous @xmath19-wave superconductor , the spectral ( quasi - particle ) density @xmath88 is finite only at the four points in the brillouin zone where the sc order parameter has nodes . in the striped superconductor
it was shown that cdw order leads to an increase of spectral weight in the nodal direction while sdw order suppresses spectral weight in this region @xcite . here
we observe that spectral weight reappears on most of the fermi surface of the normal conducting state ( fig .
[ fig : impfree_qspace ] ( b ) ) . in the near nodal direction
two peaks appear which are fingerprints of both cdw and sdw order with a dominance of the latter . comparing to ref .
@xcite , one can attribute the peak around @xmath123 in the first quadrant of the brillouin zone to spin order , while the peak at @xmath124 results from the charge order .
these features bear resemblance to arpes data found in la@xmath4nd@xmath10sr@xmath6cuo@xmath7 and lsco by zhou _
figure [ fig : impfree_qspace ] ( c ) shows the pair density @xmath90 which accentuates the states contributing to superconductivity .
the maxima in @xmath88 around @xmath125 which originate from the cdw order @xcite correspond to a suppression of the pairing in fig .
[ fig : impfree_qspace ] ( c )
. @xmath90 is maximal around @xmath126 where the pairing amplitude is largest .
these regions resemble those of a homogeneous @xmath19sc but stretching out less far into the nodal direction .
in contrast , @xmath90 is strongly suppressed at wavevectors @xmath127 where the spin density is strong and exhibits local maxima around @xmath128 .
the unidirectional character of the spin - density wave is clearly seen in @xmath129 ( see fig .
[ fig : impfree_qspace ] ( d ) ) which relates to the large amplitude oscillations of the staggered magnetization in real space
. @xmath129 dominates regions in momentum space which @xmath90 would occupy in a non - magnetic homogeneous @xmath19sc .
it extends far along the fermi arcs , crossing the nodal points .
@xmath129 has large intensities in the anti - nodal regions , i.e. , it competes with superconductivity .
though maxima of @xmath129 are clearly separated from those of the pair density @xmath90 , both quantities coexist in large parts of momentum space . in the presence of
a _ single _ strong impurity with potential strength @xmath130 , stripes are pinned as is inferred from the order parameter pattern shown in fig .
[ fig : impfree ] ( b ) , where the impurity is located at the center .
impurity - pinned stripe solutions are robust against a change of the initial conditions , in contrast to the impurity - free stripe solution ( ( fig . [ fig : impfree ] ( a ) ) .
thus we expect fluctuating stripes in real materials to be pinned by impurities and to become static .
horizontally aligned stripes emerge as well in the magnetization and the electron density .
the impurity pins a channel of reduced pairing - amplitude ( fig .
[ fig : impfree ] ( b ) ) , which minimizes the loss of pairing energy . on the other hand , inhomogeneities in the order parameter result in a charge density redistribution in the absence of particle - hole symmetry @xcite . in fact , the channels of reduced pairing amplitude collect electrons , thereby shifting the electron density in these channels towards half - filling . this in turn favors the emergence of antiferromagnetic order in the regions of enhanced electron density .
that is why the impurity pins one of the ridges of the staggered magnetization .
altogether , electron - rich stripes coincide with strongly magnetized stripes of reduced pairing amplitude .
the characteristics of the impurity - pinned stripes are similar to those of the impurity - free case , except that the impurity - pinned stripes emerge independently of the initial conditions .
thus the momentum space quantities shown in fig .
[ fig : impfree_qspace ] for the impurity - free stripe solution are essentially the same for the impurity - pinned stripes . summarizing , we observe for sufficiently large on - site repulsion @xmath131 , as in the impurity - free case , the coexistence of a mdsc , a sdw , and a cdw with wavelength @xmath132 pinned by a single non - magnetic impurity .
real - space quantities of stripes pinned by a single strong non - magnetic impurity have already been investigated by chen and ting @xcite .
they obtained similar results for the @xmath23-model at @xmath133 hole doping which is assumed to be close to optimal doping .
we also checked that the periodicity of the stripes for a single impurity at @xmath134 remains the same as for @xmath3 .
the vortex - pinned stripes shown in fig . [ fig : stripes_u3_field ] display essentially the same properties as those for the impurity - free and the impurity - pinned stripes .
the amplitudes of the mdsc , cdw , and the sdw are marginally enhanced as compared to the impurity - pinned stripes because a vortex penetrating the superconductor acts as a far stronger perturbation than a single pointlike impurity . just like the impurity , the vortices pin the ridges of the sdw .
stripes of reduced pairing run through the elliptically deformed vortices and thereby save condensation energy .
the af stripes are further attracted by the vortex since vortex cores in @xmath19-wave superconductors strongly magnetize due to a spin - dependent splitting of the andreev bound - state @xcite .
correspondingly , the vortex pins an electron - rich stripe with a filling shifted towards 1/2 . from the fourier transforms of the real - space quantities we obtain the same wavelength of the mdsc ( @xmath135 ) , the cdw ( @xmath136 ) , and the sdw ( @xmath137 ) at @xmath134 and at @xmath8 .
similar results for coexisting cdws and sdws were observed in ref .
@xcite within the @xmath23-model for vortex - pinned stripes above a critical @xmath23 at @xmath133 hole doping .
hence the wavelengths of these unidirectional density waves are identical in the impurity - free and the impurity- and vortex - pinned stripes and robust against a change of hole doping or pairing interaction strength , as we verified by explicit calculations .
fig3a.pdf ( -3,59)*a * fig3b.pdf ( -3,59)*b * fig3c.pdf ( -3,59)*c * fig4a.pdf ( -3,59)*a * ( 16,65)order parameter ( 22,71)@xmath138 fig4d.pdf ( -3,59)*d * ( 16,65)order parameter ( 22,71)@xmath139 + fig4b.pdf ( -3,59)*b * ( 15,65)electron density fig4e.pdf ( -3,59)*e * ( 16,65)electron density + fig4c.pdf ( -3,59)*c * ( 8,65)staggered magnetization fig4f.pdf ( -3,59)*f * ( 8,65)staggered magnetization fig5a.pdf ( -3,59)*a * ( 16,65)order parameter ( 22,71)@xmath138 fig5d.pdf ( -3,59)*d * ( 16,65)order parameter ( 22,71)@xmath139 + fig5b.pdf ( -3,59)*b * ( 15,65)electron density fig5e.pdf ( -3,59)*e * ( 16,65)electron density + fig5c.pdf ( -3,59)*c * ( 8,65)staggered magnetization fig5f.pdf ( -3,59)*f * ( 8,65)staggered magnetization for the modeling of cuprates , such as bscco or those based on the parent compound la@xmath13cuo@xmath7 , their dopant disorder has to be taken into account . in the following
we consider explicitly intrinsic disorder and compare our results with experimental data in zero and finite magnetic field .
for modeling dopant disorder we set the impurity concentration equal to the hole doping @xmath140 .
moreover , we assume that the dopants impurity potential decreases with increasing hole doping due to enhanced screening @xcite . in fig . [
fig : doping_dep ] and fig .
[ fig : doping_dep_struc ] we show exemplary results for two specific impurity configurations which exhibit the characteristic behavior of the disordered stripe phase .
we have verified that other impurity configurations yield density waves with the same wavelength as discussed here . for a coloumb repulsion
@xmath141 we find quasi - unidirectional stripes at @xmath8 , slightly deformed by the presence of the non - magnetic impurities ( see fig .
[ fig : doping_dep ] ( d)-(f ) ) .
hole - rich paths ( see fig . [
fig : doping_dep ] ( e ) ) coincide with anti - phase domain walls , seen in the staggered magnetization profile in fig .
[ fig : doping_dep ] ( f ) , and with stripes of strong superconductivity as observed in the sc order parameter fig .
[ fig : doping_dep ] ( d ) .
although the suppression of the electron density on the impurity sites dominates the charge pattern as seen in fig .
[ fig : doping_dep ] ( e ) , closer inspection reveals the existence of hole - rich channels ( colored orange in fig .
[ fig : doping_dep ] ( e ) ) , where @xmath57 is reduced compared to the average electron density . additionally , the lines of zero staggered magnetization ( blue lines in fig . [ fig : doping_dep ] ( f ) ) , which coincide with the maxima of the sc order parameter , also serve as a guide to the eye .
the @xmath19-wave order parameter ( fig .
[ fig : doping_dep ] ( d ) ) varies in the strong - disorder limit with much larger amplitudes as compared to the previous cases , but it never changes sign . from the analysis of different disorder configurations
we conclude that the hole - rich paths emerge only close to @xmath8 hole doping and sharpen upon approaching @xmath8 .
simultaneously the density modulations as a whole become quasi unidirectional .
we find that , by shifting the hole doping towards 1/8 , the system is driven into the stripe state .
moreover , we observe that the wavelength of the density modulation decreases by enhancing the hole doping from @xmath134 to @xmath142 ( cf fig . [
fig : doping_dep_struc ] ) .
this characteristic , which is also observed in neutron scattering experiments @xcite , is not found in the impurity - free systems .
experimentally , sdws with wavelengths @xmath143 and @xmath144 were inferred from the incommensurabilities .
while this model calculation predicts exactly the same wavelength for the sdw at @xmath8 , the wavelength at @xmath145 is slightly larger , that is @xmath146 , than in the experiment .
note that a decrease in hole density either reduces the amplitudes of the density waves or enhances the wavelength .
the latter is valid in this model calculation .
strongly disordered systems support the former , where the random impurity sites facilitate the setup of additional hole channels .
disorder allows the stripe pattern to adjust more easily to the average hole density , whereas in the clean case , the stripe pattern is tied to the underlying lattice .
all real - space quantities verify horizontally ( or vertically ) oriented stripes at @xmath8 .
the periodicity of these patterns is extracted from the fourier transformed ( ft ) quantities as shown in fig .
[ fig : doping_dep_struc ] .
intriguingly , the ft data in the strong disorder regime for @xmath8 ( fig .
[ fig : doping_dep_struc ] ( d)-(f ) ) peak in general at the same wavevectors as for the perfectly unidirectional stripes discussed for clean systems .
the magnetic structure factor ( fig .
[ fig : doping_dep_struc ] ( f ) ) shows two dominating peaks at the incommensurate wavevectors @xmath147 with incommensurability @xmath109 . the ft electron density ( fig .
[ fig : doping_dep_struc ] ( e ) ) and the ft sc order parameter ( fig .
[ fig : doping_dep_struc ] ( d ) ) exhibit similar patterns with two broad maxima around wavevectors @xmath148 with @xmath112 .
the main difference to the clean stripe solutions is that the peaks in fig .
[ fig : doping_dep_struc ] ( d ) , ( e ) are broadened , and a low - intensity substructure is observable .
obviously , the slight deviation from the characteristics of perfect unidirectional stripes is caused by the finite impurity concentration .
we infer that coexisting mdsc , cdw , and sdw with wavelength @xmath149 and @xmath150 persist at @xmath8 even in the regime of strong disorder .
the wavelengths of the density modulations change with doping as is recognized by comparison of 1/10 and 1/8 hole doping . in fig .
[ fig : doping_dep_struc ] ( a)(c ) the ft sc order parameter , the ft electron density , and the magnetic structure factor are displayed for @xmath134 .
the maxima deviate clearly from those at @xmath8 . at @xmath134
the density modulations oscillate with wavelengths @xmath151 and @xmath152 .
in contrast to _ weakly _ disordered systems the incommensurabilities @xmath153 and @xmath154 in the strong disorder limit are sensitive to doping and increase with growing hole concentration .
fig6a.pdf ( -3,59)*a * fig6b.pdf ( -3,59)*b * fig6c.pdf ( -3,59)*c * fig6d.pdf ( -3,59)*d * the momentum space quantities @xmath89 , @xmath88 , @xmath90 , and @xmath129 ( see fig .
[ fig : disorder_qspace ] ) in the strong disorder limit resemble the results obtained for the impurity - free stripe solutions ( see fig .
[ fig : impfree_qspace ] ) .
disorder slightly enhances the anisotropy , best visible in @xmath129 .
importantly , in the disordered system continuous fermi arcs appear in the near nodal direction ( see fig . [
fig : disorder_qspace ] ( b ) ) .
this agrees with arpes measurements in lbco @xcite , which has a strong dopant disorder . due to the impurity potentials , through which the hole channels run , the cdw is stronger here as compared to the previously discussed cases , which results in the redistribution of spectral weight to the nodal directions @xcite .
altogether , we find that a sdw , a cdw , and a mdsc coexist at @xmath8 hole doping in strongly disordered systems such as dopant disordered lsco . the density - modulations are pinned and stabilized , just as for a single impurity or a vortex , yet slightly deformed by the strong disorder .
the doping dependence of the incommensurabilities @xmath155 and @xmath156 is similar to those observed in experiments @xcite .
fig7.pdf characteristically all the identified stripe solutions exhibit a full gap in the density of states ( dos ) ( see fig .
[ fig : dos_stripes ] ) . for comparison also the typical v - shaped gap of a homogeneous @xmath19-wave superconductor
is shown in fig .
[ fig : dos_stripes ] . in the presence of stripes
a full gap opens around the fermi energy .
the full gap is a generic property of stripes in @xmath19-wave superconductors which persists also in the strong disorder limit .
the full gap originates from a finite extended @xmath157-wave contribution in the striped @xmath19-wave superconductor and the sdw gap present in the strongly magnetized stripes . in a homogeneous @xmath19-wave superconductor the bond order parameter @xmath56 changes sign between horizontal and vertical bonds which reflects the @xmath19-wave symmetry of the energy gap @xmath158 .
thus the summation of @xmath56 over the vertical _ and _ horizontal bonds connecting to a given lattice site gives zero in a homogeneous @xmath19sc .
due to the strong anisotropy of the stripe solutions , however , this summation yields a finite value here , which characterizes extended @xmath157-wave superconductors . in a @xmath19sc coexisting with a sdw
the gap is generically not centered around the fermi energy , which is the reason why the full gap is asymmetric in all stripe solutions shown in fig .
[ fig : dos_stripes ] .
in addition , particle - hole symmetry is already broken by the constant random impurity potentials with @xmath159 .
experimentally , particle - hole anisotropy was observed recently by angle - resolved photoemission spectroscopy ( arpes ) in the pseudogap phase of bi-2201 @xcite which is attributed to competing orders and contrasted to homogeneous superconductivity .
the appearance of a full gap was also found by loder _
_ @xcite . obviously , the gap is largest in the unperturbed stripe solution ( @xmath160 ) .
this is due to the fact that with the impurity- and field - induced bound states , spectral weight is shifted into the energy gap .
fig8a.pdf ( -9,126)*a * fig8b.pdf ( -9,126)*b * fig8c.pdf ( -9,126)*c * fig8d.pdf ( -9,126)*d * the @xmath21-model presented by the hamiltonian eq . was introduced in ref .
@xcite as an alternative description for the coexistence of superconductivity with magnetic order . in the homogeneous case ,
the groundstate solution of the @xmath21-model for @xmath3 is a mdsc state similar to fig .
[ fig : impfree](a ) , ( c ) .
the main difference is that the af stripes are nearly maximally magnetized , i.e. the sdw oscillates with an amplitude @xmath162 , while @xmath163 goes to zero at the center of the af stripes without changing sign .
thus magnetic order is dominating in the @xmath21-model , whereas superconductivity is reduced to straight , quasi one - dimensional lines on the domain walls between the af stripes .
this mdsc solution is degenerate with a `` pure '' pdw , which features a periodic sign change in @xmath163 . both , the `` pure '' pdw and the mdsc have identical pair densities @xmath164 ( c.f . ref .
@xcite ) . in the strong disorder limit the @xmath23- and the @xmath21-model
show a severely different behavior .
the solutions of the @xmath21-model can be divided into two regimes .
if the scattering strength @xmath165 of the impurities is weak ( i.e. @xmath166 ) , the af stripes remain straight , as shown in fig .
[ fig : v ] .
the magnetic energy gain is dominant and prevents the superconducting stripes to wind around the impurities , which would allow to gain further condensation energy .
moreover , impurities in the straight one - dimensional sc channels strongly suppress superconductivity .
this effect culminates in the disappearance of entire sc stripes , if the impurity density in their near vicinity becomes too large ( fig .
[ fig : v ] ( a ) ) .
one of the neighboring af domains ( see fig . [
fig : v ] ( c ) ) spreads over such a metallic line expelling the holes , which collect in the remaining superconducting stripes ( fig .
[ fig : v ] ( b ) ) .
in this situation the af domain - wall boundaries are shifted locally , according to the impurities positions , but the af spin stripes remain largely straight ( fig . [
fig : v ] ( c ) ) .
the antiferromagnetic stripes are almost half filled and all sites within the af stripes are nearly fully polarized , in contrast to the small amplitude modulations seen in fig .
[ fig : doping_dep ] for the @xmath23-model .
thus the superconducting stripes in the disordered @xmath21-model have a filling below 1/4 because the holes collect mainly in the remaining sc stripes .
similar to the sdw , also the cdw in the @xmath21-model modulates with much larger amplitudes as compared to the @xmath23-model .
if the impurity potentials @xmath165 become larger than @xmath17 , we expect that the impurity - pinning of stripes is energetically favorable as long as a mdsc state prevails , similar to the @xmath23-model ( c.f . fig .
[ fig : doping_dep](a)(c ) ) .
however , we observe an overall disappearance of superconductivity accompanied by a global phase separation .
all holes accumulate where the impurity concentration is largest and antiferromagnetism weakest , while the surrounding region turns into a half - filled af insulator ( fig .
[ fig : v ] ( d ) ) .
the same occurs if weak impurities are numerous enough to destroy sufficiently many metallic ( superconducting ) lines so that the remaining ones can not accommodate all the holes in the system .
this behavior is characteristic for the strong - coupling limit described by the @xmath21-model : if the effectively one - dimensional striped system is forced to become two - dimensional because of disorder , the system divides into half - filled and empty regions featuring global phase separation .
we found that the @xmath23-model allows for two - dimensional superconductivity in coexistence with a stripe order that adapts flexibly to disorder , while the solutions of the @xmath21-model remain quasi one - dimensional .
materials in which three - dimensional superconductivity coexists with stripe order , such as lbco @xcite or the rare - earth doped cuprate la@xmath4nd@xmath5sr@xmath167cuo@xmath7 with @xmath168 and @xmath169 @xcite , are certainly not as one - dimensional as the solution of the @xmath21-model . on the other hand
the @xmath21-model reproduces the quasi one - dimensional characteristics of nd - doped lsco around @xmath170 , quite well @xcite , where the material becomes more anisotropic with increasing doping . as argued in ref .
@xcite , the models described by @xmath65 and @xmath76 are the weak and strong coupling limits of a @xmath17@xmath18 like model , and the cuprates are most likely found somewhere in the intermediate coupling regime .
the static , unidirectional cdws and sdws observed in lbco and lnsco @xcite are often ascribed to the structural phase transition towards the ltt phase , where the ltt specific buckling pattern of the cuo@xmath171 octahedra induces an @xmath42-@xmath172 anisotropy @xcite .
our results imply that the strong dopant disorder , which is present in these substances , further stabilizes the stripe state .
the results within the @xmath23-model in the strong disorder limit reproduce typical properties of these substances well , such as the fermi arc reconstruction in lbco @xcite or lsco @xcite and the doping dependence of the wavelength of the sdw as inferred from neutron scattering experiments in lsco @xcite .
though spin stripes are observed in lsco at @xmath3 , an ordering of the charges has so far _ not _ been detected by neutron scattering experiments @xcite .
as lsco does not exhibit an ltt phase we suggest that the experimentally observed spin stripes @xcite are induced by the dopant disorder .
the expected concomitant charge stripes are not found by neutron scattering experiments possibly because the charge modulations are weak and may therefore be concealed by the reduction of the electron density below the dopant sites .
ybco , which has neither an effective dopant disorder nor a ltt phase , exhibits dynamic instead of static stripes @xcite .
this agrees with the results of the @xmath23-model for an impurity - free system as discussed . in the presence of a magnetic field charge stripes become static , while spin stripes are absent @xcite .
in this case the magnetic field induced charge stripes erase superconductivity .
we find a similar behavior in the @xmath23-model when we turn off superconductivity deliberately .
the magnetization is greatly suppressed , while a dominant charge order survives .
the models we investigated in this article allow to explain a wide range of features of striped cuprate superconductors , which were found experimentally .
notably , the material specific characteristics of stripe phases in various cuprates can be modelled within the @xmath23-model . on the other hand
, we found that af correlations in the @xmath21-model tend to be stronger and the superconducting stripes are one - dimensional which reproduces some characteristics of nd - doped lsco quite well .
specifically , we found the coexistence of pdw s , sdw s , and cdw s with wavelengths @xmath173 in a @xmath19-wave superconductor for sufficiently large @xmath23 in the @xmath23-model .
impurity - free , ( single ) impurity - pinned , and vortex - pinned _ straight _ stripe solutions exist over a broad doping range in the @xmath23-model .
disorder dissolves the strict one - dimensional stripe order into meandering stripe - like patterns . only close to @xmath8 hole doping , which is ideal for stripe formation , a unidirectional stripe order reemerges in strongly disordered systems .
in the @xmath21-model stripes are rigid and remain virtually undistorted in the presence of weak disorder .
superconducting stripes react very sensitively to disorder which may even remove some of the superconducting channels .
the remaining superconducting stripes exhibit strictly one - dimensional superconductivity appearing between straight af stripes .
strong disorder , however , leads to global phase separation into dominantly antiferromagnetic regions with non - magnetic puddles absorbing the holes .
the overall identical characteristics of the impurity- , vortex - pinned , and ( strong ) disorder - pinned stripes indicate that the stripes are not generated by impurities or fields
. instead , stripe states , which are energetically close to homogeneous solutions in unperturbed systems , are pinned and stabilized by inhomogeneities which are connected to charge redistributions . a generic feature of striped @xmath19-wave superconductors in both the @xmath23- and the @xmath21-model is the opening of a full gap , induced by an extended @xmath157-wave contribution . without local probes for the superconducting order parameter
it is not clear if pair - density modulations accompany the charge- and spin stripe order . . | * stripe phases are observed experimentally in several copper - based high-@xmath0 superconductors near 1/8 hole doping .
however , the specific characteristics may vary depending on the degree of dopant disorder and the presence or absence of a low - temperature tetragonal phase . on the basis of a hartree - fock decoupling scheme for the @xmath1-@xmath2 model
we discuss the diverse behavior of stripe phases .
in particular the effect of inhomogeneities is investigated in two distinctly different parameter regimes which are characterized by the strength of the interaction .
we observe that small concentrations of impurities or vortices pin the unidirectional density waves , and dopant disorder is capable to stabilize a stripe phase in parameter regimes where homogeneous phases are typically favored in clean systems . the momentum - space results exhibit universal features for all coexisting density - wave solutions , nearly unchanged even in strongly disordered systems .
these coexisting solutions feature generically a full energy gap and a particle - hole asymmetry in the density of states . * |
soft glassy materials ( sgms ) present a theoretical concept for fluids that can be solid - like or liquid - like at rest , but all posses a microstructure with strongly interacting building blocks @xcite .
an extreme class of these are the yield stress fluids , which by virtue of their internal structure behave as solids under an imposed mechanical stress below the yield point , yet flow like fluids once the stress is increased beyond this limit @xcite .
in addition to their extensive utilization in commercial products as _
e.g. _ toothpastes , paints and process suspensions @xcite , yield stress fluids are also of profound theoretical interest due to their rich and complex rheological behavior .
indeed , the peculiarities of flow response are also witnessed in _
shear banding _
, where the flowing fluid exhibits a spatially banded structure , each band possessing a distinct viscosity associated with a unique shear rate @xcite .
this banding can appear as a true steady - state structure @xcite or as _ transient shear bands _
( tsbs ) , which , while possibly extremely long lived , eventually are replaced by the homogeneous steady - state @xcite .
state - of - the - art observations find transient shear banding of simple yield stress fluids very robustly in different experimental geometries @xcite .
furthermore , even though non - thixotropic , such fluids display time- and shear rate or stress dependent response to step shear , where the relaxation time apparently follows a power - law of both applied stress and shear rate @xcite .
steady - state shear - banding has been described as a property arising from the non - monotonic nature of the fluid s intrinsic flow curve @xcite .
this allows for a mechanically unstable flow regime , where the negative slope in the constitutive curve permits the existence of a non - unique correspondence between shear rates and shear stresses .
thus , in this special regime the fluid may separate into two bands of either identical shear stress ( gradient banding ) or identical shear rates ( vorticity banding ) @xcite .
such a steady state scenario is unexpected for fluids known to have monotonic intrinsic flow curves @xcite . against this reasoning , transient
shear banding has recently been experimentally observed @xcite , and theoretically predicted even in fluids possessing monotonic flow curves @xcite .
numerous reasons for this type of banding have been proposed . in their studies , involving two models , the fluidity model@xcite and spatially resolved version of the soft glassy rheology ( sgr ) model @xcite , moorcroft and co - workers @xcite predicted transient shear banding resulting from a mechanical instability in the start - up flows .
this was clearly seen in the instantaneous ( stress - time ) constitutive curve as a stress overshoot and subsequent negative slope of the curve as it regresses towards the steady state . in a more recent work ,
the same authors derive a more rigorous expression for the onset of this mechanical instability in multiple flow scenarios @xcite .
transient shear banding in amorphous solids has been successfully modeled by shi and falk with shear transformation zone ( stz ) theory , in which the plastic deformations occur in specific zones , activated by an effective temperature @xcite . at present , hinkle and falk are extending this framework to study transient shear banding in simple yield stress fluids ( ysf ) @xcite .
these studies properly explain the movement and limited life - time of the shear bands , however , they leave open detailed origin of the band initiation . to study this aspect ,
a slightly different approach was proposed in @xcite , where a phenomenological , simple scalar @xmath0-model is utilized . in this case , the local microstructure of a time - dependent fluid is encoded in the @xmath0 parameter , which refers to the local immobilized volume fraction of the fluid . in the framework of this model , transient shear banding can initiate due to the small stress gradients that are induced by the rotational couette flow geometry , often utilized in experimental studies .
these stress variations generate shear rate heterogeneities , which in turn result in the non - uniform disintegration of the underlying fluid structure , observed as the formation of a tsb @xcite .
however , in experimental geometries , such as the cone - and - plate , where tsb is also found , the stress heterogeneity is considered to be negligible . in this work ,
we consider the role of fluid inertia to initiate transient shear bands during the start - up , by solving the navier - stokes equations for incompressible flow simultaneously to the evolution of a scalar field variable describing the complex fluid structure .
the purpose is to show the effect of accelerating flow to the transient shear band formation , neglected earlier . as reviewed above , to generate transient shear bands , some models pre - initialize by hand a non - homogeneous structure profile @xcite and others rely on the geometrical stress heterogeneity @xcite . while especially the latter mechanism compares reasonably to most experimental setups ( couette , parallel plates ) , its relevance in the shear band initialization may be questioned in others ( cone - and - plate ) .
our results suggest a plausible mechanism initializing transient shear bands even under perfectly homogeneous initial conditions due to the flow acceleration mechanism . utilizing a computational fluid dynamics ( cfd ) approach ,
we apply the @xmath0-model as in ref .
@xcite in a homogeneous flow ( planar couette ) scenario , where , as mentioned , any and all stress heterogeneities are due to inertial effects alone .
we emphasize here , that , as will be explained later , the key concept in this model is the fact that the structure breakdown is proportional to the shear rate , a fact based on advanced experiments @xcite .
thus , the qualitative behavior shown here , is not influenced by this particular model , but is general for all the models and real physical materials that have this property .
the paper is organized as follows : first , we detail the theoretical framework and the implementation .
then , we proceed to comment on the relevant results obtained with our approach and finally , the paper finishes with concluding remarks .
as the aim here is to examine tsbs in a time - dependent fluid , a necessary prerequisite for modeling transient shear - banding theoretically involves coupling a structural model describing the internal structure of the fluid to the subsequent flow dynamics .
additionally , as the focus is on providing a minimal example that presents qualitative evidence of transient shear - banding driven by inertial effects , this structural model is chosen from a class of simple phenomenological models based on the abstract scalar structural parameter @xcite .
the specific @xmath0-model addresses the internal structure of the fluid through the structural parameter @xmath0 , describing the immobilized volume fraction present in the fluid @xcite .
this model allows fine - tuning the fluidization exponent to match the experimental values of various complex fluids , while other similar scalar models ( such as the @xmath1-model by coussot _ et al_. @xcite ) inherently produce a fixed fluidization exponent , therefore limiting their applicability for this purpose . indeed , to the best of our knowledge
, the @xmath0-model is the simplest one possessing this crucial feature .
furthermore , the fluidization behavior of more elaborate models , such as the sgr - model @xcite , are presently not reported . in the @xmath0-model , the progression of the immobilized volume fraction , @xmath0
, in time can be due to shear ( constructive and destructive ) and shear - independent motion of the structural elements ( constructive ) , reflected in the model as two superimposed kernels .
the time - evolution equation for @xmath0 is @xcite @xmath2 where @xmath3 ( @xmath4 ) is a kinetic constant for the shear growth ( destruction ) , @xmath5 is the magnitude of the strain rate ( shear rate ) and both @xmath6 and @xmath7 describe the sensitivity of @xmath0 towards shearing .
furthermore , @xmath8 , @xmath9 and @xmath10 describe the growth of @xmath0 due to shear - independent effects .
again pursuing a minimal example , setting @xmath11 in eq .
and defining the sample history by fixing the initial immobilized volume fraction @xmath12 allows for a minimal inspection of an internally structured fluid with a steady - state @xmath0 of @xmath13 , with the structure breakdown rate set by @xmath4 and proportional to the powers of the dimensionless shear rate @xmath14 .
the proportionality @xmath15 gives the minimum volume fraction reached by a physical system .
for instance in an aggregating colloidal suspension , this can be thought to describe the monomer volume fraction or in a microgel this could be the volume fraction at the maximally compressed state of the sponge like elements .
we define @xmath16 .
the parameter @xmath17 determines the system sensitivity to shear ; a larger number means longer relaxation time .
the @xmath0-parameter can be incorporated into the flow quantities , as done here , by the empirical krieger - dougherty relation @xcite @xmath18 where @xmath19 is the viscosity of the fluid , @xmath20 describes the viscosity of the suspending matrix and @xmath21 denotes the jamming volume fraction , which describes a completely jammed configuration .
this gives an additional constraint to the kinetic constants so that @xmath22 .
the quantities used in eq . are set to @xmath23 , @xmath24 ( water ) and @xmath25 @xcite .
the final step in the model development is to include eq . in the navier - stokes equations that describe the flow field completely .
the incompressible navier - stokes equations read @xmath26 and @xmath27 where eq .
is incorporated in the latter equation , physically implying the conservation of momentum . in these equations , @xmath28 is the velocity field , @xmath29 is the pressure , @xmath30 is the density of the fluid and @xmath31 describe bodily forces ( _ e.g. _ gravity ) acting on the fluid ( here , @xmath32 ) .
additionally , @xmath33 is the kinematic viscosity of the fluid , readily obtained from eq . applying the relation @xmath34 .
the inertial effects , generally omitted in rheological modeling , which usually deals with creeping flow conditions , are included in the convection and acceleration terms on the left - hand side of eq . .
transient shear - banding emerges as a property in the fluidization of an internally structured fluid experiencing spatial stress variations and subsequent shear rate inhomogeneities , leading to spatially non - uniform structural disintegration . as the viscosity is assumed a function of the structural integrity ( here modeled by coupling eqs . and ) , this is experimentally seen as viscosity bands across the measurement gap .
accordingly , to isolate any sources other than the inertial contributions for such variations , the planar couette geometry shown in fig .
[ fig : geometry ] is utilized in this work .
the fluid , confined between the two parallel plates , is subjected to simple shear as the upper plate moves and the lower one remains stationary . in this scenario ,
the stress is uniform over the gap between the plates , ensuring that all stress variations are due to inertial effects alone .
this approach is implemented in a computational fluid dynamics ( cfd ) environment .
the flexible openfoam software package @xcite that includes optimized numerical solvers , meshing , and visualization tools is applied and extended to account for the scenario presented here .
the simple shear geometry of fig .
[ fig : geometry ] is meshed to a 2-d 60@xmath3560 grid with periodic boundary conditions at the open boundaries .
the practical implementation extends the capabilities of nonnewtonianicofoam solver to allow for time - dependent viscosity .
the system of differential equations , is solved using the pcg ( pcbicg ) linear solver with dic ( dilu ) preconditioning and smoothing , setting the absolute tolerance close to the machine precision of the 64-bit system ( @xmath36 ) .
the kernel constants of the @xmath0-model in eq .
[ eq : phimodel ] were set to @xmath37 , @xmath38 , @xmath39 and @xmath40 .
additionally , the gap width @xmath41 was set to @xmath42 mm . for brevity
, the following results refer to the normalized gap ( scaled by @xmath43 ) when necessary .
the qualitative results are independent of the selection of these parameters .
however , varying the @xmath17 shifts the fluidization curves ( see fig .
4 ) in the horizontal direction . changing the ratio @xmath15 to a smaller value improves the contrast of the shear band edge , i.e. makes the kink in the velocity profile steeper , and shifts the fluidization curves in the vertical direction having similar effect as increasing the @xmath12 ( fig .
the navier - stokes equation is characterized ( at steady state ) by the reynolds number , which is represented here by the shear rate .
in addition in the relaxation of the model depends on the shear rate exponent @xmath6 and the initial condition @xmath12 .
thus , this set of parameters define the tsb and is studied in what follows . in fig .
[ fig : fig1 ] ( @xmath44 , @xmath45 ) , the linear velocity profile is displayed in the homogeneous , planar couette start - up flow scenario . in experiments ,
these 1-d pictures are accessed by _
e.g. _ ultrasonic speckle velocimetry @xcite and , in general , are used to identify shear localization effects .
observing the evolution of the velocity in the flow direction allows to monitor the shear localization here , as well . as demonstrated in the example scenario of fig .
[ fig : fig1 ] , shear banded flow profile develops in the course of a few seconds from the on - set of the flow . here
we emphasize that no external initial disturbance to the model is required and the shear localization is induced by the stress inhomogeneity due to the initial acceleration of the fluid , which interacts with the viscosity evolution as discussed in detail for the creeping flow case in ref .
@xcite . ,
@xmath39 , @xmath46 and @xmath44 and the shear rate is fixed at @xmath45 .
the stress variations induced by the inertial effects result in the formation of a tsb , seen here as the initial shear localization towards the shearing ( outer ) plane .
gradually , the profile evolves over time towards the steady - state solution ( @xmath47s ) and the tsb vanishes.,scaledwidth=49.0% ] with these model parameters , the tsb persist for up until @xmath48 s where complete fluidization and the steady - state flow profile is reached .
this certainly is a time - scale comparable to the experiments , where long lasting transient banding was observed @xcite . note that since the mechanism here is based on the flow model interaction with the time - dependent shear thinning of the fluid , any model having this property will potentially experience shear banding induced by the same effect .
experimentally this behavior is present in systems showing microstructural disintegration due to shear , where the time - dependence is a power - law ( typically of exponents in the range of 2 - 4 ) of the shear rate or shear stress @xcite .
conversely , if the structural decomposition is linear ( power - law exponent @xmath6 @xmath49 1.0 ) , the inertial effects and the subsequent stress variations are incapable of generating a tsb .
this is illustrated in fig .
[ fig : fig3 ] , where the model parameters are identical to those in fig .
[ fig : fig1 ] aside from the exponent of the @xmath0-model ( @xmath50 ) . here
, due to the linear dependence between @xmath0 and @xmath5 , the structural breakdown is uniform and the shear localization does not persist up to the time scales observed in fig . [
fig : fig1 ] .
instead , the fluidization time is simply established by the viscous time scale @xmath51 ( @xmath52 denoting the gap width ) , as for a simple ( newtonian ) liquid .
judging by this expression , the high viscosity of the jammed initial configuration should yield an extremely low fluidization time . indeed , as seen in fig .
[ fig : fig3 ] , the initial fluidization occurs within @xmath53 seconds from the moment the flow commences .
, @xmath39 , @xmath44 and the shear rate is fixed at @xmath45 while the exponent of the @xmath0-model is now reduced to @xmath50 . here ,
the stress variations induced by the inertial effects are unable to produce a tsb due to the highly linear breakdown of the underlying fluid structure.,scaledwidth=49.0% ] on the other hand , in the model , shear localization occurs only on certain conditions : the formation of a tsb is largely dependent on the model parameters as in our earlier work @xcite , where both the values of sample history variable @xmath12 and the kinetic exponent @xmath6 are able to predict the transient shear banding .
however , here , the imposed shear rate @xmath54 also influences this process , since it is this quantity that provides the necessary stress heterogeneity in the beginning of the flow .
this point is illustrated in fig .
[ fig : fig2 ] , where the fluidization time @xmath55 ( here , the time the tsb persists ) is represented as a function of the applied shear rate @xmath54 .
for each @xmath56 , @xmath55 was determined from the linear velocity profile ( as shown in fig .
[ fig : fig1 ] ) as follows : two separate linear curves were fitted to the profile , one spanning the area @xmath57 and the other @xmath58 . vs. applied shear rate @xmath54 on a log - log scale .
a sufficient applied shear rate is required for shear localization and nonuniform fluidization , seen as a discrete ramp in the results .
the fluidization time decays rapidly by increasing @xmath54.,scaledwidth=49.0% ] once the slopes ( @xmath59 and @xmath60 ) of these curves reached a certain threshold ( here : @xmath61 ) , the profile was determined to having reached the homogeneous flow state .
the results are displayed in fig .
[ fig : fig2 ] , where it is observed that a sufficient value of @xmath56 ( and accordingly , sufficient stress heterogeneity ) is required for transient shear banding , otherwise an uniform breakdown of the underlying structure is witnessed ( @xmath62 ) .
once this @xmath56 is reached , the time required for complete fluidization decays rapidly .
moreover , if the structural decomposition is linear ( @xmath50 ) , the breakdown is uniform regardless of @xmath56 .
note , that the discrete ramp in the fluidization times should not be interpreted literally as the transformation from strong shear localization ( tsb ) to homogeneous flow is of continuous nature .
however , to quantify the fluidization times , we have resorted to the procedure introduced above , which essentially imposes a cut - off criterion ( and the discrete jump ) to the results .
additionally , a complementing point of view on tsb formation is provided by various temporal plots . in fig .
[ fig : fig4 ] , the apparent shear stress @xmath63 , measured at the shearing plane , is displayed as a function of time @xmath64 for various applied shear rates @xmath54 .
as the flow progresses with increasing time , the microstructure of the fluid disintegrates and the local jammed volume fraction at the shearing plane decreases , leading to a monotonic decrease of the effective viscosity as well , as the two are linked by eq . .
this , in turn , leads to a monotonic decrease in the shear stress , as observed in the figure .
furthermore , the time at which the stresses reach their steady - state values seems to concur well with the fluidization time observed in fig .
[ fig : fig2 ] as one would expect . , @xmath39 , @xmath65 and @xmath66 ) .
as time progresses , the structure breaks down , and a lower stress is required to maintain the applied shear rate @xmath54 . if the magnitude of @xmath54 is sufficient for forming a tsb , the corresponding stress - time curve exhibits a distinct kink ( as seen here for @xmath54 = and @xmath54 ) , signaling the beginning of the shear localization . in the event of weak or nonexistent localization , this instability is not observed.,scaledwidth=49.0% ] another temporal plot is illustrated in fig .
[ fig : fig5 ] , where the shear band edge is tracked over the normalized gap ( @xmath67 ) as a function of time @xmath64 . in this plot , the value @xmath68 corresponds to a scenario where the flow is completely localized in the region near the shearing plane and @xmath69 where the tsb vanishes . for consistency ,
the shear band edge @xmath70 was determined by the linear fitting scheme introduced earlier ( see the commentary on fig . [
fig : fig2 ] ) . the edge location was determined as the intersection point of the two linear curves as long as the steady - state condition ( @xmath71 ) ) was not met . in the framework of the @xmath0-model ,
the shear band edge travels at a varying ( non - constant ) velocity , dependent on the applied shear rate @xmath54 .
in addition , the fluidization process is gradual up to a critical point , at which the fluid suddenly yields completely .
this is observed in fig .
[ fig : fig5 ] as the discrete jump of the edge location to the value @xmath69 .
the exact location of this jump is dependent on the applied shear @xmath56 .
the behavior witnessed here should be compared with the corresponding earlier results @xcite , where the shear band edge also proceeds in a highly non - linear manner . however , the critical point is not observed , and the edge travels independent of the applied shear rate @xmath54 for all @xmath6 .
the reason for these differences lies in the fundamentally different source of the stress gradient , that serves as the catalyst for transient shear banding . in ref .
@xcite , this stress variation is induced by the geometry and is essentially time - independent . here
, the stress variation is a ( time - dependent ) result of the finite fluid inertia , which is directly influenced by @xmath54 .
this leads to a weaker localization of the fluid flow at low @xmath54 and strong localization at high @xmath54 .
indeed , the blue curve in fig . [ fig : fig5 ] ( @xmath72s@xmath73 ) bears a strong resemblance to the results in ref .
@xcite , while lower values of @xmath54 yield a more convoluted profile : the velocity of the shear band edge increases at first , but then decreases and finally diverges as the critical point is reached .
thus , rather than reaching the static plate , the shear band disperses suddenly as the stress gradient has diminished over time and is unable to sustain the banded profile .
, @xmath39 , @xmath65 and @xmath44 ) , calcucated for various applied shear rates.,scaledwidth=49.0% ] finally , fig .
[ fig : fig6 ] provides a semilogarithmic `` phase '' diagram , in which the regions of localized and uniform shear are displayed .
the boundary indicates the minimum triggering value of @xmath54 , at each @xmath6 required to yield significant heterogeneous structure relaxation , triggered by the start - up stress overshoot due to the fluid inertia . above this value , a tsb forms , and below it
the coupling between the structure relaxation and the local shear heterogeneity during the start - up inertia is not strong enough to trigger the tsb . at this regime
the structure relaxes homogeneously , manifested by uniform shear .
the boundary between these two regimes moves to lower shear rates , as the relaxation exponent is increased . as ref .
@xcite explains , in the framework of this model , a kinetic exponent on the order of @xmath74 would be required to match the duration of the tsb to experiments .
unfortunately , using here an exponent as high as this would not be permitted due to numerical and computational limitations ( the number of required iterations for convergence ) . extrapolating the monotonically decreasing trend based on the computed values of fig .
[ fig : fig5 ] to @xmath75 hints that the minimum value for @xmath54 should be well below the typical ( minimum ) experimentally used shear rates ( @xmath76 ) .
therefore , we are inclined to argue based on the model , tsb formation and the subsequent localized flow dynamics due to start - up inertia only should be recovered trivially for all real - world complex fluids .
this has the important implication for the experiments that tsb should appear in all simple yield stress fluids with such fluidization exponents irrespective of the measuring geometry .
( dependent on @xmath6 ) , above which the start - up inertia induces a tsb and the corresponding banded profile is retrieved .
below this value , the flow remains decidedly uniform .
as seen here , this value declines rapidly as a function of @xmath6 .
note that the @xmath6 values for real - world complex fluids are on the order of @xmath74 as easily verified in ref .
therefore , extrapolating the minimum value for @xmath54 using this @xmath6 suggests that shear localization should occur trivially due to the start - up inertia for any real complex fluids in the experimentally accessible shear rates ( typically @xmath76 ( 1/s)).,scaledwidth=49.0% ]
earlier theoretical work proposes a scenario , where the transient shear banding is initialized by gradients in stress @xcite , or inhomogeneity of the sample @xcite .
the present study brings a new mechanism , which shows that transient shear banding can result even due to the start - up stress gradients caused by accelerating flow .
as we show here , depending on the linearity of the fluids viscosity response , and its initial state , this effect can be either negligible or very significant . when the initial viscosity of the fluid is small , and/or the time - dependence of the fluid is close to linear the stress gradient is not able to induce any inhomogeneities during the start - up flow .
however , according to our simulations , when the fluid at rest is in a high viscosity state and shows time - dependence under constant shear rate with a large fluidization exponent the start - up stress inhomogeneities initialize long lasting transient shear bands at all experimentally accessible shear rates .
this occurs despite the fact that the influence of the navier - stokes instability is multiple orders of magnitude faster compared to the actual tsb .
however , it creates a small but non - negligible inhomogeneity in the structure , which is amplified over time into a complete tsb .
an example of such material is the carbopol , a typical example of so called simple yield stress fluids , for which the fluidization exponent is larger than 2 @xcite and the viscosity diverges at rest .
our findings give a reasonable explanation for the initialization of transient shear banding even in experimental geometries exhibiting negligible stress heterogeneities , such as the cone and plate one .
this suggests that stress heterogeneities should be anticipated in all strongly shear thinning fluids during the start - up phase of rheological experiments .
this work was supported by the academy of finland through the comp center of excellence and the project number 278367 .
the simulations were performed using the computer resources within the aalto university school of science `` science - it '' project .
27ifxundefined [ 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 * * , ( ) link:\doibase
10.1039/c1sm05607 g [ * * , ( ) ] @noop * * , ( ) link:\doibase 10.1039/c1sm05740e [ * * , ( ) ] @noop * * , ( ) link:\doibase 10.1103/physreve.66.051501 [ * * , ( ) ] link:\doibase 10.1039/b812394 m [ * * , ( ) ] @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * ( ) | for quite some time non - monotonic flow curve was thought to be a requirement for shear banded flows in complex fluids .
thus , in simple yield stress fluids shear banding was considered to be absent .
recent spatially resolved rheological experiments have found simple yield stress fluids to exhibit shear banded flow profiles .
one proposed mechanism for the initiation of such transient shear banding process has been a small stress heterogeneity rising from the experimental device geometry . here ,
using computational fluid dynamics methods , we show that transient shear banding can be initialized even under homogeneous stress conditions by the fluid start - up inertia , and that such mechanism indeed is present in realistic experimental conditions . |
the standard model of electroweak theory is succesful in explaining most of the phenomena we observed till now .
the neutrino mass is the only indication we have that requires physics beyond the standard model .
it has been established that the neutrino sector must be extended to accommodate the observed mass - squared difference for the neutrinos in the atmospheric , solar , as well as in the laboratory experiments .
the observed neutrino masses are too small to treat them at par with charged fermions .
the most natural explanation for the smallness of the neutrino masses comes from the see - saw mechanism @xcite .
one introduces the right - handed neutrinos in the model that allows the usual dirac masses for the neutrinos to be of the order of charged lepton masses .
since the right - handed neutrinos are singlets , it is now possible to allow very heavy majorana masses @xmath0 for right - handed neutrinos .
this introduces the scale of lepton number violation in the model and in turn induces tiny majorana masses for the left - handed neutrinos .
when majorana masses of the right - handed neutrinos are introduced , they break lepton number explicitly .
however , it is more natural to associate any new scale of the theory with a symmetry whose breaking produces neutrino masses . in a forthcoming article we propose [ 11 ] , that the majorana masses of the right - handed neutrinos originate from the spontaneous breaking of lepton number , which leads to a massless goldstone boson , the majoron .
however , since the residual global symmetry is not exact , one goldstone boson picks up small mass producing ultra low mass pseudo nambu - goldstone bosons ( pngb ) .
an early proposal , along these lines , was the axion , which was introduced to solve the strong cp problem through the peccei - quinn symmetry @xcite .
the axion is still being searched for as a candidate for dark matter @xcite . in strong interactions , it is the breaking of symmetries within qcd that generates a large part of the proton s mass , which is relatively large compared to the masses of current quarks .
the result is the appearance of low mass pions the pngb s of the spontaneously broken chiral symmetry .
this concept of pngb was applied @xcite to the general formalism to a long - range force and phase transitions @xcite occurring at a late epoch of the universe , thus developing quintessence models contemporaneously with other authors @xcite .
neutrino physics has also found applications of quintessence model , in which the quintessence or the acceleron are formed as condensates and couple to neutrinos varying their masses @xcite .
this scenario has some interesting consequences @xcite . in this review
we point out that the pngb in models of majorana neutrinos can play the role of the acceleron , introducing an exponential potential , and giving rise to a long - range force that may be of of phenomenological importance @xcite .
finally , the model generates a lepton asymmetry through the decay of heavy neutrinos to light ones and higgs particles .
we demonstrate here the idea behind the pngb model of majorana neutrinos with a two generation example .
the theory possesses a global @xmath1 symmetry .
the breaking of this symmetry spontaneously sets a large scale in the theory far above the scale of electroweak symmetry breaking .
there are two right - handed neutrinos @xmath2 and @xmath3 , which are singlets under the standard model gauge groups and transform under the global symmetry as @xmath4 and @xmath5 respectively .
the global symmetry prevents any majorana mass of the right - handed neutrinos .
we also introduce two singlet scalar fields @xmath6 and @xmath7 transforming under the global symmetry as @xmath8 and @xmath9 respectively , whose interactions with the right - handed neutrinos are given by @xmath10 the vacuum expectation values ( @xmath11s ) of these scalars will give majorana masses to the right - handed neutrinos and there will be two goldstone bosons . as we shall demonstrate below , the dirac masses of the neutrinos at the electroweak scale will break one combination of the global symmetry softly giving a small mass to one of the goldstone bosons , which now becomes a pseudo nambu - goldstone boson ( pngb ) of the model .
the other goldstone boson , the majoron , remains massless . since the majoron couples only to singlet fields , the theory is consistent with phenomenology of known particles .
after the fields @xmath12 acquire @xmath11s , we can express them in terms of their @xmath11s @xmath13 and decay constants @xmath14 ( we assume their decay constants to be same , @xmath15 ) and the nambu - goldstone bosons @xmath16 as @xmath17 at this stage it is possible to make phase transformations to the fields @xmath18 that eliminate the @xmath16 , implying that both goldstone bosons are massless
. the self interactions of the goldstone bosons , given by @xmath19 also do not contain any term that can give masses to the goldstone bosons .
we now write down all the yukawa terms involving the neutrinos , including the dirac mass terms @xmath20 we introduced the dirac mass @xmath21 and some scaling parameters @xmath22 in order to write down the dirac mass terms that break the @xmath23 global symmetry softly .
we included all the phases which contribute to cp violation .
rephasing of the fields @xmath24 and rephasing of the cp phases , leads to the full mass matrix @xmath25 where @xmath26 .
thus we finally have one cp phase @xmath27 and one combination of the fields @xmath28 , which becomes the pngb .
the other combination of fields @xmath29 correspond to the invisible singlet majoron and remains massless by decoupling from the theory .
this conclusion is independent of the choice of phase transformation .
we shall now explicitly demonstrate how such @xmath1 global symmetry breaking soft dirac mass term may appear in the theory after the electroweak symmetry breaking .
if we assign the @xmath1 quantum numbers @xmath30 then the usual higgs doublet @xmath31 with the assignment @xmath32 ( for this to give masses to the charged fermions ) can give only the diagonal dirac mass terms .
we need two more higgs doublets @xmath33 and @xmath34 in order to generate the complete dirac mass we discussed above .
the mass term of the lagrangian now becomes @xmath35 these dimension-4 terms do not break the global symmetry .
however , after the electroweak symmetry breaking , when all the doublets @xmath31 and @xmath36 acquire @xmath11s , the global symmetry is broken .
the part of the lagrangian with all neutrino mass terms is : @xmath37 after rephasing of the fields it reduces to equation ( 5 ) . to find out the mass of the pngb
, we consider the effective potential generated by the interactions of the scalar field through the mass terms .
the colemann - weinberg potential for @xmath38 is computed through the leading loop in fig .
1 . it has the remarkable property that the symmetry structure of the theory makes the loop finite .
the reason is that the @xmath38 field could be eliminated if any of the vertices is set to zero .
this diagram is also invariant under any rephasing of the fields , which can be confirmed by observing that the most general phase transformation of the neutrinos , @xmath39 transforms @xmath40 so that @xmath41 and hence the diagram remains invariant .
the explicit calculation gives @xmath42 which has the minima at @xmath43 .
expanding around one of the minima , we can write down a mass term @xmath44 thus the mass of the @xmath38 field is now @xmath45 the symmetry at the scale of @xmath46 protects the mass of the pseudo nambu - goldstone boson , so an explicit soft breaking of the symmetry at the scale @xmath21 can generate a mass of the order of @xmath47 .
this light pngb ( @xmath48 ) can generate a long - range force and also become the acceleron field to explain the smallness of dark energy .
we consider next the structure of neutrino masses in this model . after
the global symmetries are broken by the @xmath11s of the scalars , the model reduces to the usual see - saw models @xcite , except for the pngb which generates a new long - range force because of its small mass .
the interactions required for the usual leptogenesis @xcite through the decays of right - handed neutrino are also present and will be discussed in the next section .
we shall now consider the effect of the light scalar field , the pngb in the model .
we obtain the interactions of the light neutrinos with the scalar field by keeping only the terms linear in pngb , given by @xmath56 the long range force introduced by the pngb in this model will have direct consequences in neutrino oscillation experiments @xcite .
the extension of the model to three generations and the implications of the new force will be discussed elsewhere [ 11 ] .
the direct connection between the neutrino masses and the baryon asymmetry of the universe is now well established . in our case ,
lepton number violating couplings were introduced in order to give masses to the neutrinos .
these couplings generate a lepton asymmetry @xcite , which will be converted to a baryon number asymmetry as the universe expands and approaches the electroweak phase transition .
in general , the generation of the lepton asymmetry of the universe requires three ingredients : 1 .
lepton number violation , which is also the source of neutrino masses ; 2 .
cp violation , which comes from the interference of tree level and one loop diagrams 3
. departure of the lepton number violating interactions from equilibrium the decays of the right - handed neutrinos violate lepton number and the yukawa couplings contain a phase @xmath27 which can give cp - even and cp - odd amplitudes .
if the right - handed neutrino mass can now satisfy the out - of - equilibrium condition , this model will be able to generate the required baryon asymmetry of the universe . in our model lepton number
is violated at tree - level in decays of the right - handed neutrinos @xmath57 the one loop diagrams , like vertex corrections and self energies @xcite will contain cp - even and -odd amplitudes . the structure of the lagrangian in eqs ( 5 ) and ( 6 ) has loops only when the higgs particles mix with each other .
let us denote by vo1 the mixing between @xmath58 and @xmath59 and similarly by @xmath60 the mixing between @xmath58 amd @xmath61 .
such mixings are generated by the quartic couplings of the potential when the scalar acquire vevs .
here we do not restrict ourselves to a specific form of the potential and represent the generic mixings by @xmath62 .
the interference between tree and vertex diagrams gives the asymmetry @xmath63 & = & % { 3 \over 16 \pi } { m_1 \over m_2 } { m^2 \epsilon % \epsilon^\prime \lambda^2 \over v_0 v_2 ( 1 + \epsilon^2)}~ \sin 2\eta
\frac3{16\pi}\frac{m_1}{m_2}\frac{m^2\lambda^2}{v_1 ^ 2+v_0 ^ 2\epsilon^2 } \left [ \right]\sin 2\eta\end{aligned}\ ] ] where @xmath64 and we assumed @xmath65 , so that the asymmetry is generated by the decays of @xmath2 . in the context of leptogenesis the lightest @xmath2 is naturally out of thermal equilibrium when it decays .
the dilusion factor depends on the relative magnitude of @xmath66 relative to the hubble constant @xmath67 when the ratio @xmath68 is small the lepton excess is large . on the other hand , for large @xmath69
the time of expansion is large relative to the lifetime of @xmath2 so that many decays and recombinations can take place in the cosmological scale @xmath70 giving a smaller excess of leptons .
finally , the sphaleron interactions convert a fraction of the produced @xmath71 asymmetry to a baryon asymmetry @xmath72 this conversion takes place from the time of leptogenesis down to the time of the electroweak phase transition . here
@xmath73 is the number of higgs doublets in our model . in summary
, it is evident that the model can generate the required amount of baryon asymmetry of the universe .
one of the most challenging questions at present is why the cosmological constant is very small but nonvanishing .
if the dark energy were due to the presence of a cosmological constant , then one expects @xmath74 where @xmath75 is the energy associated with a particle or a field .
the observed cosmological constant coresponds to an @xmath76 ev , which is very small for the scales of most elementary particles .
it is of the order of neutrino masses and it has recently been proposed that the dark energy tracks the neutrino density @xcite . if the neutrino masses vary as functions of a light scalar field ( called the acceleron ) the dark energy density may track the neutrino masses .
this would then explain why the scale of cosmological constant is the same as the scale of neutrino masses . in the present model ,
the pngb introduces a spatial dependence into the majorana mass m1 @xmath77 the first term @xmath78 has an exponential functional dependence on the scalar field .
the effective neutrino mass then varies explicitly with the acceleron field @xmath79 , as required by models of mass varying neutrinos @xcite .
a large majorana scale is a necessary ingredient of the see - saw mechanism . at this very high energy there
may exist global symmetries which are broken in order to create the majorana masses .
remnants of the global symmetry survive in the low energy interactions . in a two generation model developed with my collaborators hill , mocioiu and sarkar @xcite , the breaking of the symmetry produces two nambu - goldstone bosons .
the introduction of dirac masses at the electroweak scale further breaks the symmetry softly giving a small and finite mass to one of the nambu - goldstone bosons and leaves the other scalar massless .
the exchange of the scalar particle introduces a new long - range force between neutrinos . finally , the model has additional attractive properties .
it describes correctly the masses and mixings of the light neutrinos .
the decays of the heavy neutrinos generate a lepton asymmetry consistent with leptogenesis .
the neutrino masses have an explicit dependence on the scalar field , which may bring density effects with the pngb playing the role of the `` acceleron '' .
i wish to thank my collaborators drs .
c.t . hill , i. mocioiu and u. sarkar for a pleasant and fruitful collaboration .
an extensive article mentioned in ref .
[ 11 ] is in preparation .
i also thank dr .
zhou and mr .
a. karavtsev for working with me on extensions of the model . the financial support of bmbf , bonn under contract 05ht 4pea/9 is gratefully ackonwledged .
finally , i thank the dfg for a travelling grant to participate at the conference .
99 m. gell - mann , p. ramond , and r. slansky , in _
supergravity _ , edited by p. van nieuwenhuizen and d. z. freedman ( north - holland , amsterdam , 1979 ) , p. 315 ; t. yanagida , in _ proceedings of the workshop on the unified theory and the baryon number in the universe _ , edited by o. sawada and a. sugamoto ( kek report no
. 79 - 18 , tsukuba , japan , 1979 ) , p. 95 ; + p. minkowski , phys .
* b 67 * , 421 ( 1977 ) ; + r. n. mohapatra and g. senjanovic , phys .
* 44 * , 912 ( 1980 ) ; + j. schechter and j.w.f .
valle , phys . rev .
* d 22 * , 2227 ( 1980 ) ; phys . rev .
* d 25 * , 774 ( 1982 ) . r.d .
peccei and h.r .
quinn , phys .
lett . * 38 * , 1440 ( 1977 ) ; phys . rev .
* d 16 * , 1791 ( 1977 ) .
s. weinberg , phys .
. lett . * 40 * , 223 ( 1978 ) ; + f. wilczek , phys .
lett . * 40 * , 279 ( 1978 ) .
k. zioutas et al . , phys .
lett . * 94 * , 121301 ( 2005 ) .
c.t . hill and g.g .
ross , phys .
* b 203 * , 125 ( 1988 ) ; nucl . phys . * b 311 * , 253 ( 1998 ) . c.t .
hill , d.n .
schramm and j.n .
fry , nucl .
* 19 * , 25 ( 1989 ) . c. wetterich , nucl .
b302 * , 668 ( 1988 ) ; + p. j. e. peebles and b. ratra , astrophys .
j. * 325 * , l17 ( 1988 ) .
r. fardon , a. e. nelson , and n. weiner , jcap * 0410 * , 005 ( 2004 ) ; + see also p. q. hung , hep - ph/0010126 ; + p. gu , x. wang , and x. zhang , phys .
rev . * d68 * , 087301 ( 2003 ) .
d. b. kaplan , a. e. nelson , and n. weiner , phys .
lett . * 93 * , 091801 ( 2004 ) ; + g. dvali , nature * 432 * , 567 ( 2004 ) .
bi , p. gu , x. wang , and x. zhang , phys .
rev . * d69 * , 113007 ( 2004 ) ; +
gu and x .- j .
bi , phys . rev . * d70 * , 063511 ( 2004 ) .
hill , i. mocioiu , e.a .
paschos and u. sarkar , _ in preparation_. m. fukugita and t. yanagida , phys . lett . * b174 * , 45 ( 1986 ) . m.a .
luty , phys .
d*45 * , 455 ( 1992 ) ; + w. buchmuller , p. di bari and m. plumacher , annals phys .
* 315 * , 305 ( 2005 ) .
m. flanz , e.a .
paschos and u. sarkar , phys . lett .
* b 345 * , 248 ( 1995 ) ; + m. flanz , e.a .
paschos , u. sarkar and j. weiss , phys . lett .
* b 389 * , 693 ( 1996 ) ; v. barger , p. huber , and d. marfatia , phys . rev . lett . * 95 * , 211802 ( 2005 ) ; + m. cirelli , m. c. gonzalez - garcia , and c. pena - garay , nucl . phys . *
b719 * , 219 ( 2005 ) . | a natural scenario for the generation of neutrino masses is the see - saw mechanism , in which a large right - handed neutrino mass makes the left - handed neutrinos light .
we review a special case when the majorana masses originate from spontaneous breaking of a global u(1)xu(1 ) symmetry .
the interactions of the right - handed with the left - handed neutrinos at the electorweak scale further break the global symmetry giving mass to one pseudo nambu - goldstone boson ( pngb ) .
the pngb can then generate a long - range force .
leptogenesis occurs through decays of heavy neutrinos into the light ones and higgs particles .
the pngb can become the acceleron field and the neutrino masses vary with the value of the scalar field . ] .
address = institut fr physik , universitt dortmund , d-44221 dortmund , germany , |
the standard @xmath3cdm cosmological model has been successfully applied to explain the main characteristics of the universe ( see e.g. * ? ? ?
* ; * ? ? ?
. in particular , numerical simulations of collisionless cold dark matter ( dm ) describe the gravitational growth of infinitesimal primordial density perturbations , probed by the cosmic microwave background anisotropies , that lead to the formation of the present - day large - scale structure of the universe .
these simulations provide us with a number of predictions about the structure of cold dm halos and their basic properties ( see e.g. * ? ? ?
although the standard cosmological model has proven to be highly successful in explaining the observed large - scale structure of the universe , it has been less successful on smaller scales .
recent data on low mass galaxies suggests that the inferred subhalo dm distributions have almost flat cores ( see e.g. * ? ? ?
* ) , in contradiction with the cuspy profile distributions predicted by numerical simulations ( see e.g. * ? ? ?
* ) , and that sub halo numbers are overpredicted at both low mass and intermediate masses ( see e.g. * ? ? ?
* ) and massive dwarf galaxy scales ( see e.g. * ? ? ?
some of the issues associated with dwarf galaxies can be addressed if the dm is more `` collisional '' ( with baryons ) than currently considered in numerical simulations .
this hypothesis favors constant - density cores with lower central densities than those coming from cold dm models , such as self interacting dm ( see e.g. , * ? ? ?
in addition , the closeness between @xmath4 and @xmath5 , usually referred as cosmic coincidence , may suggest a profound link between the dark and the ordinary sectors . indeed , although the two sectors have different macroscopic properties , the total amount of dm observed today could be produced in the early universe by a mechanism identical to baryogenesis , and therefore an asymmetry between dm particles and their antiparticles would be expected .
a detailed account of current progress in asymmetric dm studies can be found in the literature ( see e.g. , * ? ? ?
all of these cosmological facts may suggest that dark and ordinary matter may have more properties in common than expected . in view of this
, it is tantalizing to imagine that the dark world could be similarly complex ( cp violating and asymmetric ) and full of forces that are invisible to us ( hidden parallel sector or mirror world ) .
the history of the early mirror universe has been studied in ref .
@xcite , while the impact of an hidden mirror sector in the cmb and lss data can be found in ref .
@xcite . for a general review on the properties of a hidden world neighboring our own ,
see e.g. refs .
@xcite . more specifically , since in our sector only long range electromagnetic force and gravity affect the dynamical evolution of virialized astrophysical objects , the physics of a complex dark sector in which the matter fields are charged under an extra @xmath0 gauge group is particularly interesting to study . indeed ,
if the mass of the new gauge boson ( dark photon ) is smaller that the typical momenta exchanged in the scattering , the phenomenology of the dark world can in itself be as complicated as that of our sector ( e.g. dark electromagnetism with quite large self - interactions ) , providing at the same time also a feeble long - range interaction between the two worlds ( thanks to the kinetic mixing of the new @xmath0 gauge boson with the photon ) .
this implies in general that the dark world is more collisional than the standard cold dm , and therefore some of the previously mentioned problems can probably be solved ( see e.g. * ? ? ?
in addition , since a long - range dm - nucleus interaction is enhanced for small momentum exchanges , this class of models can also relax the tension between the positive results of direct detection experiments ( the annual modulation observed by dama @xcite and cogent @xcite , and the hints of an observed excess of events on cresst @xcite and cdms - si @xcite ) , and the constraints coming from null results ( e.g. cdms - ge @xcite , xenon100 @xcite and very recently lux @xcite ) .
the phenomenology of long - range dm - nucleus interactions in direct dm searches has been studied in refs .
@xcite . in this paper
, we investigate how a dm - baryon long range interaction ( dmlri ) , induced by the kinetic mixing of a new @xmath0 gauge boson and the photon , affects the evolution of the sun .
the results obtained are then confronted with helioseismology data .
the helioseismic data that we use was obtained by several international collaborations , such as the solar and heliospheric observatory ( soho ) mission and the birmingham solar oscillations network ( bison ) observational network @xcite .
furthermore , we also discuss the constraints on the main parameters of long - range dm particle interactions which can be obtained from direct dm search data .
for a fix value of the kinetic mixing parameter @xmath6 and a dm mass @xmath7 gev . in blue
is shown the scattering cross section responsible for the capture process
@xmath8 , while in red the one entering in the energy transport mechanism @xmath9 computed at the present time ( @xmath10 k ) . ] in most of the classical models ,
one often assumes a symmetric dark sector in which the dm scattering off of baryon nuclei is done by a contact - like interaction . in our study , we will instead focus on a class of asymmetric dm models in which the interaction between dm particles and target nuclei is mediated by a light messenger .
if the typical momenta exchanged in the scattering is bigger than the mass of the mediator , a long range interaction then arises . a specific realization of this kind of picture is offered by particle physics models where a new @xmath0 hidden gauge boson @xmath11 ( dark photon ) possesses a small kinetic mixing @xmath12 with the photon .
in this case , the interaction between a nucleus with mass @xmath13 and electric charge @xmath14 ( @xmath15 is the number of protons in the baryon nucleus and @xmath16 the electrical charge ) and a dm particle with mass @xmath17 and dark charge @xmath18 ( @xmath19 and @xmath20 are the equivalent quantities of @xmath15 and @xmath16 in the dark sector ) is described in the non - relativistic limit by the following yukawa potential ( see e.g. , * ? ? ? * ) , @xmath21 where @xmath22 is the electromagnetic fine structure constant and the parameter @xmath23 measures the strength of the dm - dark photon coupling . here
@xmath24 is the mass of the dark photon that acts like an electronic cloud which screens the charges of the particles involved in the scattering . since both @xmath25 and @xmath12 are unknown , we define @xmath26 and we treat it as a free parameter of our model together with @xmath17 and @xmath24 . before moving on
we have however verify whether the symmetric component of @xmath27 in the dark sector is annihilated away or not .
in general , since for this simple model the dominant annihilation channel is given by @xmath28 , we can put a lower bound on the parameter @xmath25 by demanding that the annihilation cross section into dark photons ( @xmath29 ) is bigger than the thermal one ( @xmath30 pb ) . in agreement with refs .
@xcite , this condition requires @xmath31 , where @xmath32 .
we shall come back to this point at the end of sec .
[ sec:3 ] because for this kind of models the dm - dm scattering can plays an important role in the capture rate and in the energy transport by dm particles in the sun .
the differential cross section , neglecting the form factor of the target nuclei , can be simply obtained by performing the fourier transform of eq .
[ vr ] and it reads @xmath33 where @xmath34 , @xmath35 is the dm - nucleus reduced mass and @xmath36 is the momenta exchanged in the interaction with @xmath37 the recoil energy . as is apparent from the dependence on the dark photon mass ,
two different regimes clearly appear : * _ point like limit _
( @xmath38 ) : in this regime the interaction is of a contact type .
indeed the differential cross section turns out to be proportional to @xmath39 which plays the same role as fermi s constant in weak interactions .
the interaction reduces to the `` standard '' spin - independent picture , apart from the fact that in this case , dm particles only couple with protons ( @xmath40 ) . *
_ long range limit _ ( @xmath41 ) : in this regime , the differential cross section acquires an explicit dependence on the momenta exchanged in the interaction and therefore a rutherford - like cross section arises ( @xmath42 ) .
this is extremely interesting because ideal experiments with very low energy threshold and light target nuclei ( e.g. the sun ) are in principle more sensitive than the ones with high threshold and heavy targets ( e.g. xenon100 and lux ) . to give a concrete example ,
once dm particles are thermalized with baryons in the center of the sun , their collisions occur with @xmath43 mev considering a dm mass of 10 gev . for direct detection experiments ,
the typical momenta transferred in the scattering is instead bigger ( @xmath44 mev ) . thanks to this fact , we expect that dm models feature a long - range interaction with ordinary matter can affect the sound - speed radial profile of the sun without being excluded by terrestrial experiments .
having the differential cross section at our disposal , we assume that the normalization of both the capture rate and the transport of energy by dm particles in the sun is controlled by the energy transfer cross section obtained by weighting eq . with ( @xmath45 ) .
this is of course a good estimator for the transport mechanism , while for the capture process the introduction of the same cross section is justified because the sun s escape velocity @xmath46 is much bigger than both the thermal velocity of baryons in the sun and the typical dispersion velocity of the dm particles in the halo @xmath47 .
in particular , we have checked that the error in the total rate is negligible , if we replace the total cross section in the rate per unit time @xmath48 given e.g. in eq .
7 of @xcite , with the energy transfer cross section computed for @xmath49 . in the born approximation ( see e.g. @xcite ) .
since we are interested in the long - range regime ( @xmath50 ) , it is easy to check that this approximation is very well satisfied . ]
, the energy transfer cross section writes , @xmath51 \ , \label{eq - sigma}\end{aligned}\ ] ] where @xmath52 , @xmath53 , and @xmath54 is the relative velocity between the dm flux and the sun .
unlike the customary dm models , in this case @xmath55 depends on @xmath54 in the long range regime .
thanks to this main novelty , one therefore expects that the typical scattering cross section in the capture process differs compared to the one entering in the transport mechanism . on a more specific level , one has : _ capture : _ in general , the infalling dm particles reach a given shell of radius @xmath56 with a velocity @xmath57 , where @xmath58 is the dm velocity at infinity with respect to the sun s rest frame
. since both the thermal velocity of baryons in the sun and the dispersion velocity of dm particles in the halo are much smaller than @xmath59 , we can assume as commented upon above that the relative velocity @xmath60 .
furthermore , since the total number of dm particles captured by the sun is independent on @xmath56 , it is a good approximation to define an average infalling dm velocity by @xmath61 here @xmath62 is the sun s mass density , @xmath63 kg is its total mass and @xmath64 cm is its radius . as explained in more detail in the next section , the capture rate
is then computed numerically considering a constant cross section @xmath65 .
_ transport : _ in this case , the typical relative velocity for the scattering is much smaller than @xmath66 being due to dm particles thermalized together with the ordinary plasma in the center of the sun .
it is then a good approximation to assume @xmath67 , where @xmath68 is the thermal speed and @xmath69 is the time - dependent temperature in the sun s core .
as explained in more detail in the next section , we compute the transport of energy numerically by considering @xmath70 .
it is worth stressing that since the solar code follows the time evolution of the sun , in the early stages the energy transport , and in turn the thermal conduction by dm particles , was much more efficient with @xmath69 at that time being smaller than the present - day central temperature . in this study , we focus on the interaction of dm with hydrogen
the most abundant chemical element in the sun s interior .
[ fig:1 ] shows an illustrative example of the dm - hydrogen energy transfer cross section @xmath55 as a function of the mass of the dark photon @xmath24 for a fixed value of the kinetic mixing parameter @xmath6 and a dm mass @xmath7 gev . on a more specific level
we show in blue the scattering cross section responsible for the capture process
@xmath8 , while in red is shown the one entering into the energy transport mechanism @xmath9 computed at the present time ( @xmath10 k ) .
we can see that if the mass of the dark photon is smaller than a few mev ( long range regime ) , the capture and the transport processes are controlled by different scattering cross sections .
it is worth pointing out that in this limit , the ratio @xmath71 is barely dependent on @xmath24 , if @xmath17 is larger than the hydrogen mass .
it instead depends on the mass of the dm particle through the thermal velocity relation and in particular for a 10 gev candidate , @xmath72 .
the main new aspect is actually given by this enhanced conduction in the inner part of the sun compared to the usual dm models .
thanks indeed to this fact , dm particles , interacting via long - range forces with ordinary matter , can produce an impact on the helioseisomology data without evading the constraints coming from direct dm search experiments . in our analysis
, we consider the observed sound speed radial profile of the sun and compare it to the theoretical prediction over a broad range of dm masses ( 4 gev@xmath73 20 gev ) , dark photon masses ( 0.1 kev @xmath74 1 gev ) and therefore our results can also be extended for lighter mediators . ] and kinetic mixing parameters ( @xmath75 ) .
with these choices , the scattering cross - section spans a large interval of values from @xmath76 to @xmath77 .
the cross - section range of interest for the sun corresponds to the values for which @xmath78 is close to the sun s characteristic scattering cross - section , @xmath79 . from the zero - age mean - sequence ( zams ) until the present age of the sun ( @xmath80 ) , @xmath81 takes values between @xmath82 @xmath83 and @xmath84 @xmath83 .
, @xmath85 is the sound speed of the ssm ( e.g. , * ? ? ? * ) and @xmath86 is either @xmath87 the observed sound speed ( green - square dots : * ? ? ? * ; * ? ? ? * ) or @xmath88 the sound speed of a dmlri solar model ( continuous curves : [ @xmath89 blue , [ @xmath90 green and [ @xmath91 red ) .
the dmlri models were computed for the parameters ( see text ) : @xmath92 ; @xmath93 and @xmath6 .
the black curve corresponds to a fiducial model with @xmath94 and @xmath95 .
note that the observational error in @xmath96 is multiplied by a factor 10 . ]
the sun , as are all stars in the milky way , is immersed in a halo of dm . as with any other star in the galaxy ,
the sun captures substantial numbers of dm particles during its evolution , but any impact on the star depends on the properties of these particles , as well as on the stellar dynamics and structure of the star . in general , for low - mass stars evolving in low density dm halos , the presence of dm inside the star changes its evolution by providing the star with a new mechanism to evacuate the heat produced in the stellar core ( e.g. , * ? ? ? * ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
this is quite different for stars evolving in dm halos of high density as occurs during the formation of the first generation of stars . in these cases , the annihilation of dm particles supplies the star with an additional source of energy capable of substantially extending the lifetimes of these stars @xcite .
the computation of the impact of dm in the evolution of the sun is done by a modified version of the cesam code , which has been widely used to compute the ssm and for modelling other stars by different research groups . in this study , we follow a procedure identical to other studies published in the literature by some of us as well as other authors ( e.g. , * ? ? ? * ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
nevertheless , there are some important differences between this study and the previous ones , which we will highlight in the remainder of this section .
as commented upon in the previous section , the impact of dm in the sun s interior depends on two major physical processes : the accumulation of dm inside the star , and the efficiency of dm in transferring energy from the core to the external layers . in any case
, the density of dm in the halo is the single major ingredient affecting the impact of dm in the sun .
as usually done in these studies , we consider that the local density of dm is @xmath97 @xcite .
the choice of this value is in part made to facilitate the comparison with other work .
still , this is a very reliable value : the most recent estimates of @xmath98 made by two independent groups have obtained values of @xmath99 @xcite and @xmath100 @xcite . in particular
@xcite argue that their new method is quite robust , and they have obtained their value at 90% confidence level . in our computation , we will also consider that the dm particles in the solar neighborhood have dispersion velocity @xmath101 ( see e.g. , * ? ? ? * ) .
the accumulation of dm in the sun s core during its evolution from the beginning of the zams until the present age ( @xmath102 ) is regulated by three physical processes : capture , annihilation and evaporation of dm particles . at each step of the evolution ,
the total number of dm particles @xmath103 , that is captured by the star is given by @xmath104 where @xmath105 , @xmath106 , @xmath107 and @xmath108 refer to the dm capture , self - capture , annihilation and evaporation rates respectively .
it is worth noticing that , unlike previous studies , in this work we resolve numerically the equation ( [ eq - n ] ) for each step of the star s evolution .
a detailed discussion about these processes can be found in the literature ( e.g. , * ? ? ?
* ; * ? ? ?
the capture rate @xmath105 is computed numerically from the expression obtained by @xcite as implemented by @xcite .
the scattering of dm particles with the baryons inside the sun is the main factor affecting the capture rate @xmath105 .
we restrict our study to the scattering of dm particles to the most abundant element , i.e. , hydrogen .
we consider that the capture rate is controlled by @xmath8 obtained by substituting @xmath54 with the averaged infalling speed @xmath66 in eq .
( [ eq - sigma ] ) as briefly explained in the previous section .
in particular , in the long - range regime ( @xmath109 or in terms of the exchanged momenta , @xmath41 ) , the dm - hydrogen energy transfer cross section responsible for the capture process is independent of @xmath17 and logarithmically dependent on @xmath24 .
it reads @xmath110 where @xmath111 is a sort of coulomb logarithm that measures the strength of the screening effect in the capture process and @xmath112 is its value for @xmath113 kev .
the other chemical elements are under - abundant , consequently their contribution for the capture of dm is negligible .
the dependence of the scattering cross - section on the parameter space of long - range dm particles is given by equation ( [ eq - sigma ] ) .
the description of how this capture process is implemented in our code is discussed in @xcite .
the evaporation rate @xmath108 is relevant only for very light particles , i.e. , particles with @xmath114 @xcite .
@xcite estimated the evaporation mass for dm particles in the sun , @xmath115 , to be such that @xmath116 .
if @xmath117 the dm particle escapes the solar gravitational field and consequently has no impact on the structure of the star .
@xcite also found that the evaporation of dm particles is completely irrelevant for @xmath118 .
it is worth noticing the fact that because the evaporation boundary has a exponential dependency on the mass and the scattering cross - section of the dm particle @xcite , it follows that if @xmath17 exceeds @xmath115 by a few percent , then the evaporation of dm particles is totally negligible .
furthermore it is also important to point out that our dm model can easily have a quite large self interaction in the long range regime with size similar to the electromagnetic scattering . if then this self interaction is attractive ( e.g. this can be obtained if the dark sector is composed by light and heavy species with a different sign of the dark charge @xmath18 ) we expect that @xmath108 can be set to zero for dm masses below 4 gev as well , in such cases the properties of the dark plasma being similar to the ordinary one ( electrons are indeed trapped in the sun ) .
we leave further discussion of this important effect for future studies . in our computation , we use @xmath108 estimated by @xcite in the regime where the sun is optically thin with respect to the dm particles .
nevertheless , we restrict our analysis to particles with mass larger than @xmath119 , for which the evaporation rate is almost negligible . once gravitationally captured by the star ,
the dm particles thermalise with baryons after a few kepler orbits around the solar centre , colliding through elastic scattering with hydrogen and other elements , and thus providing the star with an alternative mechanism for the transport of energy .
the relatively efficiency of the dm energy transport in relation to the radiative heat transport depends on the knudsen number , @xmath120 where @xmath121 is the free mean path of the dm particle inside the star and @xmath122 is the characteristic radius of the dm distribution @xcite given by @xmath123 where @xmath124 gev/@xmath125 is the today s density of the sun s core .
depending on the value of @xmath126 , the transport of energy by dm is local ( conductive ) or non - local , which corresponds to @xmath127 or @xmath128 . in the conductive regime
the effective luminosity carried by dm particles is proportional to @xmath129 , while in the non - local one to @xmath130 .
the maximal luminosity is instead achieved when @xmath131 ( see e.g. fig .
11 of @xcite ) . here
@xmath132 is the number density of dm particles captured in the sun which is given by @xmath133 $ ] where @xmath134 ( see e.g. @xcite ) .
we assume that the thermal conduction by dm particles is controlled by @xmath9 obtained by substituting @xmath54 with the thermal speed @xmath135 in eq .
( [ eq - sigma ] ) as briefly explained in the previous section . unlike the cross section responsible for the capture process , in the long range
limit @xmath9 depends on the dm mass via the thermal velocity @xmath135 .
it explicitly reads @xmath136 where @xmath137 emulates the screening effect in the energy transport and @xmath138 is its value for @xmath113 kev , @xmath7 gev and @xmath139 .
once computed the energy transfer cross section the average mean free path in the long - range regime , can be estimated as @xmath140 where @xmath141 is the average number density of baryons in the sun and @xmath142 .
depending on the free parameters of the model scanned in our analysis , the energy transfer cross section covers a broad range of values ( @xmath143 ) . in this case the corresponding range of the knudsen number is ( @xmath144 ) .
both mechanisms of energy transport are considered in this study @xcite . in the case of the non - local regime , we follow the numerical prescription of @xcite rather than the original one proposed by @xcite . we note that in the long - range limit , a rutherford - like dm - baryon interaction can significantly enhance the energy transport ( @xmath145 ) compared to the usual picture ( for the standard spin - independent and spin - dependent cases , one always has @xmath146 ) . indeed ,
even if the capture cross section is around @xmath147 @xmath83 ( this choice correspond to @xmath148 ) , the dm particles can have a mean free path around @xmath149 , providing then the maximum luminosity carried by dm particles . for the standard contact interaction and for the same cross section , the mean free path is much longer and therefore the luminosity is reduced being in this case the transport non - local .
this is the main new element , and in particular , as we will see in sec .
[ cconstraints ] , this class of models with enhanced energy transport can solve the so - called solar abundance problem without being excluded by direct searches experiments .
a similar conclusion can also be found in ref .
in particular they implement the formalism of @xcite in order to properly account for both velocity and exchanged momenta dependencies in the differential cross section . although their method is more refined compare to the one we are using they did not incorporate their results in a solar simulation software yet .
the impact of dm in the sun is studied by inferring the modifications that dm causes to the sun s structure and to the solar observables . in the following , the standard solar model ( ssm
; e.g. , * ? ? ?
* ; * ? ? ?
* ) is used as our model of reference , which predicts solar neutrino fluxes and helioseismology data consistent with current measurements . the excellent agreement obtained between theory and observation results from the combined effort between the fields of helioseismolgy and solar modelling , a collaboration extended by several decades which lead to a high precision description of physical processes present inside the sun @xcite .
this was very relevant in the case of the physical processes related with microscopic physics , including the equation of state , opacities , nuclear reactions rates , and microscopic diffusion of helium and heavy elements .
a detailed discussion about current predictions of the ssm and their uncertainties can be found in the literature ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) . in fig .
[ fig:2 ] we compare the sound speed profile of ssm with the sound speed computed by an inversion technique from helioseismology data @xcite .
the green square dots correspond to the relative sound speed difference @xmath150 , where @xmath151 and @xmath96 are the sound speed from ssm and helioseismic data .
@xmath152 is smaller than 2% throughout the solar interior , above 20% and below 90% of the sun s radius .
although agreement between @xmath151 and @xmath96 is very good , a discrepancy remains between the present ssm and helioseismic data , from which there is no obvious solution @xcite .
it is worth noticing that the quality of the sound speed inversion is highly reliable , as most of the helioseismic data has a relative precision of measurements larger than @xmath153 .
contrarily , in the sun s inner core below @xmath154 , the seismology data available is quite sparse and consequently the sound speed inversion is less reliable ( cf .
[ fig:2 ] ) .
as pointed out by @xcite the inversion of the sound speed profile in the sun s inner core is limited by the low number acoustic frequencies measured ( see table 1 in @xcite and references therein ) , as well as by the weak sensitivity of the eigenfunctions of global acoustic modes to the structure of the sun s core .
this difficulty can only be overcome with the positive detection of gravity modes .
equally , in the most external layers of the sun , the inversion of the sound speed profile is not possible , mainly due to the fact that the inversion technique breaks down ( acoustic oscillations are no - longer adiabatic ) , as a complex interaction occurs between convection , magnetic fields and acoustic oscillations @xcite .
accordingly , for the purpose of this study , we choose to consider the theory - observation uncertainty to be of the order of 4% rather than 2% . in the remainder of the article we will refer to this value as the _ ssm uncertainty _ , meaning the undistinguished uncertainty related to the physical processes of the ssm or helioseismogy sound speed inversion .
the dmlri solar models were obtained in an identical manner to the ssm , by adjusting the initial helium @xmath155 and the mixing length parameter @xmath156 in such a way that at the present age ( @xmath157 ) , these solar models reproduced the observed values of the mass , radius and luminosity of the sun , as well as the observed photospheric abundance ratio @xmath158 , where @xmath159 and @xmath15 are the mass fraction of hydrogen and the mass fraction of elements heavier than helium , respectively .
[ fig:2 ] shows a comparison between ssm and different dmlri models .
the different continuous lines correspond to the squared sound speed difference @xmath160 where @xmath161 is the sound speed of dmlri solar models .
these models are obtained for a fiducial value of @xmath6 and different values of @xmath17 and @xmath24 .
the most important point about fig .
[ fig:2 ] is the fact that there are some dmlri solar models that can resolve the current discrepancy with helioseismology data , as @xmath162 reproduces the observed discrepancy @xmath163 . in dmlri models ,
the dm impact is most visible in the core of the star where the dm particles accumulate .
however , because the solar models are required to have the current observed values of radius and luminosity , a decrease of the production of nuclear energy in the sun s core due to the reduction of the central temperature ( caused by the thermalisation of dm with baryons ) , is compensated by an increase of the sound speed in the radiative region . in fig .
[ fig:2 ] we show an illustrative dmgri solar model with benchmark parameters : @xmath94 , @xmath95 and @xmath6 ( _ black curve _ ) . moreover , all the dmlri solar models have an identical impact behaviour on the solar structure , however , based upon the parameters @xmath17 and @xmath24 it is possible to distinguish three sets of models : @xmath164 dmlri models for which the squared sound speed difference is larger than the _ ssm uncertainty _ ( _ red curves _ ) ; @xmath165 dmlri models for which the agreement with the helioseismic data is better than the current ssm ( _ green curves _ ) ; @xmath166 dmlri models for which the squared sound speed difference is smaller than the _
ssm uncertainty _
( _ blue curves _ ) .
although , there is a large set of dmlri models ( _ red _ and _ green curves _ ) for which @xmath162 tends to agree @xmath163 , as for those models for which the central value of @xmath161 varies more than @xmath167 , the central temperature will change for an identical amount .
consequently , the solar neutrino fluxes of these models become strongly in disagreement with the current neutrino flux observations ( e.g. , * ? ? ?
* ; * ? ? ?
therefore , we make the conservative option in this preliminary study to only consider models for which the central temperature does not change very much from the ssm ( _ green curves _ dmlri models ) . fig .
[ fig:3 ] shows the maximum sound speed difference @xmath168}$ ] in the full parameter space of dmlri models ( @xmath169 ) once the constraint @xmath170 is imposed .
the represented percentages are performed for @xmath171 . on a more specific level in the right panel of fig .
[ fig:3 ] we project the parameter space in the ( @xmath172 ) plane by choosing @xmath113 kev . in the central plane
the ( @xmath173 ) plane is shown for @xmath7 gev , while on the left panel , the parameter space is projected in the ( @xmath174 ) plane keeping fix @xmath175 . once @xmath176 is larger than the _ ssm uncertainty _
it is reasonable to exclude all dmlri models , since they produce a large impact on the sun s core sound speed profile ( _ red regions _ ) . on the other hand ,
if @xmath176 is in the range @xmath177 the agreement with the helioseismic data , as commented upon above , is improved ( _ green regions _ ) .
for instance , we find that dm particles with a mass in the range 4 gev8.5 gev coupled with ordinary baryons via a kinetic mixing parameter @xmath12 bigger than @xmath178 , produce a very large impact on the sun s core in the long range regime ( @xmath24 smaller than a few mev )
. therefore they can be excluded as possible dm candidates . on the other hand , we can see that dm particles with a mass of the order of @xmath179 , a kinetic mixing parameter @xmath180 and a mediator with a mass smaller than a few mev improve the agreement between the best solar model and the helioseismic data .
this is quite interesting since direct dm searches experiments , as we shall see in the next section , either do not , or barely exclude , these kinds of dm models with long range interactions with baryons .
furthermore , it is worth noticing that for very light dark photon ( @xmath24 smaller than few kev ) , and for @xmath181 the transport occurs in the conductive regime . in this case , as we can see in the figure , our results are independent on @xmath24 , because the effective luminosity carried by dm particles in the conductive regime is proportional to @xmath182 .
this is important because , as we will see in the next section , the complementary constraints coming from supernova observations and astrophysical or cosmological arguments are very week for very light mediators .
in this section , we present the complementary constraints which are relevant for dmlri models .
a first class of them come from terrestrial direct detection experiments . in particular , since the interactions considered in our work is spin - independent , we only compute the constraints coming from the xenon100 and lux experiments .
the statistical analysis used to treat the datasets can be found in @xcite where the authors have provided a complete sets of numerical tools for deriving the bounds in direct searches in a completely model independent way . without entering in the details of this work we can compute the bounds in the relevant parameter space of our model following the main steps summarized in sec . 6 of @xcite .
in particular , as pointed out in the first three steps ( 1a-1c ) , we have to identify the non - relativistic operator and its coefficient associated with a yukawa - type interaction given in eq . .
since this interaction is spin independent the operator is simply the identity , while the non - relativistic coefficient is given by @xmath183 ( see e.g. @xcite for details ) .
thanks now to ready - made scaling functions provided in the webpage of @xcite , the bounds on the parameter space of our model can be obtained by following the last two steps ( 2a-2b ) . in fig .
[ fig:4 ] we show the constraints in the relevant parameter spaces of our model .
as we did in the previous section , we project the parameter space in the ( @xmath184 ) plane ( left panel ) , ( @xmath185 ) plane ( central panel ) and ( @xmath174 ) plane ( right panel ) .
the dark blue ( red ) lines refer to the constraints coming from xenon100(lux ) , while the different hatching ( dotted , dashed - dotted , dashed ) indicates the three values of the third direction in the parameter space that we kept fixed ( @xmath186(1 , 0.1 , @xmath187 0.01 ) gev on the left panel , @xmath188(6 , 10 , 14 ) gev on the central panel and @xmath189 on the right panel ) .
the areas of the parameter space above the first and second plots , and the ones below the third plot are excluded .
similar constraints can also be found in ref .
we can see that in the long range regime ( mediator masses below roughly 10 mev ) , the constraints becomes independent of @xmath24 . indeed in this case , the interaction is rutherford - like and therefore the differential cross section in eq . , which sets the normalization of the total number of events in certain experiments , solely depends on the exchanged momentum @xmath190 .
by virtue of this fact , we can just use the left plot of fig .
[ fig:4 ] and compare it directly with the results shown in fig .
[ fig:3 ] .
we can see that in the relevant dm mass range that affects helioseismic data ( 420 ) gev , the constraints , coming from the lux experiments , exclude dmlri models that are coupled with baryons through kinetic mixing parameters larger than roughly ( @xmath191 ) .
therefore , as is apparent , long - range spin - independent dm - baryon interactions can easily improve the agreement between the best solar model and the helioseismic data without being excluded by direct detection experiments for dm masses in the range ( 4 - 12 ) gev .
indeed , as we have already pointed out , the sun is an ideal experiments for dm models which posses an enhanced cross section with baryons for small momentum exchanges .
a third class of constraints which instead solely depend on the characteristics on the dark sector itself are those coming from self interactions . indeed , since for this type of model , the dm - dm scattering is not suppressed by @xmath12 , the self
interactions , especially in the long range limit , can easily reach large values , affecting the dynamics of virialized astrophysical objects . in refs .
@xcite , the constraints on @xmath24 coming from the observations of a few elliptical dm halos have been presented .
these constraints are in general very stringent for dm masses below 10 gev : in this case , in fact , the time over which energy is transferred in the system is extremely rapid and therefore a spherical dm halo tends to form in contradiction with observations . in particular , dark photon masses below roughly 100 mev are excluded
. however , the derivation of this last class of constraints is very uncertain , both from the theoretical and experimental side , because we do not really know how virialized astrophysical objects are formed in the presence of a dm sector with long range interactions . indeed , since the self - interaction needed to change their dynamics is in general of the order of the thompson scattering ( @xmath192 @xmath83 ) , from trivial dimensional analysis of such large cross sections , the following rough estimate yields that the self - interaction is of a long range type in most of the virialized astrophysical objects under the assumption that the dm - dark photon coupling is of the order of @xmath193 . in this case
, the probability to radiate a dark photon in the scattering process is different from zero and therefore it might well be possible that the dm sector is dissipative just like our sector .
the time over which energy is transferred in the system is no longer a good indicator , since the relevant quantity that describes the dynamical evolution of the system is now the cooling time : in particular , for a dm sector composed of heavy and light species , the dissipation time due to the soft emission of dark photons ( dark bremsstrahlung ) can be smaller than the age of the virialized astrophysical objects ( see e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) . in this case
, the system is no longer stable and in general it starts to collapse . by virtue of this fact
, we do not consider this last class of constraints , since a more dedicated and careful analysis also involving numerical simulations is clearly needed .
we have examined how dm - baryon long range interactions can affect the sun s sound speed radial profile . the phenomenological approach used in our analysis lets us explore all the parameter space from the contact regime to the long - range regime .
we find that dm particles lighter than 8.5 gev coupled with ordinary baryons through a kinetic mixing parameter @xmath194 bigger than @xmath178 , produce a very large impact on the sun s core in the long range regime ( @xmath24 smaller than few mev ) .
therefore they can be excluded as possible dm candidates . however , solar models for which the dm particle has a mass of @xmath179 and the mediator a mass smaller than @xmath195 improve the agreement with helioseismic data . nevertheless , when the mass of the dark photon is larger than @xmath196 the impact on the sun s structure is very small , being in this case the interaction of a contact - type ( standard spin - independent picture ) .
in particular , the results obtained here reveal that dm models featuring a long - range interaction with ordinary matter can affect the sound - speed radial profile and in turn probably solve the so - called solar abundance problem without being excluded by terrestrial experiments ( e.g. lux and xenon100 ) in which the scattering cross section is suppressed by the strong inverse dependence on @xmath190 .
the sun is in fact an ideal dm detector for the type of particle considered here , since it measures the entire nuclear recoil energy spectra in the scatterings . to summarize , in this work we have obtained two main results .
firstly , for the first time , a dm - baryon velocity dependent total cross - section has been implemented in solar simulation software .
secondly , but even more importantly , our analysis shows that dm particles with a mass of 10 gev and a long range interaction with ordinary matter mediated by a very light mediator ( below roughly a few mev ) , can have an impact on the sun s sound speed profile without violating the constraints coming from direct dm searches .
our results are valid if the parameter @xmath25 which controls the self - interaction is of the order of @xmath197 .
larger values of this parameter , as we have shown in the last paragraph of sec .
[ sec:4 ] , can dramatically reduced the effective luminosity carried by dm particles in the sun and therefore our results can not be applied in that case . furthermore , as commented upon in sec .
[ cconstraints ] , it might well be possible that a dark sector with long range forces can dissipate a relevant amount of energy through the emission of dark photons .
a dissipative dark sector is extremely interesting because it can offer a rich array of new ideas ranging from the detection of primordial dark radiation to new possibilities for dm dynamics in virialized astrophysical objects .
we would like to thank marco taoso for useful discussions .
this work was supported by grants from `` fundao para a cincia e tecnologia '' and `` fundao calouste gulbenkian ''
. this research has been supported at iap by erc project 267117 ( dark ) hosted by universit pierre et marie curie - paris 6 , and at jhu by nsf grant oia-1124403 . | assuming the existence of a primordial asymmetry in the dark sector , we study how dm - baryon long - range interactions , induced by the kinetic mixing of a new @xmath0 gauge boson and the photon , affects the evolution of the sun and in turn the sound speed profile obtained from helioseismology .
thanks to the explicit dependence on the exchanged momenta in the differential cross section ( rutherford - like scattering ) , we find that dark matter particles with a mass of @xmath1 , kinetic mixing parameter of the order of @xmath2 and a mediator with a mass smaller than a few mev improve the agreement between the best solar model and the helioseismic data without being excluded by direct detection experiments
. in particular , the lux detector will soon be able to either constrain or confirm our best fit solar model in the presence of a dark sector with long - range interactions that reconcile helioseismology with thermal neutrino results . |
let @xmath0 be an oriented surface of genus @xmath12 with @xmath17 punctures , obtained by removing @xmath13 points @xmath18 from the closed oriented surface @xmath19 of genus @xmath12 .
an _ ideal triangulation _ of @xmath0 is a triangulation of the closed surface @xmath20 whose vertex set is exactly @xmath18 .
an euler characteristic argument shows that any ideal triangulation has @xmath21 edges .
two ideal triangulations are considered the same if they are isotopic .
in addition , we require that each ideal triangulation @xmath22 is endowed with an indexing @xmath23 , @xmath24 , , @xmath25 of its edges by the index set @xmath26 .
let @xmath27 denote the set of isotopy classes of such indexed ideal triangulations @xmath22 .
the set @xmath27 admits a natural action of the group @xmath28 of permutations of @xmath29 elements , acting by permuting the indices of the edges of @xmath22 .
namely @xmath30 for @xmath31 if its @xmath32th edge @xmath33 is equal to @xmath34
. another important transformation of @xmath27 is provided by the _ @xmath32th diagonal exchange map _ @xmath35 defined as follows .
the @xmath32th edge @xmath36 of an ideal triangulation @xmath37 is adjacent to two triangles .
if these two triangles are distinct , their union forms a square @xmath38 with diagonal @xmath36
. then @xmath39 is obtained from @xmath22 by replacing the edge @xmath36 by the other diagonal @xmath40 of the square @xmath38 , as illustrated in figure [ fig : diagexch ] . by convention , @xmath41 when the two sides of @xmath36 belong to the same triangle ; this happens exactly when @xmath36 is the only edge of @xmath22 leading to a certain puncture of @xmath0 .
the reindexings and diagonal exchanges satisfy the following relations : 1 .
the _ composition relation _ : @xmath42 for every @xmath43 , @xmath44 ; 2 .
the _ reflexivity relation _ : @xmath45 ; 3 .
the _ reindexing relation _ : @xmath46 for every @xmath47 ; 4 .
the _ distant commutativity relation _ : if @xmath36 and @xmath48 do not belong to a same triangle of @xmath49 , then @xmath50 ; 5 .
the _ pentagon relation _ : if three triangles of an ideal triangulation @xmath37 form a pentagon with diagonals @xmath36 , @xmath48 , and if @xmath51 denotes the transposition exchanging @xmath32 and @xmath52 , then @xmath53 + the pentagon relation is illustrated in figure [ fig : pentagon ] . the following two results of r. penner @xcite ( although the first one may have an earlier history ) are crucial in the construction of the quantum teichmller space .
we refer to @xcite for their proof .
[ thm : penner1 ] given two ideal triangulations @xmath54 , there exists a finite sequence of ideal triangulations @xmath55 , @xmath56 ,
, @xmath57 such that each @xmath58 is obtained from @xmath59 by a diagonal exchange or by a reindexing of its edges .
[ thm : penner2 ] given two ideal triangulations @xmath54 and given two sequences @xmath55 , @xmath56 , , @xmath60 and @xmath61 , @xmath62 , , @xmath63 of diagonal exchanges and reindexings connecting them as in theorem [ thm : penner1 ] , these two sequences can be related to each other by successive applications of the following moves and of their inverses . 1 .
use the composition relation to replace
, @xmath59 , @xmath64 , @xmath65 , by , @xmath59 , @xmath66 , where @xmath43 , @xmath67 .
2 . use the reflexivity relation to replace
, @xmath59 , by , @xmath68 , @xmath69 , @xmath59 , .
3 . use the reindexing relation to replace
, @xmath59 , @xmath70 , @xmath71 , by , @xmath59 , @xmath72 , @xmath73 ,
where @xmath74 .
4 . use the distant commutativity relation to replace
, @xmath59 , by , @xmath68 , @xmath75 , @xmath76 , @xmath77 , @xmath59 , where @xmath78 are two edges which do not belong to a same triangle of @xmath59 . 5 .
use the pentagon relation to replace
, @xmath59 , by @xmath59 , @xmath75 , @xmath79 , @xmath80 , @xmath81 , @xmath82 , @xmath59 , where @xmath78 are two diagonals of a pentagon of @xmath59 .
w. thurston associated to an ideal triangulation @xmath22 a certain system of _ shear coordinates _ for the teichmller space @xmath1 .
see @xcite for the dual notion of length coordinates , and @xcite@xcite@xcite for details on these shear coordinates .
consider a complete hyperbolic metric @xmath83 .
it is well - known that the ends of the complete hyperbolic surface @xmath84 can be of two types : finite area _
cusps _ bounded on one side by a horocycle ; and infinite area _ funnels _ bounded on one side by a simple closed geodesic .
this can also be expressed in terms of the _ convex core _
@xmath85 of @xmath86 , which is the smallest non - empty closed convex subset of @xmath86 , and is bounded in @xmath0 by a family of disjoint simple closed geodesics .
the cusp ends of @xmath86 are those which are also ends of @xmath85 , while each funnel end of @xmath0 faces a boundary component of @xmath85 .
note that the interior @xmath87 is homeomorphic to @xmath0 , by the homeomorphism which is uniquely determined up to isotopy by the property that it is homotopic to the inclusion map .
it is convenient to enhance the hyperbolic metric @xmath88 with some additional data , consisting of an orientation for the boundary @xmath89 of its convex cores .
let the _ enhanced teichmller space _
@xmath90 consist of all isotopy classes of hyperbolic metrics @xmath91 enhanced with an orientation of @xmath92 . since the convex core @xmath93 depends continuously on the metric @xmath94 , the enhanced teichmller space @xmath14 inherits from the topology of @xmath1 a topology for which the natural projection @xmath95 is a branched covering map .
consider an enhanced hyperbolic metric @xmath96 together with an ideal triangulation @xmath22 .
each edge @xmath36 specifies a proper homotopy class of paths going from one end of @xmath97 to another end . this proper homotopy class
is also realized by a unique @xmath94geodesic @xmath98 such that each end of @xmath98 , either converges towards a cusp end of @xmath0 , or spirals around a component of @xmath99 in the direction specified by the enhancement of @xmath94 .
the union of @xmath99 and of the @xmath98 forms an @xmath94geodesic lamination @xmath12 contained in the convex core @xmath85 .
the enhanced hyperbolic metric @xmath100 now associates to the edge @xmath36 of @xmath22 a positive number @xmath101 defined as follows .
the geodesic @xmath98 separates two triangle components @xmath102 and @xmath103 of @xmath104 .
identify the universal covering of s , endowed with the metric @xmath94 , to the hyperbolic plane @xmath105 .
lift @xmath98 , @xmath102 and @xmath103 to a geodesic @xmath106 and two triangles @xmath107 and @xmath108 in @xmath105 so that the union @xmath109 forms a square @xmath110 in @xmath105 . in the standard upper half - space model for @xmath105 , let @xmath111 , @xmath112 , @xmath113 , @xmath114 be the vertices of @xmath110 in such a way that @xmath106 goes from @xmath111 to @xmath112 and , for this orientation of @xmath106 , @xmath113 , @xmath114 are respectively to the right and to the left of @xmath106 for the orientation of @xmath115 given by the orientation of @xmath0 .
then , @xmath116 note that @xmath101 is positive since the points @xmath111 , @xmath113 , @xmath112 , @xmath114 occur in this order in the real line bounding the upper half - space @xmath105 .
the real numbers @xmath101 are the _ exponential shear coordinates _ of the enhanced hyperbolic metric @xmath117 .
the more standard shear coordinates are just their logarithms @xmath118 , but the @xmath101 turn out to be better behaved for our purposes .
there is an inverse construction which associates a hyperbolic metric to each system of positive weights @xmath101 attached to the edges @xmath36 of the ideal triangulation @xmath22 : identify each of the components of @xmath119 to a triangle with vertices at infinity in @xmath105 , and glue these hyperbolic triangles together in such a way that adjacent triangles form a square whose vertices have cross - ration @xmath120 as above .
this defines a _ possibly incomplete _ hyperbolic metric on the surface @xmath0 .
an analysis of this metric near the ends of @xmath0 shows that its completion is a hyperbolic surface @xmath121 with geodesic boundary , and that each end of an edge of @xmath22 either spirals towards a component of @xmath122 or converges towards a cusp end of @xmath121 . extending @xmath121 to a complete hyperbolic surface without boundary and identifying this surface to @xmath0
, we now have found a complete hyperbolic metric @xmath94 on @xmath0 whose convex core is isometric to @xmath121 .
in addition , the spiralling pattern of the ends of @xmath22 provides an enhancement for the hyperbolic metric @xmath94 .
see @xcite or @xcite for details .
the @xmath101 then define a homeomorphism @xmath123 between the enhanced teichmller space @xmath14 and @xmath124 . for the enhanced hyperbolic metric @xmath96 associated to the exponential shear parameters @xmath101 , the geometry of @xmath94 near the @xmath52th puncture @xmath125
is completely determined by the _ exponential length parameter _
@xmath126 where @xmath127 is the number of end points of the edge @xmath36 that are equal to @xmath125 .
the end of @xmath0 corresdonding to @xmath125 is a cusp for @xmath94 exactly when @xmath128 .
otherwise , this end faces a boundary component of @xmath85 which is a closed geodesic of length @xmath129 , and the orientation of this boundary component defined by the enhancement of @xmath94 coincides with the boundary orientation exactly when @xmath130 .
the exponential shear coordinates associates a parametrization @xmath131 to each ideal triangulation @xmath37 ( endowed with an indexing of its edges ) .
we now investigate the coordinate changes @xmath132 associated to two ideal triangulations . if @xmath133 is obtained by reindexing the edges of @xmath22 by @xmath134 , it is immediate that @xmath132 is just the permutation of the coordinates by @xmath43 . for a diagonal exchange ,
the expression for @xmath135 depends on the possible identifications between the sides of the square where the diagonal exchange takes place .
[ prop : diagexch ] suppose that the ideal triangulations @xmath22 , @xmath136 are obtained from each other by a diagonal exchange , namely that @xmath137 .
label the edges of @xmath22 involved in this diagonal exchange as @xmath36 , @xmath48 , @xmath138 , @xmath139 , @xmath140 as in figure [ fig : diagexch ] .
if @xmath141 and @xmath142 are the exponential shear coordinates associated to the same enhanced hyperbolic metric @xmath117 , then @xmath143 for every @xmath144 , @xmath145 and : case 1 : : if the edges @xmath48 , @xmath138 , @xmath139 , @xmath140 are distinct , then @xmath146 case 2 : : if @xmath48 is identified with @xmath138 , and @xmath139 is distinct from @xmath140 , then @xmath147 case 3 : : ( the inverse of case 2 ) if @xmath48 is identified with @xmath140 , and @xmath138 is distinct from @xmath139 , then @xmath148 case 4 : : if @xmath48 is identified with @xmath139 , and @xmath138 is distinct from @xmath140 , then @xmath149 case 5 : : ( the inverse of case 4 ) if @xmath138 is identified with @xmath140 , and @xmath48 is distinct from @xmath139 , then @xmath150 case 6 : : if @xmath48 is identified with @xmath138 , and @xmath139 is identified with @xmath140 ( in which case @xmath0 is a @xmath151times punctured sphere ) , then @xmath152 case 7 : : ( the inverse of case 6 ) if @xmath48 is identified with @xmath140 , and @xmath138 is identified with @xmath139 ( in which case @xmath0 is a @xmath151times punctured sphere ) , then @xmath153 case 8 : : if @xmath48 is identified with @xmath139 , and @xmath138 is identified with @xmath140 ( in which case @xmath0 is a once punctured torus ) , then @xmath154 this immediately follows from the combinatorics of cross - ratios . for instance , let us focus on case 4 .
the other cases are similar .
isometrically identify @xmath0 , endowed with the metric @xmath94 , to the quotient of the upper half - plane model for @xmath155 under an action of the fundamental group @xmath156 .
lift the square @xmath157 bounded by @xmath48 , @xmath138 , @xmath158 and @xmath140 to a square @xmath159 bounded by lifts @xmath160 , @xmath161 , @xmath162 and @xmath163 , and with diagonal @xmath164 .
the fact that @xmath158 means that there exists a covering translation @xmath165 such that @xmath166 .
label the corners of the square @xmath110 clockwise as @xmath111 , @xmath114 , @xmath112 , @xmath113 , starting from the corner @xmath167 . then , by definition , @xmath168 and @xmath169 since the component @xmath170 goes from @xmath171 to @xmath113 .
the geodesic @xmath161 is the diagonal of a square with vertices @xmath112 , @xmath111 , @xmath114 and a fourth vertex @xmath172 outside of @xmath110 .
then @xmath173 with respect to the preimage @xmath174 of @xmath175 , @xmath176 is the diagonal of a square with vertices @xmath111 , @xmath113 , @xmath114 and @xmath172 .
consequently , @xmath177 the same argument shows that @xmath178 . for @xmath179 , there is a new twist because @xmath180 sends @xmath181 to @xmath160 , so that @xmath160 is the diagonal of a square with vertices @xmath182 , @xmath112 , @xmath183 and @xmath184 .
switching from @xmath22 to @xmath175 now changes two vertices of this square in the sense that , with respect to @xmath185 , @xmath186 is the diagonal of a square with vertices @xmath187 , @xmath114 , @xmath188 and @xmath189 .
therefore , @xmath190 and , since @xmath191 respects cross - ratios , @xmath192 the fact that @xmath143 for @xmath193 is obvious from the definition of the shear coordinates .
using theorem [ thm : penner1 ] , an immediate corollary of proposition [ prop : diagexch ] is the following .
for two ideal triangulations @xmath22 , @xmath194 , the coordinate change map @xmath195 is rational . in particular
, there is a well - defined notion of rational function on @xmath14 .
namely , the ( partially defined ) function @xmath196 is _ rational _ if , for an arbitrary ideal triangulation @xmath22 , @xmath197 is a rational function in the usual sense . if @xmath198 denotes the algebra of such rational functions , every ideal triangulation @xmath22 specifies an algebra isomorphism @xmath199 from @xmath198 to the algebra of rational fractions @xmath200 . in particular , at the rational function algebra level , two ideal triangulations @xmath22 , @xmath194 induce a coordinate change isomorphism @xmath201 defined by @xmath202 .
the goal of this section is to quantize the enhanced teichmller space @xmath203 by defining a deformation of the algebra @xmath198 , depending on a parameter @xmath204 .
fix an ideal triangulation @xmath37 .
the complement @xmath119 has @xmath205 spikes converging towards the punctures , and each spike is delimited by one @xmath36 on one side and one @xmath48 on the other side , with possibly @xmath206 . for @xmath32 , @xmath207 ,
let @xmath208 denote the number of spikes of @xmath119 which are delimited on the left by @xmath36 and on the right by @xmath48 , and set @xmath209 note that @xmath210 can only belong to the set @xmath211 , and that @xmath212 .
it turns out that the coefficients @xmath210 are related to the expression of the weil - petersson symplectic form of @xmath203 in the exponential shear coordinates , but this plays only a historical rle in this paper .
the _ chekhov - fock algebra _ associated to the ideal triangulation @xmath22 is the algebra @xmath213 defined by generators @xmath214 , @xmath215 , @xmath216 , @xmath217 ,
, @xmath218 , @xmath219 , with each pair @xmath220 associated to an edge @xmath36 of @xmath22 , and by the relations @xmath221 this is an iterated skew laurent polynomial algebra . in particular , it is a noetherian ring and a right ore domain , so that we can introduce its _ fraction division algebra _ @xmath222 .
see for instance @xcite .
the algebra @xmath222 consists of all formal fractions @xmath223 with @xmath224 , @xmath225 and @xmath226 , and two such fractions @xmath227 and @xmath228 are identified if there exists @xmath229 , @xmath230 such that @xmath231 and @xmath232 . in practice ,
the chekhov - fock algebra @xmath233 consists of all laurent polynomials in variables @xmath214 , @xmath216 , , @xmath218 satisfying the skew - commutativity relations ( [ eqn : skewcom ] ) .
its fraction division algebra @xmath222 consists of all rational fractions in the variables @xmath214 , @xmath216 , , @xmath218 satisfying the same skew - commutativity relations .
in particular , when @xmath234 , @xmath235 and @xmath236 respectively coincide with the laurent polynomial algebra @xmath237 $ ] and the rational fraction algebra @xmath238 . the general @xmath233 and @xmath222 can be considered as deformations of this case .
the algebras @xmath233 and @xmath222 strongly depend on the ideal triangulation @xmath22 . to define a triangulation independent deformation of the algebra @xmath239
, we need to generalize the coordinate change isomorphism of ( [ eqn : nonquantcoordch ] ) to this non - commutative context , by introducing appropriate algebra isomorphisms @xmath240 .
there is no geometry to guide us here , so we will do this stepwise , using theorems [ thm : penner1 ] and [ thm : penner2 ] . to ease the exposition , we will denote by @xmath241 , @xmath242 ,
, @xmath243 the generators of @xmath244 associated to the edges @xmath245 , @xmath246 , ,
@xmath247 of @xmath175 , and by @xmath214 , @xmath216 , , @xmath218 the generators of @xmath222 associated to the edges @xmath23 , @xmath24 , , @xmath25 of @xmath22 .
in particular , the isomorphism @xmath248 will be completely determined once we specify the images @xmath249 .
[ prop : quantumdiagexch ] suppose that the ideal triangulations @xmath22 , @xmath136 are obtained from each other by a diagonal exchange , namely that @xmath137 .
label the edges of @xmath22 involved in this diagonal exchange as @xmath36 , @xmath48 , @xmath138 , @xmath139 , @xmath140 as in figure [ fig : diagexch ] .
then there is a unique algebra isomorphism @xmath250 such that @xmath251 for every @xmath252 , @xmath253 and : case 1 : : if the edges @xmath48 , @xmath138 , @xmath139 , @xmath140 are distinct , then @xmath254 case 2 : : if @xmath48 is identified with @xmath138 , and @xmath139 is distinct from @xmath140 , then @xmath255 case 3 : : ( the inverse of case 2 ) if @xmath48 is identified with @xmath140 , and @xmath138 is distinct from @xmath139 , then @xmath256 case 4 : : if @xmath48 is identified with @xmath139 , and @xmath138 is distinct from @xmath140 , then @xmath257 case 5 : : ( the inverse of case 4 ) if @xmath138 is identified with @xmath140 , and @xmath48 is distinct from @xmath139 , then @xmath258 case 6 : : if @xmath48 is identified with @xmath138 , and @xmath139 is identified with @xmath140 ( in which case @xmath0 is a @xmath151times punctured sphere ) , then @xmath259 case 7 : : ( the inverse of case 6 ) if @xmath48 is identified with @xmath140 , and @xmath138 is identified with @xmath139 ( in which case @xmath0 is a @xmath151times punctured sphere ) , then @xmath260 case 8 : : if @xmath48 is identified with @xmath139 , and @xmath138 is identified with @xmath140 ( in which case @xmath0 is a once punctured torus ) , then @xmath261 by inspection , these formulas are compatible with the skew - commutativity relations ( [ eqn : skewcom ] ) . for instance , in case 4 , @xmath262 , @xmath263 and @xmath264 in @xmath265 , and @xmath266 \bigl [ ( 1+qx_i)(1+q^3x_i ) x_j \bigr ] = q^4\bigl [ ( 1+qx_i)(1+q^3x_i ) x_j \bigr ]
\bigl [ x_i^{-1 } \bigr]\\ \bigl [ x_i^{-1 } \bigr ] \bigl [ ( 1+qx_i^{-1})^{-1}x_k \bigr ] = q^{-2}\bigl [ ( 1+qx_i^{-1})^{-1}x_k \bigr ]
\bigl [ x_i^{-1 } \bigr]\\ \bigl [ ( 1+qx_i)(1+q^3x_i ) x_j \bigr ] \bigl [ ( 1+qx_i^{-1})^{-1}x_k \bigr ] \qquad\qquad\qquad\qquad\qquad\qquad \\ \qquad\qquad\qquad \qquad\qquad\qquad = q^2\bigl [ ( 1+qx_i^{-1})^{-1}x_k \bigr ] \bigl [ ( 1+qx_i)(1+q^3x_i ) x_j \bigr]\end{gathered}\ ] ] in @xmath222 since @xmath267 , @xmath268 and @xmath269 .
it follows that these formulas extend to a unique algebra homomorphism @xmath270 .
extending @xmath271 to the fraction division algebra @xmath244 will require a little care .
indeed , for a formal fraction @xmath272 with @xmath224 , @xmath273 , we want to define @xmath274 . for this
we need @xmath275 to be non - zero . in other words , we need to show that @xmath271 is injective . for this , we take advantage of the reflexivity of the above formulas . exchanging the rles of @xmath22 and @xmath175
, we also have an algebra homomorphism @xmath276 .
note that the elements of @xmath277 are all polynomials in the @xmath278,with @xmath279 , 2 , , @xmath29 and in the quantity @xmath280 .
after checking the skew - commutativity relations and because @xmath281 , we can therefore extend @xmath282 to @xmath283 by defining @xmath284 .
we now have a composition @xmath285 of two algebra homomorphisms . by inspection of the formulas defining these homomorphisms , @xmath286 for every generator @xmath287 of @xmath288 .
it follows that @xmath289 is the inclusion map . in particular ,
it is injective , which proves that the first homomorphism @xmath270 is injective .
as indicated before , this enables us to extend @xmath270 to a map @xmath248 by setting @xmath290 if @xmath224 , @xmath273 .
it can be shown that this map @xmath291 is an algebra homomorphism .
see ( * ? ? ?
* chap . 9 ) .
a symmetric argument provides an algebra homomorphism @xmath292 . from the fact that @xmath293 for every generator @xmath287 of @xmath288 , we conclude that @xmath294 is the identity . by symmetry
, it follows that @xmath295 is an isomorphism , with inverse @xmath296 .
when @xmath175 is obtained from @xmath22 by an edge reindexing , we also have the immediate relation .
[ prop : quantumreindex ] suppose that the ideal triangulations @xmath22 , @xmath136 are obtained from each other by an edge reindexing , namely that @xmath297 for some permutation @xmath298 .
then there exists a unique isomorphism @xmath299 such that @xmath300 for every generator @xmath301 of @xmath244 .
the following facts are straightforward from the definitions .
[ prop : quantumrelations ] the isomorphisms @xmath248 defined by propositions [ prop : quantumdiagexch ] and [ prop : quantumreindex ] satisfy the following properties . 1 . composition relation : @xmath302 for every @xmath43 , @xmath303 .
reflexivity relation : @xmath304 3 . reindexing relation : @xmath305 for every @xmath306 .
distant commutativity relation : @xmath307 if the edges @xmath36 , @xmath48 of @xmath22 do not belong to a same triangle . dealing with
the pentagon relation will require more efforts .
the goal of this section is to show that the isomorphisms @xmath291 constructed in the previous section are compatible with the pentagon relation satisfied by the diagonal exchanges @xmath308 .
the main tool to achieve this is borrowed from @xcite .
it consists of a certain order 5 automorphism of the quantum torus .
the quantum torus @xmath309 is the algebra defined by the generators @xmath310 , @xmath311 and by the relation @xmath312 .
set @xmath313 , @xmath314 and inductively define @xmath315 from this definition we deduce the following relations .
@xmath316 this provides us with a few combinatorial tricks to deal with noncommutativity .
one of them is that the products @xmath317 and @xmath318 both commute with @xmath319 .
also , the two equations of ( [ eqn:1+qu ] ) deform to : @xmath320 what we have accomplished is decomposing the polynomial factors @xmath321 into monomials .
this decomposition will turn out to be very convenient later on .
[ lem : order5 ] @xmath322 for every @xmath323 .
this immediately follows from iterated applications of ( [ eqn : cycle ] ) : @xmath324 and @xmath325 -- consider a pentagon cycle of geodesic laminations @xmath326 , @xmath327 , @xmath328 , @xmath329 , @xmath330 , @xmath331 as in figure [ fig : pentagon ] .
we want to show that @xmath332 where the @xmath333 are defined by proposition [ prop : quantumdiagexch ] , and @xmath334 by proposition [ prop : quantumreindex ] .
let @xmath335 , @xmath336 , @xmath337 , @xmath338 , @xmath339 denote the sides of the pentagon , and let @xmath36 and @xmath48 be the diagonal edges of @xmath326 , all labelled as in figure [ fig : pentagon ] .
note that the @xmath335 , @xmath336 , @xmath337 , @xmath338 , @xmath339 are edges of all the @xmath59 , and in particular are associated to generators @xmath340 , @xmath341 , @xmath342 , @xmath343 , @xmath344 of @xmath345 .
let @xmath346 , @xmath347 , @xmath348 , @xmath349 , @xmath350 denote the respective images of @xmath340 , @xmath341 , @xmath342 , @xmath343 , @xmath351 under @xmath352 .
we do the same thing with diagonals , but with an additional twist .
the diagonals of the pentagon in @xmath59 are always its @xmath32th and @xmath52th edges , but their relative configuration with respect to each other alternates . in particular
, the factor @xmath353 occurring in the skew - commutativity relations ( [ eqn : skewcom ] ) is equal to @xmath354 .
for this reason , we let @xmath319 and @xmath355 be the respective images of @xmath356 and @xmath357 under @xmath352 , where @xmath358 is the transposition exchanging @xmath32 and @xmath52 .
this is specially designed so that @xmath359 for every @xmath360 .
in addition , as one moves from @xmath59 to @xmath58 , the diagonal exchange is always performed on the edge corresponding to @xmath319 . from the definition of @xmath361 ,
the elements @xmath319 , @xmath362 satisfy the induction relation ( [ eqn : cycle ] ) .
in particular , it follows from lemma [ lem : order5 ] that @xmath322 and @xmath363 .
the induction formulas for the other elements @xmath346 , @xmath347 , @xmath348 , @xmath349 , @xmath364 depend on whether there are identifications between the sides of the pentagon .
we first consider the case of an _ embedded _
polygon , with no identification between its sides .
[ prop : chefock ] the pentagon relation @xmath332 is satisfied in the case of an embedded pentagon .
set @xmath365 to simplify the notation .
we need to show that @xmath366 for every @xmath360 .
the property is immediate for those @xmath367 which correspond to edges outside of the pentagon . by definition of the @xmath319 , @xmath362 and by lemma [ lem : order5 ] , @xmath368 and @xmath369 .
this proves the property for @xmath370 and @xmath371 . for the sides of the pentagon , @xmath372 , @xmath373 , @xmath374 , @xmath375 and @xmath376 . from the definition of @xmath377 in proposition [ prop : quantumdiagexch ] , @xmath378 using ( [ decomp2 ] )
, @xmath379 combining this relation with @xmath380 , we get that @xmath381 , @xmath382 , @xmath383 , @xmath384 and @xmath385 as required .
we now have to worry about possible identifications between the sides of the pentagon .
note that it suffices to prove the property for any cyclic permutation of the @xmath59 .
this reduces the analysis to 6 possible cases . 1 . no identification ( embedded pentagon ) ; 2 .
@xmath386 , and the other sides are distinct ; 3 .
@xmath387 , and the other sides are distinct ; 4 . @xmath386 and @xmath388 ; 5 .
@xmath386 and @xmath389 ; 6 .
@xmath387 and @xmath390 .
we already considered case 1 . in the other cases ,
note that the identifications between the sides of the pentagon have no impact on the images of @xmath370 , @xmath371 .
we therefore only need to consider @xmath340 , @xmath341 , @xmath342 , @xmath343 , @xmath391 .
* case 2 * : @xmath392 , and the other sides are distinct . using ( [ eqn : threeterm ] ) and ( [ eqn:1+qu ] ) , @xmath393 the argument for @xmath342 , @xmath343 , @xmath391 is identical to that of proposition [ prop : chefock ] .
* case 3 * : @xmath394 , and the other sides are distinct . using ( [ eqn : threeterm ] ) and ( [ eqn : commutation ] ) , @xmath395 and @xmath396 then @xmath397 it follows that @xmath398 .
the argument for @xmath341 , @xmath343 , @xmath391 is identical to that for case 1 .
* case 4 * : @xmath392 and @xmath388 .
the fact that @xmath399 is proved as in case 2 .
shifting indices by 2 and rotating the picture , the argument of case 2 gives @xmath400 .
we can then backtrack to @xmath401 by using the 5periodicity of the @xmath319 .
this proves that @xmath402 . the remaining generator @xmath391 is treated as in case 1 .
* case 5 * : @xmath392 and @xmath403 .
again @xmath399 as in case 2 , and @xmath404 as in case 1 .
the fact that @xmath402 follows from case 3 after shifting indices by 2 as above .
* case 6 * : @xmath394 and @xmath405 .
this again follows from a combination of the arguments of cases 1 , 2 and 3 .
in conclusion , we have proved : [ prop : pentagon ] the pentagon relation @xmath332 is satisfied in all cases .
we can now state and prove the main result of this paper .
[ thm : main ] there is a unique family of algebra isomorphisms @xmath299 defined as @xmath22 , @xmath194 ranges over all pairs of ideal triangulations , such that : 1 .
@xmath406 for every @xmath22 , @xmath175 , @xmath407 ; 2 . @xmath291 is the isomorphism of proposition [ prop : quantumdiagexch ] when @xmath175 is obtained from @xmath22 by a diagonal exchange ; 3 .
@xmath291 is the isomorphism of proposition [ prop : quantumreindex ] when @xmath175 is obtained from @xmath22 by an edge reindexing .
use theorem [ thm : penner1 ] to connect @xmath22 to @xmath175 by a sequence @xmath55 , @xmath56 , ,
@xmath57 where each @xmath58 is obtained from @xmath59 by a diagonal exchange or by an edge reindexing , and define @xmath291 as the composition of the @xmath408 provided by propositions [ prop : quantumdiagexch ] and [ prop : quantumreindex ] .
theorem [ thm : penner2 ] and propositions [ prop : quantumrelations ] and [ prop : pentagon ] show that this @xmath408 is independent of the choice of the sequence of @xmath59 .
the uniqueness immediately follows from theorem [ thm : penner1 ] .
the _ quantum ( enhanced ) teichmller space _ of @xmath0 can now be defined as the algebra @xmath409 where the relation @xmath410 is defined by the property that , for @xmath411 and @xmath412 , @xmath413 note that the definition is specially designed so that , when @xmath234 , there is a natural isomorphism between @xmath414 and the algebra @xmath198 of rational functions on the enhanced teichmller space @xmath14 .
the _ cusped teichmller space _ @xmath415 is the set of isotopy classes of hyperbolic metrics on @xmath0 for which all ends are of cusp type .
since such a metric admits a unique enhancement , @xmath415 is a natural subspace of the enhanced teichmller space @xmath14 . consider the thurston parametrization @xmath123 of @xmath14 by the shear coordinates @xmath101 associated to the edges @xmath36 of an ideal triangulation @xmath416 .
recall that we associated to the @xmath52th puncture @xmath125 of @xmath0 the exponential length parameter @xmath126 where @xmath127 is the number of end points of the edge @xmath36 that are equal to @xmath125 .
we observed in
[ sect : shear ] that @xmath417 corresponds under @xmath418 to the set of those @xmath419 such that @xmath128 for every @xmath420 , , @xmath13 .
it is also natural to consider the product @xmath421 .
note that @xmath422 since every edge @xmath36 has two end points ( so that @xmath423 for every @xmath32 ) .
in particular , @xmath424 is identically 1 on the image of @xmath425 under @xmath418 since the @xmath101 are all real positive .
it can be shown that there exists an isomorphism @xmath429 sending the first @xmath430 elements @xmath431 , @xmath432 , , @xmath433 to @xmath434 , @xmath435 , , @xmath436 and @xmath437 to @xmath438 , respectively .
see for instance @xcite .
it follows that the isomorphism @xmath439 defined by @xmath418 induces an isomorphism @xmath440 where @xmath441 is the ideal generated by the elements @xmath442 and @xmath443 .
note that it is important to include @xmath443 to make sure that @xmath441 is prime .
it suffices to verify this for the transposition @xmath448 .
exchanging @xmath360 and @xmath449 in the product @xmath450 of @xmath451 results in multiplying this element by @xmath452 .
however , in the exponent @xmath453 of the @xmath204factor , @xmath454 is replaced by @xmath455 , so that this exponent increases by @xmath456 @xmath457 .
consequently , the two contributions cancel out and @xmath451 remains invariant .
the argument is similar for @xmath437 .
the @xmath204factor in the definition of @xmath437 and the @xmath451 is traditionally known as the quantum ordering in the physics literature .
it is better explained in a situation when we can write @xmath458 , where the @xmath459 are symbols with central commutators @xmath460 .
it then follows from the campbell - hausdorff formula that @xmath461 if @xmath462 . in particular , this makes the invariance under reindexing immediate .
we will verify the property case - by - case , according to the type of the diagonal exchange . to help distinguishing the quantities associated to @xmath22 from those associated to @xmath175
, we will label with primes any data associated to @xmath175 .
thus , @xmath467 is generated by the @xmath301 , and @xmath468 is the element associated to the @xmath52th puncture .
the diagonal exchange specifies two distinct triangle components of @xmath119 .
let @xmath469 be the closure of these two triangles in @xmath0 .
abstractly , @xmath469 is obtained from the square @xmath38 where the diagonal exchange takes place by identifying some of its sides according to the case we are considering .
note that distinct vertices of @xmath470 can correspond to the same puncture @xmath125 of @xmath0 . to simplify the notation ( by saving the letter @xmath52 ) , we focus attention on the element @xmath431 associated to the first puncture @xmath471 . since ( the argument of ) proposition [ prop : invarreindex ] guarantees that the order in which we write its generators does not matter , we can write @xmath472 where @xmath473 , @xmath474 ,
@xmath475 occur in this order as one goes counterclockwise around the puncture @xmath471 .
similarly , @xmath476 . after a possible cyclic permutation of the @xmath477 , those touching the vertex @xmath478 of @xmath470 correspond to an interval @xmath479 .
let the _ contribution _ of @xmath478 to @xmath431 be @xmath480 the contribution @xmath481 of @xmath478 to @xmath482 is similarly defined .
the element @xmath431 may contain the contribution of several vertices of @xmath470 .
it is immediate that the part of @xmath431 which does not come from these contributions is invariant under the diagonal exchange .
therefore , we only need to show that @xmath291 sends @xmath483 to @xmath484 for every vertex @xmath478 of @xmath470 .
because of the symmetries of the figure and of the fact that this case is its own inverse , it suffices to show this for an arbitrary corner of @xmath485 , for instance the upper right corner .
if @xmath478 contributes to @xmath431 ( and @xmath482 ) , @xmath486 it follows that @xmath487 in these cases , @xmath470 has three distinct vertices .
the first one corresponds to the upper left and lower right vertices of @xmath38 .
if this vertex @xmath478 contributes to @xmath431 , @xmath488 then , @xmath489 there are two vertices in @xmath470 .
this first one corresponds to the two left corners of @xmath38 .
if this vertex contributes to @xmath431 , @xmath492 then , @xmath493 the argument is similar for the second vertex of @xmath470 , corresponding to the right corners of @xmath38 .
proposition [ prop : qandhinvariant ] shows that @xmath437 and the @xmath451 are well - defined elements of the quantum enhanced teichmller space @xmath495 .
it is not too hard to see that these elements are central .
it is proved in @xcite that they generate the center of @xmath495 . by analogy with the non - quantum case
, we define the _ quantum cusped teichmller space _ @xmath496 as the quotient algebra @xmath497 where @xmath441 is the 2sided ideal generated by @xmath498 and the @xmath499 .
as indicated at the beginning of this section , the algebra @xmath500 on the cusped teichmller space @xmath415 is naturally isomorphic to @xmath501 when @xmath234 . in @xcite , we define invariants of a diffeomorphism @xmath502 of @xmath0 by considering certain representations @xmath503 of the polynomial core ( defined in that paper ) of the quantum enhanced teichmller space @xmath504 .
these representations are associated to the ( unique ) hyperbolic metric on the mapping torus @xmath505 , where @xmath410 identifies @xmath506 to @xmath507 . in particular ,
a property of @xmath503 is that @xmath508 and @xmath509 .
in other words , this representation @xmath503 is actually a representation of the polynomial core of the quantum cusped teichmller space . | we consider the quantum teichmller space of the punctured surface introduced by chekhov - fock - kashaev , and formalize it as a noncommutative deformation of the space of algebraic functions on the teichmller space of the surface . in order to apply it in 3-dimensional topology , we put more attention to the details involving small surfaces .
let @xmath0 be an oriented surface of finite topological type , with at least one puncture .
a quantization of the teichmller space @xmath1 of @xmath0 was developed by l. chekhov and v. fock @xcite and , independently , by r. kashaev @xcite ( see also @xcite ) as an approach to quantum gravity in @xmath2 dimensions .
this is a deformation of the @xmath3algebra of functions on the usual teichmller space @xmath1 of @xmath0 , depending on a parameter @xmath4 , in such a way that the linearization of this deformation at @xmath5 corresponds to the weil - petersson poisson structure on @xmath1 . in this paper , we develop a slightly different version of this quantization , which has a more algebraic flavor .
it essentially is the image under the exponential map of the quantization of chekhov - fock - kashaev .
the original quantization was expressed in terms of self - adjoint operators on hilbert spaces and made strong use of the holomorphic function @xmath6 called the quantum dilogarithm function . our exponential version enables us to leave the realm of analysis and to focus on the algebraic aspects of the construction . from a mathematical point of view
, the main benefit of the quantum teichmller space @xmath7 which we construct here is that it admits a rich finite - dimensional representation theory .
this representation theory is investigated in @xcite , where we show that it is strongly connected to 3dimensional hyperbolic geometry through the space of representations @xmath8 .
in particular , we use this connection to construct in @xcite quantum invariants for diffeomorphisms of @xmath0 .
the present paper is devoted to laying down the foundations of the theory , by showing that the quantum teichmller space @xmath7 is well - defined as an object in non - commutative algebraic geometry .
our construction parallels that of chekhov and fock .
we start from thurston s exponential shear coordinates for teichmller space , associated to an ideal triangulation of @xmath0 .
a fundamental property of these exponential shear coordinates is that , as one shifts from one ideal triangulation to another , the corresponding coordinate changes are rational . in particular , there is a well - defined notion of rational function on @xmath1 . in general ,
the quantization of a space is a deformation of the algebra of continuous functions on this space .
here we construct a deformation of the algebra @xmath9 of all rational functions on @xmath1 .
the main strategy is , first to define a deformation of the expression of @xmath9 in the set of shear coordinates associated to an ideal triangulation , and then to construct appropriate `` coordinate change isomorphisms '' to make the construction independent of any choice of ideal triangulation .
the technical challenge is to check that these coordinate change isomorphisms are compatible with each other , which is accomplished in [ sect : chefock]-[sect : quantumteich ] .
many aspects of this construction are already implicitly or explicitly present in the work of chekhov - fock .
in particular , the algebraic coordinate change isomorphisms originate from the symmetries of the quantum dilogarithm function @xmath10 .
our main contribution is to systematically develop the theory from an algebraic point of view , and to carefully check the details of the construction . from a technical point of view
, this leads us to be a little more careful with non - embedded diagonal exchanges , and to explicitly describe the formulas which occur in these cases .
so far , we have avoided defining the teichmller space of the surface @xmath0 .
the reason is that there are several competing definitions , leading to different objects .
one possibility is to define the teichmller space @xmath1 as the space of isotopy classes of complete hyperbolic metrics on @xmath0 , possibly with infinite area .
this is a manifold - with - corners , of dimension @xmath11 where @xmath12 is the genus of @xmath0 and @xmath13 is the number of its punctures .
this space can be expanded to the _ enhanced teichmller space _ @xmath14 defined in [ sect : shear ] , which has the same space of rational functions as @xmath1 .
this is the set - up of @xcite and of the first part of this article .
however , a more commonly used object is the _ cusped teichmller space _ @xmath15 , consisting of all isotopy classes of finite area complete hyperbolic metrics on @xmath0 ; this is a manifold of dimension @xmath16 .
we construct a quantization of this cusped teichmller space in the last section [ sect : quantumcusped ] of the article .
there is one property which we would like to mention before closing this introduction .
the original definition of the quantum teichmller space was grounded in the geometry of hyperbolic metrics on @xmath0 . however
, a recent result of hua bai @xcite shows that it is intrinsically tied to the combinatorics of the harer - penner simplicial complex of ideal triangulations @xcite .
indeed , bai proved that the coordinate change isomorphisms considered in
[ sect : chefock ] are the only ones which satisfy a certain number of natural conditions .
this emphasizes the algebraic and combinatorial nature of the quantum teichmller space .
acknowledgements : we would like to thank francis bonahon , leonid chekhov and bob penner for their help and encouragements . |
for many materials , the density functional theory ( dft)@xcite in the local - spin - density approximation ( lsda)@xcite provides a good description of their ground - state properties .
however , problems arise when the dft - lsda approach is applied to materials with ions that contain incomplete @xmath4- or @xmath5-shells , such as transition - metal oxides or heavy fermion systems . for example , most transition - metal oxides are wide - gap antiferromagnetic insulators,@xcite and the dft - lsda predicts them to be either metals ( feo and coo ) or small - gap semiconductors ( mno and nio).@xcite the failure of the dft - lsda can be traced to the mean - field character of the kohn - sham equations as well as to the poor description of strong correlation effects within the homogeneous electron gas .
the strong correlation effects are responsible for the breakdown of the dft - lsda description of the electronic structure of these compounds . in order to provide a better description of these effects ,
the mott - hubbard picture has been introduced.@xcite in the mott - hubbard picture of nio , the @xmath4-@xmath4 coulomb interaction splits the ni @xmath4 sub - bands into the so - called lower and upper hubbard bands .
the upper hubbard band has mostly ni @xmath6 character , while the top of the valence band is of @xmath7 character , leading to a mott - hubbard gap of @xmath4-@xmath4 type .
however , o @xmath8 x - ray absorption@xcite as well as x - ray photoemission and bremsstrahlung isochromat spectroscopies@xcite on li@xmath9ni@xmath10o have shown that the additional hole has mainly oxygen character .
in contrast to the mott - hubbard model , the energy band gap caused by the ni @xmath11 correlations is therefore of the charge - transfer type between the occupied oxygen @xmath3 and the ni @xmath11 empty states . on the other hand , localized approaches , @xcite such as the local cluster scheme based on the configuration interaction method or the anderson impurity model , in which transition - metal ions are treated like an impurity in an oxygen @xmath3 host ,
predict a well - defined band gap of 5.0 ev .
however , the oxygen @xmath3 band dispersion observed in angle - resolved photoemission spectroscopy@xcite can not be described by these methods because the lattice effects are neglected .
several attempts have been made to include the missing correlation effects in dft - lsda .
the generalized gradient approximation ( gga),@xcite which takes into account the radial and angular gradient corrections , can only open a small band gap.@xcite the self - interaction correction ( sic)@xcite eliminates the spurious interaction of an electron with itself from the conventional dft - lsda method . compared to lsda , the band gap and the magnetic moments are significantly increased . however , the band gap still is too small , and the sic - lsda method predicts a larger energy band gap for nio than for feo and coo , in contradiction to experiment.@xcite the crystal - field orbital polarization introduced by norman @xcite to determine the magnetism and insulating band gap of transition - metal oxides is promising but underestimates both the spin magnetic moment and the band gap
. another promising approach for correlated materials is the so - called local density approximation ( lda ) plus the multi - orbital mean - field hubbard model ( lda+u)@xcite which includes the on - site coulomb interaction in the lsda hamiltonian . after adding the on - site coulomb interaction to the lsda hamiltonian , the potential becomes spin- _ and _ orbital - dependent .
because a larger energy cost is associated with fluctuations of the @xmath4-occupancy , the orbital - dependent potential reduces the fluctuations of the @xmath4-occupancy , resulting in a better justification of a mean - field approach .
lda+u , although it is a mean - field approach , has the advantage of describing both the chemical bonding and the electron - electron interaction .
the question regarding the best value for the coulomb repulsion parameter @xmath0 is , however , still under debate .
the @xmath0 parameter for nio obtained from a constrained lda calculation is about 7 to 8 ev , and this is the value generally used in lda+u calculations .
a similar value of @xmath0 has been obtained from a constrained lda calculation for bulk fe,@xcite even though a much smaller value had been expected because of the metallic screening in fe .
the authors argued@xcite that the higher value could be an artifact due to the poor screening within the atomic sphere approximation ( asa ) , and that within a full - potential calculation a much smaller value of less than 4 ev would be expected .
in contrast , an unpublished full - potential linear muffin - tin orbital method calculation by alouani and wills@xcite clearly shows that the value of @xmath0 for bulk fe is even slightly larger than the asa value .
furthermore , bulut , scalapino , and white@xcite showed that the renormalization of the coulomb interaction depends on the type of model used .
as lda is not a diagrammatic method , it is not known which type of renormalization is the most appropriate for the lda+u model . in this work we shed light on this problem by treating the coulomb repulsion parameter @xmath0 as adjustable parameter , and by investigating how the electronic and optical properties depend on its value .
we show that for an intermediate value of @xmath12 ev , good agreement with the measured ground - state antiferromagnetic magnetic moment and optical properties is obtained .
we also show that the o @xmath3 character near the top of the valence states is enhanced for a larger value of @xmath0 .
our calculation seems to indicate that the nature of the band gap at intermediate @xmath0 is a mixture of charge transfer and mott - hubbard type , and that it becomes almost purely of the charge - transfer type for higher values of @xmath0 .
our calculations are based on the projector augmented wave ( paw ) method,@xcite an efficient all - electron method without shape approximations on the potential or electron density to avoid uncertainties due to the asa approach .
based on a car - parrinello - like formalism,@xcite the paw method allows complex relaxations and dynamical properties in strongly correlated systems to be studied .
our implementation of lda+u within the paw method is described in detail .
furthermore we discuss possible extensions of the existing method that will enhance its applicability .
the paper is organized as follows : in section [ sec : two ] we present those aspects of the paw formalism that are needed for the implementation of the lda+u method . in section [ sec : three ] we present and discuss the lsda and lda+u ground - state properties of nio , and in section [ sec : four ] we study the optical properties of nio , namely the imaginary part of the dielectric function , and compare the results to experiment .
the paw method developed by one of us @xcite combines ideas of the pseudo - potential ( pp ) and the linear augmented plane wave ( lapw ) methods .
it is applicable to all elements of the periodic table .
the nodal behavior of the wave function is correctly described and , as in the pp method , the forces on the ions are easily expressed . in the paw method ,
the all - electron ( ae ) crystal wave function is constructed from a pseudo ( ps ) wave function and atom - like functions localized near the nuclei .
the ps wave function @xmath13 coincides with the crystal ae wave function @xmath14 in the interstitial region , i.e. outside the atomic regions . inside the atomic regions
@xmath15 , called augmentation regions , the wave function is almost atom - like because the effect of the surrounding crystal is small .
therefore , a natural choice is to use solutions @xmath16 of schrdinger s equation for the isolated atom , the so - called ae partial waves , as a basis set for the augmentation region . here
@xmath17 is a global index for the atom @xmath18 , the angular momentum @xmath19 , the magnetic quantum number @xmath20 , and the index @xmath21 , the energy for which schrdinger s equation is solved . to link the expansion in atom - like functions near the nuclei to the ps wave function
, we introduce a set of auxiliary functions @xmath22 , so - called ps partial waves , which are centered on the atom and coincide per construction with the corresponding ae partial waves @xmath16 outside their augmentation regions : @xmath23 the coefficients @xmath24 of the expansions in ae and ps partial waves are chosen such that the ps partial wave expansion latexmath:[$\sum_\lambda @xmath13 inside the augmentation region . for this purpose
we introduce so - called projector functions @xmath26 such that @xmath27 and therefore @xmath28 for the hilbert space spanned by the ps partial waves @xmath22 .
thus we identify the expansion coefficients with @xmath29 . equation ( [ eq : pawident ] ) results in the biorthogonality condition @xmath30 for the projector functions , which moreover are chosen to be localized within the corresponding augmentation region . with these conditions
, the ae bloch wave function @xmath31 can be obtained from the ps wave function @xmath32 as @xmath33 \langle\tilde{p}_\lambda|\tilde\psi\rangle \ , .
\label{eq_psi_paw}\ ] ] the first term represents the ps wave function defined over the entire space , which is equal to the ae wave function in the interstitial region , and which is expanded in plane waves .
the second term is the ae partial wave expansion , which describes the correct nodal behavior of the wave function in the augmentation region @xmath34
( @xmath35 ) .
the third term eliminates the spurious contribution of the ps wave function in the augmentation region .
note that eq .
( [ eq : psexpand ] ) holds only approximately if the set of ps partial waves is not entirely within the augmentation regions .
however , this has the advantage that only those contributions of the ps wave function will be removed that are also replaced by ae partial waves . as a result
, the ae wave function converges rapidly with the number of partial waves used and , moreover , it is continuous and differentiable for every truncation of the partial - wave expansion .
expectation values of any sufficiently local operator @xmath36 are obtained as @xmath37 note that the double sum is diagonal in the site indices @xmath38 .
this equation is exact for a complete set of ps partial waves and rapidly attains the converged result if incomplete .
the paw method provides the freedom to represent a zero operator in the form @xmath39 by any operator @xmath40 entirely localized within the augmentation region .
equation ( [ eq : pawzero ] ) has the same range of validity as eq.([eq : pawoperator ] ) does and allows a further acceleration of the convergence by using a well - chosen operator @xmath40 and adding the corresponding zero operator to the expression for the expectation value . for transition - metal oxides ,
the @xmath4-orbitals are well localized and keep a strong atom - like character .
even though lda provides a good approximation for the average coulomb energy of the @xmath4-@xmath4 interactions , it fails to describe correctly the strong coulomb and exchange interaction between electrons in the same @xmath4-shell .
the main intention of lda+u is to identify these atomic orbitals and to describe their electronic interactions as strongly correlated states .
the other orbitals are delocalized and considered to be properly described by the lda .
the procedure is to eliminate the averaged lda energy contribution of these atom - like orbitals from the lda total energy functional @xmath41 , and to add an orbital- and spin - dependent correction .
the total energy within the lda+u method then has the form @xmath42 \ , , \label{eq : etot}\end{aligned}\ ] ] where @xmath43 is the average occupation of the @xmath4-shell for each spin direction as obtained from the @xmath4-orbital occupancies @xmath44 .
@xmath0 and @xmath45 are the coulomb self - energy and the exchange parameter , respectively .
the expressions @xmath46 are the four - center matrix elements of the screened coulomb interaction @xmath47 .
an additional requirement of the lda+u approach is that the additional energy is applied only to the valence electrons , which are re - optimized while constrained to remain orthogonal to the core states .
the rationale behind the definition of the orbital occupations is to project the density matrix onto one consisting of a restricted set of localized orbitals @xmath48 , where each orbital index @xmath49 stands for an angular momentum and @xmath18 for an atomic site .
here we focus in particular on the @xmath4-orbitals of ni .
the projected density matrix operator has the form @xmath50 the density matrix elements @xmath51 in the localized basis set can be obtained from the condition that the expectation values of a set of projection operators @xmath52 are optimally reproduced by the localized basis set , i.e. the weighted square deviation @xmath53 ^ 2\end{aligned}\ ] ] of the projections should be minimized .
here we choose the projection operators acting on the ni @xmath4-orbitals in analogy to previous implementations@xcite as @xmath54 where the site index @xmath18 refers to a particular ni site , and @xmath55 is the spherical harmonic centered at site @xmath18 .
the step functions @xmath56 are unity for @xmath57 and zero otherwise .
@xmath58 are the spherical harmonics for @xmath59 .
( note that @xmath60 . )
the radius @xmath61 for ni corresponds to an atomic sphere radius , and we have chosen @xmath62 a@xmath63 .
the localized orbitals @xmath64 have been chosen to be identical to those of the spheridized , non - spin - polarized atoms .
we used an atom in the @xmath65 configuration of ni .
for such a localized ni @xmath4-orbital @xmath64 with magnetic quantum number @xmath20 , we obtain @xmath66 independent of @xmath20 and @xmath18 .
thus we obtain the orbital occupations @xmath67 from the minimal condition for @xmath68 , which in this case are independent of the weights @xmath69 . in the paw method
we obtain the orbital occupations directly from the ps wave functions @xmath70 as @xmath71 using the pseudo - version @xmath72 of the projection operator @xmath73 .
consistent with the paw formalism , the small correction @xmath74 .
the construction is analogous that of the projector functions in the paw method , but the purpose is different : namely , the goal is to decompose a wave function into local orbitals such that @xmath75 holds for any wave function @xmath14 that can be represented exactly as a superposition of the local orbitals @xmath76 .
this requirement is fulfilled whenever @xmath77 .
the projector functions can be obtained from a set of functions @xmath78 , which are linearly independent and localized within an augmentation region as @xmath79 after multiplication with the inverse of @xmath80 .
if we now use @xmath81 we obtain the matrix elements in the representation of localized orbitals directly as @xmath82 it is now a simple matter to obtain the corresponding pseudo - projection operator @xmath83 this expression has a number of advantages : ( i ) it is the most general expression because any operator can be expressed in a fully separable expression,@xcite and , unlike eq .
( [ eq : projector ] ) , it is not limited to a semi - local form .
( ii ) the step function in eq.([eq : projector ] ) introduces a potential that is discontinuous , whereas the operator suggested here can be constructed such that the resulting potential is continuous and differentiable , while still being fully localized in a well - defined region .
( iii ) the separable form of the operator renders the evaluation numerically convenient .
in particular , it is conceivable to evaluate @xmath84 explicitly if necessary , whereas the corresponding operation using the semi - local projection operator would be computationally prohibitive .
( iv ) we can systematically construct projection operators that specifically act on a well - defined portion of the hilbert space : by enlarging the set of functions @xmath85 to include also orbitals that should _ not _ be represented in the restricted set , they are excluded by simply setting the corresponding matrix elements of the density matrix @xmath86 to zero .
the four - center integrals used in the expression of the lda+u total energy are defined as @xmath87 where @xmath88 is the screened coulomb interaction between two electrons .
if we choose localized @xmath4-orbitals that are described by an atomic @xmath4-wave function @xmath89 with magnetic quantum number @xmath20 ( where @xmath90 are the spherical harmonics for @xmath59 ) and furthermore assume that the static dielectric function @xmath91 is constant in space , we can exploit the multipole expansion of @xmath92 : @xmath93 here @xmath94 and @xmath95 denote the smallest and largest values of @xmath96 and @xmath97 , respectively . under these assumptions
we can transform eq.([eq_u_ijkl_def ] ) into @xmath98 where @xmath99 are the angular momenta quantum numbers of @xmath76 , @xmath100 the gaunt coefficients , and @xmath101 the so - called screened slater s integrals . because of the special properties of the gaunt coefficients , only @xmath102 and @xmath103 contribute to the coulomb integrals : @xmath104 the parameters @xmath0 and @xmath45 are identified with averages of the coulomb and exchange integrals , which are related to the slater integrals @xmath105 and @xmath103 by the properties of clebsch - gordan coefficients , eq .
( [ eq_u_j_clebsch ] ) : @xmath106 the dielectric constant and therefore the coulomb and exchange parameters @xmath0 and @xmath45 are not known a priori .
usually , they are obtained from a constrained dft calculation.@xcite here , we are interested in how sensitively the results depend on the choice of coulomb parameters , and which coulomb parameters will provide the best agreement with reality as probed by optical absorption . therefore , we adopt the general form for the four - center integrals as function of the @xmath0 and @xmath45 suggested by the arguments provided above , and perform calculations for different @xmath0 values , namely @xmath107 ev@xmath108 and @xmath109 ev@xmath108 . in the following we suppose that @xmath110 . because the results are fairly insensitive to the exchange parameter @xmath45 , we have adopted @xmath111 ev@xmath108 from previous constrained lda calculations.@xcite the third mandatory relation is obtained from the work of degroot et al.,@xcite who determined the ratio @xmath112 for transition - metal oxides to be between 0.62 and 0.63 .
we therefore adopt a ratio @xmath113 . the pseudo - hamiltonian operator @xmath114 which acts on the ps wave functions , is obtained as the derivative of the total energy functional with respect to the two - center pseudo - density matrix operator @xmath115 .
the non - lda contribution of the lda+u hamiltonian is then obtained as the product of the derivative @xmath116 of the non - lda contribution to the total energy and the projection operator @xmath117 , which is the derivative of the occupation with respect to the two - center pseudo - density matrix operator .
thus we obtain @xmath118 where @xmath119 n^{t,\sigma}_{m_3,m_4 } \nonumber\\ & - & \sum_{\sigma ' } \left [ u ( n^{t,\sigma'}-\frac{1}{2 } \delta_{\sigma,\sigma ' } ) - \delta_{\sigma,\sigma ' } j ( n^{t,\sigma'}-\frac{1}{2 } ) \right ] \delta_{m , m ' } \ , .
\label{eq_pot_u}\end{aligned}\ ] ] the lda contribution of the pseudo - hamiltonian has the usual form:@xcite @xmath120 \left \langle \widetilde{p}_{\lambda_2 } \right| \ , .
\label{eq_h_lda}\end{aligned}\ ] ] for the lda+u calculation , we first performed a self - consistent lsda calculation using the all - electron paw method , and then used the self - consistent potential to construct the lsda hamiltonian for a large number of @xmath121-points in the brillouin zone .
next , the hubbard correction is added to the lsda hamiltonian as given by eq.([eq_h_ldapu ] ) , and the new hamiltonian iterated until the occupation numbers @xmath122 have converged .
we did not use the so - called second - variation procedure for self - consistent lda+u,@xcite in which the lsda potential is updated , because it failed to yield any improvement on calculations similar to ours.@xcite
the ground state of nio has been calculated using the paw method , and the density of states ( dos ) is calculated from the self - consistent paw potential using the tetrahedron method for the brillouin - zone integration.@xcite figure [ lsda_dos ] presents the atom - resolved dos in the augmentation region .
as can be seen , lsda produced an antiferromagnetic insulating ground state with a small band gap .
the lsda magnetic moment is 0.95 @xmath123 , and mainly due to the ni@xmath124 band splitting .
this value is much smaller than the experimental value ( 1.641.90 @xmath123 , refs . ) .
the calculated band gap of about 0.1 ev is also much smaller than the experimental one ( 3.04.0 ev ) .
the most interesting feature of our lsda dos is that the @xmath4-states of ni dominate the region in the vicinity of the band gap , and that the top of the valence state is of ni@xmath124 type for the first and ni@xmath125 for the second spin .
this electronic structure suggests that the band gap is of mott - hubbard type .
hence , this lsda picture of nio disagrees completely with experiment .
it is surprising that the quasi - particle calculation within the so - called gw approximation , performed by aryasetiawan and gunnarsson,@xcite produced results qualitatively similar to lsda except for an increased band gap of 6 ev and an increased magnetic moment of 1.6 @xmath123 .
the quantitative change is the reduction of the o @xmath3 bandwidth by almost 1 ev .
however , a recent self - consistent model gw calculation by massida _ et al._,@xcite in which the dynamic effects were neglected , produced other results than those of aryasetiawan and gunnarsson.@xcite it was argued by massida _
_ that the results of the former gw calculation are not quite self - consistent , presumably because of the additional non - local ad - hoc potential that is adjusted to the gw calculation in each self - consistent step .
the main difference between the two reported gw model calculations is that the latter calculation@xcite produced ( i ) a spreading of the ni @xmath4-states over the entire valence bandwidth , ( ii ) a vanishing gap between the o @xmath3 and ni @xmath11 and , most importantly , ( iii ) an enhancement of o @xmath3 states at the valence band maximum .
the latter effect attributes the origin of the band gap mainly to a charge transfer gap because this gap is now between the o @xmath3 and ni 3@xmath4 conduction states .
next , we will show that this latter finding is in agreement with the results of the lda+u model .
we have used our implementation of the lda+u model to determine the ground - state electronic structure of nio .
although it is common practice to use the @xmath0 extracted from a constrained lda calculation , we adopt a different point of view here .
as stated in the introduction and in agreement with recent results reported in the literature,@xcite we believe that the value of @xmath0 extracted from constrained lda is not the best possible choice .
therefore we have determined the electronic structure of nio for an intermediate @xmath0 of 5 ev as well as for a larger value of 8 ev .
figure [ ldapu_dos ] shows our calculated lda+u dos for @xmath2 and 8 ev .
the energy band gap is found to be 2.8 and 4.1 ev , respectively .
the total antiferromagnetic spin moment is 1.73 and 1.83 @xmath123 , respectively .
our dos obtained for @xmath1 ev agrees well with previous lda+u calculations.@xcite however , the dos for @xmath0 = 5 ev is in better agreement with the gw model calculation of massida _
et al_.@xcite the top of valence band is reinforced by the o @xmath3 states , rendering the band gap a mixture of charge - transfer type and ni @xmath126-like excitations . in agreement with previous lda+u calculations ,
the spin majority ni @xmath127 states are pushed towards lower energies , and the energy difference between @xmath128 and @xmath129 is about 11 ev for @xmath1 ev and 8.6 ev for @xmath130 ev . here again the gw model yields a value of about 9 ev for this splitting , in good agreement with our results for @xmath0 = 5 ev .
for insulators , the imaginary part of the macroscopic dielectric function is obtained within the random phase approximation ( rpa ) in the long wave length limit without local - field effects as@xcite @xmath131 here @xmath132 are the interband transition matrix elements , @xmath133 is the zero - temperature fermi distribution , @xmath134 is the cell volume , @xmath135 denotes the conduction - band and @xmath136 the valence - band index . in the case of a local potential ,
the interband transition matrix elements are given by @xmath137 where the matrix elements @xmath138 are calculated using the paw crystal wave functions @xmath139 described by eq.([eq_psi_paw ] ) : @xmath140 \langle\tilde{p}_{\lambda'}|\tilde\psi_{c}^{\bf k}\rangle \ , , \label{matelements}\ ] ] in the most general case , i.e. where the potential is nonlocal as in lda+u , a nonlocal contribution has to be added to the interband transition matrix elements.@xcite the full derivation is given in the appendix by eq .
( [ eq_mcvk_def ] ) . for nio the nonlocal contribution to the matrix elements
is found to be small , i.e. of a few percent .
figure [ lsda_e2 ] shows the imaginary part of the dielectric function calculated within the lsda .
the resulting optical spectrum is not in agreement with experiment ( see fig .
[ ldapu_e2 ] ) .
in particular , the optical gap is considerably underestimated and the first structure has a much higher intensity compared to experiment .
conversely , for @xmath0 = 5 ev , our calculated imaginary part of the dielectric function within the lda+u is in a better agreement with experiment , as shown in fig .
[ ldapu_e2 ] .
the optical band gap and the oscillator strength of the first excitation peak are in excellent agreement with experiment . however , at higher photon energies the agreement with experiment is only qualitative , which is not expected owing to the mean - field approximation of this simple model .
a much higher value of @xmath0 , i.e. 8 ev , produces a much larger optical gap in contrast to experiment . in agreement with our conclusion
that a much smaller value of @xmath0 is required to describe nio , dudarev and coworkers@xcite also found that @xmath141 ev reproduces the lattice parameter and the measured electron - energy loss spectra .
it is too early to draw a definitive conclusion about the excited states of nio , as our lda+u model is a mean - field - like model in which excitonic effects are not included . to our knowledge
, no calculations of excitonic effects have so far been attempted .
our results represent the first investigation of the low - lying excited states of nio that considers all the subtleties of chemical bonding _ and _ strong electron - electron interaction .
the interband transitions are responsible for the first structure in the optical spectrum of nio , located between 4.1 and 4.4 ev .
we found that 40.2% of the contribution results from the transition from band 15 ( second highest occupied band ) to band 17 ( lowest empty band ) , 36.2% from the transition 16@xmath14218 , and 15.9% from the interband transition 16@xmath14217 . to analyze the character of the initial and final states of the interband transitions , the band structure dispersion along some of the high - symmetry directions is shown in fig .
[ bs_dos ] together with the dos of the states that give rise to the first optical peak .
the arrows between the parallel bands indicate the interband transitions from the initial to the final state responsible for the first structure in the optical spectrum .
figure [ charge_if ] shows the charge density plot of the initial and final states @xmath143 for bands 16 ( highest occupied band ) and 18 ( second lowest empty band ) at point @xmath144 located between the high - symmetry points k and u , where the optical matrix element value is among the largest .
it also shows that the initial state is of mixed o @xmath3 and ni @xmath11 character , whereas the final state mainly is of ni @xmath127 character .
the optical transitions then are between the o @xmath3 and the ni @xmath11 states , resulting in excitations of the charge - transfer type .
our interpretation differs from that of fujimori and minami,@xcite who used a configuration interaction within the metal - ligand cluster and claimed that the @xmath145 charge - transfer transitions are the origin of fundamental edge .
the drawback of the cluster calculation is that in reality the oxygen @xmath3 orbitals are delocalized and therefore not well described in a small cluster . on the other hand , an earlier , band - structure - based interpretation by messick and coworkers@xcite assigned the peak to one electron interband transitions associated with ni @xmath11 to the ni @xmath146 states .
this interpretation is not correct either because the ni @xmath146 are far above the top of the valence states ( greater than 6 ev ) , and only quadrupolar interband transitions are permitted between the @xmath11 and @xmath146 , which substantially reduces the peak intensity .
we have presented a new implementation of lda+u model based on the paw method,@xcite which is an all - electron method without any shape approximation to the potential or the charge density .
we tested the method on nio and obtained results that are in good agreement with previous lda+u calculations and a recent gw model calculation.@xcite in particular , we obtained the correct antiferromagnetic insulating ground state of nio .
we discussed the results in terms of the strength of the hubbard interaction @xmath0 .
the optimum value of @xmath0 has been determined by comparison with the experimental dielectric function as well as with the ground state properties .
we observed a large enhancement of the o @xmath3 character at the top of the valence state , resulting in a more charge - transfer than ni @xmath126 lsda -type band gap .
the calculated antiferromagnetic moment is in good agreement with experiment .
our calculated dielectric function for an intermediate value of @xmath0 , namely 5 ev , is in good agreement with experiment .
the low - lying , strong structure in the optical spectrum has been assigned to an interband transition from o @xmath3 states at top of the valence band to the ni @xmath127 states at the conduction - band bottom .
hence the origin of the first optical peak is due to a charge - transfer excitation .
our calculation is supported by a recent lda+u calculation by dudarev and coworkers,@xcite who also argue that a much smaller value of @xmath0 than the one obtained from constrained lda calculation is needed to describe the electron energy loss spectra and the equilibrium lattice parameter .
it should be interesting to apply this method to other transition - metal oxides and check the applicability of lsda+u for producing excitation energies .
part of this work was done during our ( o.b . and m.a . )
visit to the ohio state university in the summer of 1998 .
we thank j. w. wilkins and j. g. lepage for useful discussions .
supercomputer time was provided by cines ( project gem1101 ) on the ibm sp2 and by the universit louis pasteur on the sgi o2000 supercomputer .
the interband transition matrix elements for a given hamiltonian are obtained as follows : @xmath147_-|\psi_{c,{\bf k}}\rangle } { \epsilon_{v,{\bf k - q}}-\epsilon_{c,{\bf k } } } \ , .\end{aligned}\ ] ] the commutator involving of the lda contribution @xmath148 to the hamiltonian @xmath149 is @xmath150_- = -\frac{1}{2}[\nabla^2,e^{-i{\bf qr}}]_- = -\frac{1}{2}(-2i{\bf
q\nabla}+q^2 ) = -{\bf qp}+o(q)^2 \ , .\ ] ] the quadratic and higher - order terms in @xmath151 can be ignored in the long - wavelength limit appropriate for optical transitions .
the commutator involving the non - lda part @xmath152 is obtained as @xmath153_-=-i{\bf q}[h^u,{\bf r}]_-+o(q)^2 = -i{\bf q}\sum_{t , m , m'}\frac{v^t_{m , m',\sigma}}{i}[p^t_{m',m},{\bf r}]_-+o(q)^2\ ] ] next we use the relation which holds for the special form of the projector operator presented in eq .
( [ eq : projector ] ) @xmath154 i.e. the projection operator changes the magnetic quantum number of a specific @xmath4-like partial wave from @xmath20 to @xmath155 and truncates it beyond the atomic sphere @xmath15 .
hence we obtain @xmath156_-|\phi_{\lambda'}^t\rangle = \langle p^{t\dagger}_{m',m}\phi_{\lambda}|\theta_{\omega_t}({\bf r}){\bf r } finally , we obtain the expression of the matrix elements for the dipole transition , with the paw lda+u formalism : @xmath157 \langle\tilde{p}_{\lambda'}|\tilde{\psi}_{c,{\bf k}}\rangle \bigg\ } \ , .\end{aligned}\ ] ] the difference between wave functions @xmath158 and @xmath159 has been ignored because it only contributes to terms that are proportional to @xmath160 , which are ignored in the long - wavelength limit .
shen , c. k. shih , o. jepsen , w. e. spicer , i. lindau , and j. w. allen , phys .
lett . * 64 * , 2442 ( 1990 ) ; z - x .
shen , j. w. allen , p. a. p. lindberg , d. s. dessau , b. o. wells , a. borg , w. ellis , j. s. kang , s - j .
oh , i. lindau , and w. e. spicer , phys .
b * 42 * , 1817 ( 1990 ) ; z - x .
shen , r. s. list , d. s. dessau , b. o. wells , o. jepsen , a. j. arko , r. barttlet , c. k. shih , j. c. huang , and p. a. p. lindberg , _ ibid .
_ * 44 * , 3604 ( 1991 ) .
v. i. anisimov , j. zaanen , and o. k. andersen , phys .
b * 44 * , 943 ( 1991 ) ; v. i. anisimov , m. a. korotin , j. zaanen , and o. k. andersen , phys . rev
. lett . * 68 * , 343 ( 1992 ) ; v. i. anisimov , i. v. solovyev , m. a. korotin , m. t. czyzyk , and g. a sawatzky , phys .
b * 48 * , 16929 ( 1993 ) ; v. i. anisimov , p. kuiper , and j. nordgren , phys .
b * 50 * , 8257 ( 1994 ) . . magnetic moment and band gap of nio within lda , lda+u ( @xmath2 ev ) and lda+u ( @xmath1 ev ) .
the radii of the augmentation regions are 1.7 and 2.1 a.u . for the oxygen and the nickel atoms , respectively .
[ cols=">,<,^,^,^,^ " , ] | the so - called local density approximation plus the multi - orbital mean - field hubbard model ( lda+u ) has been implemented within the all - electron projector augmented - wave method ( paw ) , and then used to compute the insulating antiferromagnetic ground state of nio and its optical properties .
the electronic and optical properties have been investigated as a function of the coulomb repulsion parameter @xmath0 .
we find that the value obtained from constrained lda ( @xmath1 ev ) is not the best possible choice , whereas an intermediate value ( @xmath2 ev ) reproduces the experimental magnetic moment and optical properties satisfactorily . at intermediate @xmath0 ,
the nature of the band gap is a mixture of charge transfer and mott - hubbard type , and becomes almost purely of the charge - transfer type at higher values of @xmath0 .
this is due to the enhancement of the oxygen @xmath3 states near the top of the valence states with increasing @xmath0 value . |
the decay constants of heavy - light p - wave mesons have a number of important uses in phenomenology @xcite . in this paper we use unquenched lattice qcd to compute the decay constants of @xmath4 heavy - light mesons , using heavy quarks with masses around that of the charm quark , and with heavy quarks in the static limit .
due to heavy - quark symmetry , there are four p - wave ( @xmath5 ) heavy - light excited mesons , noted for short @xmath6 , organised in two doublets , one doublet carrying a total angular momentum @xmath7 and the other @xmath8 for the light quarks . in the charm sector , we use the notation that the scalar @xmath9 meson is @xmath10 when the light quark is an up or a down quark and @xmath11 when it is a strange quark . in the static limit of hqet ,
the members of each of the two doublets are degenerate in mass .
the low mass of the @xmath12 meson , recently discovered by babar @xcite and confirmed by other experiments , relative to the quark model predictions of godfrey and isgur @xcite was originally a puzzle .
the quantum numbers of @xmath12 are consistent with @xmath13 = @xmath4 , but this needs confirmation @xcite .
there are some speculations that this state may be a molecule @xcite .
the theoretical studies to understand the @xmath12 have recently been reviewed by colangelo et al .
@xcite and by swanson @xcite .
the lattice qcd results for the mass of the @xmath11 state were tantalising close ( within large errors ) to the experimental mass of the @xmath12 state @xcite .
however , the lattice calculation of the @xmath11 meson in @xcite did not agree with the mass of the @xmath12 .
as discussed by ukqcd @xcite a hadronic state will couple to many different interpolating operators made out of quarks , anti - quarks and glue , with the same quantum numbers as the hadron . in order to say
whether a hadronic state is more like a molecule or @xmath14 state , the lattice calculation needs to also determine the amplitude for a hadronic state to be in a specific configuration of quarks and anti - quarks .
one possible way to determine the quark distribution of a hadronic state is to look at observables like form factors , the most basic being the decay constant .
for example , in a simple picture of a molecular state , the decay constant should be suppressed relative to that of a bound state of quark and antiquark .
an analysis of non - leptonic decays of the @xmath15 meson using the factorisation hypothesis suggested that the decay constant of the @xmath12 could be significantly smaller than that of the pseudo - scalar heavy - light meson @xcite .
this motivates a lattice calculation of the decay constant of the @xmath4 heavy - light meson .
the @xmath16 decays are also relevant to clarify the so called ` @xmath17 vs. @xmath18 puzzle ' @xcite indeed , the heavy to heavy @xmath19 transitions , with @xmath20 , are parametrised by the generalised isgur - wise functions @xmath21 , where @xmath22 .
several theoretical considerations using independent methods ( sum rules derived from qcd @xcite , covariant quark model @xcite , lattice qcd @xcite , experimental data combined with naive factorisation @xcite ) , suggest that the @xmath8 states dominate , with respect to the @xmath7 states in @xmath23 transitions , i.e. @xmath24
. nevertheless this prediction is in contradiction with some of the currently available experimental data .
the solution for this puzzle still requires new inputs from both theory and experiment . from the theoretical side , the determination of the decay constant
, @xmath25 of the @xmath10 meson is needed to evaluate the decay width of two ( out of the three ) classes of @xmath16 decays , when naive factorisation is assumed @xcite .
a recent summary @xcite of results for the decay constant , showed that estimates for @xmath25 lay between 122 and 417 mev .
this summary also reported that the decay constant of @xmath4 heavy - light mesons in the static limit was higher than at the charm mass .
this required confirmation , because the heavy quark mass dependence was obscured by model dependence .
some factorisation schemes @xcite for non - leptonic decays of @xmath15 mesons use the static limit as the leading term , hence computing the decay constant of the p - wave static - light meson is also important for this reason .
the evaluation of the @xmath26 corrections to the heavy quark limit of the various quantities describing @xmath19 transitions is crucial since simple estimates tend to suggest the presence of large corrections @xcite .
there is a huge effort in lattice gauge theory that aims to compute the decay constant of the heavy - light pseudo - scalar mesons , because they are crucial for determining ckm matrix elements @xcite .
the decay constant of the heavy - light vector meson has been computed in @xcite in order to test the hqet scaling relations .
there has been no published work ( apart from a private communication reported in @xcite ) on using lattice qcd to compute the decay constants of p - wave heavy - light mesons .
the study of the wave - function of p - wave static - light mesons @xcite , using quenched lattice qcd , did not extract a decay constant .
ukqcd has studied the charge and matter radial distributions of p - wave static - light mesons @xcite . in our calculations
, we used the non - perturbatively improved clover action with clover coefficient of 2.0171 for the light quarks . the wilson gauge action was used .
the lattice volume was @xmath27 and @xmath28 .
there were two flavours of sea quarks .
the @xmath29 values are tabulated with the results .
the sea quark masses span from the strange quark to one third of the strange quark mass .
the details of the light hadron spectroscopy can be found in @xcite .
the plan of the paper is that in section [ se : prop ] , we first discuss our results using clover fermions with quark masses around the charm value .
next in section [ se : static ] , we present data using static heavy quarks . in the final section [ se : concl ]
we compare our results to other determinations and make some remarks about the heavy quark mass dependence of the decay constants .
we have used the unquenched data from @xcite to study the decay constant of the @xmath4 meson around the charm mass with clover quarks . the data set used the sea @xmath29 = 0.1350 and three valence light quark masses using @xmath29-values of 0.1340 , 0.1345 and 0.1350 .
the @xmath29 values for the heavy quarks were : 0.113 , 0.119 and 0.125 . using @xmath30 = 0.5 fm ,
the inverse lattice spacing is 1.88 gev @xcite .
the sea quark mass was kept fixed at the mass of the strange quark , hence this is a partially quenched calculation .
the data sample included 394 gauge configurations separated by 20 trajectories .
adjacent data were binned together . a smearing matrix of order two , using local and fuzzed basis states , was fitted to a factorising fit model @xcite .
the decay constant of the @xmath10 meson can be defined by equation , @xmath31 where @xmath32 is the vector current with two non - degenerate quark flavours .
this uses the @xmath33 structure of the charged weak quark current and parity conservation .
we used a unit gamma matrix between the heavy and light quark fields as the interpolating operator for the @xmath10 state .
unfortunately , we did not measure the lattice correlators with @xmath34 at the sink and @xmath35 at the source that would be required to use equation [ eq : vectordefn ] directly . in principle due to the number of gamma matrices
, there are 256 different local meson correlators .
many lattice qcd codes do not compute every possible correlator , because some correlators are zero by symmetries . the correlator with a @xmath34 at the sink and a unit matrix at
the source is zero for degenerate quarks because of charge conjugation .
colangelo et al .
@xcite define the decay constant ( @xmath36 ) using equation [ eq : decaydefncol ] .
@xmath37 where @xmath38 is the mass of the charm quark .
equation [ eq : decaydefncol ] can be related to equation [ eq : vectordefn ] since the divergence of the vector current for non - degenerate quarks is proportional to the scalar density .
the derivation of equation [ eq : decaydefncol ] from equation [ eq : vectordefn ] assumes that the mass of the light quark can be neglected relative to that of the charm quark . from a lattice qcd perspective , it is more natural to define the decay constant using equation [ eq : decaydefn ] .
@xmath39 the determination of the mass of the charm quark can have quite large systematic errors at a fixed lattice spacing @xcite , so we prefer the definition of the decay constant with no explicit factor of the charm mass . to convert from our normalisation of the decay constant @xmath40 to that of @xmath36 , we note that for this data set ukqcd obtained @xmath41 = @xmath42 gev @xcite , using a quark mass defined using the fermilab heavy quark formalism @xcite .
the matrix element is related to the coupling amplitude @xmath43 in the fit to the @xmath4 to @xmath4 meson correlator .
@xmath44 a simple linear fit model was used to extrapolate and interpolate in the quark masses .
we first interpolated to the strange quark mass @xcite , or extrapolated to the light quark mass ( at @xmath45 ) .
the decay constants were then interpolated to the value of the mass of the charm quark @xcite . in figure
[ fig : meff ] we report an effective mass plot for the local - local and fuzzed - local correlators used in the analysis . in figure
[ fig : massdependlight ] we plot the bare decay constant as a function of the bare light quark mass in physical units with mass of the heavy quark fixed .
similarly , figure [ fig : massdepend ] shows the bare lattice decay constant as a function of the heavy quark mass , with the light quark mass fixed at the mass of the strange quark .
the dependence of the decay constant on the mass of the quarks is very mild .
we discuss some of the issues in the chiral extrapolations in section [ se : static ] .
= 0.113 and light @xmath29 = 0.1350 . ] at the fixed heavy quark mass with @xmath29 = 0.113 , as a function of the bare light quark mass in physical units .
the estimate of the decay constant at the mass of the strange quark is also shown . ] at the strange quark mass , as a function of the bare heavy quark mass in physical units .
the vertical line is the estimate for the charm quark mass . ] to convert the lattice number into the @xmath46 scheme we use tadpole improved perturbation theory to one loop order ( the required expressions are listed in @xcite ) . @xmath47 where @xmath48 is the fourth root of the plaquette .
the mass dependent improvement factor @xmath49 was multiplied into the scalar current .
we used the one loop expression for @xmath50 @xcite , @xmath51 and the same prescription for the coupling as used in ukqcd s estimate of the mass of the charm quark @xcite .
we use @xmath52 to determine @xmath53 . in lattice units we obtain @xmath54 = 0.280(70 ) and @xmath55 = 0.270(90 ) . using @xmath30 = 0.5 fm , and @xmath53 from equation [ eq : zstad ]
, we obtain @xmath25 = 360(90 ) mev and @xmath56 = 340(110 ) mev . in principle
, we should estimate the effect of the lattice spacing dependence and higher order terms in the perturbative expansion , as we did when we estimated the mass of the bottom quark @xcite .
however , because the statistical errors are of the order of 30% , the statistical errors will clearly dominate .
for example , a typical estimate of the systematic error due to the different ways of determining the lattice spacing in a lattice calculation with these parameters would be to vary the value of @xmath30 between 0.5 fm and 0.55 fm .
this systematic error is a 10 % effect .
we now discuss the central value of @xmath30 used .
when @xmath30 , a number derived from the heavy quark potential , was introduced by sommer @xcite , he quoted a value of @xmath30 about @xmath57 fm .
the value of @xmath30 can also be determined from lattice calculations .
as reviewed in @xcite , unquenched calculations with similar parameters to those used here compute a value of @xmath30 between 0.5 and 0.55 fm .
the issue is not so clear because the hpqcd collaboration , using gauge configurations from the milc collaboration , obtain a value of @xmath30 = 0.469(7 ) fm . as ukqcd has argued @xcite , for calculations with these parameters it better to use a value of @xmath30 closer to 0.5 fm because this is the value determined from similar calculations .
the milc / hpqcd value for @xmath30 is with in 10% of 0.5 fm .
this part of the work is a continuation of ukqcd s study of static - light mesons .
the static - light meson spectrum was reported in @xcite , and then used to extract the mass of the bottom quark @xcite .
here we use the extended data set that was used to look for chiral logs in the @xmath58 decay constant @xcite .
in addition to using the standard static action of eichten and hill , we also report data using one of the static actions developed by the alpha collaboration @xcite with improved signal to noise ratio .
the exact static actions are described in @xcite .
here we focus on the meson doublet with total light quark momentum of @xmath7 , we call this the @xmath59 static - light meson @xcite . as corrections to the static limit
are included the @xmath59 state will split into a @xmath4 state and a @xmath60 state . in this paper , the decay constant of the @xmath59 static - light meson
is called @xmath61 . the @xmath61 decay constant
is defined by the matrix element in equation @xmath62 where @xmath63 is the vector current @xcite .
the simple spin structure of the static quark means the same decay constant is obtained if in equation [ eq : decaydefn ] the vector current ( with @xmath64 ) is replaced with a scalar current .
the @xmath61 matrix element is extracted from the amplitudes in the two point correlator @xmath65 where @xmath66 is the interpolating operator for the @xmath59 static - light meson and ground state dominance is shown in equation [ eq : fitmodel ] . in practise
we use a basis of smearing functions to do variational smearing @xcite .
the @xmath67 amplitude is related to the @xmath61 decay constant @xmath68 where @xmath69 is a perturbative matching factor that we discuss below .
this is an equivalent definition to the one used for the pseudo - scalar heavy - light decay constant ( @xmath70 ) .
it is traditional to match the results of a lattice static - light calculation to continuum qcd via two steps @xcite .
qcd is matched to the continuum static theory .
the lattice static theory is then matched to the static continuum theory .
the matching of the static continuum theory to the static - lattice theory has been done by eichten and hill @xcite and by borrelli and pittori @xcite for the clover action .
this matching was done for a heavy - light current with arbitrary gamma matrix ( @xmath71 ) . for the definition of the decay constant in equation [ eq : decaydefn ] ,
we do the perturbative matching assuming @xmath72 .
the matching between the continuum static theory and the static lattice theory is via @xmath73 .
@xmath74 where @xmath75 where the values of the constants are @xmath76 , @xmath77 , @xmath78 , @xmath79 @xcite , @xmath80 , and @xmath81 @xcite .
@xmath82 when @xmath83 .
the matching of the continuum static theory to continuum qcd was done using @xcite @xmath84 where @xmath85 is the mass of the bottom quark .
the equivalent expression for equation [ eq : zcontstat ] , for arbitrary @xmath71 matrix , is in @xcite .
the matching factor @xmath69 in equation is the product of @xmath86 with @xmath87 to leading order in the square of the coupling @xmath88 .
the improvement coefficients have not been calculated for the p - wave decay constant and we will therefore not include them in our analysis .
we used a simple boosted coupling ( @xmath89 and @xmath90 ) to compute the perturbative matching factors .
we use the mass ( @xmath91 mev ) for the mass factor in equation [ eq : zlstaticamp ] .
the experimental spectrum of the p - wave b mesons is not very well determined at the moment @xcite .
lattice calculations predict that the lowest p - wave meson will be roughly 400 mev above the s - wave states .
so we add 400 mev to the mass of the @xmath92 state @xcite .
this corresponds to a 4% effect on the decay constant .
we prefer not to use the hqet scaling law of the decay constant to quote numbers for the decay constant of the @xmath0 charm - light meson , as is done by jugeau et al .
@xcite , because it is known that @xmath93 corrections to the static limit of the decay constant of the pseudo - scalar heavy - light meson are large .
the unrenormalised data for the amplitudes are in table [ tab : staticbareresults ] .
the final column shows the value for @xmath61 in the static limit in physical units .
the value of @xmath30 = 0.5 fm was used to convert the data into physical units . as the renormalisation factor for the alpha static action has not yet been determined
, we ca nt use that data to quote a physical number .
what is disappointing is that the alpha static action does not produce any reduction in the statistical errors over the standard eichten - hill static action in our case . for the static - light pseudo - scalar meson decay constant
the statistical errors were significantly smaller for the alpha static action than the eichten - hill action @xcite .
in @xcite it was shown that the @xmath94 data set corresponded to sea quarks with mass around the strange quark mass .
hence we quote the @xmath2 decay constant as @xmath3 mev . .
amplitudes and decay constants for the @xmath95 static - light mesons .
further information about the lattice parameters is in @xcite .
[ cols="^,^,^,^,^",options="header " , ] [ zzz ] kurth and sommer @xcite have noted that there are potential problems with extrapolating in the heavy - quark mass unless the continuum limit has been taken .
also see kronfeld @xcite for a discussion of the problems combining static and propagating heavy - light data . given these theoretical concerns , the data is consistent with the view that the decay constant of the static - light meson is larger than that of the @xmath0 charm - light meson , if we use definition of the decay constant in equation [ eq : vectordefn ] ( our value of @xmath96 is in table 2 ) , but the errors need to be reduced for a definitive statement .
jugeau et al .
have collected together a number of different calculations of the @xmath0 decay constant of a charm - light meson . to compare our results to other calculations we use the normalisation in equation [ eq : vectordefn ] .
hence we multiply our result by @xmath97 = 1.27/2.317 and compare to other calculations in table 2 .
there is reasonable agreement between different determinations . to reduce the errors on the lattice results requires a calculation of similar effort to that done to compute the decay constants of the heavy - light pseudo - scalar mesons @xcite .
we thank alex dougall for help . the gauge configurations were generated on the cray t3e system at epcc supported by epsrc grant gr / k41663 , pparc grants gr / l22744 and ppa / g / s/1998/00777 .
this work has been supported in part by the eu integrated infrastructure initiative hadron physics ( i3hp ) under contract rii3-ct-2004 - 506078 .
we are grateful to the ulgrid project of the university of liverpool for computer time .
one of the authors ( cm ) wishes to thank pparc for the award of a senior fellowship .
p. colangelo , g. nardulli , a. a. ovchinnikov , and n. paver , phys .
b269 * , 201 ( 1991 ) , p. colangelo , g. nardulli , and n. paver , phys . lett . *
b293 * , 207 ( 1992 ) , babar , b. aubert _
et al . _ ,
lett . * 90 * , 242001 ( 2003 ) , hep - ex/0304021 , s. godfrey and n. isgur , phys . rev . *
d32 * , 189 ( 1985 ) , s. godfrey and r. kokoski , phys . rev . *
d43 * , 1679 ( 1991 ) , particle data group , s. eidelman _
et al . _ ,
* b592 * , 1 ( 2004 ) , t. barnes , f. e. close , and h. j. lipkin , phys . rev . *
d68 * , 054006 ( 2003 ) , hep - ph/0305025 , p. colangelo , f. de fazio , and r. ferrandes , mod .
a19 * , 2083 ( 2004 ) , hep - ph/0407137 , e. s. swanson , ( 2006 ) , hep - ph/0601110 , ukqcd , a. dougall , r. d. kenway , c. m. maynard , and c. mcneile , phys . lett . *
b569 * , 41 ( 2003 ) , hep - lat/0307001 , rbc , s. ohta , h. lin , and n. yamada , pos * lat2005 * , 096 ( 2005 ) , hep - lat/0510071 , g. s. bali , phys . rev . *
d68 * , 071501 ( 2003 ) , hep - ph/0305209 , ukqcd , a. m. green , j. koponen , c. mcneile , c. michael , and g. thompson , ( 2003 ) , hep - lat/0312007 , a. datta and p. j. odonnell , phys . lett . *
b572 * , 164 ( 2003 ) , hep - ph/0307106 , d. s. hwang and d .- w .
kim , phys .
b606 * , 116 ( 2005 ) , hep - ph/0410301 , n. uraltsev , ( 2004 ) , hep - ph/0409125 , f. jugeau , a. le yaouanc , l. oliver , and j. c. raynal , phys . rev . * d72 * , 094010 ( 2005 ) , hep - ph/0504206 , i. i. bigi _ et al .
_ , ( 2005 ) , hep - ph/0512270 , n. uraltsev , phys . lett . *
b501 * , 86 ( 2001 ) , hep - ph/0011124 , a. le yaouanc , l. oliver , o. pene , and j. c. raynal , phys . lett .
* b365 * , 319 ( 1996 ) , hep - ph/9507342 , d. becirevic _ et al .
_ , phys . lett . *
b609 * , 298 ( 2005 ) , hep - lat/0406031 , b. blossier , a. le yaouanc , v. morenas , and o. pene , phys . lett .
* b632 * , 319 ( 2006 ) , hep - lat/0507024 , m. j. dugan and b. grinstein , phys .
b255 * , 583 ( 1991 ) , v. lubicz , nucl .
suppl . * 140 * , 48 ( 2005 ) , hep - lat/0410051 , m. okamoto , pos * lat2005 * , 013 ( 2005 ) , hep - lat/0510113 , c. bernard _
et al . _ ,
phys . rev . * d65
* , 014510 ( 2002 ) , hep - lat/0109015 , a. duncan , e. eichten , and h. thacker , phys . lett . *
b303 * , 109 ( 1993 ) , ukqcd , a. m. green , j. koponen , and c. michael , aip conf . proc .
* 756 * , 369 ( 2005 ) , hep - lat/0412002 , ukqcd , a. m. green _ et al .
_ , pos * lat2005 * , 205 ( 2005 ) , hep - lat/0509161 , ukqcd , c. r. allton _
et al . _ ,
phys . rev . * d65 * , 054502 ( 2002 ) , hep - lat/0107021 , ukqcd , c. r. allton _
_ , phys . rev .
* d70 * , 014501 ( 2004 ) , hep - lat/0403007 , a. dougall , c. m. maynard , and c. mcneile , jhep * 01 * , 171 ( 2006 ) , hep - lat/0508033 , a. x. el - khadra , a. s. kronfeld , and p. b. mackenzie , phys . rev . * d55 * , 3933 ( 1997 ) , hep - lat/9604004 , ukqcd , d. j. hepburn , nucl .
119 * , 338 ( 2003 ) , hep - lat/0209036 , m. gockeler _ et al .
. suppl . * 53 * , 896 ( 1997 ) , hep - lat/9608033 , t. bhattacharya , r. gupta , w .- j .
lee , and s. r. sharpe , phys . rev .
* d63 * , 074505 ( 2001 ) , hep - lat/0009038 , s. sint and p. weisz , nucl .
suppl . * 63 * , 856 ( 1998 ) , hep - lat/9709096 , ukqcd , c. mcneile , c. michael , and g. thompson , phys . lett . *
b600 * , 77 ( 2004 ) , hep - lat/0408025 , r. sommer , nucl . phys .
* b411 * , 839 ( 1994 ) , hep - lat/9310022 , ukqcd , c. mcneile and c. michael , jhep * 01 * , 011 ( 2005 ) , hep - lat/0411014 , alpha , m. della morte _ et al .
_ , phys . lett . *
b581 * , 93 ( 2004 ) , hep - lat/0307021 , e. eichten and b. hill , phys . lett . *
b234 * , 511 ( 1990 ) , e. eichten and b. hill , phys . lett . * b240 * , 193 ( 1990 ) , a. borrelli and c. pittori , nucl . phys . *
b385 * , 502 ( 1992 ) , jlqcd , s. aoki _
lett . * 91 * , 212001 ( 2003 ) , hep - ph/0307039 , a. s. kronfeld and s. m. ryan , phys .
b543 * , 59 ( 2002 ) , hep - ph/0206058 , d. becirevic , s. fajfer , and s. prelovsek , phys . lett . *
b599 * , 55 ( 2004 ) , hep - ph/0406296 , t. mehen and r. p. springer , phys . rev .
* d72 * , 034006 ( 2005 ) , hep - ph/0503134 , ukqcd , c. mcneile , c. michael , and g. thompson , phys . rev . * d70 * , 054501 ( 2004 ) , hep - lat/0404010 , s .- l . zhu and y .- b .
dai , mod .
* a14 * , 2367 ( 1999 ) , hep - ph/9811449 , c. aubin _ et al .
lett . * 95 * , 122002 ( 2005 ) , hep - lat/0506030 , cleo , m. artuso _ et al .
. lett . * 95 * , 251801 ( 2005 ) , hep - ex/0508057 , m. wingate , t. degrand , s. collins , and u. m. heller , phys .
lett . * 74 * , 4596 ( 1995 ) , hep - ph/9502274 , alpha , m. kurth and r. sommer , nucl . phys . *
b623 * , 271 ( 2002 ) , hep - lat/0108018 , a. s. kronfeld , ( 2002 ) , hep - lat/0205021 , | we review some decays that require knowledge of the decay constants of @xmath0 heavy - light mesons .
we compute the decay constants of p - wave heavy - light mesons from unquenched lattice qcd , with two degenerate flavours of sea quarks , at a single lattice spacing .
the lightest sea quark mass used in the calculation is a third of the strange quark mass . for the charm - strange meson
we obtain the decay constant : @xmath1 mev using our normalisation conventions .
we obtain the @xmath2 ( static - strange p - wave ) decay constant as @xmath3 mev . |
the statistical description of a fully developed hydrodynamic turbulence is based on the kolmogorov hypotheses @xcite on the self - similarity of the velocity fluctuations of different scales .
however , the kolmogorov - obukhov analysis @xcite does not provide a rigorous mathematical definition of the `` fluctuation of scale @xmath0 '' . in the literature this is tacitly understood as fourier components with wavenumbers approximately equal to the inverse scale @xmath1 and the analysis is performed in wavenumber space .
this definition meets global characteristics of the fully developed isotropic turbulence , but , being based on the fourier transform , is essentially nonlocal and therefore hardly applicable to such important properties of the fully developed turbulence as the coherent structure formation . to catch the local properties of the turbulent velocity field ,
the decomposition into localized wave packets and wavelets have been performed by many authors @xcite .
most of the wavelet applications to turbulence are restricted either to an analysis of the measured turbulent fields with the `` wavelet microscope '' , capable of simultaneous analysis of the same velocity field at different resolution @xcite , or to a numerical solution of the navier - stokes equations ( nse ) in the wavelet basis @xcite .
the application of the continuous wavelet transform with the derivatives of the gaussian taken as basic wavelets was already used in the analytical study of the nse @xcite , in particular , in research on energy dissipation .
however , at least to the authors knowledge _ the wavelet decomposition has not been yet applied to the stochastic iterative solutions of the navier - stokes equation or in the framework of the field theory approach to statistical hydrodynamics _ @xcite .
our interest in extending the stochastic hydrodynamics approach by wavelet - defined random processes is stimulated by recent developments in application of the field theory and renormalization group methods to the fully developed isotropic turbulence , see e.g. @xcite for a short summary .
the `` field theoretic methods '' are taken to mean the stochastic diagram technique , the functional integral representation of the characteristic functional and the renormalization group methods .
the latter , being inherited from the theory of critical behavior in equilibrium phase transitions clearly demonstrates the need for a scale - dependent random measure in the field theoretic approach to turbulence .
such measures , having been already in use for phenomenological studies of multifractal behavior and itermittency @xcite , have not been yet used in analytical approach of stochastic hydrodynamics .
the aim of this paper is to extend the wavelet representation of the stochastic navier - stokes equations in such a way that the probability distribution of the stirring force , used to compensate the energy dissipation , is defined for the wavelet coefficients of random force ( for the scale components of forcing ) . in our description
the velocity field wavelet coefficients @xmath2 attain the kolmogorov meaning of local velocity fluctuations of scale @xmath0 at a given point @xmath3 ( a particular case of the difference of two gaussians or delta - functions was considered in the literature @xcite ) . defining the random force in the space of wavelet coefficients
we get an extra analytical flexibility : there are random processes in the space of wavelet coefficients with different correlation functions whose images under inverse wt coincide in the space of common random functions . tuning the random force correlation function in the space of wavelet coefficients , in stochastic hydrodynamics formalism
, we get rid of loop divergences ; for a special type of narrow - band forcing the contributions to the response and the correlation functions are explicitly calculated .
it is shown that the kolmogorov hypotheses , the statistical closures of moment equations , the stochastic hydrodynamics approach and the wyld diagram technique are naturally reformulated in multiscale ( wavelet ) formalism .
besides , the consideration of random processes depending on scale explicitly @xmath4 gives a possibility of the perturbation expansion converging without introducing an ultra - violet ( uv ) cutoff wavenumber , and an extra contribution to the kolmogorov energy dissipation term @xmath5 .
the rest of this paper is organized as follows . in _ section 2
_ we review the methods of solution of the nse by the fourier decomposition and by the wavelet decomposition . in _ section 3
_ the stochastic hydrodynamics approach is reformulated for the random processes explicitly depending on scale .
a regularization of the perturbation expansion for a random force acting at a single scale is presented .
energy dissipation rate and energy flux in the multiscale formalism are considered in _
section 4_. _ section 5 _ presents a generalized form of the kolmogorov hypotheses formulated in a multiscale framework . in _ conclusion
_ we discuss some perspectives of the method .
in analytical studies of the hydrodynamic turbulence the basic role is played by the navier - stokes equations : @xmath6 our consideration of hydrodynamic turbulence , based on the navier - stokes equations , assumes a fully developed homogeneous isotropic turbulence far from any boundaries .
although the adequacy of the stochastic nse to fully developed turbulence still remains an open problem , a significant progress has been achieved in studying simple models , proving the kolmogorov spectrum from basic principles @xcite and calculating the anomalous scaling corrections @xcite .
turbulence in an incompressible fluid considered in these settings is more frequently studied in a wavenumber space rather than in a real space .
the advantage of wavenumber space is a very simple form of the laplacian this enables to eliminate the pressure term from the nse .
the price paid for this simplification is the non - locality of the fourier transform that completely hides all information related to the real space distribution of the velocity field .
the discrete fourier transform used in numerical simulations also imposes periodicity on the system .
the usefulness of the wavenumber representation stems from two basic reasons : the existing of fast fft algorithms and pseudospectral methods @xcite and the direct experimental interpretation of the power spectra density of velocity fluctuations @xcite . for the incompressible
fluid the pressure term can be eliminated from the nse by the substitution @xmath7 in the wavenumber space representation @xmath8 both the inverse laplacian operator @xmath9 in and the incompressibility condition @xmath10 are simplified and the nse system becomes a system of integro - differential equations @xmath11
, \quad p_{ij}(\vk ) = \delta_{ij } - \frac{k_i k_j}{\vk^2}.\end{aligned}\ ] ] the system of equations allows for perturbative calculations , statistical closures of moment equations . , but is not local in the coordinate space and therefore is of little use when studying the effects locally produced by fluctuations .
the system of equations is complete and , being correctly solved numerically , gives reliable results .
this however requires taking into account a tremendous number of fourier modes , that is hardly bearable even for modern supercomputers ; alternatively certain block - averaging procedures in wavenumber space can be applied , the spectral reduction @xcite .
the idea of studying turbulence using the space - scale or wavenumber - scale representation is not a new one .
the band - pass filtering of the turbulent signals was used to study intermittency long ago @xcite .
later the prototype of wavelet cascade model was given in @xcite .
the wavelet transform ( wt ) is known to be an excellent local tool , widely used in a turbulence data analysis @xcite and numerical simulations @xcite .
wt in real space , as a particular type of filtering @xcite , reveals the formation of the coherent structures @xcite and is useful to study the local energy dissipation effects related to filament formation @xcite .
wt in @xmath12 space enables to enhance the idea of band - pass filtering by separating the contributions of different scales . in this paper
we use continuous wavelet transform ( cwt ) to derive the equations for the fluctuations of different scales .
this analytical consideration goes along with the filtering approach @xcite , when considering moment closures for a fully developed isotropic turbulence . _
the crux of our approach is the application of continuous wavelet transform in the stochastic hydrodynamics framework : the wt is applied to both the velocity field and the random force stirring the turbulence_. working in the space of wavelet coefficients , rather than in a real space of velocity fields , we can define the random stirring force , which is essentially used in rg calculations , see @xcite in a way that allows for getting rid of loop divergences in the stochastic perturbation expansion of the velocity field statistical momenta @xcite . for simplicity
we restrict ourselves with the homogeneous isotropic turbulence and the isotropic wavelets @xmath13 . in this case , the wavelet transform of the velocity field @xmath14 , taken with respect to the basic wavelet @xmath15 , and the corresponding reconstruction formula are @xmath16 we perform the wavelet transform only in the spatial argument of the velocity field because we need the _ spatial _ resolution .
using the @xmath17 norm instead of @xmath18 , we provide wavelet coefficients @xmath19 with the same @xmath20 dimension as the velocity field @xmath21 itself .
the wavelet coefficients @xmath19 are referred to hereafter , as the components of the velocity field corresponding to scale @xmath22 ; @xmath23 is referred to as an analyzing function used to measure the scale components . for practical calculations
it is often convenient to express wavelet transform ( [ dcwt],[icwt ] ) in @xmath24 representation taking the fourier transform in the spatial argument . in fourier form the direct and inverse wt
are : @xmath25 where @xmath26 is the fourier transform of the velocity field @xmath27 ( the minkovski - like notation @xmath28 is used . ) therefore , the wavelet transform can be considered as a frequency filter that conveys the harmonics with typical wavenumbers of order @xmath29 and is localized close to point @xmath30 . the only restriction imposed on the basic wavelet @xmath31 to make the wavelet transform invertible the admissibility condition is the finiteness of the normalization constant @xmath32 : @xmath33 for a real - valued basic wavelet @xmath15 we can restrict the integration to the positive frequencies only @xmath34 if the basic wavelet is also isotropic @xmath13 , we get @xmath35 where @xmath36 is the area of the unit sphere in @xmath37 .
in this paper we will assume @xmath38 integration and the isotropic real wavelets .
thus , the decomposition of the velocity field with respect to the basic wavelet @xmath31 takes the form @xmath39 we drop the integration limits @xmath40 hereafter . substituting the wavelet
transform into the system of the component equations , we yield the system of equations for the scale components @xmath41 : @xmath42 let us derive statistical closures for the scale components .
for this purpose we take eq . and
its complex conjugate @xmath43 multiply the first equation by @xmath44 , sum up over the vector index @xmath45 and take the statistical averaging @xmath46 .
doing so , we get @xmath47 applying the same procedure to the second of the equations and summing up the results at coinciding time arguments @xmath48 , we get the moment equation @xmath49 which is different from its plane wave counterpart only by extra scale indexes and extra integrations in scale logarithms @xmath50(octaves ) . to express the third order moments in via the second moments
, we must substitute @xmath51 where @xmath52 is the response function .
the difference from the standard plane - wave approach @xcite is that additionally to the summation over vector indices we have to sum up over octaves to integrate over @xmath50 in each scale variable .
so the statistical closures can be reproduced for the scale components . in the zero - th order approximation ( with no interaction term : @xmath53 )
the bare response function is given by @xmath54aa_0}(\vk , t\!-\!s ) = \delta_{il } \int_{-\infty}^\infty \frac{\delta(a - a_0)a_0}{-\imath\omega+\nu\vk^2}e^{-\imath\omega(t - s)}\frac{d\omega}{2\pi } = \delta_{il } \delta(a - a_0)a_0 e^{-\nu\vk^2|t - s|}. \label{brf}\ ] ] the full response function , in view of the component equations , satisfies the integro - differential equation @xmath55 the substitution of into gives a relation between the second and the forth order moments of the scale components : @xmath56 the forth order moments @xmath57 can be further decomposed into the sum of all pairs @xmath58 using a stochastic perturbation expansion .
the stochastic hydrodynamics approach consists in introducing random force in the navier - stokes equations and calculating the velocity field momenta @xmath59 using the stochastic perturbation theory , pioneered by wyld @xcite , or the functional integral formalism . in a coordinate representation ,
the stochastic nse is written in the form @xmath60 the random force correlator @xmath61 should obey certain conditions to make the resulting theory physically feasible and the perturbation expansion suitable for analytical evaluation .
first , the energy injection by random force should be equal to the energy dissipation ; secondly , the forcing should be essentially infrared ( ir ) , it should be localized at large scales ; third , it is desirable to have a parameter to control the ir divergences ( when the size of the system tends to infinity ) . to exclude the pressure term in using the incompressibility condition ,
the fourier representation is used @xmath62 the random force correlator is usually taken in the form @xmath63 where the function @xmath64 has a suitable power - law behavior . in the simplest , but not very feasible physically , case @xmath65 , we deal with the white noise , @xmath66-correlated in both space and time .
what is most realistic physically , is to have a random force concentrated in a limited domain in @xmath67-space @xmath68 .
this case , however , is difficult to evaluate analytically in the perturbation theory @xcite . in the multiscale approach
we are going to present , we solve this problem by constructing a noise acting in a limited domain of scales @xmath22 in @xmath24 space
. in the @xmath24 representation , having excluded the pressure by standard means of the orthogonal projector , the eq .
leads to a system of integro - differential equations for the scale components @xmath69 : @xmath70 now we face the problem of appropriate choice of the force correlator @xmath71 it was shown in the previous paper @xcite that the @xmath12 representation provides an extra analytical flexibility in constructing random processes with desired correlation properties in a coordinate space . for instance , the random process given by wavelet coefficients with the correlation function @xmath72 possesses the same correlation properties in @xmath37 coordinate space as the white noise does
. casting @xmath73 in terms of @xmath74 by means of we get @xmath75 that coincides with the correlation function of the white noise .
the direct wavelet transform of the white noise @xmath76 apparently leads to another result @xmath77 which is different from and explicitly depends on the basic wavelet @xmath31 .
physically , the scale - dependent processes obeying and , respectively , describe quite different processes : fluctuations of the former type are mutually correlated only for coinciding scales ( @xmath78 ) , while for the latter case all fluctuations are correlated .
exactly as in the standard wavenumber space approach @xcite , we can generalize the @xmath66-correlated force assuming its variance to be dependent on both the scale the wave vector : @xmath79 .
taking into account we deal with the incompressible fluid in @xmath80 dimensions , we can put down a general form of the desired force correlator @xmath81 the @xmath66-correlated random force in the wavenumber space does not provide an adequate description of hydrodynamic turbulence , for it gives an energy injection in all scales , small and large . in physical settings , the fluid is usually stirred at a predetermined scale , or in a narrow range of scales , comparable to the size of the system . as a simplest model of such a forcing , we can consider a force acting on a single scale @xmath82 by choosing @xmath83 now let us turn to the perturbative calculations . the stochastic diagram techniques for the component fields @xmath84 stems from eq . and is a straightforward generalization of the wyld diagram technique for the fourier components @xmath85 : @xmath86 where @xmath87 is the bare response function for fourier component . to keep the scales and wavevectors on the same footing and
make the notation covariant in that sense , we can rewrite using the response functions bearing scale indices explicitly .
thus , we get @xmath88 the feynman expansion for the scale component fields @xmath69 can be derived either from or . iterating the eq .
once , we get the one - loop contribution to the response function : @xmath89.\end{aligned}\ ] ] as usual , we assume the random force to be gaussian , with all odd correlators @xmath90 vanish identically . following @xcite , we introduce a formal parameter of the perturbation expansion @xmath91 in the interaction term ( @xmath92 ) . in the final end
the initial value @xmath93 should be restored .
the validity of considering @xmath91 as a small parameter of perturbation expansion is justified by renormalization group methods @xcite ; see also @xcite for recent developments and generalizations .
in the zero - th order of perturbation expansion the response function does not depend on scale and coincides with @xmath94 for all scale components : @xmath95 in the @xmath96 and the next orders of perturbation expansion the standard stochastic diagram techniques , used by many authors @xcite , is reproduced with the difference that : ( i ) each vertex , each response and correlation function attain scale superscripts ; ( ii ) integration over octaves @xmath50 is performed over all pair - matching scale indices .
mathematically , this means that each diagram line corresponding to the plane wave component @xmath97 in standard techniques now attains an extra wavelet factor and becomes @xmath98 .
the feynman graphs and their topological factors of course remain the same . here , for bookkeeping reasons , we present only one loop 1pi contributions to the response , see fig .
[ resp1l : pic ] , and correlation , see fig .
[ cor1l : pic ] , functions of stochastic hydrodynamics . in the first order in the force correlator @xmath99 ( one - loop contribution ) we substitute the scale components @xmath100 in the r.h.s . of by the zero - th order solutions ,
perform necessary scale integrations in @xmath101 and average over the random force : @xmath102 the corresponding diagram is shown in fig .
[ resp1l : pic ] .
the factor 4 accounts for two possible ways to expand the nonlinear term multiplied by two possible ways of taking the random force averaging .
after substituting the random force correlator , the using of the explicit form of the interaction vertexes , and integrating over the scales , this leads to @xmath103 eq .
means that for our special type of the scale - dependent random forcing all internal parts of the diagrams , that do not carry the scale indices , can be evaluated by substitution of the effective force correlator into the standard diagrams drawn in wavenumber space . in this way we can easily evaluate the perturbative corrections to the usual response function @xmath104 for fourier components ( see appendix ) , and hence evaluate the turbulent corrections to viscosity , produced by scale - dependent force .
similarly , we can evaluate the contributions to the correlation functions @xmath105 . in the one - loop approximation , using and the zero - th order approximation , we get @xmath106 after the integrations over @xmath107 , identically to the response functions calculations , with the random force , we get the one - loop contribution to the correlation function , shown in fig . [ cor1l : pic ] : @xmath108 expanding the wavelet factors @xmath109 in each vertex and integrating over all matching scale arguments , we get the one - loop correction to the correlation function @xmath110 the evaluation of the one - loop contribution to the correlation function is easily performed for the above mentioned narrow - band force correlators .
the contribution takes the form @xmath111 @xmath112 where the trace of the one - loop tensor structure @xmath113 is given in appendix .
the integration in frequency argument in the limit of zero frequency ( @xmath114 ) , is not different from that in stochastic hydrodynamics in the wavenumber space and gives @xmath115 . \label{c2e1}\end{aligned}\ ] ] as an example , let us present the one - loop contribution to the effective pair correlator , calculated for the case of single - scale forcing with the basic wavelets from the gaussian vanishing momenta wavelet family @xmath116 with @xmath117 being the eulerian gamma - function .
for the single - scale forcing this gives an effective force correlator in the wavenumber space @xmath118 straightforward calculation leads to : @xmath119 q^{d-1}dq \sin^{d-2}\theta d\theta.\end{aligned}\ ] ] the integration over the angle variable @xmath120 can be performed explicitly . with the calculations presented in appendix , we get the effective pair correlator for @xmath121 in the large - scale limit @xmath122 : @xmath123
the energy dissipation rate per unit of mass of an incompressible viscous fluid is given by the navier - stokes equations @xmath124 using wavelet decomposition for velocity field we get @xmath125 where @xmath126 is the dissipation connection for the scale components . in symbolic form , the contribution of the fluctuations of all scales @xmath127 to the mean dissipation of energy per unit of mass can be written as @xmath128 where @xmath129 is the viscosity between @xmath127 and @xmath130 scales .
being well localized in both the real and the wavenumber space , the analyzing function @xmath31 perceives the interaction of the components of the same or close scales stronger than the contributions of the significantly different ones @xmath131 . for the daubechies wavelets ( the orthogonal wavelets with compact support )
often used for numerical simulation of turbulence , the viscosity connection coefficients can be found elsewhere @xcite .
for a qualitative estimation of the behavior of the viscosity connection as a function of the scale ratio @xmath132 of the interacting scale components , let us consider the vanishing momenta wavelet family of gaussian wavelets , considered by lewalle in a wavelet - based analysis of energy dissipation @xcite .
the viscosity connection can be then evaluated analytically : @xmath133 the main contribution to energy dissipation comes from the terms with coinciding or closed arguments @xmath134 . in this limit ( with @xmath135 , taken for simplicity ) we get @xmath136 introducing the ratio @xmath137 , we can study the behavior of the viscosity connection as a function of scale ratio @xmath138 the plot of the viscosity connection @xmath139 as a function of the scale ratio @xmath137 , for the first three wavelets ( @xmath140 ) of the family , is shown in fig .
[ vmfcon : pic ] .
as it can be seen , regardless the number of vanishing momenta @xmath141 , the dissipation term has a maximum at coinciding scales ( @xmath142 ) . for @xmath140 related to the viscosity connection for gaussian type wavelets.,width=302 ] for this reason , if a discrete wavelet decomposition is used instead of continuous one , it is sufficient to keep the two main terms in the energy dissipation : the equal - scale interaction and the neighboring scale interaction @xmath143 the first term is the standard viscosity term , the last is the kraichnan nearest scales interaction .
the nonlinear energy transfer between neighboring scales can be evaluated by considering the wavelet connections corresponding to nonlinear term @xmath144 of the navier - stokes equation . to keep with the wavelet turbulence cascade models @xcite ,
we restrict ourselves with a discrete wavelet transform with binary scale step , @xmath145 : @xmath146 without loss of generality the mesh size is set to unity @xmath147 . in our consideration of the dwt representation
applied to hydrodynamic turbulence , in contrast to many schemes applied for numerical simulation of the nse @xcite , we have no _ a priory _ arguments to assume a mutual orthogonality of the basis functions @xmath148 . to keep the wavelet decomposition unique ,
the orthogonality of basic functions is not required .
it is sufficient if the set of basic functions forms a _ frame _
, @xmath149 such that @xmath150 if @xmath151 , the frame is called a _ tight frame_. assuming the basic functions @xmath152 form a frame , and restricting ourselves with the case of incompressible fluid , we cast the nse system in the form @xmath153 with greek letters used for the coordinate indices , the bold face for the vector subscripts being dropped , and the summation over all pair - matching indices assumed .
the component fields @xmath154 are the functions of time only ; so we deal with a typical cascade model .
our goal at this point is to derive the energy transfer between the components of @xmath155-th scale and the next small @xmath156-th one .
let us define the energy of the @xmath155-th scale pulsations as @xmath157 with the assumed unit normalization of the basic function @xmath158 .
the contribution of the nonlinear term of the nse to the time derivative of @xmath159 is @xmath160 for the orthogonal wavelets , that are most often used in numerical simulations @xcite , only the terms of coinciding scales @xmath161 survive in the r.h.s . of .
the energy flux from the @xmath155-th scale to the next @xmath156-th scale is then proportional to @xmath162 , in exact accordance to the kolmogorov phenomenological theory @xcite . in more general terms of nonorthogonal basic functions , the next term in the r.h.s .
of is proportional to @xmath163 .
this term can be interpreted as the material derivative @xmath164 of the mean energy @xmath165 of the small scale fluctuations travelling along the stream of large scale velocity @xmath166 .
the energy transfer terms analogues to have been already considered in the orthogonal wavelet formalism by c.meneveau @xcite .
they can be obtained directly from the component equations , by multiplying by @xmath167 .
this leads to the energy transfer in @xmath24 space @xmath168 given in @xcite .
the kolmogorov theory of the locally isotropic turbulence is formulated in terms of relative velocities @xmath169 the probability distribution of relative velocities hardly can be studied by the fourier transform , in case the velocity field @xmath170 is not homogeneous .
according to kolmogorov @xcite , the turbulence in space - time domain @xmath171 , is referred to as a stationary turbulence if , for any fixed @xmath172 , the distribution of relative velocities @xmath173 is stationary and isotropic .
physical assumptions on the locally isotropic turbulence were formulated in terms of the first and second kolmogorov hypotheses , that reside on the definition of the reynolds number .
this is not a quite rigorous mathematical definition .
we shall show that the kolmogorov hypotheses are the statements about the behavior of the scale components of velocity field .
first , we have to note that the definition of the reynolds number is consistent within the multiscale framework .
in fact , by definition @xmath174 where @xmath175 are said to be `` pulsations of the scale @xmath0 '' , the rigorous definition of those can be given by virtue of wavelet components , considering @xmath31 as an apparatus function used to measure the pulsations .
going further , we find the wavelet components to be identical to velocity increments , in case @xmath31 is the haar wavelet : @xmath176 now let us consider the kolmogorov hypotheses @xcite : [ [ h1-the - first - hypothesis - of - similarity ] ] h1 : the first hypothesis of similarity + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + for the locally isotropic turbulence with high enough re the pdfs for the relative velocities are uniquely determined by the viscosity @xmath177 and the mean energy dissipation rate @xmath178 .
[ [ h2-the - second - hypothesis - of - similarity ] ] h2 : the second hypothesis of similarity + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + under the same assumption as for h1 the turbulent flow is self - similar in small ( but still @xmath179 ) scales in the sense that @xmath180 using the taylor frozen flow hypothesis , and therefore considering one - dimensional pulsations @xmath181 , according to the definition of wavelet coefficients , with @xmath182 we get @xmath183 for small values of @xmath0 we can approximate @xmath184 or , taking into account the statistical homogeneity of the flow , we get @xmath185 so , the power - law behavior , viz @xmath186 is just a particular case of a local regularity of wavelet coefficients , with the haar function being used as a basic wavelet . as it was shown in general settings @xcite , the wavelet coefficients @xmath187 $ ] of a square - integrable function @xmath188 , which has the lipshitz - hlder exponent @xmath189 at the point @xmath190 ,
behave as @xmath191| \sim a^h$ ] inside the cone @xmath192 for any admissible wavelet @xmath31 which satisfies the regularity condition @xmath193 the condition is rather loose , and in physical settings one can always assume that it holds for any analyzing function used to measure the pulsations of scale @xmath0 .
so , the second kolmogorov hypothesis can be formulated as follows : [ [ h2-generalized - second - kolmogorov - hypothesis - of - similarity ] ] h2 : generalized second kolmogorov hypothesis of similarity + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + under the same assumption as for h1 the turbulent flow is self - similar in small ( but still @xmath179 ) scales in the sense , that the pulsations of the turbulent velocity defined as @xmath194 where @xmath23 is any analyzing function satisfying the admissibility condition and the regularity condition , have the following power - law behavior @xmath195 for all spatial points @xmath196 occupied by turbulent media .
in this paper we have a stochastic hydrodynamics approach to the nse based on a wavelet decomposition of both the velocity field and the stirring force used to balance the energy dissipation . in spite of a good deal of papers devoted do different choices of the stirring force for the nse in wavenumber space see . @xcite and references therein
what is essentially new in our approach is the definition of random force by the correlation function of its wavelet coefficients . establishing the force correlator in the space of wavelet coefficients ,
we have got it easy to get rid of loop divergences in the stochastic perturbation expansion , having at the same time desired physical properties of the forcing ( energy injection at a given scale ) .
therefore , a new uv - finite framework is constructed for statistical hydrodynamics .
this is a technical framework for analytical evaluation of the statistical characteristics of turbulent fluctuations , such as their correlation and response functions , by means of continuous wavelet transform . deriving physical consequences , such as energy cascade between scales or scale - dependent corrections to response function
, we specially did not touch the renormalization group aspects @xcite of the problem and multifractal formalism @xcite .
in fact , both are related .
the former is a generalization of the description of hydrodynamic turbulence in terms of differential equations to the description in measure settings .
partially , the relation of wavelets and rg in turbulence description is discussed in @xcite , and will be studied in more details in connection to multifractal properties of hydrodynamic turbulence . besides , the possible comparison with the rg based classification of asymptotic regimes of isotropic turbulence @xcite can be considered as another perspective of the proposed method . the study of hydrodynamic turbulence by methods of stochastic differential equations and those of quantum field theory has at least half a century history .
regardless phenomenologically clear and widely accepted kolmogorov @xcite theory of fully developed turbulence , still there are discussions on the preference of either differential equations , or field theoretic methods based on renormalization group , or multifractal approach to describe the turbulence in an incompressible fluid . as it concerns the physical interpretation of the velocity field wavelet coefficients , by this paper we intended to say that stochastic nature of spatially extended hydrodynamic turbulence prescribes a certain kind of `` wave - particle dualism '' to the turbulent phenomena , in a sense , that the answer we get depends on the basic functions used to describe the turbulence .
if the basis of plain waves was chosen , there are no fair reasons to comply about @xmath197 behavior or paradoxes : what we get is what we set . alternatively ,
if we want to have an analytical description of the spatially extended turbulence compatible with the kolmogorov phenomenology of local turbulent pulsations of given scales , we need to set a functional basis that respects the scale locally .
this is the wavelet decomposition .
the fourier transform , being essentially nonlocal , apparently does not fit the above mentioned requirements , but the windowed fourier transform , or wave packet decomposition used by v.zimin @xcite and t.nakano for the analysis of turbulence does , and possibly there is only a technical difference between our approach and that of nakano @xcite . however , it is important to emphasize that the incorporation of the basic function @xmath31 into consideration makes us to admit that the definition of the local fluctuations of a given scale is not completely objective and depends on means of observation .
as it was shown , the kolmogorov hypotheses ( k41 ) were easily rewritten in the wavelet framework for a multiscale description of turbulence .
the author is thankful to drs .
m.hnatich , j.honkonen , m.jurcisin and o.chkhetiani for useful discussions .
this work is supported in part by russian foundation for basic research , project 03 - 01 - 00657 .
to calculate the one - loop diagram in the response function , we introduce the following tensor structure @xmath198 where the summation over all dummy indices is assumed . the following notation for the orthogonal projector and the vertex
is used ( @xmath199 ) : @xmath200 .\ ] ] after all convolutions in matching pairs of indices , substituting @xmath201 , where @xmath202 is the cosine of the angle between @xmath67 and @xmath203 , we get @xmath204 after substitution @xmath205 in the numerators and some algebraic simplification @xmath206 \label{rfts}\ ] ] where @xmath207 to calculate the whole one - loop integral contribution to the response function , the tensor structure @xmath208 is multiplied by the integral over the frequency component @xmath209 using this structure we easily get the 1pi one - loop contribution to the response function @xmath210 \hat\eta_s(k ) .
\label{l2int } \end{aligned}\ ] ] in @xmath211 we always assume @xmath212 , with @xmath213 being the polar angle @xmath214 and @xmath215 being the azimuthal angle .
let us evaluate the integral in @xmath211 @xmath216 \hat\eta_i(k ) \label{l2resp } \end{aligned}\ ] ] here @xmath217 means diagonal matrix , where the first two terms give the transversal contribution to viscosity , and the last term , giving the longtitudal contribution , vanishes identically : @xmath218 finally , after integration in angle variable @xmath120 , we get @xmath219 where @xmath220.\end{aligned}\ ] ] similarly , for the tensor structure of the one - loop contribution to the response function , we evaluate the tensor structure of the one - loop contribution to the pair correlator @xmath221 after algebraic simplification @xmath222 the trace of this tensor structure , @xmath223 , required for the energy spectra evaluation , is equal to @xmath224 .
\label{c2trace}\ ] ] in the important case of @xmath211 this gives @xmath225 .
\label{c2t}\ ] ] the integral in frequency argument of the product of squared response functions in the integral gives in the limit of zero frequency @xmath114 : @xmath226 in the important case of @xmath211 , with the mexican hat wavelet ( @xmath227 ) taken for definiteness , the integral over the angle variable @xmath228 can be evaluated analytically for the single band random force .
the angle integration in gives @xmath229 where @xmath230.\end{aligned}\ ] ] in the limit of small wave numbers @xmath231 this gives @xmath232 therefore the effective correlator tends to @xmath233 k. gawedzki , a. kupianinen , and m. vergassola .
a program on developed turbulence .
erwin schrdinger international institute for mathematical physics , 2002 .
http://www.esi.ac.at/programs/turbulence2002.html .
orszag and g.s . patterson .
numerical simulation of turbulence . in m. rosenblatt and c.v .
van atta , editors , _ statistical models and turbulence _ , volume 12 of _ lecture notes in physics _
, pages 127147 .
springer , 1972 .
a. latto , h.l .
resnikoff , and e. tenenbaum .
the evaluation of connection coefficients of compactly supported wavelets . in _
french - usa workshop on wavelets and turbulence _ , new york , 1991 .
springer - verlag . | we present a multiscale description of hydrodynamic turbulence in incompressible fluid based on a continuous wavelet transform ( cwt ) and a stochastic hydrodynamics formalism . defining the stirring random force by the correlation function of its wavelet components ,
we achieve the cancellation of loop divergences in the stochastic perturbation expansion .
an extra contribution to the energy transfer from large to smaller scales is considered .
it is shown that the kolmogorov hypotheses are naturally reformulated in multiscale formalism . the multiscale perturbation theory and statistical closures based on the wavelet decomposition
are constructed . |
cold atoms provide a new platform in which one can explore long standing open questions of strongly correlated systems in condensed matter physics @xcite .
in particular , two - component fermi mixtures in an optical lattice provide an ideal realization of the fermionic hubbard model , where two species of fermions corresponding to the spin @xmath0 interact with an on - site repulsion .
this model is relevant for understanding properties of electrons in several classes of novel quantum materials including oxides and high-@xmath1 superconductors @xcite .
there are currently many efforts to probe the low - temperature physics of such a model , with experiments already demonstrating the mott insulating behavior expected in this model @xcite .
an important feature of strongly correlated ultracold atoms is that traditional probes used in solid state physics are often not readily available .
one thus needs to understand how experimental techniques appropriate for atomic ensembles can provide information on many - body states . in this paper
we focus on understanding lattice modulation experiments with fermions in optical lattices .
the technique of lattice modulation was originally introduced for bosonic systems and absorbed energy was measured as a function of modulation frequency @xcite . measuring energy absorption , however , can not be done accurately enough for strongly interacting fermions .
thus an extension of this technique was proposed , @xcite and implemented @xcite , in which the number of doubly occupied sites created by the lattice modulation was measured .
recent experiments successfully reached the regime of weak perturbations in which the number of doublons created scales quadratically with the modulation amplitude ( fermi s golden rule ) and the modulation pulse duration @xcite . while theoretical understanding of such experiments with bosons is now relatively complete @xcite , the case of fermions turned out to be more challenging .
the main difficulty is the presence of excitations at very different energy scales : high - energy charge excitations , so - called doublons and holons , that have energies set by the on - site repulsion @xmath2 and fermion hopping strength @xmath3 , and magnetic excitations that have energies of the order of superexchange energy @xmath4 .
understanding the interplay of charge and spin degrees of freedom in the hubbard model is a long standing problem in condensed matter physics @xcite . in the special case of half filling and fully disordered spin states , analysis of lattice modulation experiments has been performed previously @xcite
. however , such analysis is not sufficient for quantitative comparison to experiments which are done with systems in a parabolic potential that have a large number of atoms outside of the incompressible mott plateaus .
furthermore such real - time dynamics at finite temperature as seen in current experiments is problematic for numerics .
monte carlo simulations @xcite suffer from the problems of analytic continuation .
density - matrix renormalization group approaches , which can deal with real time dynamics , are so far limited to one dimension .
therefore , a comparison of dynamical quantities to experiments is a non - trivial theoretical subject . in this paper
we develop another analytical approach to calculate finite - temperature dynamics , that is , the doublon production rate ( dpr ) , based on the slave - particle technique @xcite .
this approach is particularly adapted to the paramagnetic phase of the hubbard model , and can be applied to any filling of the band and finite temperatures whose region is relevant to ongoing experiments .
it provides a remarkable agreement to the experiments and allows for potential extensions .
this paper is organized as follows .
we define the system hamiltonian , and introduce the slave - particle representation in sec .
[ sec : slave - part - repr ] . in sec .
[ sec : diagrammatic - analys ] , the spectral functions of the slave particles are evaluated , and in addition the spectral function of the original fermionic atom is also obtained . in sec .
[ sec : dpr ] , we proceed with the estimation of the doublon production rate based on the obtained spectral functions in sec . [ sec : diagrammatic - analys ] , and the analytic formula of the dpr spectrum is given . in sec . [
sec : comparison - experiments ] , the obtained analytic formula is extended to an inhomogeneous system in a trap by the local density approximation ( lda ) , and a comparison with the experiment in ref .
@xcite is implemented .
finally , sec .
[ sec : summary ] is devoted to the summary .
we consider the hubbard model @xmath5 with @xmath6 [ eq : hubbard_model ] where @xmath7 and @xmath8 are , respectively , a creation and number operator of a spin-@xmath9 fermions at a @xmath10th site .
we use the following slave - particle representation : @xmath11 where @xmath12 , @xmath13 and @xmath14 are , respectively , creation operators of a slave boson of spin-@xmath9 state ( schwinger boson ) , holon and doublon at a site @xmath10 .
they satisfy the ( anti-)commutation relations , @xmath15=\delta_{i , j}\delta_{\sigma,\sigma'}$ ] and @xmath16 .
the enlarged hilbert space is projected onto the physical one by the following constraint at every site : @xmath17 equation ( [ eq : sf - sb ] ) allows us to rewrite the hamiltonian ( [ eq : hubbard_model ] ) as @xmath18 , \label{eq : tunnelinghamiltonian } \\ & h_{\rm at } = \sum_{j } \left [ \epsilon^{\rm d}_{j}d^{\dagger}_{j}d_{j } + \epsilon^{\rm h}_{j } h^{\dagger}_{j}h_{j } + \sum_{\sigma}\epsilon^{\rm b}_{j } b^{\dagger}_{j\sigma}b_{j\sigma } -\lambda_j \right ] , \label{eq : atomichamiltonian}\end{aligned}\ ] ] where @xmath19 and @xmath20 mean the hopping of slave bosons and the creation of spin singlet pair , respectively .
the local potentials of a slave boson , holon , and doublon are , respectively , defined as @xmath21 , @xmath22 , and @xmath23 .
the constraint ( [ eq : constraint ] ) is implemented via the lagrange multiplier @xmath24 .
we start with the atomic limit ( @xmath25 ) . then the kinetic term ( [ eq : tunnelinghamiltonian ] ) which describes the scattering among slave particles vanishes .
since the atomic hamiltonian ( [ eq : atomichamiltonian ] ) is quadratic , the atomic propagators at @xmath10th site are easily obtained as @xmath26 where @xmath27 and @xmath28 are the matsubara frequency for bosons and fermions , respectively .
note that the atomic limit propagators of slave bosons are independent of the spin .
this means that the atomic limit exhibits spin - incoherent paramagnetism .
hereafter we set @xmath29 .
let us suppose the mean - field ( mf ) @xmath24 to be determined by the atomic limit .
namely , the self - consistent equation for @xmath24 corresponds to the statistical average of the constraint ( [ eq : constraint ] ) in the atomic limit : @xmath30 where the prefactor @xmath31 comes from the spin degrees of freedom . @xmath32 and @xmath33 are , respectively , the fermi and bose distribution functions .
one can expect that if the effect of the kinetic energy @xmath34 is small , that is , at relatively high temperature compared to the kinetic energy , the validity of this treatment should be guaranteed .
we thus use the mf assumption for the lagrange multiplier : @xmath35 . simultaneously the local potentials are also replaced by the homogeneous ones : @xmath36 where @xmath37 .
corresponding to the mf treatment for the lagrange multiplier , the atomic propagators also become independent of sites : for example , the replacement of @xmath35 leads the slave boson propagator @xmath38 where the site - independent propagator is defined as @xmath39 let us solve the self - consistent equation ( [ eq : sce - lambda ] ) . for @xmath40 ,
( [ eq : sce - lambda ] ) can be simplified , and solved analytically as follows : @xmath41 as discussed below , eq .
( [ eq : sce - lambda ] ) is numerically solved , and we compare the numerical result with eq .
( [ eq : mf - lagrange - multiplier ] ) .
the obtained @xmath42 leads the estimation of the slave particle densities .
the density of each slave particle is given by the mf solution @xmath42 : @xmath43 for a slave boson , @xmath44 for a holon and @xmath45 for a doublon .
the temperature and chemical potential dependency of them are shown in fig .
[ fig : slave - density ] . in the temperature region shown in fig .
[ fig : slave - density ] , the results analytically given by eq .
( [ eq : mf - lagrange - multiplier ] ) are in precise agreement with ones given by the numerically solved @xmath42 .
it means that in such a regime we may always employ eq .
( [ eq : mf - lagrange - multiplier ] ) as a solution of the mf self - consistent equation ( [ eq : sce - lambda ] ) . on the other hand ,
the densities in the atomic limit can be exactly calculated as shown in appendix [ sec : app2 ] . here
we take the exact densities to be @xmath46 , @xmath47 and @xmath48 , and their temperature and chemical potential dependency is also shown in fig . [
fig : slave - density ] .
the exact result allows us to discuss the temperature and chemical potential regime justifying the slave particle approach . from the comparison made in fig .
[ fig : slave - density ] , the slave particle technique is found to agree with the exact result in the temperature and chemical potential region where @xmath49 and @xmath50 . in other words the slave particle technique
is expected to be reasonable when the system is near a mott insulator , and such temperature as a benchmark is below @xmath51 .
we now consider a finite but small hopping by taking the infinite series of diagrams produced by the scattering @xmath34 among the slave particles , based on the non - crossing ( nc ) approximation @xcite .
this approximation can be also regarded as a certain type of high temperature series expansion ( htse ) @xcite , but in our formalism the wick theorem is still applicable due to the quadratic hamiltonian ( [ eq : atomichamiltonian ] ) , and a particular infinite series of kinetic energy perturbation can be taken .
based on the nc approximation the self - energy diagrams constructed by full propagators , shown in fig .
[ fig : non - crossing - diagram ] , are considered .
are taken for large @xmath2 , the diagram ( b ) would be irrelevant .
the holon self - energy is also given by the same type of diagrams . ]
since one is in a temperature regime higher than the anti - ferromagnetic exchange @xmath52 , we apply the spin - incoherent assumption to the slave boson propagator .
namely , the slave boson propagators in the diagrams of fig .
[ fig : non - crossing - diagram ] are replaced by the atomic one ( [ eq : mean - field - spinon - propagator ] ) , @xmath53 the doublon and holon propagators are left full ones : namely , by using the dyson equation the full propagators are written as @xmath54 unlike the standard mf theory , the dynamical fluctuation of the slave bosons is retained here , which is necessary to describe the doublon - holon excitation . the nc diagram fig .
[ fig : non - crossing - diagram ] ( b ) couples the self - consistent equations of the doublon and holon self - energy .
however , the contribution is negligibly small because it is a far off - shell diagram in this case such as mott state .
consequently the self - consistent equations of the self - energy fig .
[ fig : non - crossing - diagram ] ( a ) are decoupled and one can obtain in momentum space , @xmath55 with @xmath56j^2}$ ] corresponding to a half band width for the holon and doublon as we will see below .
@xmath57 is a coordination number , and @xmath58 is the total site number of the system .
note that due to the momentum dependence of the right - hand side the self - energy should be given as a local quantity : @xmath59 .
consequently the propagator also turns out to be local : @xmath60 .
thus the self - consistent equation ( [ eq : sce - selfenergy ] ) is easily solved through the dyson equation ( [ eq : dyson - equation ] ) as follows : @xmath61 through the analytic continuation the doublon spectral function , which is equivalent to the density of state ( dos ) in this case , is also obtained . in this approximation ,
a semi - circle type dos is formed : @xmath62 one can also obtain the self - energy and spectral function of the holon in the same way .
the forms are the same as what is obtained by the replacement @xmath63 in eqs .
( [ eq : selfenergy ] ) and ( [ eq : doublon - spectralfunction ] ) .
the holon spectral function is found to reasonably reproduce the result of brinkman and rice on single hole dynamics in a mott insulator @xcite .
the chemical potential dependency of the band width @xmath64 is shown in fig .
[ fig : band - width ] . as a function of @xmath65 for @xmath66 .
the chemical potential dependence of the band width for @xmath67 ( solid ) , @xmath68 ( dashed ) , @xmath69 ( dashed - dotted ) , and @xmath70 ( dotted ) is also shown in the inset . ] using the representation ( [ eq : sf - sb ] ) , via the matsubara green s function of the original fermion , the dos ( spectral function ) is represented as @xmath71 { \cal a}_{\rm h}(\epsilon_{\rm b}-\omega ) \nonumber \\ & \quad + \biggl[b(\epsilon_{\rm b})+f(\epsilon_{\rm b}+\omega)\biggr ] { \cal a}_{\rm d}(\epsilon_{\rm b}+\omega).\end{aligned}\ ] ] as expected , the doublon and holon spectral functions , @xmath72 and @xmath73 , give the upper and lower hubbard band , respectively .
the spectral function as a function of @xmath74 and the chemical potential is shown in fig.[fig : spec.func . ] . for @xmath66 and @xmath75 ( top panel ) , @xmath76 for @xmath77 , and @xmath75 ( middle panel ) and @xmath74 for @xmath77 and @xmath66 ( bottom panel ) .
we calculate the dpr induced by the amplitude modulation of an optical lattice potential using the above formalism .
the amplitude modulation of the optical lattice potential , @xmath78 , modifies both @xmath3 and @xmath2 as @xmath79 $ ] and @xmath80 $ ] , where @xmath81 and @xmath82 are dimensionless modulation perturbation parameters and given as a function of @xmath83 .
however , it is possible to map the two parameter modulations problem to single parameter one of either @xmath3 or @xmath2 @xcite .
namely , the modulation perturbation to be discussed here can be written as @xmath84 where @xmath85 . within the second - order perturbation in terms of @xmath86 , the dpr defined as the time averaged growth rate of _ atoms forming doublons _
is given as @xcite @xmath87 where @xmath88\rangle$ ] .
as shown in appendix [ sec : app1 ] , we represent the correlation function @xmath89 without vertex corrections in the slave - particle description .
in addition , it does not affect crucially in the insulator with the large mott gap .
] , @xmath90 ^ 2 \mathcal{g}_{\mathrm{h}}(\tau)\mathcal{g}_{\mathrm{d}}(\tau ) -[\bar{\mathcal{g}}_{\mathrm{b}}(\tau)]^2 \mathcal{g}_{\mathrm{h}}(-\tau)\mathcal{g}_{\mathrm{d}}(-\tau ) \ } , \label{eq : h_k - correlation - function0}\end{aligned}\ ] ] where @xmath91 ( @xmath92 or @xmath93 ) is a two - particle green s function of a doublon ( @xmath94 ) and holon ( @xmath95 ) .
the diagrams corresponding to the terms in eq .
( [ eq : h_k - correlation - function0 ] ) are illustrated in fig .
[ fig : kinetic - correlation - function ] . without the vertex correction ,
the two - particle propagators are contracted to a single particle propagators by the wick expansion : @xmath96 . through the fourier
transform of @xmath97 and analytic continuation , one can straightforwardly obtain the real - time kinetic energy correlation function in frequency domain . as a result ,
the imaginary part of the correlation function is given as @xmath98 [ { \cal a}_{\rm h}(\nu){\cal a}_{\rm h}(\nu-\omega ) \nonumber \\ & \quad + { \cal a}_{\rm d}(\nu){\cal a}_{\rm d}(\nu-\omega ) ] + 2\sinh(\epsilon_{\rm b } ) \int\!\!\frac{d\nu}{2\pi } [ b(2\epsilon_{\rm b } ) \nonumber \\ & \quad + f(\nu ) ] \ { [ f(2\epsilon_{\rm b}-\nu)-f(2\epsilon_{\rm b}-\nu+\omega ) ] \nonumber \\ & \quad \times{\cal a}_{\rm d}(\nu){\cal a}_{\rm h}(2\epsilon_{\rm b}+\omega-\nu ) -(\omega\rightarrow -\omega ) \}. \label{eq : h_k - correlation - function}\end{aligned}\ ] ]
let us compare our result ( [ eq : h_k - correlation - function ] ) with the experimental data .
we employ a set of the parameters evaluated in the @xmath99k atom experiment @xcite : the hopping @xmath100 ( hz ) and the interaction @xmath101 ( hz ) . in terms of the optical lattice potential ,
the depth , modulation rate and lattice constant are , respectively , taken to be @xmath102 where @xmath103 is a recoil energy , @xmath104 and @xmath105 ( nm ) .
the lattice modulation is translated into @xmath106 in hopping modulation .
the lda is used to take into account the effect of the harmonic trap potential @xmath107 whose frequency is @xmath108 ( hz ) . in the lda , we replace the chemical potential of the homogeneous case by the local one , @xmath109 where @xmath110 is self - consistently determined to give the total trapped atom number @xmath111 . in our framework , temperature is treated as a free parameter so that we determine the temperature by a fit of the dpr spectrum intensity at @xmath112 , which is obtained in the experiment . and the temperature is determined from the crossing point to the experimental data ( dotted line ) , which is @xmath113 . ]
the temperature dependence of the dpr spectrum at @xmath112 is shown in the inset in fig .
[ fig : dpr ] , and @xmath114 in this system is determined .. ] the determined temperature is in the region to justify the slave - particle approach , which is discussed in sec .
[ sec : diagrammatic - analys ] , and the theory is thus expected to work well .
furthermore , using the obtained temperature , we plot the dpr spectrum in fig . [
fig : dpr ] .
the agreement is remarkably good .
in addition to giving access to the line shape it means that via our theory one can use the shaking method as a good thermometer in low temperature regime , since the curve giving the amplitude versus temperature ( the inset in fig .
[ fig : dpr ] ) is reasonably smooth and steep . to check this point we compare in our case the temperature determined by the fitting of the shaking curve with other estimates from entropy calculations @xcite and
find that the two results are perfectly consistent .
in this paper , we have described the charge excitation of strongly correlated fermionic systems in the spin - incoherent paramagnetic regime by a slave - particle representation and diagrammatic approach from the atomic limit .
this method allows us to take the finite temperature and trapping into account .
based on the spectrum functions of the doublon and holon , the analytical form of the dpr spectrum as a second - order response of the optical lattice modulation has been given , and extended to the homogeneous system of the trapped atom cloud by using the local density approximation .
in addition , substituting the parameters evaluated in the experiment @xcite , a comparison with the experiment has been made .
although temperature has not been directly measured in experiments , it has been determined as an optimization parameter which is controlled to fit the experimental data .
the result has been in agreement with the experimental data , which shows that one can use the lattice modulation spectroscopy as a thermometer .
our method has potential extension such as su(@xmath58 ) higher symmetric atom systems realized in alkaline - earth - metal atom experiments .
@xcite we are grateful to c. berthod , t. esslinger , d. greif , n. kawakami , a. koga , l. pollet , l. tarruell , and t. uehlinger for fruitful discussions . in addition , we thank d. pekker for contributions at an early stage of this work and t. esslinger s group for making their data available .
ed acknowledges support from the darpa ole program , harvard - mit cua , nsf grant no .
dmr-07 - 05472 , afosr quantum simulation muri .
this work was supported by the swiss national foundation under manep and division ii .
in order to clarify the parameter regime to justify the slave - particle approach , we make a comparison with the exact calculation in the atomic limit .
then we focus on the single site occupations as a function of temperature and chemical potential , which is identical to the densities of the slave boson , holon , and doublon .
let us perform the atomic limit calculation .
we consider the atomic hamiltonian ( [ eq : atomic - hamiltonian ] ) . then the partition function is calculated as @xmath115 where @xmath116 and @xmath117 . to calculate the densities of the slave boson , holon , and doublon
, we introduce the projection operators . in the original fermion picture ,
the projection operators of the slave boson , holon and doublon state are , respectively , written as @xmath118 , @xmath119 , and @xmath120 .
they obviously obeys the constraint @xmath121 . taking the statistical average of these projection operators , we obtain the following exact result of slave - particle densities : @xmath122 [ eq : atom - limit - calc ] on the other hand , in the case of the slave - particle approach combined with the mf solution of the lagrange multiplier @xmath42 ,
the slave - particle densities are given as @xmath123 [ eq : mf - slave ] where @xmath124 , @xmath125 , and @xmath126 , and @xmath42 is a solution of the mf constraint equation ( [ eq : sce - lambda ] ) .
for @xmath40 the mf lagrange multiplier @xmath42 is analytically obtained as shown in eq .
( [ eq : mf - lagrange - multiplier ] ) .
the comparison of the exact result ( [ eq : atom - limit - calc ] ) and the mf approach ( [ eq : mf - slave ] ) is shown in fig .
[ fig : slave - density ] .
here we show the calculation of the kinetic energy correlation function @xmath127 in detail , and derive eq .
( [ eq : h_k - correlation - function0 ] ) given in sec .
[ sec : dpr ] . using the expression of the kinetic energy hamiltonian ( [ eq : tunnelinghamiltonian ] ) , the correlation function @xmath127 is straightforwardly written as @xmath128 [ h_{l}^\dagger(0)h_{m}(0)-d_{l}^{\dagger}(0)d_{m}(0 ) ] \rangle \nonumber \\ & \quad + \langle t_{\tau}a_{ij}^{\dagger}(\tau)a_{lm}(0 ) \rangle \langle t_{\tau}d_{j}(\tau)d_{m}^{\dagger}(0 ) \rangle \langle
t_{\tau}h_{i}(\tau)h_{l}^{\dagger}(0 ) \rangle \nonumber \\ & \quad + \langle t_{\tau}a_{ij}(\tau)a_{lm}^{\dagger}(0 ) \rangle \langle t_{\tau}h_{i}^{\dagger}(\tau)h_{m}(0 ) \rangle \langle
t_{\tau}d_{j}^{\dagger}(\tau)d_{l}(0 ) \rangle \ } , \label{eq : kinetic - energy - correlation } \end{aligned}\ ] ] where the correlation functions between the operators @xmath129 and @xmath130 have been supposed to be zero because the slave bosons do not condense in this case .
the autocorrelations for @xmath129 and @xmath130 should be finite , and in the calculation the spin - incoherent assumption ( [ eq : spin - incoherent - assumption ] ) is applied in the same way as in sec .
[ sec : diagrammatic - analys ] . as a result
, the autocorrelations are calculated as follows : @xmath131 as seen in sec .
[ sec : diagrammatic - analys ] the doublon and holon propagators turn out to be local : namely , @xmath132 we substitute eqs .
( [ eq : ff - correlation])- ( [ eq : doublon - propagator ] ) into eq .
( [ eq : kinetic - energy - correlation ] ) , and then the correlation function is rewritten as @xmath133 \nonumber \\ & \quad - [ \bar{\mathcal{g}}_{\mathrm{b}}(-\tau)]^2 \mathcal{g}_{\mathrm{h}}(\tau ) \mathcal{g}_{\mathrm{d}}(\tau ) - [ \bar{\mathcal{g}}_{\mathrm{b}}(\tau)]^2 \mathcal{g}_{\mathrm{h}}(-\tau ) \mathcal{g}_{\mathrm{d}}(-\tau ) \ } , \end{aligned}\ ] ] where the following two particle correlation functions have been defined as @xmath134 29ifxundefined [ 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 * * , ( ) link:\doibase 10.1103/physrevlett.92.130403 [ * * , ( ) ] @noop * * , ( ) link:\doibase 10.1103/physrevlett.106.145302 [ * * , ( ) ] @noop * * , ( ) @noop * * , ( ) link:\doibase 10.1103/physrevlett.97.050402 [ * * , ( ) ] @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) link:\doibase 10.1103/physrevlett.102.065301 [ * * , ( ) ] link:\doibase 10.1103/physreva.84.021607 [ * * , ( ) ] @noop * * , ( ) link:\doibase 10.1103/physrevb.81.155118 [ * * , ( ) ] link:\doibase 10.1103/physrevb.46.6328 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.104.180401 [ * * , ( ) ] http://stacks.iop.org/1367-2630/11/i=10/a=103033 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.105.190401 [ * * , ( ) ] @noop * * , ( ) link:\doibase 10.1103/physrevlett.105.030402 [ * * , ( ) ] | we study theoretically lattice modulation experiments with ultracold fermions in optical lattices .
we focus on the regime relevant to current experiments when interaction strength is larger than the bandwidth and temperature is higher than magnetic superexchange energy .
we obtain analytical expressions for the rate of doublon production as a function of modulation frequency , filling factor , and temperature .
we use local density approximation to average over inhomogeneous density for atoms in a parabolic trap and find excellent agreement with experimentally measured values .
our results suggest that lattice modulation experiments can be used for thermometry of strongly interacting fermionic ensembles in optical lattices . |
murphy s law is not a law in the formal sense ; instead it is an interesting satirical statement : _ if something can go wrong then it will definitely go wrong _ @xcite .
the origin of the statement is unknown ( even murphy himself ! ) , yet it is not very hard to find it relevant in various situations , either in its rudimentary form or in the form of some derivative .
popular science often relates this statement to the second law of thermodynamics , which states that the entropy of the universe will always increase with time @xcite .
the analogy between these two seemingly very dissimilar statements seems to crop up from the idea that left to themselves , things tend to get disorganized with time and murphy comes into the picture , albeit in a modified form , stating that bad will eventually turn worse . in the realm of philosophy these statements may seem connected , however from a scientific viewpoint they are clearly unrelated , as ( _ i _ ) one is a universal law , whereas the other , a rather ambiguous statement ; ( _ ii _ ) the second - law strictly talks about energy dissipation in physical processes and their irreversibility , whereas murphy s statement ( ` law ' is a misnomer in this case ) is just a popular imagination , and does not even remotely relate to any of the physical or mathematical concepts to rely upon .
surprisingly , given these discrepancies , we still find it very interesting when the law is formulated as a mathematically equivalent statement . in this paper
, we try to _ actually _ relate murphy s statement with the second law of thermodynamics , and draw inspirations from energy and entropy .
we check the correctness of the statement from a scientific standpoint , by first constructing a mathematically equivalent statement using concepts from mathematical logic and probability theory .
next , we understand the physical essence of the terms ` rightness ' and ` wrongness ' from the context of energy and time , using the second law and principle of least action . finally , we relate our findings to check the truthfulness of murphy s statement .
we observe that what initially seemed to be completely unrelated statements - the second law and murphy s statement - _ indeed _ have some interesting tales to tell .
murphy s statement is more of a psychological statement having philosophical connotations of ` rightness ' and ` wrongness ' .
a statement like this is not only hard to evaluate but even more challenging to prove ( or disprove ) .
the first step in any proof strategy lies in the rigorous formulation of the statement which is under scrutiny , and hence it is necessary that we try to formulate this rather anecdotal philosophy into a logical one .
doing so lets us set limits to the premise of the statement , as well as to its domain of application .
however , in order to proceed we need to make some underlying assumptions .
the idea of using propositional logic to build a mathematically coherent statement , even before attempting to prove or disprove it , is a necessity .
whereas , assigning an additional element of probability to the statement(s ) is actually non - intuitive , because ( _ i _ ) murphy s speaks about occurrence of events and in mathematics , occurrence is always associated with _ chance _ ; ( _ ii _ ) the presence of the intensifier , _ ` definitely ' _ , in the linguistic construction of the statement itself , further strengthens our intuition of associating the statement(s ) with finite values of probability . logically speaking , murphy s statement can be broadly divided into two distinct components : * 1*. ( _ if _ ) something can go wrong , and * 2*. ( _ then _ ) it will definitely go wrong .
in propositional logic , this can be viewed as a causal chain of statements , @xmath1 , where @xmath2 is the first component , @xmath3 the second , and @xmath2 causes @xmath3 .
such an argument is known as _ modus ponnens _
@xcite which asserts that the occurrence of @xmath3 is dependent upon the occurrence of @xmath2 , or if @xmath2 happens , @xmath3 definitely happens
. it can be represented mathematically as @xmath4 .
it was noted earlier that the above statements @xmath1 have an inherent probability associated with them .
the presence of an intensifier , ` definitely ' further affirms the occurrence of the latter half of the statement , irrespective of the magnitude of the chances of occurring of the former half .
mathematically , if the probability of occurrence of @xmath2 is _ finite _ ( remember , the probability that @xmath5 occurs is also finite ) then @xmath3 occurs _ almost surely _ , _
i.e. _ , with probability one ( note that , even though we are interested in the happening of something which is not desired ) . what we earlier stated as a _
modus ponnens _ , interestingly transforms into _ modus tollens _ once we associate finite probabilities to the occurrence of the events ( _ modus ponnens _ is the law of affirming by affirming , therefore we affirm @xmath2 in order to justify @xmath3 , and murphy s statement affirms @xmath3 by associating finite chances of occurrence of @xmath2 ) @xcite .
according to the law of contrapositive : ( _ modus tollens _ ) , @xmath6 , or if probability of occurrence of @xmath2 is _ finite _ then @xmath3 occurs _ almost surely _ and @xmath3 did not occur _ almost surely _ enforces the fact that @xmath2 did not have a finite chance of occurring in the first place , or @xmath7 @xcite .
this is the form of murphy s statement that we will adhere to throughout this paper .
also , in the following sections of this paper , we will analyze murphy s statement using fundamental laws of physics , the least action principle and the second law of thermodynamics .
it may seem puzzling at first as to how energy and entropy come into the picture , but a deeper inspection will reveal the intimate connection that murphy s statement holds with respect to a system , process and entropy . the connection between murphy s statement and thermodynamics is the central theme of this paper . as proposed earlier , when we talk about murphy s statement we refer to the probability of occurrence of any event in space and time .
we thus make a transition from a system - theoretic perspective to a process - theoretic one , and once we do that we inevitably include the second law into the picture .
where , thermodynamics is a branch of science dealing with energy interactions with systems , energy conversion and energy degradation ; it has also given birth to one of the most fundamental laws of nature , known as the second law or the entropy principle @xcite .
the second law deals with processes in nature , and acts as a standard test to justify the occurrence or nonoccurrence of any phenomenon .
every process ( phenomenon ) in nature can be visualized as a flow of energy . when this flow is spontaneous , _ i.e. _ , the driving force behind the flow is the presence of gradients ( like , energy , temperature , pressure , concentration , etc . ) , then such a flow is realized with an equivalent increase in entropy @xcite .
the flow of energy during any arbitrary process when coupled with the dimension of time gives rise to the notion of _ action _ , formally represented as @xmath8 , where the pair , @xmath9 is the generalized position - momentum pair , @xmath10 is the time , @xmath11 is the change in energy along a specific path ( in this case @xmath12 ) , @xmath13 and @xmath14 are the kinetic and potential energies respectively and @xmath15 , the lagrangian @xcite .
there are several interpretations of the action principle , for example , the laws of physics when formulated in terms of the action principle identify any natural process as the one in which energy differences are leveled off in the least possible time ( maupertius formulation ) @xcite ; among all existing possible paths , the path of least action is the one along which a natural process must proceed ( hamiltonian formulation ) @xcite . on the energy landscape , there is a unique trajectory that directs ( energy ) flows along the lines of steepest descent .
a classical example of this principle can be observed in case of the brachistochrone problem .
the principle of least action , thus dictates the existence of a unique path out of numerous possible trajectories ( see figure 1 ) . while some are thermodynamically feasible , others are not . from a thermodynamic perspective decrease in action
is equivalent to an increase in entropy ( we restrict ourselves to non self - organizing systems and processes ) , or along a trajectory , both the quantities can be represented along mutually orthogonal directions but with opposite signs , @xmath16 @xcite .
murphy s statement vaguely mentions processes , but stresses more on their qualitative aspect of ` rightness ' and ` wrongness ' .
it does not talk about the system , per se , undergoing the process , nor does it go into the details of the mechanism underlying the process ; it basically concentrates only on the qualitative aspects of the outcomes .
in order to test the correctness of the statement , it is absolutely necessary to look into the details of the process , as focusing on the finer details of the process will enable us to justify the outcomes .
a deeper inspection is needed when we talk about processes and trajectories ( pathways to achieve a particular outcome during a process ) .
when we talk about murphy s statement , we must invariably refer to processes and the trajectories ( pathways ) along which these processes occur ( flow ) . in order to model a physical phenomenon into a thermodynamic one , we must explicitly distinguish between a system , surroundings and the set of processes that the system undergoes . in the algebraic formulation of variational thermodynamics ,
these terms are defined in a detailed , mathematically rigorous yet an abstract fashion @xcite .
roughly speaking , a system is a set of all those elements which are of current interest , and there exists a boundary that separates the system from its surroundings , where surroundings ( relative to the system ) refer to everything else in the universe except the system and the system boundary .
a process is a change of state of a subset of system elements or the system itself in time .
mathematically , a process can be represented as a map that captures the changes incurred by the system or a subset of a system .
we define a system undergoing a process by a tuple , ( @xmath17 ) , where the elements of @xmath18 are the system elements / constituents / agents , and the elements of @xmath19 are processes @xcite .
we identify an event as a _ spontaneous _ process in space - time if it proceeds along the path that reduces action and increases entropy @xcite .
the principle of least action is that tool which helps us in finding that particular trajectory out of a set of uncountable possibilities along which an event must proceed . using the standard notation
, we define action for an arbitrary element in a system undergoing an arbitrary process , as @xmath20 , where @xmath21 , the set of all system elements in @xmath18 , and @xmath22 , the set of all arbitrary processes , @xmath23 @xcite . for any arbitrary process , @xmath24 there exist numerous possibilities or trajectories , @xmath25 .
we define the set @xmath26 to be the set of all such trajectories , @xmath25 for the corresponding process , @xmath24 @xmath27 and @xmath28 ( @xmath29 , @xmath30 , and @xmath31 are index sets ) .
it is now crucial to define what exactly is meant by ` right ' and ` wrong ' events as this is the central idea of murphy s statement .
we define two subsets of trajectories , @xmath32 and @xmath33 , such that both @xmath34 .
the trajectories which are thermodynamically feasible form the set , @xmath32 , and the ones which are not , form the set , @xmath33 . therefore , the elements of @xmath26 are either labeled by @xmath35 or @xmath36 ( for representation purpose only ) .
the right trajectories are denoted by @xmath37 , and the wrong ones by @xmath38 . clearly , the subsets , @xmath32 and @xmath33 partition the set , @xmath26 , so @xmath39 and @xmath40 ( see figure 2 ) .
it is not very hard to see that the tuple , ( @xmath41 ) forms a @xmath42-algebra be a collection of sets of @xmath26 , then ( @xmath33 , @xmath32 ) @xmath43 , ( @xmath44 , @xmath45 ) @xmath43 , @xmath46 , @xmath47 , and finally @xmath48 and @xmath49 .
so , ( @xmath41 ) is a @xmath42-algebra . ] where @xmath50 is a collection of sets of @xmath26 @xcite . we can define a probability measure , @xmath51 , such that @xmath52 and @xmath53 , then the triple , ( @xmath54 ) forms a probability space . , and the action @xmath55 ( for @xmath56 particle ) undergoing @xmath57 process along @xmath58 trajectory , @xmath25 @xcite.,scaledwidth=40.0% ]
as stated above , every element in @xmath26 is a possible trajectory , and it is not surprising that the set is unbounded because there exist infinitely many possible trajectories or sequence of trajectories for a given outcome ( feynman s path integral formulation argument ) @xcite .
furthermore , we associate some _ finite _ ( however small or large ) probability to every trajectory ( trajectories are basically the worldlines of a particular outcome ) .
thus , without any loss of generality , we can assume that the set , @xmath33 , is _ countably _ infinite , whereas the set , @xmath32 , is finite .
it is even more interesting to note that the set , @xmath33 is , in fact , _
uncountable_. since the elements of the set @xmath33 represent all those trajectories which are thermodynamically infeasible , we can make any arbitrary combination of @xmath59 ( or a mix of @xmath59 and @xmath60 , because it does nt matter if a process is thermodynamically feasible in a later stage if it is infeasible in the very first instant ! ) , and always come up with a new infeasible trajectory , say @xmath61 . as the set @xmath33 is infinite , we come up with an argument similar to cantor s diagonalization argument for uncountability of reals @xcite , and prove that @xmath33 is uncountable because for every increment in @xmath62 , we can always find a new element , @xmath63 based on the combination of the diagonal elements ( in this case , @xmath64 ) and this will happen infinitely many times , _
i.e. _ , @xmath65 for the physical explanation of the above argument , we consider the case of a chemical reaction where substrates are converted into products . in order that the reaction is thermodynamically feasible
, a minimum energy barrier ( energy hill ) called activation energy , @xmath66 , has to be surpassed @xcite .
those trajectories along which this energy threshold is successfully realized can be visualized as the ` right ' trajectories and vice - versa .
even the so called ` right ' trajectories have further thermodynamic constraints , so basically , we can refine our intuition and come up with a finite number of possibilities , or maybe in some exceptional cases , _ atmost _ countable ` right ' possibilities and _ always _ uncountable ` wrong ' possibilities .
let @xmath67 be an action map such that , @xmath68 , in this case , @xmath69 , and @xmath70 $ ] .
the function , @xmath67 , takes every trajectory , @xmath25 from the set @xmath26 , and calculates the action along that trajectory to the one along which the magnitude of action is maximum , which forms the transformed set , @xmath71 .
in fact , the map , @xmath72 is a homeomorphism are the trajectories in @xmath26 .
the action map is similar to the pay - off function in game theory @xcite and sigma - profiles in multi - agent complex system modeling @xcite .
] , @xmath73 ( see figure 2 ) . from our previous arguments ,
there are certain trajectories which are absolutely infeasible ( although having a finite probability , however small ) , @xmath74 and few feasible ones , @xmath75 .
thus , the transformed sample space , @xmath71 would take up all the values in the compact set , @xmath76 \subset \mathcal{r}$ ] .
let @xmath77 be a collection of subsets of @xmath71 , then the tuple @xmath78 forms a @xmath42-algebra .
the interval , @xmath79 $ ] is borel - measurable and the probability measure , @xmath80 in this case is the lebesgue measure ; the triple , @xmath81 forms a probability space with @xmath82 , and @xmath83 . from our previous argument based on the countability and uncountabilty of ` right ' and ` wrong ' trajectories , we can infer that the ` right ' trajectories in @xmath84 will cluster around @xmath85 $ ] , and would be countable in number , whereas the ` wrong ' trajectories would cluster around @xmath86 $ ] , and would be uncountably many .
the subset , @xmath32 in the transformed sample space , @xmath71 is denoted by @xmath87 .
similarly , the subset @xmath33 is denoted by @xmath88 , and @xmath89 and @xmath90 . on assigning the probability measure to the sets , @xmath91 and @xmath92 , we observe @xmath93 , since @xmath87 is countable , and @xmath94 , as @xmath88 is uncountable @xcite .
what we just proved above is that no matter what , the probability that a ` wrong ' trajectory is chosen is one , whereas the probability of choosing a ` right ' trajectory is zero , which in fact , supports murphy s statement ! and @xmath71.,scaledwidth=70.0% ] there is a fallacy in the above argument .
when we say there is a finite probability of any event to happen in some way , there is _ also _ a finite probability that the event might happen in some other way , that we and also murphy s statement ignore . as with any natural process , it can happen in infinitely many ways ( as seen above ) , and the _ few _ so - called ` correct ' trajectories get ` lost ' in the multitude of the uncountably many ` wrong ' ones , in spite of having finite chances of occurring .
we define two sequences , @xmath95 and @xmath96 .
the sequence , @xmath97 lies in the neighborhood of @xmath98 , and the sequence , @xmath99 , in the neighborhood of @xmath100 ( @xmath101 and @xmath102 ) . from our initial assumption , @xmath103 and @xmath104
are both finite for _ every _ @xmath62 in the respective sequences , @xmath105 and @xmath106 . based on our definition for ` right ' and ` wrong ' events , the terms in the sequences
, @xmath107 and @xmath108 will again be infinitely many .
following our intuitive idea , we can easily foresee the probability , for every element in the sequence , @xmath107 takes the form , @xmath109 , for every @xmath110 ( @xmath111 is not a distribution function , strictly because for some processes , the probability might decay , as @xmath112 , while for others , as @xmath113 , and so on ) .
we can see that however small the probability for any process to occur along a specific trajectory , @xmath107 be , it is finite .
since the probability for every trajectory is decaying with an increment in @xmath62 , the summation of the probabilities , @xmath114 ( because as we go nearer to the trajectories in the neighborhood of @xmath90 , their probability will decrease at an even faster rate , making the probability of the sum of infinitely many such trajectories finite ) . from the borel - cantelli lemma
, we can say that the probability of occurrence of these trajectories infinitely many times will be zero , and there are infinitely many of them @xcite .
so , the probability of their occurrence is zero .
conversely , for the sequence , @xmath97 , the probability distribution takes the form of @xmath115 , where @xmath116 for some @xmath117 , while the probability is _ exactly _ @xmath118 for finitely many trajectories .
thus , the summation of probabilities in this case , @xmath119 diverges because @xmath120 diverges . from the converse of the lemma ,
the probability of occurrence of these trajectories infinitely often will be one , and there are infinitely many of them and that such a trajectory ( from the set , @xmath32 ) will be chosen _
almost surely_. we reflect on our findings based on the logic behind murphy s statement , @xmath6 and observe that although ` wrong ' trajectories have finite probability , they do not occur , which is a contradiction to the proposition that , @xmath7 ( with finite probabilities ) .
murphy s statement often serves as an anecdote in many real - life circumstances . due to its affinity towards the negative aspects of any situation ,
it has often been misinterpreted to hold a resemblance with the second law of thermodynamics . in this paper
, we actually tried to relate both these statements and found quite striking results . as we mentioned earlier
, murphy s statement holds several philosophical connotation , s and appears as a generalized take on various things happening around us all the time . in order to test its scientific validity
, we need to construct it as a chain of mathematically consistent statements and derive some logical inference from them .
any proof is invalid if the statements to be proved or disproved are themselves logically incoherent and mathematically not sound .
once this is achieved ( in section 2.1 ) , we need to outline a proper proof strategy .
this is where the complexity associated with murphy s statement increases , the reasons being : ( _ i _ ) it deals with physical , observable processes , ( _ ii _ ) the outcomes of the processes could be anything , and depending upon external perturbations the outcomes and even the processes may change their course of action , and ( _ iii _ ) every outcome is plausible , _ i.e. _ , every process has a finite chance of occurring and along numerous pathways leading to numerous finite possible outcomes . in order to deal with these uncertainties , we made assumptions like : any process or any outcome always has some energy associated with it , and any observable outcome can be achieved in several ways , which we call trajectories or pathways . in order to deal with a complex situation like this , we look for inspiration in the most fundamental laws of nature , and thereby propose that those outcomes , which are thermodynamically feasible shall be achieved along those trajectories that minimize action and maximize entropy . using mathematical tools from probability theory and logic , we claim to disprove murphy s statement in this paper . during the course of disproving the statement , we clearly showed the fallacy which is often the mis - interpreted truth behind the statement ( section 3.1 ) .
we saw that there always exists an equivalent finite chance of an event to occur in different ways when it occurs in a certain way , which murphy s statement generally ignores .
we would further like to add that the intensifier , ` definitely ' in the construction of murphy s statement enforces the fact that if something can go wrong , it will go wrong for sure .
but we have seen earlier that the probability of something happening along the wrong way , as well as along the right way is finite , which raises a situation of _ post - hoc _ fallacy in the statement .
this type of logical fallacy is known as _ post - hoc ergo propter hoc _ @xcite .
a similar situation is observed in the case of the law of large numbers and the central limit theorem . according to the law of large numbers ( lln ) , given sufficiently large number of trials , the average of the results thus obtained , shall converge ( almost surely ) to the expected value or mean of the experiment .
similarly , the central limit theorem ( clt ) states that the arithmetic mean of a sufficiently large number of iterates of independent random variables ( each with a well - defined mean and variance ) will be approximately normally distributed , regardless of the underlying distribution @xcite .
both clt and lln seem to miss out on a crucial piece of information , _
i.e _ , the information of the _ individual _ outcomes or the _ individual _ distributions respectively !
thus , murphy s statement seems to be a far - fetched generalization of clt and lln in daily life .
however , we must not forget that murphy s statement has a physical essence associated with it ( _ process _ ) , and therefore needs to be analyzed on a different scale .
since it also deals with large number of outcomes and events , a statistical approach does sound good but a statistical formulation often fails to capture certain fine - tuned characteristics of the system . our methodology , thus holds strong promises in the study of complex systems , where every motion of a particle counts and little perturbations may drive the system from an organized state to a state of chaos @xcite . during the formulation of murphy s statement
, we also focused more on the process and the trajectories , and less on the system and its constituting elements .
it is not necessary for the system or a system element to be a physical particle , rather it is immaterial and irrespective of the physical definition ( or topology ) of the system for our reasoning to hold . by a process
, we refer to the change in state of the system , which is the same as the change in outcome(s ) due to the system undergoing a process in time .
as far as our assumption - any process is associated with energy flows on the energy landscape - holds good , the central idea of the paper will hold true .
further , we mathematically prove that nature orders the occurrence of processes and their outcomes based on energy and entropy considerations . thus a reinterpretation of murphy s statement would read as : irrespective of all the possible outcomes and their qualitative aspects of ` rightness ' and ` wrongness ' anything that shall happen will definitely happen along the way that must minimize action .
the author would like to thank the reviewers for their valuable feedback highlighting the merits and drawbacks of this paper .
their constructive criticism and suggestions have made this paper much more informative and interesting to read .
the author would also like to thank his colleagues , sheetal surana and catherine nipps , for their inputs and lengthy discussions on the topic of this paper . | murphy s law is @xmath0 a law in the formal sense yet popular science often compares it with the second law of thermodynamics as both the statements point toward a more disorganized state with time . in this paper , we first construct a mathematically equivalent statement for murphy s law and then disprove it using the intuitive idea that energy differences will level off along the paths of steepest descent , or along trajectories of least action . * pacs * : 89.20.-a , 89.75.-k , 05.70.-a , 02.30.-f + * keywords * : murphy s law , principle of least action , second law of thermodynamics , probability measure |
quasi - one dimensional material formed of weakly coupled chains are known to posses low temperature instability towards a state with an incommensurate static modulation of the electronic charge and of the lattice displacement known as charge density wave ( cdw ) phase .
cdws behave as elastic systems @xcite provided coulomb interaction are screened .
when coulomb interaction are not sufficiently screened elasticity becomes non - local @xcite and disorder becomes long ranged correlated .
an important question is whether the competition of the non - local elasticity with the long - range random potential results in the complete destruction of the crystalline order , or whether some remnants of the underlying periodic structure remain observable .
a powerful experimental technique able to characterize the structural properties of cdw with impurities is afforded by x - ray diffraction . in this paper
we derive the static displacement correlation functions and x - ray intensity of the cdw with charged impurities and we highlight the similitude of the latter to the x - ray intensity of smectic - a liquid crystals subjected to thermal fluctuations .
finally , we discuss the experimental significance of our result and suggest that the smectic - like correlations should be observable in experiments on @xmath2 .
we consider an incommensurate cdw , in which the electron density is modulated by a wave vector @xmath3 incommensurate with the underlying crystal lattice . in this phase , the electron density has the following form @xcite : @xmath4 where @xmath5 , is the average electronic density . the second term in eq .
( [ eq : rhoexp ] ) is the long wavelength density and corresponds to variations of the density over scales larger than @xmath6 .
the last oscillating term describes the sinusoidal deformation of the density at a scale of the order of @xmath7 induced by the formation of the cdw with amplitude @xmath8 and phase @xmath9 .
the low energy properties of the cdw are described by an elastic hamiltonian for the fluctuations of the phase @xmath9 .
when the coulomb interactions are unscreened this hamiltonian becomes non - local .
choosing the @xmath10 axis aligned with @xmath11 the total hamiltonian in three dimensions reads @xcite : @xmath12 \nonumber\\end{aligned}\ ] ] where @xmath13 is the fermi velocity and @xmath14 is the number of chains per unit surface that crosses a plane orthogonal to @xmath3 .
the velocity of the phason excitations parallel to @xmath3 is @xmath15 with @xmath16 the effective mass of the cdw and @xmath17 the mass of an electron . @xmath18
denote the phason velocities in the transverse directions .
the lengthscale @xmath19 is defined by : @xmath20 where we have assumed for simplicity an isotropic dielectric permittivity @xmath21 of the host medium and @xmath22 is the charge of the electron .
when a non - zero concentration of charged impurities is present , the hamiltonian becomes : @xmath23 where @xmath24 is the density of impurities at position @xmath25 .
we assume gaussian distributed impurities with @xmath26 , where @xmath27 is the measure of the disorder strength . using the expression of the density in ( [ eq : rhoexp ] ) ,
the interaction with disorder comprises two parts , one that describes the interaction with long wavelength fluctuations of the density ( forward scattering ) , and a second part that describes interactions of impurities with density fluctuations of wavelength @xmath7 ( backward scattering ) . in @xcite
we have shown that even in the case of long range disorder , the backward scattering terms behave essentially like their short ranged ( neutral impurities ) counterparts . in that case
, it is known that in three dimensions , the contribution of the backward scattering to the phase correlations grows slowly as @xmath28.@xcite the contribution of the forward scattering term is obtained from the hamiltonian : @xmath29 , \nonumber \\
\end{aligned}\ ] ] and reads in the limit of @xmath30 : @xmath31 where @xmath32 is a momentum cutoff and : @xmath33 as a result , in the presence of charged impurities , the forward scattering terms generate the leading contribution to the phase correlations .
it is well known that the presence of a cdw in a compound is associated with the appearance of two satellites at positions @xmath34 around each bragg peak.@xcite the intensity profiles of these satellites give access to the structural properties of the cdw . the main contribution to the x - ray satellite intensity
is given by @xmath35 , which is symmetric under inversion around the bragg vector @xmath36.@xcite the intensity @xmath37 is given by the following correlation function : @xmath38 where @xmath39 is an average x - ray scattering amplitude and @xmath40 is the amplitude of the lattice modulation induced by the presence of the cdw . using the result given in eq.([eq : correlator ] ) we extract the small @xmath41 behavior of the intensity peak : @xmath42 the intensity @xmath43 is divergent for @xmath44 but is finite for @xmath45 , i.e. for strong disorder .
we note that these intensities are remarkably similar to those of a disorder - free smectic - a liquid crystal@xcite at positive temperature .
in fact , the expression of the exponent @xmath46 eq .
( [ eq : kappa - def ] ) is analogous to the expression ( 5.3.12 ) in @xcite , with the disorder strength @xmath27 playing the role of the temperature @xmath47 in the smectic - a liquid crystal .
let us turn to an estimate of the exponent @xmath46 appearing in the intensities to determine whether such smectic - like intensities are indeed observable in experiments .
a good candidate is the blue bronze material @xmath48 which has a full gap so that coulomb interactions are unscreened at low temperature .
we use the parameters of @xcite : @xmath49 and a relative permittivity of @xmath50 , so that @xmath21 is equal to the permittivity of the vacuum . for the doped material @xmath51
, we find that the disorder strength can be expressed as a function of the doping and obtain : @xmath52 for the crystal parameters , @xmath5318.25 , @xmath547.56 , @xmath559.86 , @xmath56117.53@xmath57 @xcite , with 20 molybdenum atoms per unit cell , and a doping @xmath58 , the disorder strength @xmath59 .
moreover , using the experimental bounds of the velocities : @xmath60 and @xmath61 , we find that @xmath46 is in the range @xmath62 $ ] .
therefore , the smectic - like order should be observable in x - ray diffraction measurements on this material . | when unscreened charged impurities are introduced in charge density wave system the long wavelength component of the disorder is long ranged and dominates static correlation functions .
we calculate the x - ray intensity and find that it is identical to the one produced by thermal fluctuations in a disorder - free smectic - a .
we suggest that this effect could be observed in the blue bronze material k@xmath0moo@xmath1 doped with charged impurities such as vanadium . |
the discovery of the c@xmath4 fullerene@xcite and the remarkable physical and chemical properties resulting from its unique topology and electronic structure have motivated an extraordinary research activity in past years.@xcite one of the recent questions of interest in this field is the possibility of producing even smaller c or si cagelike clusters and to synthesize novel solids using them as building blocks .
thus , several experimental and theoretical studies have been performed in order to elucidate the complex mechanisms of formation of these nanostructures.@xcite in the particular case of c@xmath0 , which is expected to be the smallest fullerene , experiments indicate that the dominant species in laser - vaporization of graphite are ring structures , while theory yields different isomers depending on the calculation method.@xcite hartree - fock ( hf ) studies predict that the ring is the lowest - energy isomer , followed by the bowl ( a substructure of the c@xmath4 fullerene ) and then by the dodecahedral cage.@xcite in contrast , the local density approximation ( lda ) to density - functional theory ( dft ) yields the cage to be more stable than the bowl and the ring , whereas in a generalized gradient approximation ( gga ) the same ordering as hf is obtained.@xcite these investigations , as well as quantum monte carlo ( qmc ) calculations,@xcite indicate that non - trivial electron correlation effects play a central role in the structural and electronic properties of these nanoclusters .
recently , prinzbach and coworkers succeeded to produce the cage - structured fullerene c@xmath0 in the gas phase starting from the perhydrogenated form c@xmath0h@xmath0.@xcite this isomer , which is not formed spontaneously in carbon condensation or cluster annealing processes , has a sufficiently long lifetime . moreover , photoelectron spectrum ( pes ) measurements have been performed from which its cage - like structure has been inferred.@xcite theoretical studies by saito and miyamoto have reproduced the pes thereby confirming the structure assignment given in experiment.@xcite in addition , experimental evidence has been provided on the oligomerization of c@xmath0 fullerenes@xcite and on the formation of a solid phase of c@xmath0 dodecahedra@xcite which , unlike c@xmath4 solids , involve strong c - c bonds between the pentagons of neighboring units .
besides its experimental realization , the topology of c@xmath0 appears to be a particularly attractive physical situation for theoretical investigations of correlated itinerant electrons in cagelike clusters with pentagonal rings .
similar studies on c@xmath5 and c@xmath4 clusters have already been performed in relation to alkali - metal - doped c@xmath4 solids like k@xmath6c@xmath4 and rb@xmath7c@xmath4,@xcite since one expects that the basic physics behind the superconducting and optical properties of these materials should be captured at the scale of an individual molecular constituent .
the smaller fullerenes have attracted a special attention in this context due to the perspective of achieving larger superconducting transition temperatures than the c@xmath4-based solids .
indeed , a number of theoretical investigations predict a significant enhancement of the electron - phonon coupling as the size of the fullerene is reduced from c@xmath4 to c@xmath8 , c@xmath9 , and finally c@xmath0.@xcite consequently , theoretical studies aiming at understanding the properties of correlated electrons in small fullerene clusters are of considerable interest .
lattice models have been used to determine low - energy properties of c@xmath4 which derive from the outermost half - filled @xmath10-electron cloud . in particular ,
the spin - density distribution on the buckyball has been analyzed in a series of papers.@xcite coffey and trugman@xcite calculated the ground - state spin configuration using a classical antiferromagnetic ( af ) heisenberg hamiltonian corresponding to the strongly correlated limit of the hubbard model .
they found that the lowest - energy spin structure shows a nontrivial noncollinear order which minimizes magnetic frustrations within each pentagonal ring , keeping strong af short - range order in bonds connecting nearest neighbor ( nn ) pentagons.@xcite the spin structure in c@xmath4 was also investigated by means of hubbard or pariser - parr - pople hamiltonians at half - band filling , taking into account on - site and inter - site coulomb interactions within the unrestricted hartree - fock ( uhf ) approximation.@xcite a common result of these investigations is the presence of a magnetic instability for a critical value @xmath11 of the on - site coulomb repulsion yielding a magnetic order which resembles that of the classical af heisenberg model .
in addition remarkable noncollinear spin arrangements and charge - density redistributions have been obtained in mean - field calculations as a function of electron and hole doping.@xcite qmc simulations on c@xmath4 at half - band filling and exact diagonalization studies of cagelike c@xmath5 support the existence of non - vanishing short - range spin correlations.@xcite however , it should be recalled that there is no experimental evidence for an spontaneous symmetry breaking . instead
, the spin - density - wave instability should be interpreted as an indication of fluctuating spin - spin correlations within the cluster .
it is therefore interesting to improve on the treatment of correlations in order to quantify the possible effects of artificial symmetry breakings on the electronic properties .
the purpose of this paper is to investigate the ground - state properties of correlated electrons on a c@xmath0 cagelike cluster by using the hubbard hamiltonian and a recently developed lattice density - functional theory ( ldft).@xcite previous applications of this approach to 1d and 2d lattices have yield quite accurate ground - state properties for all band fillings and interaction regimes .
a particularly interesting feature in the present context is that ldft does not necessarily involve a symmetry breaking in order to account for the effects of electron correlation and localization in the strongly interacting limit.@xcite thus , it appears as an appropriate means of improving on uhf calculations .
moreover , besides the aspects specifically related to c@xmath0 and its topology , the present calculations should be of interest as a finite cluster application of new density - functional approaches to lattice fermion models.@xcite the body of the paper is organized as follows . in sec . [
sec : teo ] the model hamiltonian and the main steps in the formulation of ldft are briefly recalled .
results for the ground - state energy , kinetic and coulomb energies , local magnetic moments , and charge excitation gap of c@xmath0 are presented and discussed in sec .
[ sec : res ] as a function of coulomb repulsion @xmath1 and for electron or hole dopings @xmath2 close half - band filling ( @xmath3 ) .
finally , sec .
[ sec : disc ] summarizes the main conclusions and discusses some perspectives of extensions in view of applications to lower symmetry structures .
we consider the hubbard hamiltonian@xcite @xmath12 on the c@xmath0 structure illustrated in fig .
[ fig : struct ] . in the usual notation
, @xmath13 ( @xmath14 ) refers to the creation ( annihilation ) operator of an electron with spin @xmath15 at site @xmath16 , and @xmath17 to the corresponding number operator .
the first term is the kinetic - energy operator with hopping integrals @xmath18 for nns and @xmath19 otherwise .
the second term takes into account intraatomic interactions by means of the on - site coulomb repulsion integral @xmath20 .
given the lattice structure and the number of atoms @xmath21 , the model is characterized by the number of electrons @xmath22 , or doping @xmath23 , and by the ratio @xmath1 . cluster . ] in order to determine the ground - state properties we consider a density - functional approach to electron correlations in a lattice , in which the fundamental variable is the single - particle density matrix @xmath24.@xcite the ground - state energy @xmath25 and density matrix @xmath26 are determined by minimizing the energy functional @xmath27 = e_k[\gamma ] + w [ \gamma]\ ] ] with respect to @xmath24 . the first term in eq .
( [ eq : e ] ) is the kinetic energy @xmath28 = \sum_{ij } t_{ij } \gamma_{ij } \;\ ] ] associated with the electronic motion in the lattice .
the second term is the interaction - energy functional , formally given by@xcite @xmath29 = { \min_{\psi\to\gamma } } \left [ u \sum_i \langle \psi [ \gamma ] | \hat n_{i\uparrow } \hat n_{i\downarrow}| \psi [ \gamma ] \rangle \right ] \ ; , \ ] ] where the constrained minimization runs over all @xmath30-particle states @xmath31 \rangle$ ] that satisfy @xmath32 | \ ;
\sum_\sigma \hat c_{i \sigma } ^{\dagger}\hat c_{j \sigma } \ ; @xmath33 .
thus , @xmath34 represents the minimum value of the interaction energy compatible with a given degree of electron delocalization or density matrix @xmath24 .
the universal functional @xmath35 $ ] , valid for all lattice structures and hybridizations , can be considerably simplified if the hopping integrals are short ranged .
for example , if only nn hoppings are considered as in the present case , the kinetic energy @xmath36 is independent of the density - matrix elements between sites that are not nns . therefore , the constrained search in eq .
( [ eq : wex ] ) may be restricted to the @xmath31 \rangle$ ] that satisfy eq .
( [ eq : rep ] ) only for @xmath37 and for nn @xmath38 .
moreover , for highly symmetric clusters like c@xmath0 , one has the same equilibrium value of @xmath24 for all nn pairs @xmath38 , and @xmath39 for all sites @xmath16 .
the interaction energy can then be regarded as a simple function @xmath40 of the density - matrix element between nns .
however , notice that restricting the minimization constraints in eqs .
( [ eq : wex ] ) and ( [ eq : rep ] ) to nn @xmath24 also implies that @xmath34 loses its universal character , since the nn map and the resulting dependence of @xmath34 on @xmath24 are in principle different for different lattice structures.@xcite two previously investigated scaling approximations to the interaction energy @xmath34 of the hubbard model shall be used for the calculations.@xcite in the first one the functional dependence is derived from the exact solution of eq .
( [ eq : wex ] ) for the hubbard dimer .
it is given by@xcite @xmath41 where @xmath42 refers to the hartree - fock energy and @xmath43 measures the degree of electron correlation in a nn bond @xmath38 . here
, @xmath44 stands for the largest possible value of the bond order @xmath24 for a given @xmath45 , band filling @xmath46 , and lattice structure .
it represents the maximum degree of electron delocalization regardless of correlations .
@xmath47 refers to the strongly correlated limit of @xmath24 , i.e. , to the largest nn bond order that can be achieved under the constraint of vanishing @xmath34 . for half - band filling @xmath48 , while for @xmath49 , @xmath50.@xcite thus , @xmath51 in an uncorrelated state , and @xmath52 in the strongly correlated limit corresponding , for example , to a fully localized or fully spin - polarized state ( nagaoka state ) .
physically , the scaling hypothesis underlying eq .
( [ eq : w2 ] ) means that the relative change in @xmath34 associated with a given change in @xmath53 is considered as independent of the system under study .
exact numerical studies of the functional dependence of @xmath34 have shown that this is a good approximation in a wide variety of 1d , 2d , and 3d lattices and band
fillings.@xcite notice that @xmath54 , @xmath55 , and @xmath56 are system specific . in practice ,
@xmath55 is approximated by the ferromagnetic fully - polarized @xmath57 which can be obtained , as @xmath54 and @xmath56 , from the single - particle electronic structure .
the dimer approximation to @xmath34 is very appealing since it combines remarkable simplicity and good accuracy.@xcite nevertheless , it also presents some limitations particularly in the limit of strong correlations at half - band filling . for @xmath58 and small @xmath24 ,
@xmath59 can be expanded as @xmath60 with @xmath61 ( e.g. , @xmath62 for a pentagonal ring and @xmath63 for c@xmath0 ) .
the exact functional @xmath34 shows the same behavior but usually with a somewhat larger coefficient @xmath64 ( e.g. , @xmath65 for a pentagonal ring and @xmath66 for c@xmath0).@xcite these discrepancies have direct consequences on the resulting ground - state properties for @xmath67 , since in this limit @xmath68 and @xmath69 , where @xmath70 is the local coordination number ( @xmath71 ) .
in order to improve the accuracy a more flexible approximation has been proposed which is also based on the scaling properties of @xmath34 , and which recovers the exact dependence on @xmath24 for @xmath72 . in this case
the interaction energy includes a fourth - order term in @xmath73 and is given by @xmath74 where @xmath75 . in a pentagonal ring @xmath76 , while in the c@xmath0 cluster @xmath77 .
these values should be compared with the dimer result @xmath78 , for which eq .
( [ eq : w4 ] ) reduces to eq .
( [ eq : w2 ] ) .
thus , the 4th - order term appears as a moderate correction to the dimer or 2nd - order approximation ( @xmath79 ) .
previous applications to 1d and 2d systems have shown that eq .
( [ eq : w4 ] ) provides a systematic improvement on the ground - state properties for all @xmath1.@xcite in the following section , eqs .
( [ eq : w2 ] ) and ( [ eq : w4 ] ) are applied to determine several electronic properties of the c@xmath0 hubbard cluster in the framework of ldft .
the model is characterized by the dimensionless parameter @xmath1 and by the doping @xmath23 .
we consider the complete range of repulsive interactions ( @xmath80 ) and single electron or hole dopings ( @xmath3 ) .
commonly accepted values of @xmath1 for the @xmath10-electrons in carbon fullerenes correspond to the intermediate range @xmath81.@xcite before discussing the results for the c@xmath0 cluster it is useful to consider a single pentagonal ring as a preliminary test on the accuracy of the method , since the pentagon constitutes the basic building block of the c@xmath0 structure , and since it is small enough so that exact diagonalizations can be easily performed .
[ fig : egsrg5 ] shows results for the ground - state energy @xmath25 of the @xmath82 ring as a function of @xmath1 . in the uncorrelated limit @xmath25 decreases monotonously with increasing number of electrons @xmath83 since the three lowest single - particle eigenvalues of the ring are negative . on the other side , for large @xmath1
, @xmath25 vanishes at half - band filling while for negative ( positive ) doping @xmath25 ( @xmath84 ) remains negative .
in fact , for @xmath85 the suppression of charge fluctuations always implies electron localization , whereas for @xmath86 a finite kinetic energy persists , even for @xmath87 , due to the delocalization of the extra electron or hole .
the ldft calculations reproduce remarkably well the crossover between weak and strong correlations as given by the exact diagonalizations .
appreciable discrepancies are found only for @xmath88 and @xmath89 , in which case the largest relative error amounts to @xmath90 for @xmath91 . in the other cases ( @xmath85 or @xmath92 ) the ldft results are practically indistinguishable from the exact ones . of the hubbard model on a pentagonal ring as a function of coulomb interaction @xmath1 for different numbers of electrons @xmath93 close to half - band filling ( @xmath94 ) . for @xmath95 ,
results for @xmath96 are shown .
the solid curves correspond to lattice density - functional theory ( ldft ) , the crosses to exact numerical diagonalizations , and the dashed curve for @xmath85 to unrestricted hartree - fock ( uhf ) calculations including noncollinear spin arrangements . ] in order to quantify the effects of electron correlations we have also performed calculations by using the fully unrestricted hartree - fock ( uhf ) approximation which corresponds to the most general single - slater - determinant wave - function and which allows for noncollinear site - dependent spin polarizations @xmath97.@xcite figure [ fig : egsrg5 ] shows the uhf results for @xmath25 of the pentagonal ring at half - band filling ( @xmath85 ) .
for @xmath98 the only solution to the self - consistent equations is non - magnetic ( i.e. , @xmath99 for all @xmath16 ) , while for @xmath100 local magnetic moments @xmath97 set in .
these increase monotonously with @xmath1 reaching saturation in the limit of @xmath101 .
concerning the magnetic order one observes that the @xmath97 are all coplanar and that the angle between nn moments is @xmath102 for all @xmath100 .
this amounts to split one parallel - spin frustration among five bonds which corresponds , as expected for half - band filling , to the solution of a classical heisenberg model with antiferromagnetic nn interactions .
one observes that uhf reproduces the @xmath1 dependence of @xmath25 qualitatively well , including the strongly correlated limit where @xmath103 .
this is achieved by localizing the electrons through the formation of increasingly large local magnetic moments as @xmath1 increases .
nevertheless , significant quantitative discrepancies with the exact numerical solution are observed already for @xmath104 , which reflect the importance of electron - correlation effects .
these can be regarded in part as the result of fluctuations between different spin configurations which can not be recovered in a single - determinant state.@xcite in contrast , ldft gives a much better description of the ground - state , particularly in the intermediate and strong interacting regimes .
uhf overestimates @xmath25 appreciably , while the differences between exact and ldft results are very small . , ( b ) coulomb energy @xmath105 , and ( c ) local magnetic moments @xmath106 of the hubbard model on a pentagonal ring at half - band filling ( @xmath107 ) .
the dashed - dotted and full curves correspond to lattice density - functional theory ( ldft ) using , respectively , eqs .
( [ eq : w2 ] ) and ( [ eq : w4 ] ) for the interaction - energy functional .
the dashed curves refer to the unrestricted hartree - fock ( uhf ) calculations including noncollinear spin arrangements , and the crosses to exact numerical diagonalizations . ]
more detailed information on the correlation effects and on the accuracy of the calculations is obtained from the kinetic energy @xmath36 , coulomb energy @xmath105 , and local square magnetic moments @xmath108 presented in fig .
[ fig : ekecrg5 ] .
ldft accurately reproduces the exact numerical solution , showing that the very good results obtained for @xmath25 are not the consequence of important compensations of errors . moreover , the differences between 2nd - order and 4th - order approximations to @xmath34 [ see eqs . (
[ eq : w2 ] ) and ( [ eq : w4 ] ) ] are quite small , typically of the order of the differences between the 4th - order calculations and the exact results .
nevertheless , one may observe that the interaction - energy functional @xmath109 provides a consistent improvement with respect to @xmath59 .
in fact taking into account the 4th - order term yields a small increase ( reduction ) of @xmath105 for @xmath110 ( @xmath111 ) as well as a small reduction of @xmath112 for @xmath111 which bring the outcome of ldft systematically closer to the exact solution.@xcite in contrast to ldft the uhf calculations do not provide a correct description of the kinetic and coulomb energies separately . on the one side , for @xmath110 ,
uhf overestimates the coulomb energy @xmath105 due to an underestimation of the local moments @xmath106 ( see fig .
[ fig : ekecrg5 ] ) . on the other side , for @xmath113 ,
both @xmath105 and @xmath112 are appreciably underestimated .
the inaccuracies in @xmath105 and @xmath112 for @xmath113 tend to cancel each other , which improves to some extent the uhf result for @xmath25 ( see fig .
[ fig : egsrg5 ] ) .
notice that uhf reproduces quite accurately the local moments for @xmath114 ( see fig .
[ fig : ekecrg5 ] ) .
however , a single - determinant state fails to incorporate correlations , and in particular the fluctuations between different equivalent spin configurations , which should restore the broken spin symmetry and which are often important in finite systems.@xcite this reflects the limitations of the broken symmetry uhf solutions , even in the most general noncollinear case.@xcite notice that in ldft no artificial symmetry breaking is required in order to account for correlation - induced localizations and the resulting effects on the ground - state properties .
the present ldft formalism with the dimer or 4th - order scaling approximations to interaction - energy functional describes correctly the electronic correlations in a pentagonal hubbard ring .
it can be therefore expected that the method should remain accurate in more complex systems with five - fold symmetry such as the c@xmath0 cluster to be discussed below .
moreover , comparison with exact results also shows that the differences between uhf and ldft results for @xmath25 , @xmath36 , and @xmath105 are a good quantitative estimation of the correlation energies and their @xmath1 dependence . of the hubbard model on the cagelike c@xmath0 cluster illustrated in fig .
[ fig : struct ] . results are given as a function of coulomb interaction @xmath1 for different numbers of electrons @xmath93 close to half - band filling ( @xmath115 ) .
in the inset @xmath96 is shown for @xmath95 .
the solid curves correspond to lattice density - functional theory ( ldft ) using eq .
( [ eq : w4 ] ) for the interaction - energy functional .
the dashed curves refer to unrestricted hartree - fock ( uhf ) calculations including noncollinear spin arrangements . ] , ( b ) coulomb energy @xmath105 , and ( c ) local magnetic moments @xmath106 of the cagelike c@xmath0 hubbard cluster at half - band filling ( @xmath85 ) .
the dashed - dotted and full curves correspond to lattice density - functional theory ( ldft ) using , respectively , eqs .
( [ eq : w2 ] ) and ( [ eq : w4 ] ) for the interaction - energy functional .
the dashed curves refer to unrestricted hartree - fock ( uhf ) calculations including noncollinear spin arrangements . ] in figs .
[ fig : egsc20 ] and [ fig : ekecc20 ] ldft and uhf results are given for several ground - state properties of the hubbard model on the cagelike c@xmath0 cluster illustrated in fig .
[ fig : struct ] . in the uncorrelated limit the ground - state energies @xmath25
are the same for @xmath3 , a consequence of the degeneracy of the single - particle spectrum . as @xmath1 increases , @xmath25 increases monotonously with a slope @xmath116 .
since the average number of double occupations is larger for larger number of electrons the degeneracy with respect to @xmath2 is removed at finite @xmath1 .
on the other side , in the strongly correlated limit , @xmath25 vanishes at half - band filling , while for @xmath88 ( @xmath95 ) @xmath25 ( @xmath117 ) remains negative .
the @xmath1 dependence of @xmath25 in c@xmath0 follows similar trends as in the pentagonal ring .
the uhf calculations reproduce the exact limits for @xmath118 and @xmath119 , and agree qualitatively with ldft for finite values of the coulomb repulsion
. however , the quantitative differences can be quite significant particularly for intermediate and large @xmath1 ( e.g. , 26% for @xmath120 ) .
the effects of correlations are more significant in the kinetic energy @xmath36 , coulomb energy @xmath105 , and local square moment @xmath108 shown in fig .
[ fig : ekecc20 ] for @xmath85 .
the differences between uhf and ldft calculations for @xmath36 and @xmath105 are similar to the corresponding differences between uhf and exact results for the pentagonal ring and can be qualitatively understood in similar terms . in the case of c@xmath0 one
may notice that the uhf results for the local square moment @xmath108 are the same as in the uncorrelated limit for @xmath121 .
therefore , they are significantly underestimated . for stronger coulomb repulsions ( @xmath122 )
@xmath108 increases rapidly reaching values that are very close to the ldft results for @xmath123 ( see fig . [
fig : ekecc20 ] ) .
this behavior is related to the @xmath1 dependence of the self - consistent solution of the uhf equations . for @xmath124
the uhf charge distribution is not uniform due to the four - fold degeneracy of the single - particle spectrum at the fermi energy .
moreover , for @xmath121 the uhf ground state shows a weakly ferromagnetic solution with very small local spin polarizations . in this way advantage
is taken from the single - particle degeneracy in order to reduce the coulomb - repulsion energy without increasing significantly the kinetic energy .
for @xmath122 the system starts to develop an af spin - density wave which coexists with a weak charge - density wave for @xmath125 . only for @xmath126 the af - like magnetic order valid for strong interactions
is definitely settled and the charge distribution becomes then uniform throughout the cluster .
an illustration of the uhf magnetic order in the large @xmath1 regime is shown in fig .
[ fig : c20sdw ] .
notice the noncollinearity of the local spin polarizations @xmath127 , which form angles of about 138 degrees between nns .
this value is smaller than the angle between nn @xmath127 in an isolated pentagonal ring ( @xmath128 degrees ) because of frustration effects between the pentagons in the c@xmath0 topology ( see figs .
[ fig : struct ] and [ fig : c20sdw ] ) .
further increase of the coulomb repulsion beyond @xmath129 only results in an increase of the size of the local moments keeping the magnetic order unchanged . in an undoped cagelike c@xmath0 cluster as obtained by using the hubbard model and the unrestricted hartree - fock approximation ( @xmath85 ) . due to spin - rotational invariance
only the relative orientations of @xmath127 are physically meaningful . the three - dimensional cluster structure ( see fig . [
fig : struct ] ) is mapped onto a plane in order to ease the visualization .
the radius of the circle on each atomic site @xmath16 is proportional to @xmath130 .
the arrows represent the projection of @xmath131 onto the @xmath132 plane which is the plane containing the outermost pentagon . shaded circles ( open circles )
indicate that the perpendicular component @xmath133 is positive ( negative ) . ]
the charge excitation gap is given by @xmath134 where @xmath25 refers to ground - state energy .
it is a property of considerable interest since it measures the low - energy excitations associated with changes in the number of electron @xmath22 and is thus very sensitive to electron correlation effects . in fig .
[ fig : gaps ] results are shown for @xmath135 of the c@xmath0 hubbard cluster at half - band filling ( @xmath58 ) .
the corresponding calculations for the pentagonal ring are reported in the inset for the sake of comparison . at half - band
filling @xmath136 for @xmath137 , despite the finite size , due to the degeneracy of the single - particle spectrum in both the ring and c@xmath0 . @xmath135 increases with increasing @xmath1 approaching the limit @xmath138 for @xmath139 , where @xmath140 is the bottom of single - particle spectrum of the cluster ( @xmath141 for the ring and @xmath142 for c@xmath0 ) .
the ldft results and the exact calculations for @xmath82 yield a linear dependence of @xmath135 as a function @xmath1 for @xmath143 .
this implies , in particular for c@xmath0 , that @xmath144 for arbitrary small @xmath1 , in contrast to the uhf result .
comparison with exact diagonalizations for @xmath82 indicates that the ldft gap is not overestimated .
the uhf calculations fail to reproduce the correct @xmath1 dependence of @xmath135 showing once more the importance of correlations .
notice that the differences between uhf and exact or ldft results are qualitatively similar for c@xmath0 and for the pentagonal ring , which indicates that the contribution of correlations to @xmath135 are comparable in both cases . of the hubbard model on the cagelike c@xmath0 cluster as a function of coulomb interaction @xmath1 .
the solid ( dashed ) curves correspond to ldft ( uhf ) . in the inset figure results
are given for the pentagonal hubbard ring .
here the crosses refer to exact diagonalizations . ]
a density - functional approach to lattice - fermion models has been applied to study the ground - state properties of the hubbard hamiltonian on a cagelike c@xmath0 cluster .
the ground - state energy , the kinetic and coulomb contributions , and the charge excitation gap have been determined as a function of @xmath1 for dopings @xmath2 close to half band filling .
the importance of electron correlations has been quantified by performing noncollinear hartree - fock calculations .
comparisons with exact diagonalizations and with uhf in the case of a pentagonal ring demonstrate the ability of ldft to describe subtle correlation effects in a very simple and systematic way .
it is therefore expected that ldft should be an efficient tool for investigating the properties of strongly correlated fermions in more complex clusters and nanostructures , which are otherwise very difficult to tackle with other accurate methods .
the present ldft study of clusters encourages the development of more flexible interaction - energy functionals in order to determine the properties of finite systems with lower symmetry . as a first step it would be interesting to extend the calculations to fullerene structures having different types of bonds , for example c@xmath4 . in these cases
the functional dependence of @xmath145 $ ] is more complex and one has to consider at least two different nn density - matrix elements , within and between pentagons , even if one assumes that the hopping integrals are the same .
the symmetry of these bonds is actually different despite the fact that all sites are equivalent .
the interaction - energy functional should be generalized following previous works on dimerized 1d chains.@xcite the domain of representability of @xmath24 and the accuracy of the scaling hypothesis could be explicitly tested for smaller similar clusters like c@xmath5 by solving levy s constrained search.@xcite furthermore , the functional dependence of @xmath34 for low - symmetry clusters deserves to be explored by taking into account charge transfers and eventual redistributions of the spin density between non - equivalent sites .
this should open the way to investigations of correlation effects in a variety of complex physical situations such as metal clusters , disordered systems , and low - dimensional nanostructures .
this work has been supported by conacyt mexico ( grant no .
j-41452 and millennium initiative w-8001 ) and by the eu growth project ammare ( contract no .
g5rd - ct-2001 - 00478 ) .
computer resources were provided by idris ( cnrs , france ) .
h. w. kroto , j. r. heath , s. c. obrien , r. f. curl , and r. e. smalley , nature ( london ) * 318 * , 162 ( 1985 ) ; w. krtschmer , l. d. lamb , k. fostiropoulos , and d. r. huffman , nature ( london ) * 347 * , 354 ( 1990 ) .
y. h. kim , i. h. lee , k. j. chang , and s. lee , phys .
lett . * 90 * , 065501 ( 2003 ) ; y. zhao , b. y. yakobson and r. e. smalley , phys . rev . lett . * 88 * , 185501 ( 2002 ) ; d. herschbach , rev . mod . phys .
* 71 * , s411 ( 1999 ) ; r. e. smalley , rev .
phys . * 69 * , 723 ( 1997 ) .
u. rthlisberger , w. andreoni and m. parrinello phys .
lett . * 72 * , 665 ( 1994 ) ; l. mitas , j. c. grossman , i. stich and j. tobik phys . rev .
lett . * 84 * , 1479 ( 2000 ) ; q. sun , q. wang , p. jena , b. k. rao and y. kawazoe , phys
lett . * 90 * , 135503 ( 2003 ) ; d. conntable , _ et al .
* 91 * , 247001 ( 2003 ) ; m.n .
huda and a.k .
ray , eur .
j. d * 31 * , 63 ( 2004 ) . for alternative density - functional approaches to lattice - fermion models
see , for instance , a.e .
carlsson , phys .
b * 56 * , 12058 ( 1997 ) ; r.g .
hennig and a.e .
carlsson , _ ibid .
_ * 63 * , 115116 ( 2001 ) ; n.a . lima , m.f .
silva , l.n .
oliveira and k. capelle , phys .
lett . * 90 * , 146402 ( 2003 ) . for simplicity
, we focus here on @xmath146 which is the relevant case for @xmath18 [ see eq .
( [ eq : ek ] ) ] .
consequently , we also consider only @xmath44 and @xmath50 .
however notice that in non - bipartite lattices as the c@xmath0 cluster the domains of representability for positive and negative values of the nn @xmath24 are different .
therefore , one would have to distinguish the cases @xmath147 from @xmath148 , since in non - bipartite lattices @xmath149 and @xmath150 .
nevertheless , as shown in ref . , the same scaling behavior of @xmath151 as a function of @xmath152 is found for @xmath148 .
the situation is analogous to what is found in dimerized systems where the domain of representability @xmath153 $ ] depends on the ratio @xmath154 between the bond orders corresponding to short and long bonds ( see ref . ) .
the expansion coefficient of the exact functional @xmath34 are here obtained from numerical exact diagonalizations for the heisenberg limit of the hubbard model ( @xmath72 ) . for extended systems or larger clusters
they can be calculated by using degenerate perturbation theory ( see ref . ) . a more detailed discussion of the accuracy of ldft and the interaction - energy functionals may be found in refs . , , and , where applications to periodic one- , two- , and three - dimensional systems are reported . | the ground - state properties of c@xmath0 fullerene clusters are determined in the framework of the hubbard model by using lattice density - functional theory ( ldft ) and scaling approximations to the interaction - energy functional .
results are given for the ground - state energy , kinetic and coulomb energies , local magnetic moments , and charge - excitation gap , as a function of the coulomb repulsion @xmath1 and for electron or hole doping @xmath2 close half - band filling ( @xmath3 ) .
the role of electron correlations is analyzed by comparing the ldft results with fully unrestricted hartree - fock ( uhf ) calculations which take into account possible noncollinear arrangements of the local spin - polarizations .
the consequences of the spin - density - wave symmetry breaking , often found in uhf , and the implications of this study for more complex fullerene structures are discussed . |
the space - time evolution of a quantum mechanical system is governed by its schrdinger equation and in its most complete form it is encoded in the propagator .
the propagator defines the probability amplitude for a particle to move from one point of the space to another in a given time .
similar to propagators in relativistic field theories , it provides a global picture of the causal structure of a quantum system which goes beyond the information contained in a single wave function .
moreover , it plays the essential role in solving the probability related cauchy problem of quantum mechanics ( qm ) .
the vast literature on qm propagators , summarized e.g. in @xcite , lists mainly explicit expressions of propagators for stationary schrdinger equations in one space dimension which are reducible to hypergeometric differential equations ( des ) or their confluent forms .
this is in strong contrast to the available methods of supersymmetric quantum mechanics ( susy qm ) @xcite which usually lead to much broader classes of exactly solvable schrdinger equations with solutions in particular expressible in terms of linear combinations of hypergeometric functions ( see e.g. @xcite ) . from the structure of
the susy induced relations between superpartner hamiltonians it is clear that via the corresponding schrdinger equations these relations should extend to relations between the associated propagators .
the main goal of the present paper is to analyze these relations between susy partner propagators and to reshape them into user friendly general recipes for the construction of new propagator classes .
our interest in such new exact propagators is less motivated by their mere existence or their technical subtleties , but rather it is in their applicability to concrete physical problems ( see e.g. @xcite ) and here especially to models with well - tailored new properties and to setups which up to now were not related to susy techniques at all ( see e.g. @xcite ) . in order to derive the corresponding technical tools we concentrate in this paper on the general approach which allows to establish the link between the propagators of any two susy partner hamiltonians .
our main idea is the following .
for a given hamiltonian @xmath0 , whose schrdinger equation as a second order differential equation is exactly solvable , the general solutions of the schrdinger equation of its susy partner hamiltonian @xmath1 can be explicitly constructed . knowing the relations between the solutions of the schrdinger equations for @xmath0 and @xmath1 one can expect to derive transformation operators relating the associated propagators . at a first glance
, this approach seems to be technically trivial .
but when one recalls that the calculation of a closed expression for a propagator is usual more difficult than the derivation of a corresponding wave function , one may expect the problem to be connected with rather nontrivial technical subtleties .
moreover , if one wants to establish the explicit link between the propagators one may imagine that sometimes the problem may even become unsolvable .
probably the first indication that the problem may have a solution was given by jauslin @xcite who constructed a general integral transformation scheme simultaneously for schrdinger equations and for heat equations , but who did nt provide a discussion of convergency and divergency of the derived expressions .
for the sake of convergency he applied his technique to the heat - equation - type fokker - planck equation only . in general , this result
may be extended via wick rotation to propagators for schrdinger equations of a free particle and a particle moving through transparent potentials .
but the question of convergency and with it the question of solvability remains to be clarified .
another indication that the problem may be solvable has been provided by refs @xcite where a similar model has been analyzed at the level of green functions of stationary schrdinger equations . in the present paper
we carefully analyze the problem along the ideas first announced in @xcite .
the material is organized as follows . in the next section [ prelim ] , we briefly recall some main facts and notations from susy qm necessary for our subsequent analysis and we sketch the definition of the propagator . section [ susy-1 ] is devoted to the basic tool of our approach the interrelation between propagators for models with partner hamiltonians which are linked by first - order susy transformations . in section [ susy - n ]
we generalize these results to polynomial supersymmetry . the general technique developed in these sections
is afterwards used in section [ applic ] to derive the propagators for susy partner models of `` a particle in a box '' , of the harmonic oscillator and of the free particle ( i.e. for models with transparent potentials ) .
section [ conclu ] concludes the paper .
the subject of our analysis will be a polynomial generalization of the simplest two - component system of witten s non - relativistic supersymmetric quantum mechanics @xcite described by the schrdinger equation @xmath2 where @xmath3 is a diagonal super - hamiltonian consisting of the two super - partners @xmath0 and @xmath4 as components @xmath5 the interval @xmath6 may be both finite or infinite . for simplicity
we restrict our consideration to a stationary setup with @xmath0 and @xmath4 not explicitly depending on time so that the evolution equation ( [ supershredinger ] ) reduces via standard substitution @xmath7 and properly chosen boundary conditions to the spectral problem @xmath8 the time evolution of the system may be described in terms of a corresponding propagator .
furthermore , we assume that the partner hamiltonians @xmath0 and @xmath4 are intertwined by an @xmath9th - order differential operator @xmath10 with the following properties : + * 1 .
intertwining relations * lh_0=h_nl h_0l^+=l^+h_n [ intertwiner ] & & l^+l = p_n(h_0 ) ll^+=p_n(h_n ) + & & p_n(x)=(x-_0) (x-_n-1)[polyn-1 ] + & & im(_i)=0 _ i_kii , k=0, ,n-1 . here
, the adjoint operation is understood in the sense of laplace ( i.e. as formally adjoint with the property @xmath11 , @xmath12 and @xmath13 ) and the roots @xmath14 of the polynomial @xmath15 play the role of factorization constants .
for simplicity we assume that the polynomial @xmath15 has only simple roots .
the intertwining relations together with the factorization rule can be represented in terms of the polynomial super - algebra @xcite [ superalgebra ] q^2=(q^+)^2=0=[q^+,h]=0qq^++q^+q = p_n(h ) with nilpotent super - charges @xmath16 although the component hamiltonians @xmath0 and @xmath1 enter the super - hamiltonian ( [ superhamiltonian ] ) in an algebraically symmetric way , we consider @xmath0 as given hamiltonian with known spectral properties and @xmath1 as derived hamiltonian with still undefined spectrum .
more precisely we assume @xmath17 to be real - valued , continuous and bounded from below bounded from below and short ranged , because we restrict our attention here to models with a finite number of bound states only . for spectral problems on the half - line a repulsive singularity at the origin not stronger than @xmath18
, @xmath19 is possible . ] so that the differential expression @xmath20 defines a sturm - liouville operator which is symmetric with respect to the usual @xmath21 inner product .
the corresponding functions @xmath22 are additionally assumed sufficiently smooth denotes the space of twice continuously differentiable functions .
] , e.g. @xmath23 , over the interval @xmath24 . moreover , we assume dirichlet boundary conditions ( bcs ) for the bound state eigenfunctions of @xmath0 , i.e a domain @xmath25 ( see , e.g. @xcite ) and the operator @xmath0 itself being at least essentially self - adjoint ( with a closure that we denote by the same symbol @xmath0 ) . as usual ,
eigenfunctions which correspond to the continuous spectrum of @xmath0 are supposed to have an oscillating asymptotic behavior at spatial infinity .
concentrating on physically relevant cases we restrict our attention to the following three types of setups : + * ( i ) : * the interval @xmath6 is finite @xmath26 so that @xmath0 has a non - degenerate purely discrete spectrum ( see e.g. @xcite ) . + * ( ii ) : * for spectral problems on the half - line @xmath27 we consider so called scattering ( or short - ranged ) potentials which decrease at infinity faster than any finite power of @xmath28 and have a continuous spectrum filling the positive semi - axis and a finite number of discrete levels ; the whole spectrum is non - degenerate . + * ( iii ) : * for spectral problems on the whole real line , @xmath29 , we consider confining as well as scattering potentials . confining potentials
produce purely discrete non - degenerate spectra ( see e.g. @xcite ) , whereas scattering potentials lead to two - fold degenerate continuous spectra filling the whole real line and to a finite number of non - degenerate discrete levels ( see e.g. @xcite ) . everywhere in the text
we choose real - valued solutions of the differential equation @xmath30 .
this is always possible since @xmath17 is supposed to be a real - valued function and we always restrict ourselves to real values of the parameter @xmath31 .
the intertwiner @xmath10 is completely described by a set of @xmath9 transformation functions @xmath32 , which may be both `` physical '' and `` unphysical '' . all solutions
@xmath33 corresponding to a spectral parameter @xmath31 outside the continuous spectrum are interpreted as `` unphysical '' . in the present paper , eigenfunctions corresponding to the continuous spectrum
are not used as transformation functions and , although physically meaningful , we excluded them from our classification scheme of `` physical '' and `` unphysical '' solutions . ]
solutions of the stationary schrdinger equation with @xmath0 as hamiltonian : @xmath34 in our case of a polynomial @xmath35 with simple roots ( i.e. @xmath36 ) the action of the intertwiner @xmath10 on a function @xmath37 is given by the crum - krein formula @xcite [ operator ] lf= with @xmath38 denoting the wronskian w = w(u_0,u_1,
,u_n-1)= | ccccu_0 & u_1 & & u_n-1 + u_0&u_1& &u_n-1 + & & & + u_0^(n-1)&u_1^(n-1)&
&u_n-1^(n-1 ) + the schrdinger equations with @xmath1 and @xmath0 as hamiltonians by the relation @xmath39 . in particular , for @xmath40 ( a first - order transformation ) ( [ operator ] ) reads [ l1 ] = lf=(-u_0x / u_0+_x)f= with @xmath41 .
furthermore , the determinant structure ( [ operator ] ) of the operator @xmath10 leads to the immediate implication that it has a nontrivial kernel space @xmath42 spanned by the set of transformation functions @xmath43 : @xmath44 the solutions @xmath45 of the equation @xmath46 are elements of the kernel space of the adjoint operator @xmath47 and can be obtained as ) in [ ap1 ] . ] [ hnsol ] v_n= n=0 , , n-1 @xmath48 where @xmath49 denotes the wronskian built as a determinant of the @xmath50 matrix with the @xmath51related column omitted @xmath52 .
the potential @xmath53 of the hamiltonian @xmath1 can be expressed as @xcite @xmath54''\,.\ ] ] in general , it is not excluded that the transformation operator @xmath10 may move a solution of @xmath0 out of @xmath55 thus transforming a physical solution of @xmath0 into an unphysical solution of @xmath1 .
moreover the inverse scenario is also possible , i.e. @xmath10 may transform an unphysical solution of @xmath0 into a physical solution of @xmath1 . in such cases the point spectrum of @xmath1 will differ from that of @xmath0 .
subsequently , we will concentrate on mild transformations @xmath10 , which leave most of the original spectrum invariant with exception of a finite number of spectral points a characteristic feature of differential intertwining operators @xmath10 leaving the boundary behavior of the solutions of the schrdinger equation unchanged . recently in @xcite a conjecture has been proven , which was originally formulated in @xcite and which states that any @xmath9th - order mild differential transformation @xmath10 can be constructed as a superposition from only first- and second - order mild transformations . in this case it is possible to show @xcite that for problems formulated over the whole @xmath56 ( for infinite values of @xmath57 and @xmath58 ) @xmath53 behaves asymptotically like @xmath59 .
therefore the operator @xmath1 is also essentially self - adjoint and `` lives '' in the same hilbert space @xmath60 as @xmath0 .
moreover , since the point spectrum of the self - adjoint sturm - liouville problems that we consider is non - degenerate there is no way to create a new discrete level at the position of an already existing discrete level and by this means to increase the geometric multiplicity of that level , @xmath61 for which a second - order intertwiner @xmath10 can map two distinct discrete levels of @xmath0 into a second - order jordan block of @xmath61 , i.e. an eigenvalue of geometric multiplicity one and algebraic multiplicity two ( for the details see ref @xcite ) . ] . according to @xcite the necessary condition for an @xmath9th - order transformation to produce an essentially self - adjoint operator @xmath1 is belonging to the point spectrum of @xmath0 the eigenfunctions @xmath62 with @xmath63 should map into corresponding eigenfunctions @xmath64 of @xmath1 with @xmath65 what via ( [ polyn-1 ] ) implies ( [ usl ] )
here , the equality takes place for those @xmath62 for which the point @xmath31 does not belong to the spectrum of @xmath1 , i.e. if @xmath66 and @xmath67 .
a more detailed analysis of corresponding sufficient conditions for spectral problems on the whole real line in case of scattering potentials and of confining potentials is given in @xcite . ]
[ usl ] ( e-_0) (e-_n-1)0 e(h_0 ) .
this criterion ensures the mildness of the transformation @xmath10 leading only to changes in maximally @xmath9 spectral points ( of the point spectrum ) . specifically , the spectrum of @xmath1 may contain @xmath68 points more and @xmath69 points less than @xmath70 , where necessarily @xmath71 .
in the present paper we will consider the following possibilities . *
the spectrum of @xmath0 is a subset of the spectrum of @xmath1 .
a new energy level may be created in the spectrum of @xmath1 if and only if the corresponding transformation function @xmath72 is such that @xmath73 , i.e. , in particular , that @xmath74 is @xmath75integrable and satisfies the dirichlet bcs . *
the spectrum of @xmath1 is a subset of the spectrum of @xmath0 .
an energy level may be removed from the spectrum of @xmath0 if and only if the corresponding transformation function @xmath72 coincides with the @xmath76eigenfunction of this level , i.e. when @xmath72 satisfies the dirichlet bcs .
* the spectrum of @xmath0 coincides with the spectrum of @xmath1 . in this case
non of the transformation functions @xmath77 nor @xmath78 should be physical , i.e. satisfy dirichlet bcs on both ends of the interval @xmath6 .
this property should be fulfilled for all transformation functions @xmath77 from which the transformation operator is built . in all cases
we assume that the transformation functions @xmath79 are linearly independent one from the other and their wronskian @xmath80 does not vanish @xmath81 . in the remainder of this section
, we briefly recall some basic properties of quantum mechanical propagators .
once , the super - hamiltonian ( [ superhamiltonian ] ) is diagonal it suffices to restrict to one - component ( scalar ) propagators .
it is well known that the propagator @xmath82 of a non - stationary schrdinger equation contains the complete information about the space - time behavior of the wave function @xmath83 evolving from an initial configuration @xmath84 @xmath85 solving in this way the probability related cauchy problem of qm .
the propagator ( integration kernel ) satisfies a differential equation with dirac delta function as initial condition [ k00 ] [ _ t - h(x)]k(x , y;t)=0k(x , y;0)=(x - y ) . for non - dissipative systems , like in our case , the propagator @xmath86 can be interpreted as coordinate representation of the unitary evolution operator @xmath87 : @xmath88 , where unitarity implies the symmetry @xmath89 .
subsequently , we will mainly work with a spectral decomposition of propagators in terms of complete basis sets of eigenfunctions [ decomp ] k(x , y;t)= _ n=0^n_p_n(x , t)_n^*(y)+ _ -^dk_k(x , t)_k^*(y ) where summation over the point spectrum and integration over the continuous ( essential ) spectrum are understood . by susy - transformations
we will only induce changes in the point spectrum of @xmath0 so that , for simplicity , we will work with decompositions over discrete sets of basis functions ( corresponding to point spectra ) keeping in mind that extensions to the continuous spectrum are straight forward .
it is clear that the defining equation for the propagator of the non - stationary schrdinger equation with super - hamiltonian ( [ superhamiltonian ] ) can be trivially decomposed as ( x , y;t)=0= + ( x , y;0)=i(x - y ) .
in this section we study the structure of propagators interrelated by first - order susy transformations .
such susy transformations are generated from a single function @xmath72 and they act as basic building blocks in chain representations of higher - order intertwiners . although any @xmath9th - order intertwiner may be represented as a chain of first order intertwiners @xcite , one has to distinguish between chains which are completely reducible within a given hilbert space @xmath60 @xcite and chains which are partially or completely irreducible in @xmath60 .
complete reducibility means that apart from @xmath0 and @xmath1 also all intermediate first - order susy related hamiltonians @xmath90 , @xmath91 are self - adjoint or essentially self - adjoint in the same hilbert space @xmath60 . in case of irreducible chains
several or all intermediate hamiltonians are non - self - adjoint in @xmath60 . below both chain types will play a role .
we note that chain representations lead to extremely simplified transformation rules for higher - order intertwined propagators and allow for very efficient calculation techniques ( see subsection [ 4.2 ] below ) . according to ( [ operator ] ) a first - order intertwiner has the form @xmath92 due to @xmath93 ( see ( [ usl ] ) ) and depending on the concrete form of the function @xmath72 the intertwiner @xmath10 may result in the following three types of relations between the spectra of the hamiltonians @xmath0 and @xmath94 ( see , e.g. , @xcite ) : + * ( i ) * for @xmath95 and @xmath96 the ground state level @xmath97 of @xmath0
is removed from the spectrum of @xmath94 , + * ( ii ) * @xmath94 has a new and deeper ground state level @xmath98 than @xmath0 , + * ( iii ) * the spectra of @xmath94 and @xmath0 completely coincide ( @xmath99 ) . + in order to create a potential @xmath100 which is nonsingular on the whole interval @xmath101 the function @xmath72 should be nodeless inside this interval .
this property is evidently fulfilled for type ( i ) relations since the function @xmath72 coincides in this case with the ground state eigenfunction @xmath102 . in the cases ( ii ) and ( iii ) the nodelessness should be ensured by an appropriate choice of @xmath72 , a choice which is always possible because of the ` oscillation ' theorem ( see e.g. @xcite ) . in case ( iii ) it implies @xmath99 . introducing the green function [ g0 ] g_0(x
, y;e)=_m=0^ of the stationary @xmath76schrdinger equation at fixed energy @xmath31 has a purely discrete spectrum .
as it was already stated in the preliminaries , a generalization to the continuous spectrum is straight forward . ]
( see , e.g. @xcite ) and its
`` regularized '' version @xmath103\ ] ] the corresponding structural relations for the propagators can be summarized in the following the propagators @xmath104 and @xmath105 of non - stationary schrdinger equations with susy intertwined hamiltonians @xmath94 and @xmath0 are interrelated with each other and with the green functions @xmath106 and @xmath107 in the following way : + type * ( i ) * relation [ trprv1 ] k_1(x , y , t)=l_xl_y_a^bk_0(x , z , t)_0(z , y , e_0)dz .
type * ( ii ) * relation [ trpr1]k_1(x , y , t)=l_xl_y_a^bk_0(x , z , t)g_0(z , y,)dz+ _ -1(x)_-1(y)^-t . type * ( iii ) * relation [ trpris2 ] k_1(x , y , t)=l_xl_y_a^bk_0(x , z , t)g_0(z , y,)dz . we start from the type * ( ii ) * relation and represent the propagator @xmath104 in terms of the basis functions @xmath108 of the hamiltonian @xmath94 ( cf .
( [ decomp ] ) ) .
we note that the explicit time - independence of @xmath94 implies a factorization @xmath109 with @xmath110 purely real - valued . expressing @xmath111 in terms of the corresponding wave functions of the hamiltonian @xmath0 , @xmath112 , with @xmath113 a normalization constant ( see , e.g. @xcite )
, we arrive at @xmath114 modulo a normalization factor @xmath9 , the wave function @xmath115 of the new ground state is proportional to the inverse power of the transformation function @xmath72 , @xmath116 .
it remains to express the time - dependent phase factor in terms of the propagator .
this can be easily done using the evident property of the bound state solutions of the schrdinger equation [ vc0 - 1 ] _
a^bk_0(x , z , t)_m(z)dz=_m(x)^-e_mt so that the previous equation reads [ vc1 ] k_1(x , y , t)= l_xl_y_-^k_0(x , z , t ) _
m=0^dz + _ -1(x)_-1(y)^-t .
the sum in this relation can be identified as the green function ( [ g0 ] ) . due to
@xmath117 this green function is regular @xmath118 and the proof for type * ( ii ) * transformations is complete .
the proof for type * ( i ) * and * ( iii ) * transformations follows the same scheme .
the formally regularized green function @xmath107 in * ( i ) * results from the fact that the ground state with energy @xmath97 is not present in the spectrum of @xmath94 so that a sum @xmath119 appears and the ground state contribution has to be subtracted from @xmath120 . in case of a type * ( iii ) * transformation a sum @xmath121 over the complete set of eigenfunctions appears in ( [ vc1 ] ) and no new state occurs .
we conclude this section by reshaping relation ( [ trprv1 ] ) for the propagator of a system with removed original ground state , i.e. of a type * ( i ) * transformed system . the corresponding result can be formulated as [ th2 ] for transformations with @xmath102
the propagator @xmath104 of the resulting system can be represented as @xmath122 first of all we recall that @xmath123 being the ground state function of @xmath0 satisfies the zero boundary conditions . to facilitate the proof of theorem [ th2 ] we need the following two lemmas .
[ lemma1 ] @xmath124 this result follows from the explicit representation of @xmath125 in terms of basis functions . on the one hand , it holds @xmath126
whereas on the other hand the kernel property ( annihilation ) of the ground state @xmath127 gives @xmath128 [ lemma2 ] let @xmath129 and @xmath130 satisfy the schrdinger equation [ eqf]h_0f(x , e):=-f(x , e)+v_0(x)f(x , e)=ef(x , e)x(a , b ) and boundary conditions [ bcf ] f_l(a , e)=0f_r(b , e)=0 . let also @xmath131 be the ground state level of @xmath0 with @xmath123 as the ground state function ( we assume that @xmath0 has at least one discrete level ) then [ lim1 ] _ ee_0 & = & - + _ ee_0 & = & [ lim2 ] where @xmath132 with @xmath133 .
first we note that according to ( [ l1 ] ) @xmath134 .
next , since both @xmath96 and @xmath135 satisfy the same schrdinger equation ( [ eqf ] ) it holds @xmath136 and hence [ l2 - 1 ] w(u , f_r)=(e - e_0)_y^bu(z)f_r(z , e)dzwhere we have used the property @xmath137 which follows from the bcs for @xmath138 and @xmath135 . via ( [ l2 - 1 ] ) we find l_yf_r(y , e)=_y^bu(z)f_r(z , e)dz and hence [ lfw ] = - where it has been used that the wronskian @xmath139 is @xmath28-independent and can be calculated at @xmath140 where @xmath141 . since the spectrum of @xmath0 is non - degenerate , the ground state function is unique up to an arbitrary constant factor and , hence , @xmath102 , @xmath142 and @xmath143 have to be proportional to each other [ fc ] f_r , l(x , e_0)=c_r , lu(x ) and for @xmath144 only the first fraction in ( [ lfw ] ) remains undetermined .
the lhospital rule gives for this limit [ lee ] _
ee_0= where the dot denotes the derivative with respect to @xmath31 .
making use of ( [ lee ] ) and [ ff ] ( b , e_0)f_l(b , e_0)=_a^b f_l^2(z , e_0)dz ( which we prove below ) relation ( [ lfw ] ) yields _ ee_0 = -f_l(x , e_0 ) and via ( [ fc ] ) it leads to the result ( [ lim1 ] ) .
the proof of ( [ lim2 ] ) follows the same lines with evident changes .
finally , it remains to derive equation ( [ ff ] ) .
this is easily accomplished by multiplying schrdinger equation ( [ eqf ] ) for @xmath145 by @xmath146 , its derivative with respect to @xmath31 by @xmath145 , and integrating their difference over the interval @xmath6 . the intermediate result _
a^bf_l^2(x , e)dx & & + = f_l(a , e)f_l(a , e)-f_l(a , e)f_l(a , e ) -f_l(b , e)f_l(b , e)+f_l(b , e)f_l(b , e ) & & reduces to ( [ ff ] ) via bc ( [ bcf ] ) and its derivative with respect to @xmath31 ( what cancels the first two terms ) and the limit @xmath131 , its implication ( [ fc ] ) and the bc for @xmath72 .
_ proof of theorem [ th2 ] .
_ for the green function @xmath147 in ( [ trprv1 ] ) we use the standard representation in terms of two linearly independent solutions @xmath148 of the @xmath76schrdinger equation introduced in lemma [ lemma2 ] ( see , e.g. @xcite ) : g(x , y , e)=/w(f_r , f_l ) where @xmath149 denotes , as usual , the heavyside step function
. then relation ( [ trprv1 ] ) takes the form @xmath150dz\end{aligned}\ ] ] where the step functions can be resolved to give k_1(x , y , t)&= & l_xl_y_a^yk_0(x , z , t)_ee_0dz + & + & l_xl_y_y^bk_0(x , z , t)_ee_0dz .
the second argument of the functions @xmath148 has been omitted for notational simplicity .
explicitly acting with the differential operator @xmath151 on the integrals with variable @xmath152boundary yields k_1(x , y , t)&= & l_x_a^yk_0(x , z , t)l_y_ee_0dz + & + & l_x_y^bk_0(x , z , t)l_y_ee_0dz whereas via lemma [ lemma1 ] the intertwiner @xmath151 and the limit @xmath153 can be interchanged to give [ prop1l ] & & k_1(x , y , t)= + & & l_x\{_a^yk_0(x , z , t)_ee_0dz+ _ y^bk_0(x , z , t)_ee_0dz}. application of lemma [ lemma2 ] leads to [ prop2l ] k_1(x , y , t)= \{- _
y^b _ 0 ^ 2(q)dq _ a^yk_0(x , z , t)_0(z ) dz .
_ a^y _ 0 ^ 2(q)dq _ y^bk_0(x , z , t)_0(z)dz } which we further reshape by expressing the integral with respect to @xmath69 over the interval @xmath154 by the difference of two integrals over the intervals @xmath6 and @xmath155 .
substitution of @xmath156 in the first term results in [ prop3l ] k_1(x , y , t)=- l_x _ a^yk_0(x , z , t)u(z ) dz + + _ a^y _ 0 ^ 2(q)dql_x _ a^bk_0(x , z , t)_0(z)dz .
the very last integral is nothing but the ground state stationary wave function @xmath157 .
therefore , since @xmath158 , we obtain the first equality in ( [ trans1l ] ) . the second equality results from applying a similar transformation to the second term in ( [ prop2l ] ) . @xmath159 the following remarks are in order .
first we have to note that the integral representation ( [ trans1l ] ) is only valid in case of first - order susy transformations which remove the ground state level .
if one wants to create a level in a problem on the whole real line one has to use a transformation function @xmath72 which diverges for @xmath160 ensuring in this way the normalizability and dirichlet bcs of the new ground state wave function @xmath161 . an attempt to calculate the propagator @xmath162 via ( [ trans1l ] ) would usually lead to a divergent integral .
the correct approach is to use ( [ trpr1 ] ) in this case .
jauslin @xcite using a different procedure obtained the same result ( [ trans1l ] ) both for removing and creating a level , but he completely ignored questions of convergence or divergence of the corresponding integrals . in concrete calculations he avoided divergent integrals by considering the heat equation only .
let us consider an @xmath9th - order ( @xmath163 ) polynomial supersymmetry corresponding to the appearance of @xmath9 additional levels in the spectrum of @xmath1 compared to the spectrum of @xmath0 .
in this case new levels may appear both below the ground state energy of @xmath0 ( reducible supersymmetry ) and between any two neighbor levels of @xmath0 ( irreducible supersymmetry , see e.g. @xcite ) .
the propagator for the transformed equation can be found in the following way .
we develop @xmath164 over the complete orthonormal set @xmath165 of eigenfunctions of @xmath1 and express all @xmath111 with eigenvalues already contained in the spectrum of @xmath0 in terms of @xmath166 , i.e. @xmath112 . the normalization constants @xmath167 for transformations fulfilling condition ( [ usl ] ) have the form @xcite : @xmath168^{-1/2}\,.\ ] ] all other eigenfunctions of @xmath1 which correspond to new levels and which are not contained in @xmath169 we keep untouched .
this yields @xmath170 here we interchanged the derivative operators present in @xmath171 and the summation . this interchange is justified because the propagators are understood not as a usual functions but as generalized functions @xcite ( which , in particular , may be regular , i.e. defined with the help of locally integrable functions ) .
it remains to express @xmath172 with the help of ( [ vc0 - 1 ] ) in terms of @xmath173 , to make use of the identity @xmath174 and to represent the sum over @xmath175 in terms of the green function @xmath176 . as a result ,
one arrives at & & k_n(x , y , t)= l_xl_y_n=0^n-1 ( _ j=0,jn^n-1 ) _
-^k_0(x , z , t)g_0(z , y,_n)dz + [ trprn ] & + & _
n=0^n- 1_n(x)_n(y)^-_nt . in this section
we need information on the intermediate transformation steps of @xmath9th - order susy - transformations which goes beyond that presented in section [ prelim ] .
therefore we start with a more detailed description of the transformation operators and solutions of the schrdinger equation at each transformation step .
let us consider a chain of @xmath9 first - order transformations @xmath177 built from operators @xmath178 which intertwine neighbor hamiltonians @xmath90 and @xmath179 as @xmath180 .
we assume all hamiltonians @xmath90 , @xmath181 self - adjoint or essentially self - adjoint in the same hilbert space @xmath60 so that the susy - transformation chain itself is completely reducible .
furthermore , we assume that at each transformation step the ground state of the corresponding hamiltonian is removed .
this means that after @xmath9 linear susy - transformations the first @xmath9 states of the @xmath76system are removed and the @xmath182st state of @xmath0 maps into the ground state of @xmath1 . in our analysis
these first @xmath182 states of @xmath0 will play a crucial role and we denote them by @xmath183 , @xmath184 .
furthermore , we use a numbering for the solutions @xmath185 of the schrdinger equations of the susy - chain hamiltonians @xmath90 , @xmath186 which is synchronized " with the level numbering of @xmath0 , meaning that a function @xmath185 is related to the spectral parameter @xmath187 .
we have to distinguish between physical solutions , which correspond to the existing bound states of @xmath90 and which have indices @xmath188 , and unphysical auxiliary solutions @xmath185 with @xmath189 which we construct below .
the ground state eigenfunction of a hamiltonian @xmath90 is given by @xmath190 and for @xmath191 it is annihilated by the susy - intertwiner @xmath178 @xmath192 the bound state functions @xmath193 , @xmath194 of @xmath179 may be obtained by acting with the susy - intertwiner [ w1 ] l_k+1,k =- u_k , k , x / u_k , k+_xl_k+1,kf= on the corresponding eigenfunctions of @xmath90 @xmath195 next , we note that the chain of @xmath196 , ( @xmath197 ) first - order transformations is equivalent to a single @xmath196th - order transformation ( [ operator ] ) generated by the transformation functions @xmath198 . furthermore , the transformation operators obey the composition rules [ lcomposition ] l_k+1,kl_k , l = l_k+1,ll=0,
,k-1 k=1, ,n-1 so that , e.g. , the second - order transformation operator @xmath199 intertwines the hamiltonians @xmath90 and @xmath200 @xmath201 the @xmath9th - order transformation operator @xmath202 is then inductively defined as @xmath203
. obviously , it annihilates the @xmath9 lowest states of the original hamiltonian @xmath0 , i.e. @xmath204 . as further ingredient for the derivation of the propagator - mapping we need the set of unphysical auxiliary functions @xmath205 , @xmath206 .
we construct them as @xmath207 , where the functions @xmath208 are the unphysical solutions of the @xmath76schrdinger equation at energies @xmath187 which are linearly independent from the eigenfunctions @xmath183 .
normalizing @xmath208 by the condition @xmath209 and integrating this wronskian gives @xmath210and finally [ ln0u ] u_n , n(x)= l_n,0u_0,n(x)_x_0^x n=0, ,n-1 . in [ ap1 ]
we show that [ norm21]u_n , n = c_n , n + c_n , n=(e_n-1-e_n)(e_n-2-e_n)
(e_n+1-e_n ) n=0, ,n-2 + u_n , n-1= = .
[ norm21a ] the ground state function of @xmath1 is obtained by acting with @xmath202 on the @xmath9th excited state of @xmath0 : [ unn ] u_n ,
n = l_n,0u_0,n = w(u_0,0, ,u_0,n-1 ) . below we will often use the abbreviations @xmath211 , @xmath212 for the wronskians ( when there is no ambiguity in the definition of the transformation functions @xmath213 in their arguments ) indicating explicitly only the dependence on the spatial coordinate . within the above framework , the @xmath9 first discrete levels
@xmath214 have been removed from the spectrum of @xmath0 by choosing the ground state functions @xmath190 of the hamiltonians @xmath90 as intermediate transformation functions .
for such a construction the transformed propagator may be calculated according to [ th3 ] let the @xmath9 first eigenfunctions @xmath215 , @xmath216 of @xmath0 be the susy transformation functions .
then the propagators @xmath164 and @xmath105 of the schrdinger equations with hamiltonians @xmath1 and @xmath0 are interrelated as [ transnl]k_n(x , y;t)&=&(-1)^nl_n,0,x _
n=0^n-1(-1)^n _ a^y k_0(x , z;t)u_n(z)dz + & = & ( -1)^n-1l_n,0,x _ n=0^n-1 ( -1)^n _ y^b k_0(x , z;t)u_n(z)dz .
[ transn2 ] the proof of these relations can be given by induction .
we start with ( [ transn2 ] ) . for @xmath104
the statement is proven in ( [ trans1l ] ) . assuming that ( [ transn2 ] ) holds for @xmath164 we verify its validity for @xmath217 .
the corresponding hamiltonians @xmath218 and @xmath1 are intertwined by the _ linear _ transformation @xmath219 so that ( [ trans1l ] ) is applicable and @xmath217 can be represented as @xmath220 replacing @xmath221 by ( [ transn2 ] ) and making use of relations ( [ norm21 ] ) , ( [ norm21a ] ) and the composition rule ( [ lcomposition ] ) gives & & ( -1)^n-1 k_n+1(x , y;t)=l_n+1,0,x + & & _ n=0^n-1 ( -1)^nc_nn^-1 _ y^bdz_z^bdqu_n , n(z)u_n , n(z ) k_0(x , q;t)u_0,n(q ) .
[ propint ] the integration region of the double integral is the upper triangle of the rectangle @xmath222 in the @xmath223plane .
we replace this double integral by the difference of two double integrals over the whole rectangle and the lower triangle , respectively , ( -1)^n-1k_n+1(x , y;t)= l_n+1,0,x_n=0^n-1 ( -1)^nc_nn^-1 + .[44 ] here , we reshape the two integrals of the type @xmath224 as follows . first we note that @xmath205 and @xmath225 are solutions of the same schrdinger equation with hamiltonian @xmath1 and therefore _
^b dq u_n , n(q)u_n , n(q ) & = & -w_b , n + & = & -w_b , n [ intuu-2 ] where @xmath226/(e_n - e_n)$ ] and where the second equality was obtained via ( [ w1 ] ) .
applying the general relation ( [ norm21 ] ) to @xmath227 leads finally to [ aux - int-2 ] _
^b dq u_n , n(q)u_n , n(q ) & = & -c_n , nu_n , n - w_b , n + & = : & -c_n , nu_n , n()-w_b , n . with ( [ aux - int-2 ] ) as substitution rule the propagator ( [ 44 ] ) takes the form [ spa ] ( -1)^n-1k_n+1(x , y;t)=-l_n+1,0,x_n=0^n-1 ( -1)^n _ y^b k_0(x , z;t)u_0,n(z)dz + + l_n+1,0,x_y^bdz k_0(x , z;t)_n=0^n-1 ( -1)^nu_0,n(z)w_n(z ) .
( the terms containing @xmath228 exactly cancelled . )
the sum [ xx1 ] s_n:=_n=0^n-1 ( -1)^nu_0,nw_n(u_0, ,u_n ) in the second term can be calculated explicitly .
comparison with the evident determinant identity 0= | ccccu_0,0 & & u_0,n-1&u_0,n + u_0,0 & & u_0,n-1&u_0,n + & & &
+ u_0,0^(n-1 ) & & u_0,n-1^(n-1)&u_0,n^(n-1 ) + = | ccccu_0,0 & & u_0,n-1&0 + u_0,0 & & u_0,n-1&u_0,n + & & &
+ u_0,0^(n-1 ) & & u_0,n-1^(n-1)&u_0,n^(n-1 ) + cccc0 & & 0&u_0,n + u_0,0 & & u_0,n-1&u_0,n + & & &
+ u_0,0^(n-1 ) & & u_0,n-1^(n-1)&u_0,n^(n-1 ) + determinant in the second line of ( [ xx3 ] ) over the elements of its first row .
hence , it holds @xmath229 representing the ground state eigenfunctions @xmath225 in ( [ spa ] ) via ( [ unn ] ) in terms of wronskian fractions we find that @xmath230 and , hence , that the second term in ( [ spa ] ) is nothing but the absent @xmath231 summand of the sum in the first term . as a result , we arrive at @xmath232 what completes the proof of ( [ transn2 ] ) .
the representation ( [ transnl ] ) follows from ( [ transn2 ] ) and the relations @xmath233 we note that in formulas ( [ transnl ] ) and ( [ transn2 ] ) only one - dimensional integrals are present . in this way
, they may turn out more convenient for concrete calculations than similar equations derived in @xcite .
furthermore , we note the following .
theorem [ th3 ] is proven for the case when the @xmath9 lowest discrete levels are removed from the spectrum of @xmath0 starting from the ground state level .
this scenario corresponds to reducible supersymmetry . in order to see which of the conditions on the transformation functions @xmath183 used for the construction of the propagator representations ( [ transnl ] ) and ( [ transn2 ] ) are indeed necessary conditions one may simply insert @xmath164 directly into schrdinger equation ( [ k00 ] ) .
it turns out that neither the condition of level deletion starting from the ground state nor a deletion of a level block without surviving levels inside is used .
this means that equations ( [ transnl ] ) and ( [ transn2 ] ) hold for any choice of transformation functions provided their wronskian does not vanish inside the interval @xmath6 , i.e. it holds for reducible as well as for irreducible susy transformation chains .
a necessary but in general not sufficient condition for the nodelessness of the wronskian is inequality ( [ usl ] ) ( for further details see @xcite ) .
strictly isospectral transformations can be generated with the help of unphysical solutions of the schrdinger equation as transformation functions . in this section ,
we extend theorem [ th3 ] to models defined over the whole real line @xmath234 by assuming transformation functions which vanish at one of the infinities @xmath235 or @xmath236 and violate the dirichlet bcs at the opposite infinities ( @xmath236 or @xmath237 ) .
in accordance with ( [ transnl ] ) and ( [ transn2 ] ) we formulate the corresponding relaxed version of theorem [ th3 ] as [ th3a ] let the transformation functions @xmath32 vanish at only one of the infinities @xmath235 or @xmath236 of the real axis @xmath56 . then the propagators @xmath164 and @xmath105 of the schrdinger equations with @xmath1 and @xmath0 as hamiltonians are related as follows : + @xmath238 @xmath239 @xmath240 @xmath241 @xmath242 [ transnllr]k_n(x , y;t)&=&(-1)^nl_x_k=0^m(-1)^n _
k_0(x , z;t)u_k(z)dz + & & + ( -1)^n-1l_x_m = m+1^n-1 ( -1)^m_y^ k_0(x , z;t)u_m(z)dz .
we have to verify that the initial condition @xmath243 and the schrdinger equations @xmath244 and @xmath245 fulfilled by the original propagator @xmath246 map into corresponding relations for the final propagator @xmath247 , i.e. that @xmath248 , @xmath249 and @xmath250 are satisfied .
we demonstrate the explicit proof for the setup with @xmath251 omitting the technically identical considerations for the other cases .
we start by noticing that the intertwiner @xmath252 maps solutions of the schrdinger equation for @xmath0 into solutions of the schrdinger equation for @xmath1 and , hence , @xmath249 is automatically satisfied .
next , we consider the initial condition @xmath253 which should be fulfilled by the r.h.s . of ( [ transnllr ] ) . with @xmath243 and @xmath254
we have from ( [ transnllr ] ) [ t4p1]k_n(x , y;0)&=&(-1)^n_n=0^n-1(-1)^n l_n,0,x[(y - x)u_n(x ) ] . in the crum - krein formula ( see ( [ operator ] ) ) [ t4p2 ] l_n,0,x [ ( x- y)u_n(x)]= we represent the derivatives @xmath255,\ m=0,\ldots , n$ ] as [ der]_x^m[(y - x)u_n(x)]= _ k=0^m-1c_k^m_x^(m - k)(y - x)u_n^(k)(x ) + ( y - x)u_n^(m)(x ) , @xmath256 .
taking into account that @xmath257 and making use of the linearity properties of determinants we reshape ( [ t4p2 ] ) as [ kn1 ] & & l_n,0,x [ ( x- y)u_n(x ) ] + & = & 1w(x)| ccccu_0(x ) & & u_n-1(x)&0 + u_0(x)& &u_n-1(x)&-(x - y)u_n(x ) + & & & + u_0^(n)(x)& &u_n-1^(n)(x)&_k=0^n-1c_k^n_x^(n - k)(y- x)u_n^(k)(x ) column we find [ t4p3 ] l_n,0,x [ ( x- y)u_n(x)]=_m=1^n(-1)^m w_nm(x)_k=0^m-1c_k^m_x^(m - k)(y- x)u_n^(k)(x ) where @xmath258 are corresponding minors .
for the verification of the relation @xmath253 we use its representation [ dlt ] _
-^k_n(x , y;0)f(x)dx = f(y ) where @xmath259 is a sufficiently smooth test function with compact support . with ( [ t4p1 ] ) and ( [ t4p3 ] ) the l.h.s . of ( [ dlt ] ) reads [ t4p4]_n=0^n-1_m=1^n_k=0^m-1(-1)^n+mc_k^m _ -^dx _ x^(m - k)(y - x ) . as next step
, we use @xmath260 and multiple integration by parts to remove the derivatives from the @xmath261functions : [ t4p5]= _ n=0^n-1_m=1^n_k=0^m-1(-1)^n - kc_k^m _ y^m - k-1 . the relation [ i d ] _
n=0^n-1 ( -1)^nw_nu_n^(j)=(-1)^n_j , n-1,j=0, ,n-1 reduces this multiple sum to [ t4p6 ] = ( -1)^n_k=0^n-1(-1)^k c_k^n and because of @xmath262 and @xmath263 ( cf .
4.2.1.3 in @xcite ) the condition ( [ dlt ] ) is satisfied .
it remains to prove that the schrdinger equation @xmath250 is fulfilled too . by explicit substitution of equation ( [ transnll ] )
we have [ t4p7]&&(_t - h_ny)k_n(x , y , t + & & = ( -1)^n l_x_n=0^n-1 ( -1)^n _ -^y _ tk_0(x , z;t)u_n(z)dz + & & + l_x_n=0^n-1 ( -1)^n_-^y k_0(x , z;t)u _ n(z)dz + & & + l_x_n=0^n-1 ( -1)^nk_0(x , y;t ) .
first , we note that due to relation ( [ i d ] ) and its derivative the last sum vanishes .
taking further into account that @xmath264 is a solution of the schrdinger equation for @xmath1 at energy @xmath187 ( cf .
( [ hnsol ] ) ) and replacing @xmath265 one reduces equation ( [ t4p7 ] ) to [ t4p8]&&(_t - h_ny)k_n(x , y , t ) + & = & ( -1)^n l_x_n=0^n-1 ( -1)^n _ -^y [ ( h_0z - e_n)k_0(x , z;t)]u_n(z)dz . integrating by parts and making use of @xmath266 , the asymptotical behavior @xmath267 , @xmath268 and relation ( [ i d ] )
one finds that the r.h.s . in ( [ t4p8 ] ) vanishes and , hence , the schrdinger equation @xmath250 is fulfilled .
the three different types of transformations considered above may be combined in various ways to produce a supersymmetry of more general type . in general , from the spectrum of the original hamiltonian @xmath0 @xmath69 levels may be removed and @xmath68 additional levels may be added , @xmath269 producing in this way the spectral set of @xmath1 .
the inequality would correspond to susy transformation chains between @xmath0 and @xmath1 which contain isospectral transformations .
for further convenience we split the spectra of @xmath0 and @xmath1 according to their transformation related contents as ( h_0)&=&\{_i,_j , e_k}+_c(h_0 ) , i=1 , , q ; j=1 , ,
n-(p+q+r ) + ( h_n)&=&\{_l,_j , e_k}+_c(h_n ) , l=1 , , p ; j=1 , , n-(p+q+r ) where the discrete levels @xmath270 and the continuous spectrum @xmath271 are not affected by the sysy transformations .
the set of transformation constants @xmath272 corresponds to @xmath68 new discrete levels @xmath273 not present in @xmath169 , @xmath69 levels @xmath274 not present in @xmath275 , @xmath276 levels @xmath277 present in both spectra and @xmath278 constants @xmath279 not coinciding with any energy level of both hamiltonians , @xmath280 .
transformations induced at constants @xmath281 are strictly isospectral .
summarizing the previous results the following expression for the propagator @xmath247 can be given [ trprnp ] k_n(x , y , t)= l_xl_y_n=0^n-1 ( _ j=0,jn^n-1)_- ^k_0(x , z , t)_0(z , y,_n)dz + + _ _ n__n(x)__n(y ) ^-_nt + _ _ n__n(x)__n(y ) ^-_nt where for @xmath282 @xmath283\ ] ] and @xmath284 otherwise .
as first application we consider a free particle in a box , i.e. the schrdinger equation with @xmath285 and dirichlet bcs at the ends of the finite interval @xmath286 .
the eigenfunctions of this problem are the well known @xmath287 , @xmath288 with energies @xmath289 .
the corresponding propagator reads @xcite @xmath290\ ] ] with @xmath291 and @xmath292 denoting the third theta function @xcite . as susy partner problem
we choose a model which we derive by removing the ground state level @xmath97 by a linear ( one - step ) susy - mapping with @xmath293 as transformation function .
this leads to the schrdinger equation with potential @xmath294 , i.e. a singular sturm - liouville problem .
the propagator of this problem can be represented via ( [ trans1l ] ) as [ kbb]k_box1(x , y , t)=-l_x_0^y(z ) dz or after explicit substitution of @xmath295 as k_box1(x , y , t)= _ 0^y(z ) dz + -_0^y(z ) dz + . here after using the property @xmath296 we integrated in ( [ kbb ] ) by parts .
here we consider the hamiltonian @xmath297 using a two - fold transformation with transformation functions @xmath298 corresponding to the second and third excited state eigenfunctions of @xmath0 we obtain a perturbed harmonic oscillator potential @xcite [ p23 ] v^(2,3)(x)=- + + 2 which for large @xmath299 behaves like the original harmonic oscillator potential , but for small @xmath299 shows two shallow minima .
for completeness , we note that the transformation functions @xmath300 and @xmath301 have nodes , whereas their wronskian @xmath302 is nodeless so that the corresponding second - order susy - transformation itself is well defined , but irreducible .
the propagator for the schrdinger equation with hamiltonian @xmath303 can be constructed from the well known propagator @xmath304}{4\sin t}}\,.\ ] ] for the @xmath76schrdinger equation ( see e.g. @xcite ) via relation ( [ transnlr ] ) .
the occurring integrals @xmath305 can be represented as derivatives with respect to the auxiliary current @xmath306 of the generating function [ ist ] s(j , x , y , t)&= & _ y^dz + & = & + & & ( 1 + ) . as a result , we obtain k^(2,3)(x , y , t ) + = ^y^2/2l_x ( -)_j=0 in terms of obvious abbreviations
. the technique can be straight forwardly generalized to second - order susy - transformations built on any eigenfunction pair @xmath307 of the hamiltonian @xmath0 .
the corresponding generalized potentials @xmath308 ( see @xcite ) read [ pkk1 ] v^(k , k+1)(x)=- 2 + 2 ^ 2++4 where @xmath309 are built from the re - scaled hermite polynomials @xmath310 . with the help of ( [ transnl ] )
the propagator is obtained as k^(k , k+1)(x , y , t ) = ^y^2/2l_x ( - ) _ j=0 with @xmath311 denoting the @xmath196th - order differential operator obtained from @xmath312 by replacing @xmath313 .
finally we note that the potentials ( [ pkk1 ] ) behave asymptotically like @xmath314 for @xmath315 and have @xmath196 shallow minima at their bottom . here
we apply our propagator calculation method to transparent potentials which are susy partners of the zero potential in models defined over the whole real axis .
we note that the propagator for a one - level transparent potential was previously calculated by jauslin @xcite for the fokker - planck equation .
the propagator for a two - level potential can be found in @xcite where the general form of the propagator for an @xmath9-level potential has been given as a conjecture . here
we will prove this conjecture .
we construct the @xmath9-level transparent potential from the zero potential with the help of the following @xmath9 solutions of the schrdinger equation with @xmath285 ( see e.g. @xcite ) [ fpfs ] & & u_2j-1(x)=(a_2j-1x+b_2j-1 ) + & & u_2j(x)=(a_2jx+b_2j)j=1,2,
,[(n+1)/2 ] as transformation functions . here , @xmath316 , @xmath317 is assumed and @xmath318 $ ] denotes the integer part of @xmath319 .
the factorization constants @xmath320 define the positions of the discrete levels ( point spectrum ) @xmath321 of @xmath322 , whereas the continuous part of the spectrum of @xmath1 fills the whole real axis .
the eigenfunctions of the discrete levels normalized to unity have the form @xcite @xmath323^{1/2 } \frac{w(u_1,u_2,\ldots , u_{n-1},u_{n+1},\ldots , u_n)}{w(u_1,u_2,\ldots , u_n)}\,.\ ] ] the propagator @xmath324 and the green function @xmath325 for the free particle are well - known @xcite [ k0free ] k_0(x , y;t)= ^ + g_0(x , y , e)=^|x - y|>0e=^2 so that according to ( [ trprn ] ) the propagator of the transformed system can be calculated as [ k1 ] & & k_n(x , y , t)= _ n=1^n ( _ j=1,jn^n ) _
-^ ( -a_n|z - y| ) + & & + _
n=1^n_n(x)_n(y)^-_nt = : k_cn(x , y , t)+k_dn(x , y , t ) . in the last line we separated contributions from the continuous spectrum , @xmath326 , from contributions from the discrete spectrum , @xmath327 .
the integral in @xmath326 can easily be calculated since the primitive of the integrand is well known ( see e.g. integral 1.3.3.17 of ref @xcite ) .
using the well studied convergency conditions of the error - function erfc from @xcite we find i(a , x , y , t):= _ -^ ( -a|z - y| ) + = + a>0t>0and with it [ kc ] k_cn(x , y;t ) = l_xl_y_n=1^n(_m=1,m n^n ) i(a_n , x , y , t ) . introducing the notation @xmath328 and abbreviating the wronskian of the @xmath329dependent transformation functions @xmath330 , , @xmath331 as @xmath211 we formulate the final expression for the propagator as [ th4 ] the propagator for an @xmath9-level transparent potential has the form [ nspp]k_n(x , y;t)= ^ + + _ n=1^n ( _ j=1,jn^n|a_n^2-a_j^2| ) ^a_n^2 t .
for the proof of this theorem we need the additional [ lm3 ] let @xmath332 be a set of @xmath9 non - coinciding complex numbers @xmath333 and for the natural numbers with zero included . ]
then the following identity holds : [ s1 ] _
_ i^n(_j=1,ji^n)=_n , n-1 .
consider the auxiliary function @xmath335 which is analytic in any finite part of the complex @xmath336-plane except for @xmath9 simple poles @xmath337 . from the residue theorem
follows @xmath338 what with the residue at infinity yields ( [ s1 ] ) .
as the next step we prove theorem [ th4 ] . without loss of generality we may set @xmath339 .
we start with the propagator component @xmath326 in ( [ kc ] ) related to the continuous spectrum .
first we note that the function @xmath340 in ( [ kc ] ) depends only via the difference @xmath341 on the spatial coordinates so that the action of @xmath342 in @xmath151 can be replaced by @xmath343 .
hence , the composition of the two @xmath9th - order transformation operators @xmath344 acts as an effective @xmath345th - order differential operator in @xmath28 [ l2x ] l_xl_y = r_0+r_1_x+ +r_2n_x^2n with coefficient functions @xmath346 .
accordingly , ( [ kc ] ) takes the form [ kcc ] k_cn(x , y;t ) = _
n=1^n(_m=1,m n^n)[r_0+r_1_x+
+r_2n_x^2n ] j_n where @xmath347 . from the explicit structure of the first derivatives of @xmath348 = ^a_n^2 t + = ^+a_n^2i(a_n ) + = ( x - y)^ + + ^a_n^2 t we find ( by induction ) the general expression of an even - order derivative & = & _ k=0^m- 1a_n^2ka_km(x , y , t ) ^ + & + & ^a_n^2 tand of an odd - order derivative & = & _ k=0^m- 2a_n^2ka_km(x , y , t ) ^ + & - & ^a_n^2 t. it holds @xmath349
subsequently we use the abbreviation [ in]i_m(a_n)= ^a_n^2 t and we will need the explicit form of the @xmath350 and @xmath351 terms in the highest - order derivative [ hd ] =
-a_n^2(n-1 ) ^ + a_n^2ni_2n .
the complete propagator component @xmath352 in ( [ kcc ] ) can now be rewritten as k_cn(x , y;t ) = _
n=1^n(_m=1,m n^n)r_0i_0(a_n ) + + _ n=1^n ( _ m=1,m n^n ) a_nr_1i_1(a_n ) + + _ n=1^n ( _ m=1,m n^n ) r_2 + + + _ n=1^n ( _ m=1,m n^n ) r_2n .
comparison with lemma [ lm3 ] shows that due to @xmath353 only a single term containing @xmath354 does not vanish .
it is the @xmath355 term in the very last sum which with @xmath356 and @xmath357 yields exactly the propagator ( [ k0free ] ) of the free particle .
all other terms containing @xmath358 , after interchanging the sums @xmath359 and @xmath360 , cancel out due to lemma [ lm3 ] .
thus , the above formula reduces to [ ksol]k_cn(x , y;t ) = ^ + _
n=1^n(_m=1,m n^n ) + . expressing @xmath361 via ( [ in ] ) in terms of @xmath362 and @xmath363 one finds contributions @xmath364 @xmath365 which can be represented as @xmath366 @xmath367 comparison with ( [ l2x ] ) shows that the sums yield simply @xmath368 $ ] . recalling furthermore that the transformation operators @xmath252 and @xmath151 are given by ( [ operator ] ) and that the transformation functions have the form ( [ fpfs ] ) we arrive after some algebra at ) is compatible with ( [ ortef ] ) .
the function @xmath369 is a solution of the initial schrdinger equation related to one of the factorization constants and it is linearly independent from the corresponding factorization solution .
therefore @xmath370 up to a constant factor is one of the bound state functions of @xmath1 given in ( [ ortef ] ) . ]
[ s0 ] l_x^a_nx= ( 1)^n(-1)^n+n-1a_n_j=1,jn^n ( a_j^2- a_n^2 ) and , hence , at [ sl ] l_xl_y^a_n(x - y)= ( -1)^na_n^2_j=1,jn^n ( a_n^2- a_j^2)^2 . substituting ( [ sl ] ) into ( [ ksol ] )
we obtain [ ksol1 ] k_cn(x , y;t ) = ^ + _
n=1^n e^ia_n^2 ta_n ( _ m=1,m n^n ) + .
a further simplification can be achieved by recalling that @xmath371 and that the parameters @xmath372 are ordered as @xmath373 . setting @xmath374 in the denominator for @xmath375 gives an additional sign factor @xmath376 .
the propagator sum @xmath377 resulting from continuous and discrete spectral components reads then [ ksol2]k_n(x , y;t ) = ^ - _ n=1^n e^a_n^2 ta_n ( _ m=1,mn^n|a_n^2- a_m^2| ) + + + _
n=1^n e^a_n^2 ta_n ( _ m=1,mn^n|a_n^2- a_m^2| ) . and via the relation @xmath378 it leads to the expression in the theorem .
a careful study of propagators for susy partner hamiltonians is given .
assuming the partner hamiltonians linked by polynomial supersymmetry of a general type we derived user friendly expressions interrelating the corresponding associated propagators .
since the propagators may be also defined in terms of continual integrals the results should be useful in exploring new classes of continual integrals .
we applied our general technique to derive propagators for transparent potentials and for a family of susy partner potentials of the harmonic oscillator .
further applications can be expected in time - evolution scenarios for particles placed in corresponding potential fields , in stochastic processes described by focker - planck equations ( see e.g. @xcite ) , in statistical physics and possibly in dynamo processes of magnetohydrodynamics @xcite .
the work has been partially supported by the grants rfbr-06 - 02 - 16719 ( b.f.s . and a.m.p ) ,
ss-5103.2006.2 ( b.f.s . ) , by the russian dynasty foundation ( a.m.p . ) and by the german research foundation dfg , grant ge 682/12 - 3 ( u.g . ) .
here we derive a representation of the unphysical auxiliary solutions @xmath205 , @xmath216 of the @xmath379schrdinger equation in terms of wronskian fractions .
we start by recalling that the functions @xmath205 are defined by acting with an @xmath9th - order ( polynomial ) susy - transformation operator @xmath202 on the unphysical solutions @xmath380 of the schrdinger equation with hamiltonian @xmath0 at energies @xmath381 .
the functions are linearly independent from their physical counterparts @xmath183 and normalized as @xmath382 .
the operator @xmath202 itself is constructed from the @xmath183 , @xmath216 as transformation functions . below we show by induction that the unphysical solutions @xmath383 defined in ( [ ln0u ] ) as @xmath384 have the following representation in terms of wronskian fractions [ norm ] u_n , n = c_n , n + c_n , n:=(e_n-1-e_n)(e_n-2-e_n )
( e_n+1-e_n)n=0, ,n-2 + u_n , n-1= = .[a2 ] for @xmath40 the first order transformation operator @xmath385 is constructed with the help of @xmath386
. therefore applying ( [ l1 ] ) with @xmath387 yields @xmath388 which obviously agrees with the statement . applying ( [ l1 ] ) with @xmath389
we obtain [ u11 ] u_1,1=l_1,0u_0,1= . in order to prove the statement for @xmath390 we build the linear susy - operator @xmath391 from @xmath392 and act with it on the unphysical solutions @xmath393 and @xmath394 . as first result
we obtain @xmath395 where ( [ l1 ] ) and ( [ u11 ] ) have been used . for the second function
we find u_2,0=l_2,1u_1,0=l_2,1=l_1,0^+u_1,1 = l_1,0^+l_1,0u_0,1 + = ( e_1-e_0)u_0,1= ( e_1-e_0 ) . here
the operator @xmath396 has been replaced by @xmath397 with the help of @xmath398 .
moreover , we have used ( [ u11 ] ) , the factorization rule @xmath399 and the schrdinger equation @xmath400 .
assuming finally that the representations ( [ norm ] ) and ( [ a2 ] ) are valid for @xmath205 and @xmath401 we prove them for @xmath402 and @xmath403 .
we start with @xmath404 . to go from @xmath1 to @xmath218 only the linear ( one - step ) susy transformation @xmath219 is required . applying ( [ l1 ] ) and combining it with the normalization condition @xmath405 and the crum - krein formula ( [ operator ] ) [ or ( [ unn ] ) ] we find the equivalence chain @xmath406 comparison with ( [ a2 ] ) shows that the proof is done .
the proof of the induction @xmath407 is less obvious . in section [ 4.2 ] the linear intertwiners @xmath408
have been build strictly incrementally from the corresponding ground state eigenfunctions @xmath409 ( of the hamiltonians @xmath410 ) as transformation functions . here ,
we need a more general non - incremental construction scheme . in order to facilitate it
, we first introduce a very detailed notation for general polynomial intertwiners @xmath411 indicating explicitly the energy levels of the transformation functions from which they are built .
based on the crum - krein formula ( [ operator ] ) we set [ l - def - expl ] l^(a_1,a_2, ,a_k)_k,0f:= w(u_0,a_1,u_0,a_2, ,u_0,a_k)a_ia_ji with @xmath412 being any energy level numbers of the hamiltonian @xmath0 .
the determinant structure of ( [ l - def - expl ] ) immediately implies the generalized kernel property l_k,0^(a_1, ,n , , a_k)u_0,n=0 and the invariance of the operator @xmath413 with regard to permutations @xmath414 l^(a_1,a_2, ,a_k)_k,0=l^(a_1,a_2, ,a_k)_k,0.[permute ] recalling that ( [ l - def - expl ] ) can be built from a chain @xmath415 of linear intertwiners @xmath416 we conclude that the ordering with regard to energy levels is inessential in the transformation chain and that we can split it into sub - chains with any permuted combination of levels ( a_1, ,a_k)=(b_1, ,b_b , c_1,
,c_c),b+c = k + l_k,0^(a_1, ,a_k)=l_b+c , b^(c_1, ,c_c)l_b,0^(b_1, ,b_b)=l_c+b , c^(b_1, ,b_b)l_c,0^(c_1,
,c_c ) . [ commute ] apart from this full commutativity of the transformations we note their associativity l_m+3,m+2^(a_3)(l_m+2,m+1^(a_2 ) l_m+1,m^(a_1 ) ) = ( l_m+3,m+2^(a_3 ) l_m+2,m+1^(a_2))l_m+1,m^(a_1)[associate ] . commutativity and associativity can be used to re - arrange a sequence of transformations in any required order . in accordance with the intertwiners
@xmath10 we denote eigenfunctions as [ u - def - expl ] u_k , n^(a_1,a_2, ,a_k):=l^(a_1,a_2, ,a_k)_k,0u_0,n , a_in , i=1 , , k. another ingredient that we need is the representation & = & ( -1)^n-1-n w_n(u_0,0, ,u_0,n-1 ) + & = & ( -1)^n-1-nl^(0, ,n-1,n+1, ,n-1)_n-1,0u_0,n + & = & ( -1)^n-1-nu^(0, ,n-1,n+1, ,n-1)_n-1,n=:v_n-1,n [ wn - frac ] which immediately follows from the crum - krein formula ( [ l - def - expl ] ) and definition ( [ u - def - expl ] ) . from the physical solution @xmath417 we build the operators holds @xmath418=l^{(n)}_{n , n-1}[v_{n , n}]$ ] so that the sign factor @xmath419 plays no role in the operator @xmath420 itself . ] l_n , n-1^(n)&:=&l_n , n-1^(n)[v_n-1,n]= -+ _ x + l_n , n-1^(n)+&:=&l_n , n-1^(n)+[v_n-1,n]= -- _ x[lnv ] and the corresponding hamiltonian @xmath421 with l_n , n-1^(n)+l_n , n-1^(n)=h_n-1^(0,
,n-1,n+1, ,n-1)-e_n .
we start the proof of the induction @xmath407 with the following transformations u_n+1,n&=&l^(n)_n+1,nu_n , n= c_n , nl^(n)_n+1,nw(u_0,0,
,u_0,n-1 ) + & = & c_n , nl^(n)_n+1,nv_n-1,n^-1 + & = & c_n , n + & = & c_n , n1u_n , nv_n-1,nl_n , n-1^(n)+u_n , n[unnproof1 ] . in order to obtain the operator product
@xmath422 we use ( [ unn ] ) , the composition rule ( [ commute ] ) and the crum - krein formula ( [ l - def - expl ] ) [ unnproof2 ] u_n , n&=&l_n,0^(0,1, ,n-1)u_0,n=
+ & = & l_n , n-1^(n)l_n-1,0^(0,1, ,n-1,n+1, ,n-1)u_0,n + & = & l_n , n-1^(n)u_n-1,n^(0,1, ,n-1,n+1,
,n-1 ) where u_n-1,n^(0,1,
,n-1 ) & = & l_n-1,0^(0,1, ,n-1,n+1, ,n-1)u_0,n + & = & [ un-1n ] .
this gives u_n+1,n&=&c_n , n1u_n , nv_n-1,nl_n , n-1^(n)+u_n , n + & = & c_n , n1u_n , nv_n-1,n l_n , n-1^(n)+l_n , n-1^(n)u_n-1,n^(0,1,
,n-1,n+1, ,n-1 ) + & = & c_n , n1u_n , nv_n-1,n ( e_n - e_n)u_n-1,n^(0,1, ,n-1,n+1,
,n-1 ) + & = & c_n+1,n [ unnproof3 ] where the last line has been obtained by expressing @xmath423 , @xmath225 and @xmath417 via ( [ un-1n ] ) , ( [ unn ] ) and ( [ wn - frac ] ) in terms of their wronskian fractions . with ( [ unnproof3 ] )
the proof is complete .
99 sukumar c v 1985 supersymmetric quantum mechanics of one - dimensional systems _ j. phys .
a : math . gen . _ * 18 * 291736 + cooper f , khare a a and sukhatme u 1995 supersymmetry and quantum mechanics _ phys .
rep . _ * 251 * 26785 + junker g 1996 _ supersymmetry method in quantum and statistical method _
( berlin : springer ) + bagchi b 2000 _ supersymmetry in quantum and classical mechanics _ ( new york : chapman and hall ) + mielnik b and rosas - ortiz o 2004 factorization : little or great algorithm ? _ j. phys .
a : math . gen . _ * 37 * 1000735 bagrov v g and samsonov b f 1995 darboux transformation , factorization and supersymmetry in one - dimensional quantum mechanics _ theor .
phys . _ * 104 * 105160 + bagrov v g and samsonov b f 1997 darboux transformation of the schrdinger equation _ phys .
* 28 * 37497 samsonov b f and pupasov a m 2006 exact propagators for complex susy partners of real potentials _ phys . lett . a _ * 356 * 2104 , quant - ph/0602218 + pupasov a m and samsonov b f 2005 exact propagators for soliton potentials _ symmetry , integrability and geometry : methods and applications ( sigma ) _ *
1 * 207 , quant - ph/0511238 levitan b m and sargsjan i s 1975 _ itroduction to spectral theory : selfadjoint ordinary differential operators _ translations of mathematical monographs ; * 39 * ( rhode island : ams )
+ levitan b m 1984 _ inverse sturm - liouville problems _
( moscow : nauka ) andrianov a a and sokolov a v 2006 factorization of nonlinear supersymmetry in one - dimensional quantum mechanics i : general classification of reducibility and analysis of third order algebra _
zapiski nauchnyh seminarov pomi _ * 335 * 22 - 49 samsonov b f 1999 new possibilities for supersymmetry breakdown in quantum mechanics and second - order irreducible darboux transformations _ phys .
a _ * 263 * 274 - 280 , quant - ph/9904009 + bagrov v g and samsonov b f 2002 on irreducible second - order darboux transformations _ russ . phys . j. _
* 45(1 ) * 27 - 33 dodonov v v and manko v i 1987 evolution of multidimensional systems .
magnetic properties of ideal gases of charged particles _ proceedings of the physical lebedev institute , ( trudy fian ) _ * 183 * 182286 | pairs of susy partner hamiltonians are studied which are interrelated by usual ( linear ) or polynomial supersymmetry . assuming the model of one of the hamiltonians as
exactly solvable with known propagator , expressions for propagators of partner models are derived .
the corresponding general results are applied to `` a particle in a box '' , the harmonic oscillator and a free particle ( i.e. to transparent potentials ) .
submitted to : _ j. phys .
a : math . theor . _ |
the stellar disc of the milky way , as well as of many nearby galaxies , is traditionally viewed as consisting of two distinct components : a thin and a thick disc ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) . the milky way s thick disc not only has a larger scale - height than the thin disc @xcite , but it is also older , more metal - poor , and enhanced in alpha elements @xcite .
its stars have a higher velocity dispersion , and lag behind in rotation compared to the thin disc @xcite .
many formation mechanisms have been proposed for the thick disc , such as being formed thick ( following gas rich mergers , as in @xcite , or via clumpy disc instabilities as in @xcite ) , thickened by minor mergers after it is formed @xcite or built from accreted stars @xcite .
some authors have also argued that radial migration could create a thick disc @xcite , although @xcite recently showed that radial migration alone is not efficient at heating discs .
discriminating between these formation mechanisms is a complex issue ; it has been suggested that different scenarios could be constrained by the shape of the discs @xcite , and the distribution of their stars eccentricities @xcite or the presence of an excess of stars at high altitudes @xcite .
it is worth noting that there may not be a clear thin - thick disc dichotomy in the milky way @xcite . using segue
spectroscopic survey data , @xcite have shown that the thin and thick disc are not two separate components , but that instead a continuity of properties is found when slicing the milky way disc into populations of similar chemical abundances .
they also find that each of these mono - abundance populations has a very simple spatial and kinematical structure @xcite .
there is a clear anti - correlation between the scale - height and scale - length of these populations , so that the youngest ones are also the coldest and the most radially extended ( assuming that age can be matched with chemical abundance , as has been shown to be very likely by @xcite for a sample of solar neighbourhood stars , and by @xcite for a simulated galaxy ) .
bovy et al .
argue this is clear evidence for inside - out formation , and evidence against mergers , which would have created distinct components instead of the observed smooth continuum .
similarly , @xcite also found a continuity of vertical velocity dispersion ( @xmath1 ) as a function of chemical abundance , with no obvious gap in the kinematics of old and young stars . in the light of these new observational results ,
numerical simulations have also been sliced into mono - abundance or mono - age populations @xcite and show interesting similarities with the structure of the milky way .
both studies find that mono - age populations are simple structures , and confirm that the anti - correlation between scale - height and scale - length arises as a consequence of inside - out formation .
@xcite perform a direct comparison with the bovy et al .
results , and show that the mono - abundance groups correspond roughly to mono - age groups .
they are unable to reproduce the thickness of the oldest stars in the milky way , and also do not find a continuity between thin and thick disc , but rather a clear bimodality between the two components . in this paper , we aim at extending this work by analysing a sample of seven simulated galaxies with different merger histories .
we aim at understanding which observed properties are generic features of disc evolution , and which ones are a clear indicator of a merger - free history .
this paper is the first of a series ; paper ii @xcite is focused on the age - velocity relation ( i.e. the relation between the ages of stars and their vertical velocity dispersion ) and a study of disc heating in simulations . throughout these papers ,
our goal is not to find a simulated galaxy that exactly matches all properties of the milky way , but rather to explore different types of formation histories and study their consequences for the structure of discs at @xmath0 .
we present our simulations in section 2 , and discuss in section 3 the simplicity ( or otherwise ) of the spatial and kinematical structure of mono - age populations . in section 3 , we discuss the conditions for an anti - correlation between scale - height and scale - length , and in section 5 the conditions for a dichotomy between thin and thick components . finally , we discuss in section 6 which properties are imprinted at birth or are due to subsequent evolution .
we also present constraints on the merger history of the milky way , and discuss the limitations of our approach .
we conclude and summarize in section 7 .
from the sample of 33 galaxies presented in @xcite , we select seven disc galaxies with a range of merger histories , from a galaxy with an extremely quiescent history in the last 89 gyr to a galaxy undergoing a late 1:4 merger . in this section
, we present the simulation technique and the galaxies , and explain how we analyse their structure . the galaxies are drawn from a larger sample of 33 simulated galaxies , described in @xcite .
we refer the reader to @xcite for a detailed description of the technique and the sample , and only describe here the main features .
the galaxies are simulated using a zoom - in technique in two steps ( the technical details are discussed in @xcite ) .
the first step consists in running a large dark - matter only simulation with the adaptive mesh refinement code ramses @xcite , and extracting merger and accretion histories for interesting dark matter haloes .
we target haloes with a mass between @xmath2 and @xmath3 m@xmath4 , and that are in relatively isolated environments . for these target haloes , not only
do we record the time , mass , velocity and spin of incoming satellites , but we also keep track of diffuse dark matter accretion , for instance along filaments . the second step is to re - simulate each merger and accretion history .
for that we use the particle - mesh code described in @xcite , with gas dynamics modelled by a sticky - particle algorithm .
our box size is 800 kpc , the spatial resolution is 150 pc , the mass resolution is @xmath5 m@xmath6 for gas particles , @xmath7 m@xmath6 for star particles ( @xmath5 m@xmath6 for star particles formed from gas during the simulation ) , and @xmath8 m@xmath6 for dark matter particles . for the galaxies studied here ,
there are typically 2 to 6 million star particles in the discs at @xmath0 .
star formation is modelled by a schmidt law with an exponent of 1.5 @xcite , using a gas density threshold of 0.03 m@xmath6pc@xmath9 ( i.e. , 1 h cm@xmath9 ) .
stellar mass loss is taken into account ( see @xcite ) , as well as kinetic feedback from supernovae explosions , where 20% of the energy of the supernovae is distributed to neighbouring gas particles .
each re - simulation starts at @xmath11 with a seed galaxy , and follows its evolution down to @xmath0 with mergers and diffuse accretion as prescribed by the cosmological simulation .
each halo that was tracked in the cosmological simulation is replaced with a galaxy made of a dark matter halo , a stellar disc and bulge , and a gas disc .
the stellar mass of the galaxies is chosen to obey the scaling relations of @xcite , and their gas fraction evolves with redshift .
diffuse accretion is modelled by assuming that gas follows dark matter , in an amount given by the cosmic baryon fraction . each diffuse dark matter particle of the initial simulation
is then replaced with a new blob of gas and dark matter particles .
the main advantage of this re - simulation technique is the relatively low computing time , permitting us to compile statistically interesting samples of galaxies .
the 33 galaxies introduced in @xcite span a large range of bulge fractions at @xmath0 , from nearly bulgeless to bulge - dominated systems .
@xcite studied how this final morphology compared to the morphology at earlier redshift , and discussed how the bulge fraction is connected with the merger and gas accretion history . in the sample ,
the fraction of barred galaxies at @xmath0 and higher redshift is realistic when compared to observations @xcite , and a detailed chemical modelling of one of the galaxies has shown an interesting match to many features found in the milky way @xcite .
note also that disc dynamics and radial migration properties have been compared with isolated galaxy simulations performed with a tree - sph code , and good agreement was found @xcite .
.properties of the simulated galaxies at @xmath0 : stellar mass within the optical radius , bulge fraction as measured with galfit @xcite , optical radius ( @xmath12 measured in @xmath13-band ) , disc scale - length ( @xmath14 ) obtained from an exponential fit to the stellar mass distribution .
the last column shows a brief description of their formation history ( mergers are denoted by a mass ratio ) [ cols="<,^,^,^,^,^",options="header " , ] as a function of the fraction of the stellar mass in a hot component ( i.e. with @xmath1@xmath15 km s@xmath16 ) .
the data points are shown for all simulated galaxies , at 2 @xmath14 ( in red ) and 3 @xmath14 ( in blue ) , and with two different ways of measuring @xmath1as in table [ tab - slopes ] ( dots : direct measures , triangles : from gaussian fits ) .
the inset shows a zoom on the bottom left part of the plot .
we find that the galaxies closest to isothermality ( i.e. an average vertical slope closest to 0 ) are the ones with the lowest contribution of stars in a hot component , although this is not directly related with their merger history ( g102 has a very low slope but a 1:15 merger at @xmath17 gyr and a high bulge fraction).,scaledwidth=45.0% ] from this section , we conclude that : * radial profiles are nearly always simple exponentials for all maps , independently of the merger history ( although younger stars have more complex profiles ) * vertical density profiles are also nearly always successfully fitted by simple exponentials . in all cases but one ( g92 ) , scale - heights increase with radius .
the absence of flaring seems connected with a very quiescent history * @xmath1increases with height above the disc plane for all our galaxies , although the relation flattens strongly ( or sometimes even reverses ) for old stars .
galaxies for which @xmath1rises with @xmath18 slowly have the lowest contribution of bulge / halo stars at the radius where @xmath1is measured , independently of their total bulge fraction or merger history .
@xcite find an anti - correlation of the scale - length and scale - height of mono - abundance populations in the milky way .
this corresponds to a situation where younger stars are in progressively thinner and more extended components .
this configuration was also found in the simulation presented in @xcite .
we examine if this is the case in our seven simulations , and what are the conditions that create such an anti - correlation .
we have already discussed in the previous section the evolution of scale - height and scale - length with age .
we found a smooth increase of @xmath19 with age for quiescent galaxies , and more complex dependence for galaxies with mergers ( at the exception of g106 ) .
the time evolution of scale - lengths was found to be more complex , with quiescent galaxies showing an increase of @xmath20 for younger stars , with the exception of g37 where @xmath20 is mostly constant for all stars younger than 6 gyr .
mergers influence the radial distribution of maps , and they redistribute old stars and gas . because of
these irregularities in @xmath19 and @xmath20 vs age , @xmath19 and @xmath20 are not necessarily anti - correlated for all galaxies . in figure
[ fig : scales ] we show @xmath19 as a function of @xmath20 for maps in the seven galaxies ( the value of @xmath19 we use here is the one measured at @xmath21 , but we verified that choosing a radius of @xmath22 did not change the results ) .
we find a variety of structures , even amongst the quiescent galaxies .
both g47 and g92 show a banana - shaped distribution of points , corresponding to an anti - correlation of @xmath19 and @xmath20 . for g37
no such structure is seen , and most points are clustered around @xmath20=68 kpc : as described before , @xmath20 does not increase with age for this galaxy .
three of the four galaxies with a merger show a common interesting feature in figure [ fig : scales ] , with populations at the time of the merger gathered along a line of positive slope , departing from the arched locus defined by the rest of the populations ( this is really pronounced for g48 and g102 , less so for g22 ) .
this diagonal feature corresponds to a correlation of @xmath19 and @xmath20 for the populations affected by the merger .
once again , g106 is different from the other active galaxies : in spite of a merger , the anti - correlation is well preserved .
contrary to g47 and g92 , where there is a gap between oldest stars ( red colours ) and the disc populations , in g106 we find that the populations affected by the merger fill that gap so that there is a continuity of scale - heights from the disc to the inner halo / very thick disc .
we conclude that : * the anti - correlation between @xmath20 and @xmath19 can be reproduced by simulations * the anti - correlation does not necessarily imply an absence of mergers .
in this section we explore if our simulated galaxies show a clear thin / thick disc dichotomy , or a continuum of scale - heights for populations of increasing age .
we have already discussed that mergers create gaps in the distribution of @xmath19 with age , which suggests that galaxies undergoing mergers would tend to host distinct thin and thick discs . however , even for galaxies with a continuous range of @xmath19 for maps , depending on the star formation history of the galaxy and the mass in each population , gaps could appear in otherwise smooth @xmath19 distributions when weighted by the mass in each map . in figure [
fig : mass ] we show for each simulated galaxy the star formation history within a 2 kpc - wide annulus at @xmath21 ( these values are computed from the stellar mass in each map , and are thus representative of the stars found at @xmath21 at @xmath0 independently of their birth location ) .
we indeed find that some maps contain very little mass .
for most galaxies , little mass is in populations older than 8 gyr . for @xmath23 gyr ,
the increasing mass in each map corresponds to the start of disc formation , and is due to a mixture of stars born in situ and of stars born in the rest of the galaxy which have migrated to @xmath21 by @xmath0 .
we find that mergers create v - shaped features in this relation ( this is particularly obvious for g22 , g48 and g102 , bottom panels in figure [ fig : mass ] ) .
a similar shape was found in figure [ fig : r0 ] for the evolution of the scale - length of the maps . as discussed in section [ sec : rcarac ] , this corresponds to star formation being limited to the central regions of the galaxies after a merger occurred .
this also explains why few of these stars are found at a radius of @xmath21 , and why discontinuities are observed in the time evolution of the mass of maps at that radius .
combined with discontinuities in the scale - height of maps because of mergers , this has the potential to create a dichotomy between a thin and thick disc . to explore that issue , we use a similar plot as the one introduced by bovy et al ( 2012a ) .
we study how much each map contributes to the total stellar mass surface density at a given radius as a function of their @xmath19 . in figure
[ fig : bimodality_250myr ] we show the contribution of each map , as well as the contributions binned as a function of @xmath19 ( in bins of 150 pc ) . to improve the statistics , we here consider maps in 250 myr age bins instead of 500 myr as previously ( the large number of particles in each map allows us to reduce the width of each age bin while still keeping enough star particles to determine @xmath19 ) .
this figure shows that there is a clear difference between galaxies with quiescent and active merger histories .
a first difference is that the scale - height for the thickest population is greater in galaxies with mergers relative to quiescent galaxies .
in addition , the three quiescent galaxies show a continuous distribution of scale - heights ( the shapes of these distributions being quite different from one galaxy to another ) , while the galaxies with mergers show some bimodal distributions , with clear gaps for g22 , g48 and g102 .
these gaps are a result of jumps in @xmath19 for populations influenced by the mergers , and match the time of coalescence of the last merger so that the `` thick '' populations are the pre - merger populations ( note that mergers also create jumps in the age - velocity relation as discussed in paper ii ) .
we find that g106 does not have such a clear gap , but still shows more bimodality than the quiescent galaxies .
we also note that even in the case of a continuous distribution of scale - heights , a simple two - component fit to the vertical density profile of stars could give the impression of a bimodality .
for instance , for g92 , which has a very regular and continuous structure , a decomposition into a thin and thick disc is possible ( but perhaps not physically meaningful ) , and gives scale - heights of @xmath24300 pc and @xmath24800 pc , relatively constant with radius out to @xmath25 .
this means that the standard decompositions into thin and thick discs are entirely compatible with a continuum of scale - heights for maps ( as also shown by @xcite for the milky way ) .
assuming that mono - age populations are similar to mono - abundance populations , we conclude that , indeed , a continuous distribution of @xmath19 as shown in the milky way is a strong indicator of a very quiescent merger history .
it is interesting to note that @xcite have performed a similar test for one of their galaxies undergoing mergers and they find a bimodality in the distribution of @xmath19 ( their figure 10 ) , supporting our basic picture .
and 8.1 gyr .
the evolution of each population is shown at regular times between birth and 13.7 gyr ( from dark blue to red curves ) . also shown
are the scale - length from an exponential fit of each profile . in both cases , soon after birth , the density profile is irregular and truncated at a small radius . with time , the profiles become smoother and the truncation moves out to larger radii .
while both populations end up with a very similar profile at @xmath26 gyr , their starting points are very different : in one case ( top panel ) a very flat profile with a large scale - length at birth , and in the other case a birth scale - length which is very close to the final one .
, title="fig:",scaledwidth=45.0% ] and 8.1 gyr .
the evolution of each population is shown at regular times between birth and 13.7 gyr ( from dark blue to red curves ) . also shown
are the scale - length from an exponential fit of each profile . in both cases , soon after birth , the density profile is irregular and truncated at a small radius . with time , the profiles become smoother and the truncation moves out to larger radii .
while both populations end up with a very similar profile at @xmath26 gyr , their starting points are very different : in one case ( top panel ) a very flat profile with a large scale - length at birth , and in the other case a birth scale - length which is very close to the final one .
, title="fig:",scaledwidth=45.0% ] we now study how stars in maps are re - distributed radially as a function of time . as an example , figure [ fig : radial_evol ] shows the time evolution of the radial surface density profile for two maps in galaxy g92 .
it illustrates some general features seen in our simulated galaxies .
soon after birth , the surface density profile of a map is irregular , following overdensities in the gas disc ( typically spiral arms and rings ) .
as stars migrate radially , the irregularities are blurred and the profile smoothens significantly .
this redistribution of mass occurs for all populations in all simulated galaxies .
another effect of migration is that , while the profile for young stars often shows a break in the outer regions , as stars move outwards , this break is either erased or moved to larger radii ( see also * ? ? ?
* ; * ? ? ?
this populates the outer disc with stars that were born in the inner regions ( this is the case for both populations shown in figure [ fig : radial_evol ] ) .
note that this expansion in the outer disc is matched by an increase of mass in the inner regions , to conserve the total angular momentum . to quantify the radial extension of the profiles , we measure how @xmath27 ( the radius enclosing 95% of stars ) changes with time for a given map .
the strongest variations are seen for g92 ; for instance for the population shown in the top panel in figure [ fig : radial_evol ] , @xmath27 grows from 13.5 kpc at birth ( @xmath28 gyr ) to 22 kpc at @xmath0 .
other galaxies usually show smaller variations of @xmath27 with time , but if @xmath27 evolves , it always increases with time .
we also find that the expansion of the stars is not limited to the first gyr after birth , but is a continuous process : for both maps shown in figure [ fig : radial_evol ] , the outer parts of the profile change continuously .
while exponential fits are not always the most informative way to describe the profiles ( particularly when they are irregular soon after birth ) , we also use them to characterize the profiles ( each profile is fitted from @xmath14 to @xmath27 ) .
we find that the measured scale - lengths are either constant with time , or decline ( which corresponds to a steepening of the profiles ) . the map shown in the bottom panel in figure [ fig : radial_evol ] corresponds to a situation where there is only little evolution of the scale - length ( from 7.1 kpc at birth to 6 kpc at @xmath0 ) .
by contrast , the map in the top panel of that figure has a very flat density profile at birth , with a sharp truncation .
as the truncation moves outwards with time , the overall profile also steepens . for maps where the scale - length declines with time , the strongest changes are often seen in the first gyr of evolution ,
as it is the case in the top panel of figure [ fig : radial_evol ] : for this sub - population of g92 , there is a significant steepening of the profile in the first gyr , and the scale - length remains nearly constant for the remaining @xmath24 5 gyr of evolution .
we thus find that radial migration has a strong influence on the density profiles of maps , and makes it hard to reconstruct birth properties from the present - day state : both maps shown in figure [ fig : radial_evol ] have similar profiles at @xmath0 , but in one case the stars were born with a very flat profile and a sharp truncation , while in the other case the scale - length did not change much with time . in both cases ,
though , migration had a strong influence on the outer disc , populating it with stars born in the inner disc .
the time evolution of the vertical velocity dispersion , and the sources of heating , are the focus of paper ii , to which we refer the reader for more details .
we here briefly summarize the findings of that paper .
we find that the vertical structure of the oldest stars ( ages greater than 89 gyr ) is mostly imprinted at birth : these stars are born hot in a violent phase at high redshift .
after that , though , stars are mostly born with a constant ( or slightly decreasing ) velocity dispersion , at least in galaxies with a quiescent merger history .
we find that the shape of the age - velocity relation is then due to heating after birth .
we find that radial migration is not the source of that heating , at least at the radius of 2@xmath14 that we studied in paper ii : at that radius , the stars born in situ mostly determine the total @xmath1 .
stars coming from the inner disc have a slightly larger @xmath1 , and stars from the outer disc a smaller @xmath1 , so that the net effect of migration on the total @xmath1is small ( see figure 7 in paper ii ) .
we discuss in paper ii that the sources of heating vary from one galaxy to another , but include heating due to disc growth , and heating due to a combination of spiral arms and bars coupled with overdensities in the disc and vertical bending waves .
mergers with a ratio of 1:10 should be quite common for milky - way mass galaxies in the last 10 gyr @xcite , and are usually thought to be quite damaging to the discs @xcite , even if their effect might have been overestimated @xcite .
indeed , @xcite argue that a recent 1:10 merger would not be in conflict with the present - day structure of the milky way , provided that the milky way contained at least 20% of gas at the time of the merger .
our simulations suggest that the structure of maps is very sensitive to these mergers ( and is thus a better tracer of the merger history than the global bulge fraction ) .
we showed that mergers in the range 1:101:15 happening in the last 7 gyr leave some clear jumps in the avr and create bimodal distributions of scale - heights .
these results , in combination with the observations presented in @xcite , would mean that the milky way has experienced no such merger since @xmath29 ( as already speculated for instance by @xcite ) .
it is however very likely that mergers played an important role in shaping the density distribution of older stars .
indeed , we find that internal evolution is not enough to produce the hottest thick disc stars , such as are found in the milky way .
this is illustrated by g92 , which shows a very regular and smooth structure for most maps , and density profiles that do not flare vertically . in that
, g92 is probably similar to the milky way .
however , the range of scale - heights found for the disc , from 200 to 800 pc for young to old stars , is slightly smaller than in the milky way , where the thickest stars have a scale - height of 1000 pc @xcite .
we also find that in g92 the oldest disc stars only reach a vertical velocity dispersion of @xmath24 30 km s@xmath16(for a galactocentric distance of 2@xmath14 , see figure [ fig : prof_sigma ] and paper ii ) , while @xcite find that the oldest local disc populations have a @xmath1of about 50 km s@xmath16 .
some violent early event seems necessary to produce the thickest and hottest disc populations ( in addition to building the halo ) , as is the case for g106 , and as is demonstrated in @xcite by comparing g106 with rave data .
this idea is consistent with the results of @xcite , where the thick disc of the milky way is found to be made of two components , the oldest one likely coming from mergers , and the rest being formed by internal mechanisms .
we have not discussed thick disc formation by clump instabilities , since this is not a mechanism that significantly affects the evolution of our selected simulated galaxies .
it is however a possible mechanism for forming thick discs .
it remains to be seen if it is consistent with the complex chemical signatures discussed in @xcite .
we study a sample of seven simulated galaxies , which we slice into stellar populations in age bins of 500 myr ( `` mono - age populations '' , or maps ) .
we measure the spatial and kinematical properties of these populations to study if and how they retain traces of the formation history of their host galaxies . in particular , we have studied the influence of mergers on the structure of maps by comparing galaxies with no significant merger in the last @xmath30 gyr ( 3 galaxies in our sample ) to galaxies with various types of mergers at low redshift ( 4 other galaxies ) .
all the studied galaxies undergo a phase of merging activity at @xmath31 , so that stars older than 9 gyr are always found in a spheroidal and centrally concentrated component .
a fraction of these stars are accreted from the merging satellite galaxies .
younger stars are mostly found in the discs , and the properties of these discs are the main focus of this paper .
we do not discuss the properties of bulges and bars , and often present quantities measured at a galactocentric radius of 2 or 3 @xmath14 , where @xmath14 is the global disc exponential scale - length .
we first find that most maps have radial surface density profiles that can be fitted by an exponential ( section 3.1 ) , with the exception of the youngest stars ( which have complex profiles , following spiral arms and rings found in the gas discs ) .
the complexity of radial profiles for young stars is actually common at all redshifts , and the final profiles result from gradual smoothing of the initial distribution because of radial migration of stars . in most cases ,
most of the radial evolution occurs in the first gyr after birth , but with ongoing changes down to @xmath0 .
an important feature of migration is that it is very efficient at populating the outer disc with stars born in the inner regions , and it has a strong impact on the overall shape of the density profiles ( section 6.1 ) . in the end
, we find that younger stars tend to have larger scale - lengths , both as a result of their properties at birth and of a steepening of the profiles with time .
mergers have a noticeable effect on radial density profiles , by both redistributing the old stars and changing the spatial distribution of newly formed stars .
we have also studied the vertical density profile and vertical velocity dispersion of maps ( sections 3.2 and 3.3 ) .
we find that the vertical density profiles can be fitted by simple exponentials .
looking at the variation of scale - height with radius , we often see an increase of the scale - height at larger radius , giving a flared geometry to the maps .
the three quiescent galaxies show the simplest structures : populations of increasing age have an increasing scale - height , and this is true at all radii .
mergers introduce some complexity , and also seem to increase the flaring of populations affected by mergers ( as expected from previous studies ) .
we similarly find that @xmath1is an increasing function of age ( see details in paper ii ) .
the increased @xmath1for older stars comes from two different origins : oldest stars are truly born kinematically hot ( following the mergers happening at high redshift ) , while for stars younger than @xmath32 gyr we find that @xmath1at birth is small and nearly constant with time .
these stars are subsequently heated to their present day @xmath1 .
we find that radial migration is only a minor source of heating ( if any ) .
we also show in paper ii that @xmath1is very sensitive to mergers , which create jumps in the age - velocity relation .
finally , we quantify the isothermality of each map by measuring @xmath1at a fixed radius , but as a function of height above the disc .
we find that @xmath1increases with height above the disc plane for all our galaxies .
galaxies closest to isothermality have the lowest contribution of bulge / halo stars at the radius where @xmath1is measured , independently of their total bulge fraction or merger history . as a result of the properties at birth , radial redistribution and vertical heating
, we find that the relation between scale - length and scale - height is complex ( section 4 ) . assuming the equivalence between mono - age and mono - abundance populations , the anti - correlation observed in the milky way @xcite is reproduced by some of our simulations ( @xcite show however that even if such an anti - correlation is found for mono - age populations , it might not be seen for mono - abundance populations ) . in any case
, the presence of an anti - correlation does not necessarily imply an absence of mergers : we both find a quiescent galaxy with no anti - correlation ( g37 ) and a galaxy with mergers but a really nice anti - correlation ( g106 ) .
we suggest that a better indicator of past mergers is the repartition of stellar mass as a function of scale - height ( section 5 ) .
we find that our quiescent galaxies show a continuous distribution of scale - heights , while a bimodality between a thick and thin component is found when mergers occur .
the continuous distribution found in the milky way by @xcite points then towards an evolution dominated by secular processes .
we also note that even in the case of a continuous distribution of scale - heights , a simple two - component fit to the vertical distribution of stars could give the impression of a bimodality .
this means that the standard decompositions into thin and thick discs are totally compatible with a continuum of scale - heights for mono - age populations .
however , our simulations do not produce the thickest , hottest components in the absence of mergers .
that would suggest that the old component of the milky way has a mixed origin , also in agreement with @xcite .
we thank the referee for a constructive report . we thank hans - walter rix , jo bovy and alex bdenbender for useful comments .
mm is supported by a humboldt research fellowship .
cf acknowledges financial support by the beckwith trust .
plain abadi m. g. , navarro j. f. , steinmetz m. , eke v. r. , 2003 , apj , 597 , 21 allende prieto c. , beers t. c. , wilhelm r. , newberg h. j. , rockosi c. m. , yanny b. , lee y. s. , 2006 , apj , 636 , 804 aumer m. , white s. d. m. , 2013 , mnras , 428 , 1055 bensby t. , feltzing s. , lundstrm i. , ilyin i. , 2005 , a&a , 433 , 185 binney j. , tremaine s. , 1987 , princeton university press bird j. c. , kazantzidis s. , weinberg d. h. , guedes j. , callegari s. , mayer l. , madau p. , 2013
, apj , 773 , 43 bond n. a. , et al . , 2010 , apj , 716 , 1 bournaud , f. , combes , f. , 2002 , a&a , 392 , 83 bournaud , f. , combes , f. , 2003 , a&a , 401 , 817 bournaud f. , elmegreen b. g. , martig m. , 2009 , apj , 707 , l1 bovy j. , rix h .- w .
, hogg d. w. 2012a , apj , 751 , 131 bovy j. , rix h .- w . , liu c. , hogg d. w. , beers t. c. , lee , y. s. 2012b , apj , 753 , 148 bovy j. , rix h .- w . , hogg d. w. , beers t. c. , lee , y. s. , zhang l. 2012c , apj , 755 , 115 bovy j. , rix h .- w . , 2013 , arxiv , arxiv:1309.0809 brook c. b. , kawata d. , gibson b. k. , freeman k. c. , 2004 , apj , 612 , 894 burstein , d. 1979 , apj , 234 , 829 chen b. , et al .
, 2001 , apj , 553 , 184 comern s. , et al . , 2011 ,
apj , 741 , 28 dalcanton j. j. , bernstein r. a. , 2002 , aj , 124 , 1328 di matteo p. , lehnert m. d. , qu y. , van driel w. , 2011 , a&a , 525 , l3 fathi k. , allen m. , boch t. , hatziminaoglou e. , peletier r. f. , 2010 , mnras , 406 , 1595 freudenreich h. t. , 1998 , apj , 492 , 495 fuhrmann k. , 1998 , a&a , 338 , 161 gilmore g. , reid n. , 1983 , mnras , 202 , 1025 gilmore g. , wyse r. f. g. , jones j. b. , 1995 , aj , 109 , 1095 haywood , m. , di matteo , p. , lehnert , m. d. , katz , d. , gmez , a. 2013 , a&a , 560 , a109 juri m. , et al . , 2008 , apj , 673 , 864 kazantzidis s. , bullock j. s. , zentner a. r. , kravtsov a. v. , moustakas l. a. , 2008 , apj , 688 , 254 kennicutt r. c. , 1998 , apj , 498 , 541 kraljic k. , bournaud f. , martig m. , 2012 , apj , 757 , 60 loebman s. r. , rokar r. , debattista v. p. , ivezi . , quinn t. r. , wadsley j. , 2011 , apj , 737 , 8 liu c. , van de ven g. , 2012 , mnras , 425 , 2144 martig m. , bournaud f. , teyssier r. , dekel a. , 2009 , apj , 707 , 250 martig m. , bournaud f. , 2010 , apj , 714 , l275 martig m. , bournaud f. , croton d. j. , dekel a. , teyssier r. , 2012 , apj , 756 , 26 martig m. , minchev i. , flynn c. , 2014 , arxiv:1405.1727 mcmillan p. j. , 2011 , mnras , 414 , 2446 minchev i. , famaey b. , quillen a. c. , di matteo p. , combes f. , vlaji m. , erwin p. , bland - hawthorn j. , 2012a , a&a , 548 , a126 minchev i. , famaey b. , quillen a. c. , dehnen w. , martig m. , siebert a. , 2012b , a&a , 548 , 127 minchev i. , chiappini c. , martig m. , 2013 , a&a , 558 , a9 minchev , i. , et al . 2014 , apj , 781 , l20 moster b. p. , somerville r. s. , maulbetsch c. , van den bosch f. c. , macci a. v. , naab t. , oser l. , 2010a , apj , 710 , 903 moster b. p. , macci a. v. , somerville r. s. , johansson p. h. , naab t. , 2010b , mnras , 403 , 1009 nemec j. , nemec a. f. l. , 1991 , pasp , 103 , 95 nemec j. m. , nemec a. f. l. , 1993 , aj , 105 , 1455 norris j. , 1987 , apj , 314 , l39 peng c. y. , ho l. c. , impey c. d. , rix h .- w . , 2002 , aj , 124 , 266 prochaska j. x. , naumov s. o. , carney b. w. , mcwilliam a. , wolfe a. m. , 2000 , aj , 120 , 2513 qu y. , di matteo p. , lehnert m. d. , van driel w. , 2011 , a&a , 530 , a10 quinn p. j. , hernquist l. , fullagar d. p. , 1993
, apj , 403 , 74 read j. i. , lake g. , agertz o. , debattista v. p. , 2008 ,
mnras , 389 , 1041 reddy b. e. , lambert d. l. , allende prieto c. , 2006 , mnras , 367 , 1329 rix h .- w .
, bovy j. , 2013 , a&arv , 21 , 61 robin a. c. , haywood m. , creze m. , ojha d. k. , bienayme o. , 1996 , a&a , 305 , 125 robin , a. c. , reyl , c. , derrire , s. , & picaud , s. 2003 , a&a , 409 , 523 ryan s. g. , norris j. e. , 1993 , aspc , 49 , 103 sales l. v. , et al . , 2009 , mnras , 400 , l61 schnrich r. , binney j. , 2009a , mnras , 396 , 203 schnrich r. , binney j. , 2009b , mnras , 399 , 1145 seabroke g. m. , gilmore g. , 2007 , mnras , 380 , 1348 soubiran c. , bienaym o. , siebert a. , 2003 , a&a , 398 , 141 stewart k. r. , bullock j. s. , wechsler r. h. , maller a. h. , zentner a. r. , 2008 , apj , 683 , 597 stinson g. s. , et al . , 2013 , mnras , 436 , 625 teyssier r. , 2002 , a&a , 385 , 337 tsikoudi , v. 1979 , apj , 234 , 842 velazquez h. , white s. d. m. , 1999 , mnras , 304 , 254 villalobos . , helmi a. , 2008 , mnras , 391 , 1806 yin j. , hou j. l. , prantzos n. , boissier s. , chang r. x. , shen s. y. , zhang b. , 2009 , a&a , 505 , 497 yoachim p. , dalcanton j. j. , 2006 , aj , 131 , 226 | we study seven simulated disc galaxies , three with a quiescent merger history , and four with mergers in their last 9 gyr of evolution .
we compare their structure at @xmath0 by decomposing them into `` mono - age populations '' ( maps ) of stars within 500 myr age bins .
all studied galaxies undergo a phase of merging activity at high redshift , so that stars older than 9 gyr are found in a centrally concentrated component , while younger stars are mostly found in discs .
we find that most maps have simple exponential radial and vertical density profiles , with a scale - height that typically increases with age .
because a large range of merger histories can create populations with simple structures , this suggests that the simplicity of the structure of mono - abundance populations observed in the milky way by @xcite is not necessarily a direct indicator of a quiescent history for the milky way .
similarly , the anti - correlation between scale - length and scale - height does not necessarily imply a merger - free history .
however , mergers produce discontinuities between thin and thick disc components , and jumps in the age - velocity relation .
the absence of a structural discontinuity between thin and thick disc observed in the milky way would seem to be a good indicator that no merger with a mass ratio larger than 1:151:10 occurred in the last 9 gyr .
mergers at higher redshift might nevertheless be necessary to produce the thickest , hottest components of the milky way s disc .
galaxies : formation - galaxies : structure - galaxies : kinematics and dynamics - methods : numerical |
wall - crossing phenomena and , more generally , the counting of degeneracy of bps states have played a central role during last twenty years of development in superstring theory and supersymmetric gauge theories .
bps states , which preserves part of supersymmetry of the underlying field theory or gravity , are supposed to be robust due to the short supermultiplet structure they come with .
this often allows us to probe otherwise inaccessible non - perturbative aspects , such as u - dualty between different string theories or even the grand idea of m - theory . with smaller
supersymmetry preserved , however , this is not always reliable , and a theory sometimes loses or gains bps states suddenly under continuous deformation of parameters , hence the wall - crossing phenomena .
@xcite arguably , the most versatile framework where the intrinsic degeneracy of bps states can be counted is the quiver quantum mechanics .
this can be motivated rather simply by considering collections of d - branes that are completely wrapped on supersymmetric cycles in calabi - yau 3-fold . for compact calabi - yau s ,
such a collection corresponds to bps black holes while for noncompact ones , to bps states .
this line of approach was initiated by douglas et al . , @xcite and extensively developed by denef s pioneering work @xcite which inspired a lot of physics and mathematics study of the wall - crossing .
@xcite nevertheless , direct and systematic counting of bps states starting with quiver quantum mechanics became available only very recently .
problem of counting wall - crossing discontinuity is often simpler than counting degeneracies themselves .
the so - called coulomb approximation , which is inspired by the multi - center picture of wall - crossing came to fruition in refs.@xcite , which is however later seen to miss a lot of physical states , when the quiver in question involves oriented loops of bifundamental chiral multiplets .
honest and comprehensive counting of witten index was finally formulated in ref .
@xcite not only for quiver theories but also for general gauged linear sigma models ( glsm ) , whose main result is phrased in terms of residue formulae in the space of gauge field expectation values .
section 3 below will summarize the result . despite such developments , one aspects of the bps
state counting remains much obscure .
namely , how the degeneracy scales as a function of the increasing charge .
this question itself is of course very familiar in the context of bps black holes , with the obvious answer being `` exponentially large , '' yet most of definite answers on such a question are found in systems where wall - crossing is either absent or rather simple .
black holes that preserve four supercharges only and are bps states in the supergravity theory that preserves eight supercharges , microstate counting in general is still unavailable . isolating the correct exponential behaviour from microstate counting based on wrapped d - branes would be most useful , but this looks rather difficult technically despite the above residue formula as the number of residue integral to do grows linearly with charge , and the number of poles to consider itself grows exponentially fast . on the other hand
, the exponential behaviour of the degeneracy turns out to be present also in certain field theory bps states , albeit to a lesser degree .
this was initially anticipated by kol @xcite on the basis of string - web picture of 1/4 bps dyons in @xmath4 field theory in four dimensions . while the former s argument is somewhat anecdotal , a more definite counting using low energy dynamics of solitons has shown how a highly charges states in field theory can be equipped with an exponentially large degeneracy .
this exponential behaviour is different from that was anticipated with black holes in the following sense . for bps black holes in four dimensional @xmath5 supergravity ,
the scaling is such that @xmath6 for a large charge @xmath7 , while for field theory bps states the anticipated scaling goes like @xmath8 for either , one must deal with quiver quantum mechanics of large rank @xmath9 , and the direct computation become exponentially difficult even with the general formulation of witten index computation , unless one finds a mitigating circumstance .
this note is an attempt to reproduce such an exponential behaviour in the simplest possible nontrivial quiver , namely the kronecker quiver , in the hope of finding more general systematics of large rank quiver dynamics . after a cursory introduction to quiver quantum mechanics
, we go on in section 3 to describe how the witten index of the kronecker quiver of rank @xmath10 and @xmath11 can be computed via diagrammatical representation of the residue formula , following the method in ref .
a useful middle step in this computation is a mutation which take @xmath12 kronecker quiver to @xmath13 kronecker quiver where @xmath1 is the common intersection number .
we briefly discuss how this duality for these quivers could be realized in this context .
especially we obtained @xmath14g_k(x ) = \frac{k}{d((k-1)d+1 ) } { ( k-1)^2d + ( k-1)\choose d-1}\ , \ ] ] which implies the scaling bahaviour @xmath15 where @xmath16 .
the result is consistent with the euler number of quiver moduli spaces which has been widely discussed in the mathematical literature @xcite and in the context of spectral network @xcite .
we close with some further speculations in section 5 . as this work was completed , a paper @xcite with some overlap has appeared in the arxiv .
the simplest setting where quiver quantum mechanics emerges is type iib theory compactified on calabi - yau 3-fold . the effective theory in the remaining four dimensions carries @xmath5 supersymmetry , and the bps state thereof are realized as d3-branes wrapped on special lagrange subcycles of the calabi - yau .
when the cycle is rigid , as with @xmath17 , the vector multiplet on the d3-brane reduces to quantum mechanical vector multiplet whose content is the same as @xmath18 vector multiplet in four dimensions .
if d3 wraps the same cycle @xmath10 times , the gauge theory is elevated to @xmath19 . for each such d3-branes on @xmath17 , we can associate a fayet - iliopoulos constant @xmath20 such that the low energy effective action carries a term @xmath21 where @xmath22 is auxiliary field in the vector multiplet .
when a pair of such wrapped d3 s , say each wrapping @xmath23 and @xmath24 times , meets at @xmath1 intersections , one find additional chiral multiplets in the bifundamental representation @xmath25 .
such a quiver , which is the simplest possible nontrivial class , is called kronecker quiver .
we have the supersymmetry constraint @xmath26 so we really have only one fi constant , say , @xmath27 . for negative @xmath28
the classical vacuum moduli space is null and so is the witten index . for positive @xmath28 , the moduli space is given as a kaehler quotient , @xmath29 the number of stable bps bound states can be obtained by calculating indices of these quiver quantum mechanics .
furthermore , the theory has @xmath30 r - symmetry , where @xmath31 came from the rotational symmetry of the three spatial direction .
we can refine the indices by turning on the fugacity @xmath32 for @xmath33 of r - symmetry . due to the pioneering work of reineke @xcite and manschot et al .
@xcite , the systematic procedure of evaluating the equivariant indices of mutually co - prime quiver @xmath34 are viable . however , since they are given in terms of very particular sum of partitions of charges , it is hard to examine the large @xmath10 behaviour with those formula .
indices for @xmath35 kronecker quivers are well - documented and also easy to compute directly . note that the complex dimension of the classical higgs moduli space when @xmath36 is @xmath37 this counting is misleading for nonprimitive cases such as @xmath38 , since the noncompact classical moduli spaces open up along the coulomb direction .
direct evaluation of the standard formula leads to fractional quantities , but with more care the true integral index can be computed . with @xmath39 ,
the only primitive case with nonnegative dimension , thus nonempty moduli space , is @xmath40 .
this case generates the most basic wall - crossing pattern , corresponding to the so - called pentagon - identity , and the index is @xmath41 with @xmath42 , there are two classes with nonempty moduli spaces .
the first is @xmath40 with @xmath43 and the second is @xmath44 with @xmath45 .
indices for these can be inferred to the well - known spectrum of @xmath46 seiberg - witten theory , and also have been computed directly by several different methods .
the answers are of course , @xmath47 and @xmath48 seiberg - witten wall - crossing formula implies that there are no other nontrivial indices in this class , meaning also that all nonprimitive kronecker quivers have null ( integral ) index for @xmath35 .
one most effective way to obtain eq .
( [ hyper ] ) is to employ the mutation map , which is beautifully motivated and described in physics and mathematics of wall - crossing .
@xcite . for general quivers with the nodes labeled with charges @xmath49 and intersection number @xmath50 the left - mutation maps @xmath51_+ \gamma_k & & \hbox{otherwise } \end{array}\right.\ ] ] where @xmath52_+$ ] is @xmath53 for positive @xmath53 and zero otherwise ,
while the right mutation maps @xmath54_+ \gamma_k & & \hbox{otherwise } \end{array}\right.\ ] ] the basic assertion is that if we change @xmath55 as well to keep the total @xmath56 fixed , the index remains unchanged .
the mutation is not allowed for any nodes , however . for wall - crossing quivers like this , there are the questions concerning which node can be mutated and which of the two mutations should be taken , and the answers to these questions are rather complicated . because kronecker quivers have only two nodes and because there is only one nonempty chamber , the allowed mutation maps are uniquely fixed in that chamber as @xmath57 under the left and right mutation respectively . with @xmath42 kronecker quivers , for instance , repeated usage of this maps @xmath58 and @xmath59 quivers to @xmath60 and @xmath61 , bringing us somewhat trivially back to eq .
( [ hyper ] ) .
in this note we are mostly concerned with @xmath62 or its mutation equivalent @xmath63 , for @xmath3 .
it turns out that @xmath64 is a sort of watershed in that the asymptotic scaling with @xmath3 is qualitatively different than those with @xmath65 .
this may imply a similar scaling behaviour for @xmath66 , which indicates that @xmath67 quivers have nontrivial bound states for @xmath3 , unlike their counterpart @xmath35 , despite the flat coulombic directions opening up .
the equivariant witten index of interest is @xmath68\ , \ ] ] where we fixed the usual sign ambiguity of the index by choosing @xmath69 .
when we do this we should take care to remove the center of mass part of the low energy dynamics , which is to say , to remove one overall @xmath70 decoupled from the rest of the dynamics .
the residue formula above gets contribution from partial choice of set of rank @xmath79 poles which are determined by the definition of the jk - residue . at a singularity where exactly @xmath80 hyperplanes ( each of which is defined by a charge @xmath81 ) meet ,
the definition can be written as @xmath82 where we allowed constant shift of the pole location for notational convenience .
this procedure encounter some technical difficulties when , at a contributing pole , more than @xmath80 such hyperplanes meet resulting in the so - called degenerate cases .
constructive procedures are known to deal with such cases , one of which we will encounter in the appendix .
we concentrate on mutually co - prime @xmath23 and @xmath24 , where our choice of @xmath73 satisfies the regularity condition where the definition of jk - residue for non - degenerate point can be safely applied .
@xcite in this note , we introduced all allowed chemical potentials , @xmath53 s , to reduce degeneracies as much as possible .
due to a powerful theorem @xcite , dependence on these chemical potential @xmath53 washes out whenever the classical moduli space is compact , which is the case for kronecker quivers with mutually co - prime @xmath83 .
degenerate singularities that can not be resolved this way will be separately addressed in the appendix .
first of all , we examine what this formula implies for the simplest case , @xmath35 . for @xmath39 , since the classical moduli space is empty , we expect @xmath84 except for @xmath85 .
the latter can be easily seen from a simple abelian residue integral . for @xmath86
, we can show that there is no charge set which contributes to the jk - residue .
first of all , in the appendix , we showed that the charge set which involves vector multiplet can contribute only when @xmath87 .
only , we conjecture that for all values of @xmath1 , when @xmath88 are mutually co - prime , singularities involving vector multiplet does not contribute to the integral . ] hence , under the assumption that @xmath88 are mutually co - prime , the residue integral gets contribution only from poles of chiral multiplets .
suppose we have a set of chiral multiplet charges such that @xmath73 is in a positive cone of the charges .
then all the charges in this set should be connected to each other to meet this assumption , and this implies that they collide at @xmath89 for all @xmath90 s and @xmath1 s .
this immediately produces @xmath91 additional zeros from the vector multiplet numerator where additional degeneracy of the pole is @xmath92 . since @xmath93 when @xmath94
, the number of zeros always exceed that of additional poles .
hence , chiral multiplets can not contribute either .
since there is no singularity which has non - vanishing residue and meets the jk condition , we can say that @xmath95 for all co - prime @xmath23 and @xmath24 as expected .
next , let us consider @xmath34 quivers with @xmath42 , whose indices count the degeneracy of bps spectra in well - known @xmath46 seiberg - witten theory . first of all , the index of @xmath96 quiver can be easily evaluated by adding two residue integrals of chiral multiplet poles , @xmath97 which correctly reproduces spin character of the four - dimensional bps vector multiplet . when @xmath98 , as can be inferred from the argument in the appendix , residue integral gets contribution from the vector multiplet , which again implies the existence of the flat direction in coulomb branch .
for co - prime @xmath23 and @xmath24 , when the index is expected to get contribution only from chiral multiplets , one can show that there are only two non - trivial cases , which are @xmath99 and @xmath100 .
suppose that we turn off all the flavor fugacities .
since all charges which contribute to the residue integral should be connected , the rank @xmath80 singularity is located at @xmath89 for all @xmath90 and @xmath1 . at this point , the order of zeros minus the order of additional poles are given by @xmath101 = ( d_1-d_2)^2 - 1 $ ] .
note that except @xmath102 and @xmath99 , the residue integral will vanish . for @xmath59 ( and equivalently
@xmath58 ) quiver , we can further show that the only non - trivial residue integral comes from a charge set where each component @xmath75 of the first node appear exactly twice .
the contributing charge set is drawn in figure [ fig:45 ] below . quiver .
this is only possible non - zero contribution , up to weyl permutation .
the number denotes assignments of flavor fugacity.,scaledwidth=40.0% ] if there exist a component @xmath75 with more than three charges are connected , one can see that the residue always vanishes or does not satisfy the jk - condition .
the residue integral at this fixed point can be readily evaluated .
if we set @xmath103 , it follows that @xmath104 are fixed locus of the above type .
then , from the vector multiplets , we have a factor @xmath105}{\sinh[((n_1-n_2)(e_1-e_2)+z)/2 ] } \prod_{\substack{m_1,m_2=1\\m_1\neq m_2}}^{d}\frac{\sinh[(m_1-m_2)(e_1-e_2)/2]}{\sinh[((m_1-m_2)(e_1-e_2)+z)/2]}\ , \ ] ] and from the chiral multiplet , we have @xmath106}{\sinh[(n - m)(e_1-e_2)/2 ] } \prod_{\substack{n , m=1\\n\neq m-1 } } ^{d-1,d}\frac{\sinh[((n - m+1)(e_1-e_2)-z)/2]}{\sinh[(n - m+1)(e_1-e_2)/2]}\ .\ ] ] note that if we multiply two contributions , all the @xmath107 factors cancel out each other .
finally we should multiply by @xmath108 which takes into account the contribution from the weyl equivalent fixed points .
we do not sum over @xmath109 , since it can be obtained from left - right flipping of the diagram , which is a part of the weyl group .
hence , we correctly reproduce @xmath110 as expected . even though general recipe for calculating equivariant indices of the kronecker quiver is well - established , actual evaluation of the residue integral is very complicated when degenerate singularities appear .
however , almost all quivers with @xmath111 turns out to contain such singularity , as can be seen from the naive pole counting after turning off the flavor fugacities .
if degenerate singularities exist , the answer depends crucially on the order of taking each residue , which makes the evaluation for the large rank case very involved .
see appendix or @xcite for how we should deal with of such singularities
. however , there are special types of two node quiver where we can explicitly show that no degenerate singularity appears even for @xmath111 .
these are the quivers of type @xmath112 .
interestingly , these quivers are known to be mutation dual to quivers of type @xmath0 , where large @xmath10 behaviour of the index has interesting physical consequences @xcite . especially , when the jk - residue gets contribution only from the non - degenerate poles , the definition of the residue integral reduces to that of ( [ jk ] ) . now
, suppose that we found a set of @xmath113 charges such that @xmath73 is in a positive cone of the charge set .
since they are all simple poles , evaluation of the residue corresponds to simple substitution of these relations defined by @xmath114 and @xmath115 , where @xmath116 s and @xmath117 s are certain linear combinations of the flavor fugacities . then the contribution of this singularity to the residue integral is in the following form : @xmath118}{\sinh[(f_p(a_i)-f_q(a_i)-z)/2 ] } \prod_{k\neq l } \frac{\sinh[(g_k(a_i)-g_l(a_i))/2]}{\sinh[(g_k(a_i)-g_l(a_i)-z)/2]}\\ & & \times \prod_{f_p\neq g_k } \frac{\sinh[(f_p(a_i)-g_k(a_i)-z)/2]}{\sinh[(f_p(a_i)-g_k(a_i))/2]}\ .\end{aligned}\ ] ] note that the pre - fector @xmath119 are cancelled by rank @xmath113 residue integral .
furthermore we multiplied @xmath120 and @xmath121 , taking into account the weyl permutation of the solutions which give rise to the same contribution .
we expect that after we add up contributions from all singularities , flavor fugacity cancels out @xcite and end up with a laurent polynomial in @xmath122 .
interestingly , the limit @xmath123 is well - defined for each of these term and the residue becomes 1 at all such fixed points .
hence , if we concentrate on the value of the witten index only , the problem reduces down to counting number of such set of charges which is in a positive cone of corresponding @xmath73 .
the following section is devoted to the evaluation of the witten index for @xmath2 quiver , by counting all such contributing poles .
from now on , we denote set of _ nodes _ by @xmath124 and @xmath125 which represent each cartans of source and sink of the kronecker quiver respectively . we will call _ contributing arrows _ as a set of rank @xmath113 chiral multiplet charges @xmath126 with specified flavor fugacities which passes jk condition and contributes non - zero value to the witten index . in this section ,
we focus on the explicit evaluation of the witten index for quivers of type @xmath2 , using the jk - residue formula derived from the path integral . in the mathematical literature ,
the euler number of moduli space of these types of quiver was extensively studied in @xcite .
our formula turns out to agree with these results .
the evaluation is done in the following procedure . for quivers of type @xmath2
, one can show following facts .
* consider a set of contributing arrows . for each @xmath90 , which labels cartan of the source
, there are exactly @xmath1 set of arrows of type @xmath126 .
* there is no degenerate contributing arrows .
* suppose that there exist two charges @xmath126 and @xmath127 with @xmath128 , for a source node @xmath90 of given contributing arrows .
then these solution does not contribute to the integral .
similarly , for a sink node @xmath129 , if there exist two charges @xmath130 and @xmath131 with @xmath128 , this solution does note contribute to the integral .
these three statements are proven in the appendix .
once we have these , the procedure of finding all contributing arrows reduces down to that of @xcite .
we can construct all sets of contributing arrows recursively , by _ gluing _ contributing arrows of the quiver with type @xmath132 , which is defined as follow . * _ gluing _ : consider two sets of contributing arrows of type @xmath133 and @xmath134 respectively .
let us denote the @xmath135 as the set of source and sink nodes of the first quiver , and @xmath136 as that of the latter quiver . by _ gluing
_ we mean that identifying two sink node @xmath125 and @xmath137 with @xmath138 , in a way that it satisfies the condition 3 .
finally we also prove the following in the appendix .
* all such configuration obtained from the gluing of @xmath132 type quiver satisfies the jk condition . then , since we have shown that each non - degenerate contributing arrows contributes 1 to the witten index , we can obtain index of the quiver just by counting number of possible ways of gluing @xmath10 number of quivers of type @xmath132 .
we briefly review the procedure below .
let us first illustrate the procedure with a simple example with @xmath64 .
suppose that we initially have @xmath139 type quiver with a flavor fugacity specified .
then , by the condition 3 , we can attach at most @xmath140 subquiver to this . by gluing @xmath10 copies of @xmath141 subquiver to this
, we obtain a quiver with @xmath142 sink nodes . hence
, if we denote @xmath143 as the number of possible configuration of such quiver with @xmath144 sink nodes , @xmath145 can be obtained from the expression of @xmath146 with @xmath147 . in the figure [ fig : mmm ] , this procedure for @xmath148 is drawn . note that it is crucial to divide by the order of symmtery group of the graph , since it yields weyl equivalent combinations , which we already have taken into account in the calculation of the witten index of each singularity .
one can easily find the expression for @xmath149 in terms of @xmath146 s with @xmath150 , @xmath151 subquivers to @xmath139 type initial quiver .
the numbers on the arrows denote the choice of the flavor fugacities of corresponding arrows .
note that it is crucial to divide by the order of symmetry group of the graph , since it yields weyl equivalent combinations , which we already have taken into account in the calculation of the witten index.,scaledwidth=90.0% ] this procedure can be straightforwardly generalized to arbitrary @xmath1 and @xmath10 . if we denote @xmath152 by the number of contributing configuration with @xmath153 sink nodes , it can be obtained via following relation .
@xmath154 where the @xmath90-th term counts the number of ways of attaching @xmath90-many subquivers with @xmath155 sink nodes in total , to the initial quiver of type @xmath139 .
each subquiver can be further decomposed into an arrow ( which is attached to the initial quiver ) and @xmath156 sub - subquivers .
when each of the latter contains @xmath157-many sink nodes respectively , it contributes @xmath158 factor to the @xmath90-th term .
finally , the combinatorial factor @xmath159 for @xmath90-th term takes into account the number of ways of choosing flavor fugacity of arrows attached to the initial quiver , divided by order of the weyl permutation of those .
furthermore , this recursion relation can be neatly summarized by introducing the generating function @xmath160 with an auxiliary variable @xmath161 . then the recursion relation ( [ 42 ] )
can be written in terms of the algebraic equation which ( [ 43 ] ) satisfies , @xmath162 finally , the actual generating function of the quivers in question , i.e. , the one obtained by gluing @xmath10 subquivers of type @xmath132 can be readily obtained in terms of @xmath163 as @xmath164 which takes into account the fact that the initial quiver has been replaced by @xmath132 quiver , yielding @xmath1 choices of initial gluing .
then @xmath165-th coefficient of @xmath166 encode the number of contributing arrows of quiver type @xmath2 .
the coefficients of generating function which satisfies such algebraic equation can be evaluated by the lagrange inversion theorem.@xcite especially , when a generating function satisfies an algebraic equation @xmath167 for a function @xmath168 , we have @xcite @xmath169(f(x)))^k = \frac{k}{n } { n \choose \frac{n - k}{b } } a^{(n - k)/b}\ , \ ] ] where we denoted @xmath170g(x)$ ] as @xmath155-th coefficient of a power series @xmath171 . using this theorem
, we can extract the general expression for the witten index , @xmath14g_k(x ) = \frac{k}{d((k-1)d+1 ) } { ( k-1)^2d + ( k-1)\choose d-1}\ .\ ] ] note that we further divided the answer by @xmath10 which corresponds to the choice of the initial quiver .
the asymptotic behaviour of @xmath172 for this expression can be also calculated , @xcite which we can write as @xmath173 recently , it was noted by galakhov et al .
@xcite that this exponentially large degeneracy of four - dimensional bps states of conformal field theory is something unexpected , since the naive dimensional analysis implies a bound for the index , @xmath174 , where @xmath175 is energy of state supported in a finite volume @xmath176 .
they showed that , by carefully examining denef s multi - center bound state formula @xcite , the radius of bps bound state increases with the mass of the state .
this corrects the above bound by @xmath177 , which is consistent with the observed scaling behaviour of quiver quantum mechanics .
one of the most interesting properties of the 1d quiver quantum mechanics is that they are expected to be invariant under certain duality , which is called the _ mutation _ equivalence . in terms of the kronecker quivers in question
, it can be phrased into the isomorphism between moduli space of @xmath71 quiver with that of @xmath178 , as shown in the figure [ fig : mu2 ] . as a simple check
, one can see the dimension of the classical moduli space of these two quivers are the same .
furthermore , we expect that , the equivariant indices of both quivers are the same , i.e. , @xmath179 which is calculated for region @xmath180 and @xmath181 respectively .
the index is trivially same for the other side of the wall , since both of them vanish . in particular
, we will examine how the relation @xmath182 can be realized , using the diagrammatic expression of the jk residue developed so far . before delving into general @xmath10 and @xmath1
, we illustrate the proof of ( @xmath183 ) for one of the simplest pair of the kronecker quivers , @xmath184 and @xmath185 in detail , where the contributing charge sets are explicitly written down .
in particular , we show that each fixed point determined by a set of hyperplane charges can be mapped to each other in a particular way .
let us denote the cartans of @xmath186 quiver by @xmath187 .
then the equivariant index of this quiver can be written in the following expression @xmath188}\right)^4 \left(\prod_{p\neq q}^2\frac{\sinh[(x_p - x_q)/2]}{\sinh[(x_p - x_q - z)/2]}\right ) \\\label{bbb } & & \times \left(\prod_{k\neq l}^3\frac{\sinh[(y_k - y_l)/2]}{\sinh[(y_k - y_l - z)/2]}\right ) \prod_{p , k , i}\left(\frac{\sinh[(x_p - y_k - a_{i}-z)/2]}{\sinh[(x_p - y_k - a_{i})/2]}\right)\ , \qquad\end{aligned}\ ] ] where @xmath77 s are flavor fugacities for number of arrows .
since they potentially carry degenerate singularities , we need constructive definition of the jk - residue which can determine the ordered set of contributing charges .
( for the definition used here , see @xcite or section 2.4.3 of @xcite ) . for this purpose , @xmath73 is chosen to be proportional to @xmath189 with positive @xmath190
. should not be at the boundary of flag defined by the sum of charges . here
, @xmath73 should be slightly shifted from @xmath191 since @xmath20 can be spanned by @xmath192 , sum of all charge sets . ]
if we carefully examine the jk - condition with this choice of @xmath73 , it turns out that the following ordered set of charges passes the jk condition , and potentially non - zero . @xmath193 with their weyl copies @xmath194 and all different different set of flavor fugacities . when @xmath195 , @xmath196 and @xmath197 , this singularity is non - degenerate . on the other hand , if @xmath198 , this singularity becomes degenerate which has two additional poles @xmath199 , @xmath200 and order two zero from @xmath201 .
s are fixed by choice of the sign of @xmath190 . ]
other than these , the residue integral vanish due to the zero s of vector multiplet .
these two different classes of singularities are illustrated in the second and the fourth diagram of figure [ hi ] respectively .
meanwhile , the equivariant index of @xmath202 quiver with three arrows can be obtained from gluing three @xmath141 quivers as was shown in the last section .
it turns out that there are two topologically distinguished contributing arrows which can be obtained .
these are also illustrated in the first and the third diagram of figure [ hi ] .
the non - equivariant index can be evaluated simply by counting number of ways of assigning flavor fugacities to each arrows up to symmetries of diagram .
this turns out to be @xmath203 and @xmath204 respectively .
appear as a subdiagram of contributing arrows of @xmath205 quiver , which has the same index.,scaledwidth=90.0% ] interestingly , one can show that these two types of diagrams are mapped respectively to the two different types of fixed points of the @xmath186 quiver .
for this purpose , we rewrite a term in the integral in ( [ bbb ] ) evaluated at ( [ fixed ] ) with @xmath206 as follow , @xmath207}\right)^4 \oint_{y_1=y_2+a_{3}-a_{1 } } \oint_{y_3=y_2+a_{2}-a_{1 } } \left(\prod_{k\neq l}^3\frac{\sinh[(y_k - y_l)/2]}{\sinh[(y_k - y_l - z)/2]}\right)\\ & & \oint_{x_2=y_2+a_{2}}\oint_{x_1=y_1+a_{1 } } \left(\prod_{p\neq q}^2\frac{\sinh[(x_p - x_q)/2]}{\sinh[(x_p - x_q - z)/2]}\right ) \prod_{p , k , i}\left(\frac{\sinh[(x_p - y_k - a_{i}-z)/2]}{\sinh[(x_p - y_k - a_{i})/2]}\right ) . ~~\qquad\end{aligned}\ ] ] note that , the integral over all @xmath75 s in the last line can be evaluated to the simple expression as @xmath208)^2\prod_{\substack{y_i , a_i\in a \\
y_j , a_j \in a'}}\frac{\sinh[(y_i - y_j + a_i - a_j - z)/2]}{\sinh[(y_i - y_j + a_i - a_j)/2]}\ .\ ] ] here we defined that @xmath209 is a subset of all combinations @xmath210 which is integrated over in the above integral ( this case @xmath211 and @xmath212 ) , and @xmath213 is the complement of @xmath209 in which case we have 7 elements .
the factor @xmath214 takes into account the weyl equivariant singular point . on the other hand ,
if we look at @xmath202 quiver side , we have @xmath215}\right)^9 \left(\prod_{p\neq q}^7\frac{\sinh[(\tilde x_p-\tilde x_q)/2]}{\sinh[(\tilde x_p-\tilde x_q - z)/2]}\right)\\\label{mmq } & & \times\left(\prod_{k\neq l}^3\frac{\sinh[(y_k - y_l)/2]}{\sinh[(y_k - y_l - z)/2]}\right ) \prod_{p , k , i}\left(\frac{\sinh[(-\tilde x_p+y_k+a_{i}-z)/2]}{\sinh[(-\tilde x_p+y_k+a_{i})/2]}\right)\ .\qquad\end{aligned}\ ] ] since all the poles are all non - degenerate , we can safely change the order of integration without affecting the result .
suppose that we have contributing arrows defined by the following assignments of flavor fugacities , and fix the integration order as specified below .
@xmath216 if we evaluate the residue integral for the first seven poles , it yields a factor @xmath217)^7\prod_{\substack{y_i , a_i\in a \\ y_j , a_j \in a'}}\frac{\sinh[(y_i - y_j + a_i - a_j - z)/2]}{\sinh[(y_i - y_j + a_i - a_j)/2]}\ , \ ] ] where the sets @xmath209 and @xmath213 are exactly the same as what is defined in the expression ( [ qqq ] ) .
again , we have a factor of @xmath218 which takes into account different fixed points obtained by the weyl permutation of @xmath219 . in this way , for a given set of charges of @xmath186 quiver which satisfies the jk condition , we get unique rank 9 singularity for @xmath202 quiver after summing over the weyl permutations for both sides , and vice versa .
note that for charge sets mapped like this , we are left with exactly the same integral over non - mutated nodes .
this procedure is summarised in figure [ fig : mu3 ] .
summing over all contributions , we end up with @xmath220 which agrees with the formula obtained in various literatures .
@xcite and @xmath186 quiver with @xmath64,scaledwidth=70.0% ] if we look at figure [ hi ] where we illustrated the mapping of each contributing arrows , it is interesting to note that the contributing arrows of @xmath184 quivers appear as a subdiagram of contributing arrows of its mutation dual quiver .
once we pick all the arrows directly attached to the glued nodes , it gives a unique diagram which corresponds to the contributing arrows of the dual quiver which has the same residue integral .
as can be seen in this example , once we classify all contributing arrows of both side of the dual pair , this procedure may be generalised to arbitrary @xmath10 and @xmath1 .
suppose that , there exist an ordered charge set @xmath221 which satisfies the jk - condition for the @xmath2 quiver .
since this is non - degenerate , we can choose an order of the charge set in a way that we could integrate over all @xmath222 s first , as was shown in the example .
given this set , we uniquely pick a subquiver which includes all the arrows directly connected to the glued node . out of this subquiver
, we associate a unique bipartite quiver as in figure 4 , with flavor fugacities assigned .
one can easily check that bipartite diagrams are the only possibility which satisfies the jk - condition with @xmath223 charges for @xmath0 quiver . then
, for non - degenerate poles of the latter , it is a direct generalization of the above example that the residue of this pole is the same as that of a pole of the original quiver . for degenerate poles of @xmath0 quivers
, it needs to be separatley confirmed that the poles which can be obtained this way saturate all set of degenerate poles of the dual quiver , after shifting fi parameter properly as in the above example .
to conclude , we would like to mention several interesting questions which follows immediately . * from the residue integral expression , we can systematically obtain the index for the kronecker quiver with non - coprime dimension vector .
it would be interesting to study the scaling behaviour for non - coprime charges @xmath224 directly . as was partly shown in the appendix ,
this type of quivers always involve particular contributing charge sets which contain poles in the vector multiplet .
this implies that the index gets contribution from the flat direction in the coulomb branch , which can be inferred from the fact that we generally get the fractional number for the poles of such type .
* one of the most interesting problem remaining is to find a large - rank scaling behaviour of quivers with oriented loop , which is closely related to counting the degeneracy of the single centered bps black hole in four - dimensional @xmath5 theory
. it would be interesting to see if we can find any quiver with loop whose large - rank behaviour is accessible through the residue integral formula . *
the index formula described in this note is given by particular sum over various residue integrals , and the formula described by mps @xcite and reineke @xcite are also written in terms of sum of various partitions .
it would be interesting to see how our construction is related to the latter two expansion .
one of the pros of using the residue integral definition is that it is rather easy to see the large rank behaviour and the mutation equivalence of quivers , as shown in the example of the last section .
it remains to check if mutation equivalence holds for other types of the tree - like quiver and quiver with loops .
the author is grateful to piljin yi for fruitful comments and reviewing the manuscript , and thanks to seung - joo lee for useful discussion .
the research was supported by the perimeter institute for theoretical physics .
research at perimeter institute is supported by the government of canada through industry canada and by the province of ontario through the ministry of economic development and innovation .
this research was made possible through the support of a grant from john templeton foundation .
the opinions expressed in this publication are those of the author and do not necessarily reflect the views of the john templeton foundation .
* _ for quivers of type @xmath225 , vector multiplet does not contribute when @xmath86 . _ + _ proof .
_ suppose that there exist a co - dimension @xmath80 singularity which contains poles from vector multiplet @xmath226 .
then since @xmath73 is taken to be positive for the first node , it should be followed by a pole of type @xmath227 or @xmath228 .
since a charge set without any chiral multiplet can not generate rank @xmath80 singularity , we need at least one pole from the chiral multiplet . thus in general
, poles that contain vector multipets have a sequence @xmath229 as a part .
however the last two relations generate the equation @xmath230 which is zero of the hypermultiplet , and the only possibility that a pole involving vector multiplet contributes is that additional poles collide and become degenerate .
let us scan all the charges possibly connected to this set . for @xmath231 ,
* * @xmath232 : generates additional zero @xmath233 . *
* @xmath234 : gives additional pole @xmath235 . *
* @xmath236 : gives order two zero from @xmath237 . * * @xmath238 : generates order two zero or go back to the same type . *
* @xmath239 : yields order two zero from @xmath240 . *
* @xmath241 : no additional poles or zero . * * @xmath242 : additional zero from @xmath243 . + from these possibilities , we see that the sequences ( [ se ] ) can contribute only when it also contains a pole of type @xmath234 .
however when @xmath244 , introducing this pole yields another relation @xmath245 which generates a zero . in order to cancel this zero again , we further introduce @xmath246 in a similar manner .
this procedure continues to introduce @xmath1 additional @xmath247 s , and we end up with the relations @xmath248 which involves exactly @xmath249 pairs of @xmath161 s and @xmath250 s . if we try to add one more cartan @xmath251 which is related to the above set by adding @xmath252 or @xmath253 , it always generates additional zeros of type @xmath254 or @xmath255 .
finally , if we add @xmath256 , this goes back to the argument of the beginning , again introduces new @xmath257 which falls into type ( [ aaa ] ) .
hence the set can not be connected to other cartans by adding additional poles . however ,
when @xmath86 , one can see that the singularity involving this set can not meet the jk condition .
this is simply because , sum of all components of any linear combination @xmath258 vanishes where @xmath259 s are charge sets appearing in ( [ aaa ] ) .
the analysis implies that the vector multiplet can not contributes for co - prime @xmath23 and @xmath24 , but they can contribute if @xmath98 , even for larger intersection number @xmath1 . * _ consider a quiver with type @xmath112 and their particular set of @xmath260 contributing arrows . in this set , for each @xmath124 , there are exactly @xmath1 arrows of type @xmath126 . _ + _ proof .
_ note that the rank of this quiver is @xmath261 . in order to satisfy the jk condition ,
there must exist a set of @xmath261 positive coefficients @xmath262 , such that @xmath263 holds . here
@xmath264 denotes a charge vector which gives a pole of type @xmath126 .
now suppose that there exist @xmath124 such that there are @xmath265 ( for @xmath266 ) charges of a form @xmath267 .
then there is at least one @xmath268 such that @xmath269 , the coefficient for charge of type @xmath126 , satisfies @xmath270 then it is clear that existence of the coefficient of this property violates @xmath268-th component of the equation ( [ a1 ] ) .
hence , for each @xmath90 , there should be at least @xmath1 vectors of type @xmath267 . furthermore , since the number of charges are exactly @xmath261 , we can say that there should be exactly @xmath1 charges of type @xmath267 for each @xmath90 . * \1 ) _ for quivers with dimension vector @xmath271 , degenerate poles do not contribute to the index . _
+ \2 ) _ suppose that there exist two charges @xmath126 and @xmath127 with @xmath128 , for any @xmath90 in a contributing arrow set .
then this set does not contribute to the integral .
similarly , for a sink node @xmath129 , if there exist two charges @xmath130 and @xmath131 with @xmath128 , this set does note contribute to the integral . _
+ _ proof . _ for non - degenerate singularities , poles of type 2 ) do not contribute due to the zeros of the vector multiplets .
however , these singularities may contribute for degenerate singularities because of additional pole orders .
note that , since the rank of the gauge group is @xmath261 and there are only @xmath1 flavors , degenerate singularity can appear only when one of the 2 ) is satisfied .
+ let us briefly summarise how we can deal with such singularities in general . consider a gauge group with rank @xmath80 .
then the jk residue gets contribution from codimension @xmath80-singularities defined by @xmath272 for @xmath273 with @xmath274 .
when @xmath275 , definition of the jk residue integral can be unambiguously written down as ( [ jk ] ) .
on the other hand , for @xmath276 , which we call _ degenerate singularity _ , the final answer depends very much on the order of taking residues , and we need an alternative definition of jk residue which clarifies this ambiguity . in the literature , there exist several equivalent definitions which could be applied for these type of singularities . @xcite among them , we are going to use the definition elaborated in @xcite , which we briefly summarize as below .
+ suppose that we have rank @xmath80 degenerate singularity with charges @xmath277 with @xmath278 colliding .
from these , we fix the order of charges _ arbitrary _ , and denote this ordered set as @xmath279 . among them , there are @xmath280 choices of picking up an ordered set with @xmath80-entry , which we denote as @xmath281 . finally , for each @xmath282 , we assign so called _ basic fraction _ , @xmath283 out of these basic fraction @xmath284 s , we can build a set @xmath285 such that @xmath286 form a rank @xmath80 basis of the basic fractions . then , from @xmath286 , one can obtain any meromorphic function in @xmath287-plane by taking derivatives and linear combinations of elements in @xmath288 .
+ furthermore , the useful fact is that , for _ any _ choice of ordering of set @xmath289 , we can find a set @xmath286 such that we can define iterated residue , @xmath290 with @xmath281 , which satisfy @xmath291 in particular , rhs of the definition ( [ asd ] ) should be regarded as iterated operation of the residue integral from the rightmost one , keeping other components of @xmath287 at generic points . with all these definitions , jk - residue for this degenerate point
can be simply written as @xmath292 where @xmath293 sgn(det @xmath282 ) . with this alternative definition of the jk - residue
, we are going to show that for quiver with type @xmath2 , there is no degenerate point which contributes to the integral .
+ as was mentioned above , for the kronecker quivers of this type , degenerate singularities from chiral multiplets can appear only when one of the following is satisfied .
* * contributing arrows contain two charges @xmath126 and @xmath294 such that @xmath295 . * * contributing arrows contain two charges @xmath126 and @xmath296 such that @xmath295 .
+ first , let us consider the first case .
suppose that there exist such @xmath124 and @xmath297 .
if both of @xmath298 are not connected to any other source nodes , then this pole can be non - degenerate , but they do not contribute since they generate the relation @xmath299 , which introduces order 2 zero from the vector multiplet numerator .
if there exist charges of type @xmath300 or @xmath301 in this charge set , they generate additional relations @xmath302 or @xmath303 .
this implies that the corresponding pole additionally collides at the singular point , so the singularity become degenerate .
+ let us consider the second possibility .
suppose that we have such @xmath304 and @xmath125 in a given set of contributing arrows .
it is rather easy to show that the second type of degenerate points never contribute even if they generate degeneracy .
suppose that there exist such @xmath125 and @xmath304 which are assigned to a same flavor charge .
then from the argument of * a2 * , we know that there exist at least @xmath156 @xmath305 which is connected to @xmath90-th source node by @xmath306 , and @xmath156 @xmath307 which is connected to @xmath308-th source node by @xmath309
. then they generate @xmath156 additional relations of type @xmath310 , and also the additional @xmath156 relations of type @xmath311 . in total , there are @xmath312 additional poles which additionally collide at this singular point .
however , they also generate at least @xmath312 zeros of type @xmath313 for @xmath314 and two more from @xmath315 .
since the order of additional zero @xmath316 always exceeds additional pole @xmath317 , this degenerate singularity never contributes .
+ finally , let us come back to the first degenerate singularity .
suppose that @xmath318 is the only relation which gives additional zero .
note that , for this case , there may exist a singularity where number of additional poles exceed or equal to the number of additional zeros .
however , one can show that they never satisfy the jk - condition .
first of all , since we have a freedom to choose the order of the charge set , we set the order of charges @xmath289 such that @xmath126 @xmath319 appear at the very first position , and immediately followed by @xmath320 @xmath321 .
then the ordered set @xmath282 contained in the set @xmath286 constructed as above procedure should be one of the following . * * @xmath322 * * @xmath323 * * @xmath324 * * @xmath325 + where @xmath326 .
note that the case @xmath327 has residue zero since this relation imposes @xmath299 which gives arise the order two zero .
there are no additional poles at this point since the other components of @xmath287 are kept generic , by the definition of the iterated residue .
finally , consider the second , third and the fourth case .
if the set @xmath282 still has @xmath1 arrows connected to the source @xmath75 , then it defines rank @xmath261 equation without any additional relations .
if this is the case , the relation imposed by the original set @xmath289 would have been overdetermined and can not define a singularity .
hence , we can conclude that there are less than @xmath1 arrows emerging from source @xmath75 in the set @xmath282 of type 2),3 ) and 4 ) .
going back to the proof of * a2 * , we concluded that this type of charge set @xmath282 can not satisfy the jk condition .
this procedure straightforwardly generalizes to the charge sets with larger number of additional zeros . * _ all configurations of contributing arrows obtained from gluing that of @xmath132 quiver satisfy the jk condition . _ + _ proof .
_ this can be proven by induction .
first , let us assume that the statement is true for @xmath112 type quiver .
it means that we have a set of @xmath261 positive coefficients @xmath262 such that the relation @xmath328 holds , where @xmath329 is a charge set with @xmath261 elements which is obtained from the gluing procedure .
the assignment of flavor symmetry @xmath330 obeys the two conditions in 2 ) of * a3*. now , we attach a subquiver of type @xmath331 to this quiver by identifying one of the sink node @xmath76 of the original quiver with that of new subquiver @xmath332 by @xmath138 .
we will see that for all such possible gluing , there exists a positive coefficient set @xmath333 with @xmath334 elements such that it satisfies @xmath335 where @xmath336 is a charge of the new quiver .
+ we can show that positive coefficient @xmath337 that satisfies ( [ aa ] ) can be easily constructed from @xmath338 of the original quiver . here
, we denoted @xmath209 as a set of all nodes of original quiver of type @xmath112 , and @xmath213 as a set of all nodes of new quiver @xmath339 .
first of all , @xmath340 , with @xmath341 , ( i.e. , when @xmath342 ) can be uniquely identified as @xmath343 .
then it follows that @xmath344 .
since all the arrows are connected , we can define numbers @xmath340 for all @xmath345 through the following procedure .
first of all , consider the @xmath346-th component of the above equation . since @xmath347 satisfies @xmath348 and @xmath344 , we can set @xmath349 for all @xmath350 . then , consider @xmath338 s for the above @xmath90 s and @xmath351 . if we look at @xmath90-th components of the equation , they imply @xmath352 with @xmath353 . next ,
consider the @xmath305-th component of the equation .
similarly , this implies that @xmath354 for all @xmath355 .
we can continue this procedure until all @xmath356 s are specified .
since there is no cycle and every nodes are connected , this procedure can uniquely fix all the coefficients @xmath340 which are by construction all positive .
finally , since the statement obvious holds for @xmath357 , we can say that for all @xmath10 , @xmath112 type quiver obtained from the gluing method satisfy the jk condition .
n. seiberg and e. witten , `` electric - magnetic duality , monopole condensation , and confinement in n=2 supersymmetric yang - mills theory , '' nucl .
b * 426 * ( 1994 ) 19 [ erratum - ibid .
b * 430 * ( 1994 ) 485 ] [ hep - th/9407087 ] .
n. seiberg and e. witten , `` monopoles , duality and chiral symmetry breaking in n=2 supersymmetric qcd , '' nucl .
b * 431 * ( 1994 ) 484 [ hep - th/9408099 ] .
f. ferrari and a. bilal , `` the strong coupling spectrum of the seiberg - witten theory , '' nucl .
b * 469 * ( 1996 ) 387 [ hep - th/9602082 ] .
k. -m .
lee and p. yi , `` dyons in n=4 supersymmetric theories and three pronged strings , '' phys .
d * 58 * , 066005 ( 1998 ) [ hep - th/9804174 ] .
m. r. douglas , b. fiol and c. romelsberger , `` stability and bps branes , '' jhep * 0509 * ( 2005 ) 006 [ hep - th/0002037 ] .
d. gaiotto , g. w. moore and a. neitzke , `` four - dimensional wall - crossing via three - dimensional field theory , '' commun .
* 299 * ( 2010 ) 163 [ arxiv:0807.4723 [ hep - th ] ] .
d. gaiotto , g. w. moore and a. neitzke , `` wall - crossing , hitchin systems , and the wkb approximation , '' arxiv:0907.3987 [ hep - th ] . j. de boer , s. el - showk , i. messamah and d. van den bleeken , `` quantizing n=2 multicenter solutions , '' jhep * 0905 * ( 2009 ) 002 [ arxiv:0807.4556 [ hep - th ] ] . j. manschot , b. pioline and a. sen , `` wall crossing from boltzmann black hole halos , '' jhep * 1107 * , 059 ( 2011 ) [ arxiv:1011.1258 [ hep - th ] ] .
j. manschot , b. pioline and a. sen , `` the coulomb branch formula for quiver moduli spaces , '' arxiv:1404.7154 [ hep - th ] .
k. hori , h. kim and p. yi , `` witten index and wall crossing , '' jhep * 1501 * ( 2015 ) 124 [ arxiv:1407.2567 [ hep - th ] ] . c. cordova and s. -h .
shao , `` an index formula for supersymmetric quantum mechanics , '' arxiv:1406.7853 [ hep - th ] . c. cordova and s. h. shao , `` asymptotics of ground state degeneracies in quiver quantum mechanics , '' arxiv:1503.03178 [ hep - th ] .
m. alim , s. cecotti , c. cordova , s. espahbodi , a. rastogi and c. vafa , `` bps quivers and spectra of complete n=2 quantum field theories , '' commun .
* 323 * ( 2013 ) 1185 [ arxiv:1109.4941 [ hep - th ] ] . | we explore vacuum degeneracy of kronecker quiver with large ranks , by computing witten index of corresponding 1d gauged linear sigma model . for @xmath0 quivers with the intersection number @xmath1 , we actually counted index of its mutation equivalent , @xmath2 , and find exponentially large behaviour whenever @xmath3 .
we close with speculation on more general ranks of kronecker quiver including the nonprimitive cases . 0.25 in .. * scaling behaviour of quiver quantum mechanics * heeyeon kim |
it is a general principle in the investigation of dynamical systems that _ hyperbolic systems _ ( also known as `` axiom a '' , following smale @xcite ) are the first class to investigate in a given setting : they exhibit the simplest behaviour , yet their study frequently leads to a better understanding of more complicated systems . in this article , we consider ( non - invertible ) dynamics in one complex variable . for rational maps ,
hyperbolic behaviour was understood , in general terms , already by fatou @xcite , though , of course , he did not use this terminology .
more precisely , a rational map @xmath1 is said to be _ hyperbolic _ if one of the following , equivalent , conditions holds ( * ? ? ?
* section 9.7 ) ( see below for definitions ) : a. the function @xmath2 is _ expanding _ with respect to a suitable conformal metric defined on a neighbourhood of its julia set ; b. every critical value of @xmath2 belongs to the basin of an attracting periodic cycle ; c. the _ postsingular set _ is a subset of the fatou set .
moreover ( * ? ? ?
* theorem 4.4 ) every hyperbolic rational map satisfies a. @xmath2 is _ stable _ ; in other words , any nearby rational map is topologically conjugate to @xmath2 on its julia set , with the conjugacy depending continuously on the perturbation.[item : stabledef ] the famous _ hyperbolicity conjecture _ asserts that condition [ item : stabledef ] is also equivalent to hyperbolicity ; this question essentially goes back to fatou .
see the final sentence of chapitre iv in @xcite , and also compare ( * ? ? ?
* section 4.1 ) for a historical discussion .
the iteration of transcendental entire functions @xmath3 also goes back to fatou @xcite , and has received considerable attention in recent years . as in the rational case , the _ fatou set _
@xmath4 of such a function is defined as the set of @xmath5 such that the iterates @xmath6 form a normal family in a neighbourhood of @xmath7 .
its complement , the _ julia set _
@xmath8 , is the set where the dynamics is `` chaotic '' .
the role played by the set of critical values in rational dynamics is now taken on by the set @xmath9 of ( finite ) _ singular values _ ,
i.e. the closure of the set of critical and asymptotic values of @xmath2 ( see section [ sec : singularvalues ] ) . in the transcendental setting due to the effect of the non - compactness of the phase space and the essential singularity at infinity
it is not clear how `` hyperbolicity '' should be defined .
accordingly , there is currently no accepted general definition ; see appendix a for a brief historical discussion .
we propose the following notion of `` expansion '' in this setting .
[ defn : expanding ] a transcendental entire function @xmath2 is _ expanding _ if there exist a connected open set @xmath10 , which contains @xmath11 , and a conformal metric @xmath12 on @xmath13 such that : 1 .
@xmath13 contains a punctured neighbourhood of infinity , i.e. there exists @xmath14 such that @xmath15 whenever @xmath16 ; 2 .
@xmath2 is expanding with respect to the metric @xmath17 , i.e. there exists @xmath18 such that @xmath19 whenever @xmath20 ; and 3 .
the metric @xmath17 is complete at infinity , i.e. @xmath21 whenever @xmath22 .
[ completecond ] we make three remarks about this definition .
first we note that the final condition holds whenever the metric @xmath17 is complete . indeed , as we shall see below
, we could require that @xmath17 is complete without changing the class of expanding functions .
we prefer instead to use the weaker condition [ completecond ] , which allows @xmath17 to be the euclidean metric .
second , one might ask , in analogy to hyperbolicity for rational maps , for the metric @xmath17 to be defined and conformal on a full ( rather than punctured ) neighbourhood of @xmath23 .
however , due to the nature of the essential singularity at infinity , such expansion can _ never _ be satisfied ; see corollary [ cor : sphericalexpansion ] below .
finally , one might require only expansion on a neighbourhood of the julia set , as a subset of the complex plane .
this is too weak a condition .
functions with this property may have infinitely many attractors , wandering domains or _ baker domains _ :
invariant domains of normality in which all orbits converge to infinity .
( compare figure [ fig : expanding ] . )
such behaviour is not compatible with the usual picture of hyperbolicity and , moreover , usually not stable even under simple perturbations .
our key observation about expanding entire functions is the following : [ thm : hyperbolicity ] if @xmath2 is expanding in the sense of definition [ defn : expanding ] , then @xmath9 is bounded .
in other words , every expanding function belongs to the _ eremenko - lyubich class _ @xmath24 introduced in @xcite , @xmath25 is a large and well - studied class . for functions @xmath26 there is a natural and well - established notion of hyperbolicity . indeed ,
already mcmullen ( * ? ? ?
* section 6 ) suggested calling an entire function @xmath2 `` expanding '' if the postsingular set @xmath27 is a compact subset of the fatou set ( see condition [ item : postsingularinf ] in theorem [ thmdef : hyperbolicity ] below ) .
the expansion property he established for such a function @xmath2 ( * ? ? ?
* proposition 6.1 ) is non - uniform , but rippon and stallard ( * ? ? ?
* theorem c ) established that definition [ defn : expanding ] holds for @xmath28 , with @xmath17 the cylinder metric , and @xmath29 sufficiently large . in (
* lemma 5.1 ) , it is shown that @xmath2 itself satisfies definition [ defn : expanding ] with respect to a suitable hyperbolic metric .
( note that the expanding property is stated in ( * ? ? ?
* theorem c ) only for points in the julia set , but holds also on the preimage of a neighbourhood of @xmath23 by a well - known estimate of eremenko and lyubich ; see below or ( * ? ? ?
* lemma 2.2 ) .
we also remark that rippon and stallard treat the more general case of transcendental _ meromorphic _ functions .
while we restrict here to entire functions , where definition [ defn : expanding ] seems particularly natural , our methods apply equally to the meromorphic case ; compare theorem [ thm : shrinkinginfinity ] and the discussion in appendix a. ) rippon and stallard @xcite mention that while it is not clear what the definition of a hyperbolic transcendental entire ( or meromorphic ) function should be it is natural to call the above functions `` hyperbolic '' in view of their expansion properties .
our results suggest that , conversely , only these entire functions should be classed as hyperbolic .
[ thmdef : hyperbolicity ] all of the following properties of a transcendental entire function @xmath3 are equivalent .
if any , and hence all , of these conditions hold , then @xmath2 is called _
hyperbolic_. a. @xmath2 is expanding in the sense of definition [ defn : expanding];[item : expanding ] b. @xmath2 is expanding in the sense of definition [ defn : expanding ] , and ( for suitably chosen @xmath13 ) the metric @xmath17 can be chosen to be the hyperbolic metric on @xmath13;[item : expandinghyperbolic ] c. @xmath26 and every singular value belongs to the basin of an attracting periodic cycle;[item : singularattracting ] d. the _ postsingular set _
@xmath30 is a compact subset of @xmath31.[item : postsingularinf ] moreover , each hyperbolic function is stable within its quasiconformal equivalence class .
quasiconformal equivalence classes form the natural parameter spaces of transcendental entire functions .
( see proposition [ prop : stability ] for the formal meaning of the final statement in the theorem . )
these classes were defined implicitly by eremenko and lyubich @xcite , and explicitly by the second author @xcite , and provide the appropriate context in which to consider stability .
it is well - known that functions in the eremenko - lyubich class have strong expansion properties near @xmath23 . indeed ,
if @xmath26 , then eremenko and lyubich ( * ? ? ?
* lemma 1 ) proved that @xmath32 observe that this quantity is precisely the derivative of @xmath2 with respect to the _ cylindrical _ metric @xmath33 on the punctured plane .
the second author @xcite showed that the converse also holds : no function outside of class @xmath25 exhibits this type of expansion near infinity . more precisely
: suppose that @xmath2 is a transcendental entire function , and consider the quantity @xmath34 then , either @xmath35 and @xmath36 , or @xmath37 and @xmath38 .
theorem [ thm : hyperbolicity ] follows from the following strengthening of theorem a. [ thm : shrinkinginfinity ] let @xmath2 be a transcendental entire ( or meromorphic ) function and @xmath39 .
let @xmath40 be an open neighbourhood of @xmath41 .
suppose that @xmath42 is a conformal metric defined on @xmath43 such that @xmath44 for all @xmath45 .
then @xmath46 note that one obtains the second half of theorem a by applying theorem [ thm : shrinkinginfinity ] to the cylindrical metric and a sequence of singular values tending to infinity .
a similar application of theorem [ thm : shrinkinginfinity ] gives theorem [ thm : hyperbolicity ] .
clearly some condition must be imposed upon the metric @xmath17 when @xmath40 contains no critical values .
indeed , if we let @xmath17 be the pull - back of the euclidean metric under @xmath2 , then fails by definition .
however , it appears plausible that such a pull - back must decay rapidly on some approach to infinity , and so the conclusion of theorem [ thm : shrinkinginfinity ] holds for certain metrics that are not complete at infinity .
the following theorem , another considerable strengthening of theorem a , shows that this is indeed the case .
[ thm : strong ] let @xmath2 be a transcendental entire ( or meromorphic ) function and @xmath39 .
if @xmath40 is an open neighbourhood of @xmath41 , then @xmath47 for all @xmath48 .
one obtains the second half of theorem a from theorem [ thm : strong ] by setting @xmath49 .
similarly , one obtains theorem [ thm : shrinkinginfinity ] with @xmath17 being the spherical metric by setting @xmath50 .
moreover , if @xmath2 is transcendental entire , then we can apply theorem [ thm : strong ] to the function @xmath51 , the singular value @xmath52 and @xmath50 . as claimed above
, this implies that the function @xmath2 can not be expanding with respect to the spherical metric , on the preimage of a full neighbourhood of @xmath23 .
[ cor : sphericalexpansion ] let @xmath2 be a transcendental entire function , and let @xmath14 .
then @xmath53 corollary [ cor : sphericalexpansion ] can also be seen to follow from wiman - valiron theory , which shows that there are points of arbitrarily large modulus near which @xmath2 behaves like a monomial of arbitrarily large degree .
it is straightforward to deduce ( [ sphexpansion ] ) from this fact .
( more precisely , the claim follows e.g. from formulae ( 3 ) and ( 4 ) in the statement of the wiman - valiron theorem in @xcite . )
we are grateful to alex eremenko for this observation .
finally , we consider the analogue of the class @xmath54 for functions analytic in a hyperbolic domain .
suppose that @xmath55 is a hyperbolic domain ( or , more generally , a hyperbolic riemann surface ) , and that @xmath56 is analytic .
we say that @xmath2 belongs to the eremenko - lyubich class , @xmath0 , if the set of singular values of @xmath2 is bounded .
we modify the definition ( [ etadef ] ) as follows : @xmath57 where the norm of the derivative is evaluated using the hyperbolic metric on @xmath58 and the cylindrical metric on the range .
[ thm : hyperbolicsurface ] suppose that @xmath58 is a hyperbolic surface and that @xmath56 is analytic and unbounded .
then , either @xmath59 and @xmath36 , or @xmath60 and @xmath38 .
for the second part of theorem [ thm : hyperbolicsurface ] , we could replace the hyperbolic metric by any complete metric on @xmath58 , as in theorem [ thm : shrinkinginfinity ] .
however , theorem [ thm : hyperbolicsurface ] as stated provides an appealing dichotomy in terms of the conformally natural quantity @xmath61 .
our proof of theorem [ thm : shrinkinginfinity ] considerably simplifies the original proof of theorem a , and can be summarized as follows .
the set @xmath62 must contain either a critical point of @xmath2 or an asymptotic curve @xmath63 whose image is a line segment in @xmath40 ; this is a classical and elementary fact , but its connection to the questions at hand appears to have been overlooked . since the @xmath17-length of @xmath63 is infinite , while the euclidean length of @xmath64 is finite , the conclusion is immediate
. this argument clearly does not apply to metrics that are not complete at infinity , and hence a more detailed analysis is required for the proof of theorem [ thm : strong ] . once again , we rely on elementary mapping properties of functions near a singular value , @xmath41 , that is not the limit of critical values .
we show that there are infinitely many , pairwise disjoint , unbounded simply - connected domains on which the function in question is univalent , and which are mapped to round discs near @xmath41 ; see corollary [ cor : disjointtracts ] . using a similar idea as in the classical proof of the denjoy - carleman - ahlfors theorem , it follows that , within some of these domains , the function must tend very quickly towards a corresponding asymptotic value .
this leads to the desired conclusion .
finally , the second part of theorem [ thm : hyperbolicsurface ] follows in the same manner as theorem [ thm : shrinkinginfinity ] .
the first part , on the other hand , can be deduced in a similar way to the proof of ( * ? ? ?
* lemma 1 ) ) , although we adopt a slightly different approach using basic properties of the hyperbolic metric .
in section [ sec : singularvalues ] we give background on the notion of singular values , and prove some basic results . in section [ sec : main ] , we deduce theorems [ thm : hyperbolicity ] , [ thmdef : hyperbolicity ] and [ thm : shrinkinginfinity ] . sections [ sec : strong ] and [ sec : hyperbolicsurface ] prove theorems [ thm : strong ] and [ thm : hyperbolicsurface ] . appendix
a contains historical remarks concerning notions of hyperbolicity for entire functions , and appendix b concerns the functions in figure 1 .
we denote the complex plane , the riemann sphere and the unit disc by @xmath65 , @xmath66 and @xmath67 respectively . for euclidean discs
, we use the notation @xmath68 if @xmath69 is a riemann surface , then we denote by @xmath70 the added point in the one - point compactification of @xmath69 . for example , if @xmath71 is a sequence of points of @xmath69 which eventually leaves any compact subset of @xmath69 , then @xmath72 .
( recall that , if @xmath69 is already compact , then @xmath70 is an isolated point of the one - point compactification .
this allows for uniformity of statements and definitions . )
closures and boundaries are always taken in an underlying riemann surface @xmath69 ; which @xmath69 is meant should be clear from the context .
suppose that @xmath69 is a riemann surface , and @xmath73 is open .
conformal metric _ on @xmath40 is a tensor that takes the form @xmath42 in local coordinates , with @xmath17 a continuous positive function .
when @xmath74 , which is the case of most interest to us , we can express the metric globally in this form , and we do not usually distinguish between the metric and its density function @xmath75 .
if @xmath76 is a locally rectifiable curve , then we denote the length of @xmath63 with respect to the metric @xmath17 by @xmath77 . if @xmath78 , then we denote the distance from @xmath7 to @xmath79 with respect to the metric @xmath17 by @xmath80 ; i.e. @xmath81 , where the infimum is taken over all curves connecting @xmath7 and @xmath79 .
note that , by definition , this distance is infinite if @xmath7 and @xmath79 belong to different components of @xmath40 . abusing notation slightly
, we define @xmath82 . by definition ,
this quantity is infinite if @xmath40 is relatively compact in @xmath69 .
if @xmath83 and @xmath84 , then we also set @xmath85 , and define @xmath86 . when @xmath87 and @xmath17 is the euclidean metric , then we write simply @xmath88 , @xmath89 and @xmath90 . if @xmath69 is a hyperbolic surface , then we write @xmath91 for the hyperbolic metric in @xmath69 and , in local coordinates , denote its density function by @xmath92 .
we would like to thank walter bergweiler and alex eremenko for interesting discussions about the possibility of strengthening and extending theorem a , which led us to discover the results presented in this article .
in this section , we first review the definitions of singular values . while we apply them mainly for meromorphic functions defined on subsets of the complex plane , we introduce them in the more general setting of analytic functions between riemann surfaces ; see also @xcite .
we do so to facilitate future reference , and to emphasize the general nature of our considerations .
in contrast to previous articles on similar subjects , we do not require iversen s more precise classification of inverse function singularities , for which we refer to @xcite , @xcite , @xcite and also @xcite .
instead , we only use some basic mapping properties of functions having singular values , which we derive here from first principles .
[ defn : singularvalues ] let @xmath69 and @xmath93 be riemann surfaces , let @xmath94 be analytic , and let @xmath95 .
a. @xmath41 is called a _ regular value _ of @xmath2 if there is an open neighbourhood @xmath40 of @xmath41 with the following property : if @xmath62 is any connected component of @xmath96 , then @xmath97 is a conformal isomorphism .
b. @xmath41 is called a _ singular value _ of @xmath2 if @xmath41 is not a regular value .
c. @xmath41 is called a _ critical value _ of @xmath2 if it is the image of a critical point .
d. @xmath41 is called an _ asymptotic value _ of @xmath2 if there is a curve @xmath98 such that @xmath99 and @xmath100 as @xmath101 .
such @xmath63 is called an _
asymptotic curve_.[item : asymptoticvalue ] the sets of singular , critical and asymptotic values of @xmath2 are denoted by @xmath9 , @xmath102 and @xmath103 respectively .
clearly @xmath104 .
on the other hand , critical and asymptotic values are dense in @xmath9 ( see corollary [ cor : asymptoticdense ] below ) , so that , in fact , @xmath105 .
equivalently to the definition above , @xmath9 is the smallest closed subset @xmath106 of @xmath93 such that the restriction @xmath107 is a covering map .
observe that any value in @xmath108 is a regular value of @xmath2 .
for example , if @xmath109 , @xmath110 and @xmath111 , then the set of singular values of @xmath2 coincides with the unit circle @xmath112 .
the following observation shows that , near any singular value , we can find inverse branches defined on round discs , and containing either critical or asymptotic values on their boundaries .
[ lem : inversebranches ] let @xmath69 be a riemann surface , and let @xmath113 be analytic , but not a conformal isomorphism . then there exists a round disc @xmath114 with @xmath115 , a branch @xmath116 of @xmath117 defined on @xmath114 , and a singular value @xmath118 such that either a. @xmath119 is a critical point of @xmath2 in @xmath69 , and @xmath41 is a critical value of @xmath2 , or [ item : critical ] b. @xmath120 . in particular , @xmath41 is an asymptotic value of @xmath2 , and there is an asymptotic curve @xmath63 that maps one - to - one onto a straight line segment @xmath64 ending at @xmath41.[item : asymptotic ] fix some point @xmath121 which is not a critical point of @xmath2 . by postcomposing with a mbius transformation
, we may assume without loss of generality that @xmath122 .
let @xmath123 be the branch of @xmath117 taking @xmath124 to @xmath125 , and let @xmath126 be the greatest value such that @xmath123 can be continued analytically to the disc @xmath114 of radius @xmath127 around @xmath124
. then @xmath128 , since @xmath2 is not a conformal isomorphism .
it follows that there is a point @xmath129 such that @xmath123 can not be continued analytically into @xmath41 .
if @xmath130 , then we can take @xmath131 , where @xmath132 is the radius of @xmath114 ending at @xmath41 , and the proof of case [ item : asymptotic ] is complete . otherwise , there is a sequence @xmath133 with @xmath134 and @xmath135 , for some @xmath136 . by continuity of @xmath2
, we have @xmath137 .
moreover @xmath138 is a critical point , since otherwise the local inverse of @xmath2 that maps @xmath41 to @xmath138 would provide an analytic continuation of @xmath123 into @xmath41 by the identity theorem .
discs as in lemma [ lem : inversebranches ] , on which branches of the inverse are defined and which have asymptotic values on their boundary , have appeared previously in the study of _ indirect _ asymptotic values in the sense of iversen ; see ( * ? ? ?
* proof of theorem 1 ) or ( * ? ? ?
* theorem 6.2.3 ) . to simplify subsequent discussions , we introduce the following terminology .
let @xmath69 and @xmath93 be riemann surfaces , and let @xmath94 be analytic .
suppose that @xmath114 is an analytic jordan domain such that @xmath139 , and that @xmath116 is a branch of @xmath117 defined on @xmath114 .
suppose furthermore that there is an asymptotic value @xmath140 such that @xmath120 .
then we call @xmath114 a _ disc of univalence _ at the asymptotic value @xmath41 .
we also call the domain @xmath141 a _ tract _ over the disc of univalence @xmath114 .
[ rmk : nicecurves ] observe that , if @xmath41 is an asymptotic value for which there exists a disc of univalence , then in particular there is an asymptotic curve @xmath63 for @xmath41 that is mapped one - to - one to an analytic arc compactly contained in @xmath93 , with one endpoint at @xmath41 .
the following observation is frequently useful .
[ obs : discrestriction ] let @xmath69 and @xmath93 be riemann surfaces , and let @xmath94 be analytic .
let @xmath142 be a domain , and let @xmath13 be a connected component of @xmath96 . then any disc of univalence of the restriction @xmath143 at an asymptotic value @xmath144 is also a disc of univalence of @xmath94 at @xmath41 .
let @xmath145 be a branch of @xmath117 as in the definition of the disc of univalence @xmath114 . then @xmath146 .
since @xmath13 was chosen to be a connected component of @xmath96 , and so @xmath147 , we deduce that @xmath148 .
[ cor : asymptoticdense ] let @xmath69 and @xmath93 be riemann surfaces , and let @xmath94 be analytic
. denote by @xmath149 the set of those asymptotic values at which there exists a disc of univalence .
then @xmath150 is dense in @xmath9 .
let @xmath39 , and let @xmath40 be a simply - connected neighbourhood of @xmath41 .
by definition , there exists a connected component @xmath62 of @xmath96 such that @xmath97 is not a conformal isomorphism . applying lemma [ lem : inversebranches ] to @xmath151 , where @xmath152 is a riemann map , we find that either @xmath40 contains a critical point of @xmath2 , or that @xmath153 , and hence @xmath97 , has a disc of univalence .
the claim follows from observation [ obs : discrestriction ] .
we also note the following well - known fact .
[ lem : logarithmic ] let @xmath69 and @xmath93 be riemann surfaces , and let @xmath94 be analytic .
let @xmath41 be an isolated point of @xmath9 , let @xmath40 be a simply - connected neighbourhood of @xmath41 with @xmath154 , and let @xmath62 be a connected component of @xmath155 .
then either a. @xmath62 is a punctured disc , and @xmath156 is a finite - degree covering map , or b. @xmath62 is simply - connected , and @xmath156 is a universal covering map . by definition
, @xmath156 is a covering map , and the only analytic coverings of the punctured disc are as stated ( * ? ? ? * theorem 5.10 ) .
we can now establish the following , which is crucial for the proof of theorem [ thm : strong ] .
[ cor : disjointtracts ] let @xmath69 and @xmath93 be riemann surfaces , and let @xmath94 be analytic , with no removable singularities at any punctures of @xmath69 .
let @xmath39 , and suppose that @xmath41 has an open neighbourhood @xmath40 with @xmath157 .
then there exist infinitely many discs of univalence @xmath158 of @xmath2 such that the corresponding tracts @xmath159 are pairwise disjoint .
if @xmath41 is not an isolated point of @xmath9 , then the claim follows immediately from corollary [ cor : asymptoticdense ] . otherwise , let @xmath160 be a round disc around @xmath41 in a given local chart , and apply lemma [ lem : logarithmic ] . by assumption ,
@xmath41 is not a critical value , and @xmath2 does not have any removable singularities .
hence , if @xmath62 is any connected component of @xmath161 , then @xmath162 is a universal covering .
we may thus choose a single disc @xmath163 that is tangent to @xmath41 , set @xmath164 for all @xmath165 , and let the @xmath159 be the infinitely many different components of @xmath166 in @xmath69 .
the following is a general result concerning behaviour of analytic functions on sets that map close to singular values .
[ prop : contraction ] let @xmath69 and @xmath93 be riemann surfaces , let @xmath94 be analytic , let @xmath39 , and let @xmath40 be an open neighbourhood of @xmath41 .
suppose that @xmath62 is a connected component of @xmath96 such that @xmath97 is not a conformal isomorphism , and that @xmath17 is a conformal metric on @xmath62 such that @xmath167 for all @xmath45 .
also let @xmath168 be any conformal metric on @xmath40 .
then @xmath169 where the norm of the derivative is measured with respect to the metrics @xmath17 and @xmath168 .
suppose , by way of contradiction , that @xmath170 then @xmath2 has no critical points in @xmath62 . by corollary [ cor : asymptoticdense ] and remark [ rmk : nicecurves ]
, there is an analytic curve @xmath171 to infinity such that @xmath63 is mapped one - to - one to an analytic arc @xmath132 compactly contained in @xmath40 .
then @xmath172 this follows immediately from proposition [ prop : contraction ] , by taking @xmath168 to be the euclidean metric .
in particular , theorem [ thm : shrinkinginfinity ] , remains true if we replace the full preimage @xmath62 by any component on which @xmath2 is not a conformal isomorphism .
let @xmath2 be a transcendental entire function , and let @xmath17 be a conformal metric , defined on an open neighbourhood @xmath13 of @xmath11 that contains a punctured neighbourhood of @xmath23 .
suppose that @xmath44 for all @xmath15 , and that @xmath2 is expanding with respect to the metric @xmath17 .
if @xmath138 is a critical point of @xmath2 , then either @xmath173 or @xmath174 .
it follows that @xmath13 contains only finitely many critical values , none of which lie in @xmath11 . by shrinking @xmath13
, we may hence assume that @xmath13 contains no critical values of @xmath2 .
we have @xmath175 .
in particular , @xmath9 is bounded .
let @xmath176 and let @xmath14 be such that @xmath177 . since @xmath79 is not a critical value of @xmath2 , if @xmath178 is a sufficiently small disc around @xmath79 , then every component of @xmath166 that is not contained in @xmath179 is mapped to @xmath114 as a conformal isomorphism . on the other hand ,
every component @xmath62 of @xmath166 that is contained in @xmath180 is also mapped as a conformal isomorphism , by proposition [ prop : contraction ] and the expanding property of @xmath2 .
observe that this proves theorem [ thm : hyperbolicity ] .
we have @xmath181 .
let us set @xmath182 then @xmath183 since @xmath13 contains a punctured neighbourhood of @xmath23 , and by assumption on @xmath17 .
let @xmath184 , and let @xmath185 be a simply - connected neighbourhood of @xmath79 such that @xmath186 .
by claim 1 , @xmath187 , and hence every component @xmath188 of @xmath189 is mapped to @xmath185 as a conformal isomorphism .
furthermore , by the expanding property of @xmath2 , we have @xmath190 .
hence we can apply the preceding observation to @xmath188 , and see that any component of @xmath191 is mapped univalently to @xmath185 .
proceeding inductively , we see that every branch of @xmath192 can be defined on @xmath185 , for all @xmath193 .
hence @xmath194 , and in particular @xmath195 , as required .
recall that every parabolic periodic cycle of a transcendental entire function , and the boundary of every siegel disc , lies @xmath196 ( * ? ? ?
* theorem 7 ) .
furthermore , the iterates in any wandering domain of @xmath2 converge uniformly to @xmath197 @xcite . since @xmath26 , @xmath2 has no fatou component on which the iterates converge uniformly to infinity ( * ? ? ? * theorem 1 ) .
so @xmath2 has no wandering or baker domains .
thus we conclude that @xmath31 is a union of attracting basins , and hence every point of @xmath198 belongs to an attracting basin .
this completes the proof that [ item : expanding ] implies [ item : singularattracting ] in theorem [ thmdef : hyperbolicity ] .
the equivalence of [ item : singularattracting ] and [ item : postsingularinf ] is well - known ; see e.g. @xcite or ( * ? ? ?
* section 2 ) . that these in turn imply expansion in the sense of definition [ defn : expanding ] , with @xmath17 the hyperbolic metric on a suitable domain @xmath13 ,
was shown in ( * ? ? ?
* lemma 5.1 ) . the final claim regarding stability
is made precise by proposition [ prop : stability ] below .
[ prop : stability ] suppose that @xmath199 is a complex manifold , and that @xmath200 is a family of entire functions of the form @xmath201 , where @xmath202 are quasiconformal homeomorphisms depending analytically on the parameter @xmath203 and @xmath2 is transcendental entire .
if @xmath204 is a parameter for which @xmath205 is hyperbolic , then @xmath205 and @xmath206 are quasiconformally conjugate on their julia sets whenever @xmath203 is sufficiently close to @xmath207 .
moreover , this conjugacy depends analytically on the parameter @xmath203 .
let us first observe that hyperbolicity is an open property in any such family .
indeed , [ item : singularattracting ] in theorem [ thmdef : hyperbolicity ] is equivalent to the existence of a compact set @xmath208 with @xmath209 ; see ( * ? ? ?
* proposition 2.1 ) . clearly , any function @xmath210 that is sufficiently close to @xmath2 in the sense of locally uniform convergence satisfies @xmath211 .
if , furthermore , @xmath212 is sufficiently close to @xmath9 in the hausdorff metric , then @xmath213 , and hence @xmath210 is also hyperbolic .
therefore , if @xmath2 belongs to any analytic family as in the statement of the theorem , then @xmath2 has a neighbourhood in which all maps are hyperbolic .
in particular , no function in this neighbourhood has any parabolic cycles , and it follows that the repelling periodic points of @xmath2 move holomorphically over this neighbourhood . by the `` @xmath203-lemma '' @xcite , it follows that the closure of the set of repelling periodic points of @xmath2 , i.e. the julia set , also moves holomorphically .
this yields the desired result .
we note that this argument uses the fact that the analytic continuation of a repelling periodic point encounters only algebraic singularities within a family as above .
this is proved in ( * ? ? ? * section
4 ) for maps with finite singular sets ; the same argument applies in our setting . alternatively
, the claim can also be deduced formally from the results of @xcite , where it is shown that , for a given compact subset of the quasiconformal equivalence class , the set of points whose orbits remain sufficiently large moves holomorphically .
[ sec : strong ] let @xmath2 be a transcendental entire or meromorphic function , let @xmath39 be a finite singular value of @xmath2 , and let @xmath40 be an open neighbourhood of @xmath41 .
we may assume that there is a disc @xmath214 around @xmath41 that contains no critical values of @xmath2 , as otherwise there is nothing to prove .
let @xmath208 be any positive integer . by corollary [ cor : disjointtracts ]
, we can find @xmath208 discs of univalence @xmath215 , having pairwise disjoint tracts @xmath216 .
let @xmath217 be the associated asymptotic values .
also , for @xmath218 , let @xmath219 be the preimage in @xmath220 of the radius of @xmath221 ending at @xmath222 .
then @xmath219 is an asymptotic curve for @xmath222 , @xmath223 is a straight line segment , and @xmath224 as @xmath225 in @xmath219 .
this is precisely the setting of the proof of ( * ? ? ?
* theorem 1 ) , and formula ( 12 ) in that paper shows that there exist an integer @xmath29 and a sequence @xmath226 of points tending to infinity on @xmath219 such that @xmath227 where @xmath228 is any positive integer with @xmath229 .
( in @xcite , the function @xmath2 is required to have order less than @xmath230 , and the values @xmath231 are assumed to be pairwise distinct . however , neither of these assumptions are required for the proof of formula ( 12 ) . )
since @xmath208 was arbitrary , the claim of the theorem follows .
the argument in @xcite is essentially the same as in the proof of the classical denjoy - carleman - ahlfors theorem : since the tracts @xmath220 are unbounded and pairwise disjoint , some of them must have a small average opening angle . by the ahlfors distortion theorem
, it follows that @xmath2 must approach @xmath222 rapidly along @xmath219 , which is only possible if the derivative becomes quite small along this curve . for the reader s convenience , we present a self - contained proof of theorem [ thm : strong ] , following the same idea .
since we are not interested in precise estimates , we replace the use of the ahlfors distortion theorem by the standard estimate on the hyperbolic metric in a simply - connected domain .
let @xmath232 be sufficiently large to ensure that each @xmath219 contains a point of modulus @xmath233 . for each @xmath29
, and each @xmath234 , let us denote by @xmath235 the piece of @xmath219 connecting @xmath7 to @xmath23 , and the complementary bounded piece by @xmath236 .
suppose that the conclusion of the theorem did not hold .
then @xmath237 for some @xmath238 and all @xmath239 .
this implies @xmath240 we now prove that , if @xmath208 was chosen large enough , depending on @xmath241 , then such an estimate can not hold for all @xmath219 .
indeed , for @xmath242 , let @xmath243 denote the angular measure of the set @xmath244 . since the @xmath220 are disjoint , we have @xmath245 we are interested in the reciprocals @xmath246 , since these allow us to estimate the density of the hyperbolic metric in @xmath247 .
indeed , for @xmath248 , it follows from the fact that @xmath247 is simply connected and ( * ? ? ?
* theorem i.4.3 ) that @xmath249 by and the cauchy - schwarz inequality , we have @xmath250 let @xmath251 and , for each @xmath29 , choose a point @xmath252 with @xmath253 .
then the total hyperbolic length of the pieces @xmath254 satisfies @xmath255 thus there must be a choice of @xmath29 and a sequence @xmath256 in @xmath219 such that @xmath257 as @xmath258 . now @xmath259 is a conformal isomorphism . since @xmath260 is a radial segment connecting the centre of @xmath221 to @xmath261
, we deduce that @xmath262 hence we see that can not hold for @xmath263 . since @xmath208 was arbitrary , this completes the proof .
we could have formulated a more general version of theorem [ thm : strong ] , where @xmath2 is an analytic function between riemann surfaces @xmath69 and @xmath93 , and @xmath17 is a conformal metric on @xmath69 that is complete except at finitely many punctures of @xmath69 , where the metric is allowed to decay at most polynomially .
let @xmath58 be a hyperbolic surface , and let @xmath264 be analytic .
first suppose that @xmath9 is unbounded . for every @xmath14 , we may apply proposition [ prop : contraction ] to @xmath2 , taking @xmath41 to be an element of @xmath9 with @xmath265 , taking @xmath17 to be the hyperbolic metric on @xmath58 , and @xmath168 the cylindrical metric on @xmath65 .
the fact that @xmath266 follows .
on the other hand , suppose that @xmath9 is bounded , and let @xmath267 .
consider the domain @xmath268 , and its preimage @xmath269 .
if @xmath62 is a connected component of @xmath270 , then @xmath97 is an analytic covering map , and hence a local isometry of the corresponding hyperbolic metrics . by the schwarz lemma , we know that @xmath271 . on the other hand , up to a normalisation factor
, the density of the hyperbolic metric of @xmath40 is given by ( * ? ? ?
* example 9.10 ) @xmath272 as @xmath225 .
hence , the derivative of @xmath2 , measured with respect to the hyperbolic metric on @xmath58 and the cylindrical metric on @xmath65 , satisfies @xmath273 as @xmath274 ( where the second term should be understood in a local coordinate for @xmath58 near @xmath7 ) .
hence @xmath275 , as claimed .
as mentioned in the introduction , the first notion of `` expanding '' entire functions , which coincides with our notion of hyperbolicity in definition [ thmdef : hyperbolicity ] , goes back to mcmullen @xcite .
as also mentioned , rippon and stallard @xcite discuss hyperbolicity in the transcendental setting , arguing that functions in the class @xmath25 satisfying [ item : singularattracting ] in theorem [ thmdef : hyperbolicity ] deserve to be called hyperbolic .
however , they left open the possibility that other functions might also be classed as `` hyperbolic '' .
mayer and urbanski @xcite gave a definition of hyperbolicity that relies on the euclidean metric .
specifically , they require both uniform expansion on the julia set and that the postsingular set be a definite ( euclidean ) distance away from the julia set . without additional requirements , this includes such functions as our examples in the introduction , which exhibit `` non - hyperbolic '' phenomena and are not stable under simple perturbations ; see appendix b. with an additional strong regularity assumption near the julia set , mayer and urbanski obtained strong and striking results concerning the measurable dynamics of `` hyperbolic '' functions in the above sense ; further properties of these functions are described in @xcite .
we are , however , not aware of any examples outside of the class @xmath25 where these assumptions are known to hold . on the other hand ,
if @xmath26 is hyperbolic in the sense of mayer and urbanski , then it is also hyperbolic in the sense of theorem [ thmdef : hyperbolicity ] .
all of these definitions are in fact formulated , more generally , for transcendental _ meromorphic _ functions . in this setting , there is a third , and strongest , notion of hyperbolicity , studied by zheng @xcite .
this requires , in addition , that infinity is not a singular value , and hence can never be satisfied when @xmath2 is transcendental entire .
an example of a function with this property is @xmath276 , for @xmath277 .
( we note that zheng also studied the definitions given by rippon and stallard and by mayer and urbanski ; we refer to his paper for further details . )
zheng s definition of `` hyperbolicity on the riemann sphere '' may be considered to correspond most closely to the case of hyperbolic rational functions .
in particular , zheng shows that a meromorphic function is hyperbolic in this sense if and only if it satisfies a certain uniform expansion property with respect to the spherical metric
. our results also apply in the setting of meromorphic functions , with appropriate modifications of definitions to correctly handle prepoles . once again , they indicate that the notion of hyperbolicity does not make sense outside of the class @xmath25 .
we recall also our earlier comment that , although our applications are to meromorphic functions on subsets of the complex plane , our basic results regarding the properties of singularities are given for analytic functions between riemann surfaces .
a generalisation of the study of transcendental dynamics into this setting is possible for the classes of `` finite type maps '' , and more generally `` ahlfors islands maps '' , suggested by epstein @xcite . if @xmath13 is a riemann surface and @xmath69 is a compact riemann surface , then an analytic function @xmath278 is called a _ finite - type map _
if @xmath9 is finite and @xmath2 has no removable singularities at any punctures of @xmath13 . in the case where @xmath279 , epstein develops an iteration theory that carries over the basic results from the theory of rational dynamics and of entire and meromorphic functions with finitely many singular values .
similarly to the case of meromorphic functions , we could again consider two different notions of hyperbolicity in this setting : one , analogous to that of zheng , where all singular values must lie in attracting basins or map outside of the closure of @xmath13 ; and a weaker notion , in analogy to that of rippon and stallard , where we allow singular values to lie on ( or map into ) the boundary of the domain of definition .
all the standard results for the case of meromorphic functions should extend to this setting also .
the larger class of _ ahlfors islands maps _ includes all transcendental meromorphic functions , as well as all finite type maps .
it is tempting to define an `` eremenko - lyubich class '' of such maps , consisting of those for which @xmath280 is a compact subset of the domain of definition @xmath13 .
our results still apply in this setting , and imply that hyperbolicity is only to be found within this class . however , it is no longer clear that functions _ within _ this class have suitable expansion properties near the boundary , and hence dichotomies such as that of theorem a break down . finding a natural class that extends both the eremenko - lyubich class of entire functions and all finite type maps appears to be an interesting problem
in this section , we briefly discuss the three functions @xmath281 mentioned in the introduction .
all three are well - studied , and have properties that are not compatible with what would normally be considered hyperbolic behaviour .
a. @xmath282 has a baker domain containing the right half - plane @xmath283 .
that is , @xmath284 , and @xmath285 for all @xmath286 ; b. @xmath287 has infinitely many superattracting fixed points , @xmath288 ; c. @xmath289 has @xmath290 , and possesses an orbit of wandering domains .
the function @xmath287 is precisely newton s method for finding the points where @xmath291 ; it was studied in detail by weinreich @xcite .
clearly it follows directly from the definition that the points @xmath292 are indeed superattracting fixed points .
the function @xmath289 is a well - known example of a transcendental entire functions with wandering domains , first described by herman ; see ( * ? ? ?
* example 2 on p. 564 ) and ( * ? ? ?
* section 4.5 ) . the fact that @xmath290 follows from the relations @xmath293 and @xmath294 .
since the points @xmath292 all belong to different fatou components for @xmath287 , the same is true for @xmath289 . since @xmath295 , they do indeed belong to an orbit of wandering domains for @xmath289 . for each @xmath296
, there is an open neighbourhood @xmath40 of @xmath297 in @xmath65 with @xmath298 and a conformal metric @xmath17 on @xmath40 such that @xmath299 is uniformly expanding with respect to @xmath17 . writing @xmath300 , the function @xmath282 is semi - conjugate to @xmath301 this is a hyperbolic entire function , since its unique asymptotic value @xmath124 is an attracting fixed point , and its unique critical point @xmath302 belongs to the basin of attraction of this fixed point . by (
* lemma 5.1 ) , the function @xmath303 is expanding on a neighbourhood of its julia set , with respect to a suitable hyperbolic metric . pulling back this metric under the semiconjugacy
, we obtain the desired property for @xmath282 .
the argument for @xmath287 and @xmath289 is analogous .
both functions are semi - conjugate to the map @xmath304 this function is not hyperbolic , as the asymptotic value @xmath124 is a repelling fixed point .
however , we note that this asymptotic value does not correspond to any point in the @xmath7-plane under the semiconjugacy .
one can hence think of @xmath305 as being hyperbolic as a self - map of @xmath306 , since the critical point @xmath302 is a superattracting fixed point . the same proof as in @xcite yields a neighbourhood @xmath307 of @xmath308 and a conformal metric on @xmath40 such that @xmath305 is expanding , and the claim follows .
let @xmath309 . then @xmath299 is not stable in the family @xmath310 ; more precisely , there exist values of @xmath203 arbitrarily close to @xmath311 such that @xmath312 and @xmath299 are not topologically conjugate on their julia sets .
we prove only the case @xmath313 ; the proofs in the other cases are very similar . for simplicity , we write @xmath314 and @xmath315 . consider the critical points of @xmath316 , @xmath317 .
observe that @xmath318 for all @xmath29 .
we claim that there are values of @xmath203 arbitrarily close to @xmath311 such that , for some sufficiently large value of @xmath29 , we have @xmath319 . in this case @xmath11 and @xmath320 are not topologically conjugate on their julia sets , since such a conjugacy must preserve the local degree of @xmath2 . in order to prove this claim , observe that we can analytically continue the repelling fixed point @xmath321 of @xmath2 as a solution of the equation @xmath322 in a neighbourhood of @xmath323 , by the implicit function theorem . that is
, if @xmath324 is sufficiently small , then there is an analytic function @xmath325 such that @xmath326 is a repelling fixed point of @xmath206 for all @xmath327 .
it can be seen that there are infinitely many zeros of @xmath2 , and that we can number a subsequence of them , @xmath328 , so that @xmath329 and @xmath330 as @xmath331 . by a further calculation
, there exists @xmath126 such that , for sufficiently large values of @xmath332 , @xmath2 ( and hence @xmath316 ) is univalent in @xmath333 . since @xmath334 as @xmath335 , uniformly for @xmath336 , we deduce by the koebe quarter theorem that @xmath337 for all sufficiently large @xmath332 , whenever @xmath338 .
suppose that @xmath29 is large .
observe that the mapping @xmath339 which takes @xmath203 to @xmath340 maps @xmath341 to @xmath342 .
hence , if @xmath332 is sufficiently large , then @xmath29 can be chosen such that @xmath343 and also @xmath344 whenever @xmath345 .
let @xmath346 be the branch of @xmath347 which maps @xmath348 to @xmath333 .
let @xmath356 be a component of @xmath357 .
then @xmath358 whenever @xmath359 .
hence we may apply rouch s theorem again , and deduce that there exists @xmath360 such that @xmath361 .
in other words @xmath362 is a repelling fixed point of @xmath205 and so lies in @xmath363 .
this completes the proof of our claim .
eremenko , _ singularities of inverse functions _ , lecture notes from the 2013 icms conference ` the role of complex analysis in complex dynamics ' .
available at http://www.icms.org.uk/downloads/complex/eremenko.pdf . | hyperbolicity plays an important role in the study of dynamical systems , and is a key concept in the iteration of rational functions of one complex variable .
hyperbolic systems have also been considered in the study of transcendental entire functions .
there does not appear to be an agreed definition of the concept in this context , due to complications arising from the non - compactness of the phase space . in this article
, we consider a natural definition of hyperbolicity that requires expanding properties on the preimage of a punctured neighbourhood of the isolated singularity .
we show that this definition is equivalent to another commonly used one : a transcendental entire function is hyperbolic if and only if its postsingular set is a compact subset of the fatou set .
this leads us to propose that this notion should be used as the general definition of hyperbolicity in the context of entire functions , and , in particular , that speaking about hyperbolicity makes sense only within the _ eremenko - lyubich class _ @xmath0 of transcendental entire functions with a bounded set of singular values .
we also considerably strengthen a recent characterisation of the class @xmath0 , by showing that functions outside of this class can not be expanding with respect to a metric whose density decays at most polynomially .
in particular , this implies that no transcendental entire function can be expanding with respect to the spherical metric .
finally we give a characterisation of an analogous class of functions analytic in a hyperbolic domain . |
one of the most remarkable results of recent observational studies of @xmath0 cen is that it shows a double main sequence ( dms ) in the color magnitude diagrams ( cmds ) of its stellar populations ( e.g. , anderson 1997 ; bedin et al .
bedin et al .
( 2004 ) proposed four possible scenarios for the origin of the dms : ( 1 ) theoretical isochrone models or data calibration are in error , ( 2 ) stars on the bluer main sequence ( bms ) of the dms are a super metal - poor ( [ fe / h]@xmath2 ) population , ( 3 ) the bms represents a very helium - rich ( @xmath3 ) population , and ( 4 ) the bms represents a background stellar population about @xmath4 kpc behind @xmath0 cen .
a number of recent investigations have suggested that the above scenario ( 3 ) ( hereafter referred to as the `` helium pollution scenario '' and abbreviated to `` heps '' for convenience ) is the most promising among the four ( e.g. , lee et al . 2005 ; norris 2004 ; piotto et al .
2005 ) . in the heps
, there were two epochs of major star formation in the history of @xmath0 cen .
the first generation of stars was formed from proto - globular cluster ( gc ) cloud(s ) to become stars on the red main sequence ( rms ) whereas the second generation of stars was formed from gas ejected from the rms .
one of the key questions related to the heps is whether and how the observationally suggested large helium enrichment ( @xmath5 ) for the bms can be obtained from the rms with normal @xmath6 ( @xmath7 ) in the star formation history of @xmath0 cen ( e.g. , norris 2004 ) .
there are three candidates for the helium overabundance ( norris 2004 ; dantona et al .
2005 ; piotto et al . 2005 ) : ( 1 ) agb stars with the masses larger than @xmath8 , ( 2 ) stellar winds associated with massive stars during their early evolutionary phases , and ( 3 ) type ii supernovae ( sne ii ) .
it is however unclear which of the above three is the most reasonable and realistic in the heps given a lack of extensive theoretical studies on the three candidates .
the purpose of this letter is to investigate the three possibilities in a quantitative manner and thereby discuss which is the most promising as the cause of the bms ( and the dms ) in the context of the heps . in this investigation , ( 1 )
the possible @xmath6 value observationally suggested for @xmath0 cen ( norris 2004 ; piotto et al .
2005 ) and ( 2 ) the observed number fraction of the bms ( bedin et al . 2004 ) are used to constrain the best possible initial mass function ( imf ) of forming stars in the heps .
we do not intend to discuss extensively the observed abundance pattern of the bms and the rms in this paper , because chemical yield tables for agb stars with helium - rich ejecta ( @xmath9 ) are not currently available .
we will discuss this point in our forthcoming papers ( bekki & norris 2005 ; bn ) .
previous theoretical studies demonstrated that if gcs lose more than 50% of their initial masses , they will disintegrate ( e.g. , geyer & burkert 2001 ) .
we also use this result as a constraint for globular cluster imf in the heps .
we demonstrate that none of the above three candidates can explain the above constraint ( 2 ) without the modeled @xmath0 cen disintegrating .
firstly we summarize the possible three `` polluters '' which can significantly enrich @xmath0 cen with helium if the second generation of stars ( i.e. , the bms ) was formed from their ejecta . in order for the observationally
suggested large value of @xmath6 ( @xmath10 ) to be explained , we need to select stellar objects whose ejecta have @xmath6 significantly higher than 0.24 .
figure 1 shows the helium mass fraction ( @xmath6 ) in the ejecta of three candidates ; ( 1 ) agb stars ( van den hoek & groenewegen 1997 , hereafter referred to as vg97 ) ; ventura & dantona 2005 , as va05 ) , ( 2 ) massive stars ( schaller et al .
1992 , s92 ) , and ( 3 ) sne ii ( woosley & weaver 1995 , ww95 ) . based on the values listed in these papers ,
@xmath6 is derived for each stellar mass .
the theoretical predictions of @xmath6 for agb stars is at most 0.28 for vg97 and 0.32 for va05 .
dantona et al .
( 2005 ) however suggested that more massive agb stars with the masses of @xmath11 can have @xmath12 ( see also lattanzio et al .
we therefore consider that massive agb stars can be the polluters in @xmath0 cen but that lower mass agb objects ( @xmath13 ) with smaller @xmath6 can not .
the observed difference in metallicity ( [ fe / h ] ) between the rms and the bms ( piotto et al . 2005 ) suggests that the above scenario ( 1 ) alone can not simply explain the metallicity difference , because metallicity of agb ejecta do not differ from those of their host stars .
therefore , full chemical evolution models including the contribution of the three polluters ( and mixing between the ejecta of the polluters and fresh gas ) need to be considered if we try to explain fully self - consistently both the helium and metallicity one between the rms and the bms .
we however discuss advantages and disadvantages of each of the three polluters in the helium enrichment process separately .
this will better permit us to disentangle the contribution of the different polluters in producing he - enhanced stars and thereby to analyze advantages and disadvantages of each scenario in explaining the helium abundance of the bms .
the entire chemical evolution history for different elements ( e.g. , @xmath14c ) in @xmath0 cen will be discussed in our future papers ( bn ) .
the first question is whether the number fraction of the second generation of stars ( i.e. , the bms fraction ) can be as large as @xmath15 ( bedin et al .
2004 ) for a reasonable imf without @xmath0 cen being disintegrated following mass loss from massive stars and sne ii . in order to estimate the mass fraction ( @xmath16 ) of agb progenitor stars with masses ranging from @xmath8 to @xmath17 in a gc with the total mass of @xmath18 , we assume an imf in number defined as @xmath19 , where @xmath20 is the initial mass of each individual star and the slope @xmath21 corresponds to the salpeter imf . the normalization factor
@xmath22 is a function of @xmath18 , @xmath23 ( lower mass cut - off ) , and @xmath24 ( upper one ) .
a is expressed as @xmath25 , where @xmath23 and @xmath24 are set to be @xmath26 and @xmath27 , respectively .
although the number fraction of low - mass stars on the bms can depend on the forms of imfs , we adopt the above imf and show the results .
this is mainly because we can show more clearly the roles of imf slopes in controlling the mass fraction of agb stars for the adopted imf in this preliminary study . the models with the imf proposed by kroupa et al .
( 1993 ) show a rather small fraction ( @xmath28 % ) of agb stars in comparsion with the models with the above imf . in this agb pollution scenario
, we assume that ( 1 ) the gas ejected from agb stars in the first generation ( i.e. , the rms ) is all used ( i.e. , a star formation efficiency of 100 % ) for the formation of the second generation ( i.e. , the bms ) and ( 2 ) the total amount of gas ejected from a agb star is the same as the stellar mass .
this second assumption is reasonable because more than 80 % of the stellar mass is ejected as agb winds for agb stars with @xmath29 ( e.g. , vg97 ) .
we also assume two different slopes of imf ( @xmath30 ) for the first ( @xmath31 ) and the second ( @xmath32 ) generations of stars in order to investigate the maximum value of @xmath33 , where @xmath34 and @xmath35 are the number of stars ( per unit mass ) with masses smaller than @xmath36 ( corresponding to stars older than @xmath37 gyr ) for the first generation and for the second one , respectively .
figure 2 shows that the smaller the imf slope @xmath31 is ( i.e. , the imf becomes more `` top - heavy '' ) , the larger the number fraction of the second generation ( @xmath33 ) becomes for a fixed @xmath32 of @xmath38 .
the ratio of the final cluster mass to its initial mass ( @xmath39 ) however also becomes smaller for smaller @xmath31 . as a result of this ,
@xmath39 of the models with @xmath40 become smaller than the threshold value ( 0.5 ) below which star clusters can be disintegrated as the result of mass loss ( e.g. , geyer & burkert 2001 ) : no models can be located within the regions with @xmath41 and with @xmath42 where @xmath0 cen can survive possible disintegration and have the bms fraction consistent with observations .
the number fraction of the second generation can not be larger than 0.16 , even for @xmath32 = @xmath43 ( i.e. , extremely bottom - heavy ) and @xmath31 = @xmath44 for which @xmath0 cen can manage to survive disintegration due to mass loss .
thus the results in figure 2 suggest that the observed fraction of the bms can not be explained simply by the agb pollution scenario for reasonable imfs .
if @xmath24 is as low as @xmath45 , @xmath0 cen can not disintegrate owing to mass loss from sne ii for any values of @xmath31 and @xmath32 .
it is found that @xmath33 is about 0.29 for the model with @xmath46 , @xmath47 , and @xmath48 .
this result suggests that if @xmath0 cen was formed with a very unusual imf ( i.e. , with no or little sne ii ) , the observed bms fraction can be reproduced without disintegration of @xmath0 cen . the above model with no or little stellar remnants ( black holes and neutron stars ) of massive stars , however , would not be consistent with the presence of neutron stars in @xmath0 cen ( e.g. , haggard et al .
2004 ) .
the helium mass fraction in the stellar winds of massive stars with @xmath49 can become as high as 0.4 with a maximum of 0.49 for @xmath50 ( schaller et al .
although there is no problem with @xmath6 in this scenario , the mass fraction of massive stars with @xmath49 can not be as large as @xmath51 for reasonable imfs ( @xmath52 ) for which @xmath0 cen does not disintegrate .
figure 3 shows that the mass fraction of the ejecta of the massive stars is at most @xmath53 for a plausible range of imf slopes .
given the fact that the observed star formation efficiency in the galaxy can be at most @xmath54 ( wilking & lada 1983 for the @xmath55 oph cloud ) , this result suggests that the observed fraction of the bms ( @xmath51 ) is unlikely to be explained by this scenario .
a more serious problem for this stellar wind scenario is that there would be no or little difference in the mean metallicity between the first generation of stars and the ejecta of these objects ( schaller et al . 1992 ) : piotto et al .
( 2005 ) demonstrated that there is a metallicity difference by @xmath56[m / h]@xmath57 between the rms and the ( more metal - rich ) bms in @xmath0 cen .
another possible problem of this stellar wind scenario is that the ejecta should be very quickly converted into stars ( within less than @xmath58 yr ) between the stellar wind phases of very massive stars and the onset of sne ii that can blow away the ejecta from @xmath0 cen .
the helium mass fraction ( @xmath6 ) in the ejecta of sne ii stars with @xmath59 for metal - poor stars with @xmath60 ranges from 0.37 to 0.45 .
although this scenario has no problems with @xmath6 , the abundance by mass for heavier elements in the ejecta is too high to be consistent with observations .
@xmath14c and @xmath61ca in ww95 are as high as @xmath62 and @xmath63 , respectively , which are higher than the solar values ( @xmath64 and @xmath65 , respectively ; ww95 ) and thus much higher than those of @xmath0 cen .
this inconsistency with the observed metallicities suggests that the bms can not be formed directly from the ejecta of sne ii of the rms .
if the ejecta of sne ii can be mixed with the primordial gas used for the formation of the rms , the abundance of @xmath14c and @xmath61ca of the bms formed from the mixed gas can be significantly reduced . however , the helium abundance would also be significantly reduced in the process of mixing .
therefore it can be concluded that this scenario is highly unlikely to explain the observed abundance of the bms .
furthermore this scenario has difficulty in explaining the observed fraction of the bms without assuming top - heavy imfs , just as the other two scenarios do .
the present study shows that the proposed three scenarios all fail to explain self - consistently the observed properties of the bms in @xmath0 cen .
the agb scenario appears to be the most promising among the three , though it remains unclear whether this scenario can explain the observed abundances other than helium in @xmath0 cen .
the most serious problem of the agb scenario is that the total mass of the ejecta of more massive agb stars from the rms is too small to be consistent with the observed fraction of the bms of @xmath0 cen . if however we relax the adopted assumption that _ all stars of the bms originate from agb ejecta of the rms initially within @xmath0 cen _ , the problem of the agb scenario can be significantly alleviated .
the problem of the bms fraction can be solved if the original total mass of the rms was larger ( by a factor of @xmath66 ) than the present mass ( i.e. , if @xmath0 cen was a super - giant gc ) and if most of the stars of the rms were removed from @xmath0 cen after their ejecta were used for the formation of the bms .
we here discuss a scenario in which _
@xmath0 cen was the nucleus of a nucleated dwarf galaxy ( de , n ) where the agb ejecta of the central field stellar populations surrounding the ( proto- ) @xmath0 cen were consumed for star formation of the bms . _ in this scenario ,
the ancient host de , n of @xmath0 cen has been completely destroyed by the galactic tidal field so that only its nucleus ( i.e. , @xmath0 cen ) is now observed ( e.g. , bekki & freeman 2003 ) .
not only the ejecta of the rms but also those from stellar populations surrounding the nucleus can be used for the formation of the bms : the above bms fraction problem should not be so serious in this scenario
. the bms can be more metal - rich than the rms if the mean abundance of the field stellar populations and gas in the dwarf is slightly higher than that of the rms .
the key question in this scenario is how much gas ejected from the central field stars can be converted into the bms without being removed by energetic outflow of agb stars in the central regions of @xmath0 cen s host de , n .
recent numerical simulations on transformation from a de , n into @xmath0 cen have suggested that @xmath0 cen s host de , n has @xmath67 mag , corresponding to the total stellar mass of 1.25@xmath68 @xmath69 m@xmath70 ( e.g. , bekki & freeman 1993 ) . if the total mass of the bms is about 30% of present - day @xmath0 cen s mass (= 5.0 @xmath68 @xmath71 m@xmath70 ; meylan et al .
1995 ) , it is necessary for only 1.2% of the stellar mass of the de , n to have been converted into the bms . since the mass fraction of more massive agb stars in the de ,
n ( @xmath72 ) for a reasonable imf can be significantly larger than the above value ( @xmath62 ) , the observed bms fraction is not a problem in this new scenario .
the central escape velocity ( @xmath73 ) of this host embedded in a massive dark matter halo with the total mass of @xmath74 is @xmath75 km s@xmath76 depending on the inner profile of dark matter halo ( bekki 2005 ) .
the derived @xmath73 is significantly larger than the observed maximum wind velocity of @xmath77 km s@xmath76 for agb stars ( e.g. , loup et al . 1993 ) .
therefore , the agb ejecta of the de , n is highly likely to be trapped within the central region of the de , n and consequently converted into new stars ( i.e. , the bms ) .
although extensive numerical simulations on the central gas dynamics of a de , n are necessary to confirm that gas ejected from field stars of the de , n can be transferred into the nuclear region and converted into new stars there , we propose that the bms can originate from gas ejected from stars within the ancient de , n whose nucleus was @xmath0 cen . the local helium abundance of the remaining gas of forming @xmath0 cen might have been significantly higher ( leading to a bms with higher helium abundance ) , if helium sedimentation due to gravitational diffusion ( which was originally proposed by fabian & pringle ( 1977 ) for a mechanism responsible for gaseous abundance gradients in clusters of galaxies ) occurred in @xmath0 cen .
chuzhoy & loeb ( 2004 ) showed that gravitational diffusion in the interstellar medium of giant elliptical galaxies can result in the increase of helium abundance by a factor of @xmath78 , where @xmath79 and @xmath80 are the gas temperature and the suppression factor ( likely to be @xmath81 ) of the gravitational diffusion by the magnetic field .
these results suggest that a significant increase of helium abundance is highly unlikely within gcs , which have the central velocity dispersions of order of 10 km s@xmath76 , corresponding to @xmath82 k.
we have shown that the observed bms fraction of @xmath0 cen can not be simply explained by any heps in which the bms was formed from ejecta of the rms . we accordingly have suggested an `` external pollution '' scenario in which the bms was formed from gas that was initially within the central region of @xmath0 cen s host galaxy .
the question yet to be answered in this scenario is how star formation could proceed within the rms when gas was transferred to the central region .
it would be possible that the bms was formed outside yet close to the rms as a star cluster and then merged with the rms in the central region of @xmath0 cen s host galaxy .
we are grateful to the anonymous referee for valuable comments , which contribute to improve the present paper . k.b . and
j. e. n. acknowledge the financial support of the australian research council throughout the course of this work . | recent observational studies of @xmath0 centauri by _ hubble space telescope _ have discovered a double main sequence in the color magnitude diagrams ( cmds ) of its stellar populations . the stellar population with the blue main sequence ( bms ) is observationally suggested to have a helium abundance much larger , by @xmath1 , than that of the red main sequence ( rms ) . by using somewhat idealized models in which stars of the bms are formed from gas ejected from those of the rms , we quantitatively investigate whether the helium overabundance of the bms can result from self - enrichment from massive agb stars , from mass loss of very massive young stars , or from type ii supernovae within @xmath0 cen .
we show that as long as the helium enrichment is due to ejecta from the rms formed earlier than the bms , none of the above three enrichment scenarios can explain the observed properties of the bms self - consistently for reasonable imfs .
the common , serious problem in all cases is that the observed number fraction of the bms can not be explained without assuming unusually top - heavy imfs .
this failure of the self - enrichment scenarios implies that most of the helium - enriched gas necessary for the formation of the bms originated from other external sources .
we thus suggest a new scenario that most of the second generation of stars ( i.e. , the bms ) in @xmath0 cen could be formed from gas ejected from field stellar populations that surrounded @xmath0 cen when it was a nucleus of an ancient dwarf galaxy . |
the ionization of an acid , ha , in condensed phases @xmath0 is clearly influenced by the nature of the solvent s in which the ionization takes place . a molecular understanding of the processes responsible for these solvent effects requires an analysis of the solvation forces that bring about the ionization and examination of how these forces depend on the properties of the solvent molecules .
the analysis is complicated by the fact that the proton is a quantum object and thus the theoretical description must account for its quantum character in determining the interactions between the proton and the solvent . in this article , we study such acid ionization process in aprotic media .
we have selected two simple prototypic molecular solvents for this study : acetonitrile ( acn ) and dimethylsulfoxide ( dmso ) . while these two molecules have similar dipolar properties they have markedly different effects on the ionization process ;
for example , the dissociation constant for hcl in dmso is several orders of magnitude greater than that in acn.@xcite furthermore , aprotic solvents have been chosen in order to avoid complications due to cooperative proton motion in the solvent ( grotthuss mechanism)@xcite that may play a role when water is the solvent.@xcite the main focus of this work is on the determination of the potential of mean force as a function of the separation between the h@xmath1 and cl@xmath2 ions .
this allows us to characterize the equilibrium solvation structure ; in particular , if solvent separated ion pairs ( ssip)@xcite can exist in either solvent so that the ionization process should be represented by @xmath3 here h@xmath4 cl@xmath2 represents a metastable configuration corresponding to a local minimum in the mean potential where solvent molecules reside between the ions and the ions may not yet be considered to be completely free ions in solution.@xcite our calculations permit us to determine properties of the solvated quantum proton and to study the differences in the solvation properties in both solvents . since the ionization process is an example of a reaction that is strongly solvent influenced , it is also of interest to see how the chemistry of ionization processes is changed in finite - size systems .
consequently , we also examine molecular clusters with linear dimensions in the nanometer range to study how the ionization process changes as a function of the cluster size .
it is well known that chemical reactions in clusters present some unique features due to the lack of translational symmetry ; for example , certain charge transfer reactions leading to neutral ion pairs take place in confined systems only above a certain threshold cluster size.@xcite acid dissociative processes occurring in aqueous clusters are also known to exhibit similar behavior where cooperative effects gradually increase , stabilizing the reaction products as the number of the components in the cluster increases.@xcite consequently , it is interesting to extend bulk studies to cluster domains to unveil peculiarities of the mechanisms that govern dissociative processes in aggregates containing small numbers of particles .
the organization of the paper is as follows : section ii gives details of the interaction potentials used in this work , along with the molecular dynamics methods used to carry out the simulations of the mean potential .
bulk phase and cluster results are contained in secs .
iii and iv , respectively .
the conclusions of this study are given in sec .
the acn [ ch@xmath5cn ] and dmso [ ( ch@xmath5)@xmath6so ] solvents considered in this study were modeled as rigid molecules with the following characteristics : acn was comprised of three sites representing a united atom model for the ch@xmath5 group , c and n units in the linear molecule with separations @xmath7 between sites in a molecule fixed by constraints@xcite at the following values : @xmath7(c - ch@xmath5)=1.46 and @xmath7(n - ch@xmath5)=2.63 .
the molecular dipole moment is @xmath8 .
the molecules interact by site - site lennard - jones ( lj ) and coulomb interactions .
the lj parameters and partial charges were taken from edwards et al.@xcite and interactions between identical sites are given by @xmath9 , @xmath10 and @xmath11 ; @xmath12 , @xmath13 and @xmath14 ; @xmath15 , @xmath16 and @xmath17 . here and below the @xmath18 values
are expressed in , the @xmath19 parameters in k and the partial charges in units of the electronic charge @xmath20 .
the usual arithmetic and geometrical means were used to determine respectively the @xmath18 and @xmath19 parameters for the cross interactions .
dmso was modeled as a rigid four - site molecule with the following internuclear separations between constituents@xcite : @xmath7(s - ch@xmath5)=1.799 , @xmath7(s - o)=1.48 , @xmath7(o - ch@xmath5)=2.657 and @xmath7(ch@xmath5-ch@xmath5)=2.685 .
the molecule is nonlinear with the following angles among the constituents : @xmath21(ch@xmath5-s - ch@xmath5)@xmath22 , @xmath21(ch@xmath5-s - o)@xmath23 and the angle that the os bond makes with the ch@xmath5-s - ch@xmath5 plane is @xmath24 .
the molecular dipole moment is @xmath25 .
the lj parameters and partial charges were taken from rao and singh@xcite and for interactions between the identical sites are given by @xmath26 , @xmath27 and @xmath28 ; @xmath29 , @xmath30 and @xmath31 ; @xmath32 , @xmath33 and @xmath34 .
parameters for cross interactions were determined as described above .
we consider a simple model for the ionization of hcl in the above solvents . in the gas phase the lowest energy potential energy surface corresponds to dissociation into neutral atoms while in solution
, the molecule dissociates into ions due to solvation forces . a full quantum calculation of the potential energy surface appropriate for the ionization in solution must account for solvent degrees of freedom .
we considered a much simpler model .
the interaction potential energy between the h and cl atoms leading to dissociation into ions was determined from a density functional calculation for a cl@xmath2 ion in the field of a point positive charge a fixed distance away . by appropriate distribution of the basis functions describing the electronic density of the complex in the neighborhood of the chlorine atom , the bare
" molecule was artificially constrained to dissociate into cl@xmath2 and h@xmath1 ions at large separations , while the charge density was allowed to redistribute at shorter separations .
the computation was performed at the generalized gradient approximation level@xcite using a flexible basis set for the cl@xmath2.@xcite results of these calculations as a function of the interionic distances @xmath35 were fitted using the following potential energy function @xmath36 where @xmath20 is the electron charge , @xmath37 kcal / mol and @xmath38 @xmath39 .
note that in this functional form the coulomb interaction does not depend on the partial charges on the ions determined from the computed electronic density .
nevertheless , in the interactions between the ions and the solvent sites , the variation of the ionic charge with the distance @xmath35 between the cl@xmath2 and h@xmath1 ions was taken into account . in this way
, the changes in the electronic density of the hcl molecule during the dissociation process was taken into account . > from our density functional calculations , the charge on the cl@xmath2 ion for @xmath40 1 could be represented approximately by the formula @xmath41 with @xmath42 @xmath39 , @xmath43 . in performing the calculations ,
we have made no attempt to reproduce with ( [ vhcl ] ) the actual value of dissociation energy of hcl in @xmath44 or the exact value of the equilibrium distance for the hcl chemical bond .
in fact , our simple approach based on density functional theory was used only as a reasonable interpolative scheme between two well defined electronic structures for reactants and products states : the covalent character of the intramolecular bond at short interatomic distances and the ionic character of the dissociated species . although this is a crude approximation , it captures the importante features of the solvation effects in different environments , especially at the larger separations of interest in this study where the species are well approximated by h@xmath1 and cl@xmath2 ions . using the charge density extracted from this calculation , the interactions between the cl@xmath2 and h@xmath1 ions and the solvent molecules were taken to be site - site lj plus coulomb interactions .
the lj parameters for these ions were taken from data on cl@xmath2 in h@xmath6o given in rossky et al.@xcite using the normal geometric and arithmetic means and the known parameters for spc water .
the values used in the present calculation are @xmath45 and @xmath46 ; @xmath47 , and @xmath48 and , as earlier , the lj parameters for cross interactions between these ions and the molecular solvent sites were determined from geometric and arithmetic means .
the proton was treated quantum mechanically using feynman s path integral formulation of quantum mechanics.@xcite in this representation the proton is represented by a ring polymer with harmonic bonds between the polymer monomers , which also interact with the other classical particles in the system.@xcite the effective potential of the system is @xmath49 where @xmath50 is boltzmann s constant times temperature , @xmath51 represents the proton mass , @xmath52 refers to the set of classical coordinates and @xmath53 to the proton monomer coordinates with @xmath54 the number of monomers @xmath55 . here
@xmath56 is the potential energy of all classical particles and @xmath57 is the interaction energy between the proton monomers and the classical particles .
we have taken @xmath58 in order to accurately represent the proton but the results were checked against simulations with @xmath59 and no significant changes in the equilibrium averages were seen .
equilibrium averages were computed from time averages over a fictitious molecular dynamics ( md ) generated by the following hamiltonian : @xmath60 where @xmath61 is the fictitious mass assigned to the proton monomers .
we have taken @xmath62 atomic units .
given this choice the oscillation period for the harmonic forces in the polymer is @xmath63 using the values given above , at @xmath64=200 k and @xmath64=293 k the period is @xmath65 fs and @xmath66 fs , respectively .
the bulk phase molecular dynamics simulations were carried out on a system composed of 142 solvent molecules plus the two ions confined to a periodic box with sides @xmath67 .
the lj interactions were cut off at half the size of the simulation box .
a quartic spline interpolation was used to make the the coulomb interactions go smoothly to zero at @xmath68 in a 1 region .
the box sizes were @xmath67=23.2 for acn and @xmath67=25.78 for dmso giving densities of 52.2 @xmath69/mol and 71.7 @xmath69/mol , respectively .
the bulk simulations were carried out at constant temperature using nos dynamics.@xcite two nos themostats were used , one for the proton polymer and the other for the remaining classical particles .
the temperatures of both thermostats were fixed at 293 k. the md integration was performed using the verlet algorithm@xcite with a time step of 2 fs .
note that with this time step there are between 50 and 100 time steps per oscillation period of the proton monomers .
the cluster results were obtained from time averages over constant energy molecular dynamics simulations .
the clusters were equilibrated for 50 ps using constant temperature ( nos ) molecular dynamics after this period the thermostat was switched off .
time averages were then determined from 4 - 5 ns constant energy md trajectories .
clusters with sizes ranging from @xmath70 to @xmath71 solvent molecules were studied .
the average temperature was @xmath72 k for the small clusters with much smaller fluctuations for the larger clusters .
our study of acid ionization will be primarily concerned with the change in character of the ionization process , as reflected in the potential of mean force or free energy for varying separation of the ions , as a function of the different solvent environment . in particular
, we will be interested in describing @xmath73 the magnitude of the resulting free energy barriers for the dissociation processes in both solvents and @xmath74 the main features that characterize the solvation structure of the complex as the reactive process takes place .
the mean potential as a function of the distance between the ions was obtained from the mean force on the ion pair.@xcite the proton position was defined as the centroid of the proton polymer,@xcite @xmath75 and thus the distance between the cl@xmath2 and h@xmath1 ions is @xmath76 with @xmath77 the position of the cl@xmath2 ion .
the mean force at a fixed ion pair displacement @xmath35 was calculated as@xcite @xmath78 where @xmath79 is the force exerted by the cl@xmath2 and the solvent particles upon the @xmath80-th proton monomer , @xmath81 represents the versor along the interionic distance and @xmath82 represent time average computed using the constrained reaction coordinate dynamics ensemble.@xcite the mean potential @xmath83 was then found by integration of the mean force using @xmath84 figure 1 shows the potential of mean force for the ion pair in bulk acn and dmso solvents . for reference
the bare " hcl potential in the absence of solvent is also shown .
the potentials have been shifted to correspond at their first minima .
( we term this the contact ion pair state ( cip ) . in this case
there is no distinction between the bound molecular species and the contact ion pair state . )
we note the pronounced differences between the acn and dmso solvation .
for example , at @xmath85 , the maximum in the dmso mean potential , the curves differ by @xmath86 . from these curves
, we can easily extract the ratio of dissociation constants @xmath87 @xmath88
\ \ \left [ \int \ ( 1-\theta_{dmso}(r ) ) \ e^{-\beta w_{dmso}(r ) } \
{ \rm d}{\bf r } \right]^2 } { \left [ \int \ \theta_{dmso}(r ) \
e^{-\beta w_{dmso}(r ) } \ { \rm d}{\bf r } \right ] \ \ \left [ \int \ ( 1-\theta_{acn}(r ) ) \
e^{-\beta w_{acn}(r ) } \
{ \rm d}{\bf r } \right]^2 } \ ] ] where @xmath89 is unity for those @xmath35 values which correspond to configurations of solvent @xmath80 exhibiting associated ion pairs ; for other values @xmath89 is zero .
equation ( [ rough ] ) can be approximated by @xmath90 where we have dropped terms of order one and have assumed that the boundary between associated and dissociated ionic states lies at the first maximum beyond the cip well .
our simulations not only predict the correct experimental trend but also the result is in rough accord with the large experimentally observed solvent effect on the acid ionization@xcite , where @xmath91_{exp } = 6.9 \ 10 ^ 6 $ ] .
finally , we note that the individual dissociation constants for both solvents differ by several orders of magnitude from the corresponding experimental values .
this is expected in view of the simplifications introduced in the design of the potentials .
in fact , differences in a few tenths of a kcal in the potentials lead to considerable variations in calculated dissociation constants . since our emphasis is on the study of relative , quantitative changes in the ionization process as a function of solvent type and environment
, we have made no attempt to tune our simple potential model to describe all quantitative aspects of the equilibrium structure .
the profiles of the potential of mean force presented in fig . 1 show that there are no clearly identifiable solvent separated ion pairs for acn , as signaled by a secondary minimum in the mean potential .
there is only a shallow minimum at @xmath92 giving a barrier between the ssip and cip states of @xmath93 kcal / mol which is within the statistical uncertainty of our calculations .
in contrast , the mean potential for dmso shows a fairly deep secondary minimum at @xmath94 with a barrier separating the cip and ssip states of @xmath95 kcal / mol . in the dmso calculation , the curve was computed from a constrained md simulation as described above .
the filled circles were determined using the formula @xmath96 where @xmath97 is the probability density of @xmath35 and @xmath98 is the uniform distribution .
the probability density @xmath97 was estimated from a histogram of the separation between the ions obtained from a long unconstrained md trajectory initiated near the mean potential secondary minimum at @xmath99 .
note that most of the probability density is confined between @xmath100 with no escape to either contact ion pairs or free ions observed in the course of the @xmath101 ns md simulation .
an idea of the average solvation structure in dmso can be obtained from fig . 2 , where we show a contour plot of the solvent charge density , @xmath102 , for an interionic separation of @xmath103 , which corresponds to ssip separation configuration .
the charge density is defined as @xmath104 where @xmath105 represents the solvent bulk density and the angular brackets @xmath106 denote an ensemble ( or time ) average over configurations with the ion pair separation at a given value of @xmath35 . in ( [ rrhhoo ] ) , we have selected a frame of reference centered on the ion pair , which is taken to lie along the @xmath107 direction in a cylindrical coordinate system @xmath108 ; @xmath109 and @xmath110 represent the cylindrical coordinates and charge of site @xmath111 in the @xmath80-th solvent molecule , respectively . the cl@xmath2 and h@xmath1 ions are denoted by semicircular shaded regions in the bottom of the figure while the @xmath112 and @xmath113 signs indicate regions of high charge density of the corresponding sign .
the picture provides information about both the locations of high solvent density as well as the average orientation of solvent molecules in the vicinity of the ion pair .
figure 2 clearly shows that the solvation structure for dmso is energetically dominated by the electrical coupling between the proton , a bare charge of small size , and the negatively charged oxygen sites of the solvent .
this is indicated by the prominent
@xmath114 " charge density maxima to the right and above the proton ; the molecular dipoles are oriented as might be expected , with the negative ends of the molecules lying near to the h@xmath1 side of the ion pair complex and the positive ends of the other solvent molecules oriented towards the cl@xmath2 side . however , the positive charge density near the cl@xmath2 ion is more diffuse indicating weaker solvation and less charge localization .
the analysis of the solvation structure in acn for a similar ion pair separation shows essentially the same qualitative features as previously described for dmso , so a comparative analysis of the charge density profiles is insufficient to understand the differences in ionic stabilization that these solvents present .
perhaps this is to be expected since the solvent contribution to the potential of mean force is the result of a subtle interplay between packing effects governed by short - range , repulsive forces related to the molecular shape , and solute - solvent dipolar forces which depend on overall distribution of molecular electronic density .
the analysis of the solvent structure in the neighborhood of the transition region between cip and ssip ion pair configurations presents interesting differences for the two solvents .
we examine the region where the steep attractive branch of the mean potential reaches its local maximum and levels off .
consider the dipolar density , @xmath115 in the @xmath116-plane . here
@xmath117 is the dipole moment of molecule @xmath80 and , as above , the average is taken with the distance between the ions fixed at @xmath35 .
figures 3(a ) and ( b ) show @xmath118 for @xmath119 for acn and dmso , respectively . for acn one observes two large peaks with opposite sign in the vicinity of h@xmath1 .
this indicates that the acn molecules have a considerable degree of orientational variability so that when the molecular dipole moment is projected onto the @xmath107 axis of the ion pair the projection can take either sign .
in contrast , the dipolar density for dmso in fig .
3(b ) shows no such effect implying that the dmso molecules are much more rigidly ordered in the first solvation shell as indicated in the schematic representation of the structure shown in the bottom parts of the figures .
the results suggest that in acn the local solvent density fluctuations are sufficiently large to promote dipolar orientations that , in average , lead to a less effective dielectric reactive field , lowering the energetic cost to pass from ssip to cip states .
however , in dmso a much more structured three - dimensional solvent network exists that hinders molecular rotations and leads to a stronger coulombic coupling between the proton and the negatively charged oxygen site and a larger free energy barrier .
compared to bulk environments , dissociation processes in small clusters present some distinctive features that may lead to significant variations in the overall dissociation energy of the ion pair .
perhaps the simplest question to be considered is how large should a cluster be to exhibit bulk - like behavior ?
the answer will depend on the particular stage of the dissociation process one is interested in describing . in this study
we focus our attention on the small - intermediate interionic distance regime ; @xmath120 , those characteristic of cip and ssip configurations seen in bulk phases . for interionic distances which are comparable to typical linear dimensions of the cluster ,
surface forces introduce new effects in the dissociation mechanisms.@xcite in fig .
4 we present the potential of mean force for hcl dissociation in clusters of acn and dmso .
we first observe that the results for the mean potential for an @xmath121 acn cluster , for interionic distances that are not too large , lie close to those for bulk solvents , indicating that even for this relatively small aggregate the solvent reactive field is comparable to that of bulk environments and that surface effects play only a minor role in determining the functional form of @xmath83 .
very small clusters exhibit quite different behavior as is evident from the results for @xmath83 for an @xmath70 acn cluster shown in the figure .
the typical bulk phase plateau regime at moderate interionic distances has disappeared since for such small aggregates there is insufficient space for independent solvation of the individual ions .
figure 4 also shows results for the mean potential for a dmso cluster with @xmath71 at temperature @xmath122 .
our results confirm that solvent separated ion pairs still persist in clusters of this size although the free energy barrier between the cip and ssip states is smaller , @xmath123 kcal / mol .
several interesting features of the dissociation mechanisms in acn which are common to bulk and clusters environments can be seen from typical snapshots of cluster configurations . in fig .
5 we present two often - encountered cluster configurations for acn clusters with @xmath71 where the ion pair separation is fixed at the transition value corresponding to the sharply rising part of the mean potential discussed above .
the acn molecules lying closest to the h@xmath1 ion are rendered in dark colors for contrast while the remaining solvent molecules are lightly shaded .
one sees that while on average four acn molecules strongly solvate the h@xmath1 ion , their orientations may be such that the projection of their dipole moments on the @xmath107 axis can be positive or negative .
in contrast , dmso clusters do not exhibit this dual dipolar orientation .
conformational equilibria between clusters of distinctive geometry occur as we approach the limiting regime where clusters split into two or more independent aggregates.@xcite the interionic distance at which these phenomena are observed dependens on the cluster size . for the very small @xmath70 acn cluster case shown in fig .
6 this occurs for an ion pair separation @xmath124 . at this ion pair separation
one sees the two types of solvent configurations ; namely , ( a ) three molecules strongly solvating the h@xmath1 ion with the fourth solvent molecule lying farther away and solvating the cl@xmath2 ion .
the other prominent configuration ( b ) occurs when all four solvent molecules strongly solvate the h@xmath1 ion .
this is a transitional distance : for smaller separations one always sees the solvation structure given by ( a ) while for larger separations one always sees that corresponding to ( b ) .
if one considers larger clusters analogous phenomena exist , however here we have to consider larger ion pair separations . as an example , fig .
7 shows results for a @xmath121 acn cluster with the ion pair separation fixed at the very large value @xmath125 .
the equilibrium structure is determined by configurations ( a ) where varying numbers of solvent molecules solvate the h@xmath1 and cl@xmath2 ions but these two solvated ions form subclusters which only weakly interact .
however , there also exist equilibrium configurations ( b ) where long strings " of solvent molecules form a link or rubber band " connecting the two ions leading to a significant attractive force .
this is a distinct cluster effect which has no counterpart in the bulk .
such structures were observed earlier for ions of like charge in water.@xcite however , we remark that in the regime of ion pair separations investigated there is no hysteresis in the mean potential indicating that the equilibrium configuration space is being sampled fully even for the rubber band " configurations shown in fig .
two examples of cluster structure with the ion pair at @xmath103 , the ssip configuration for dmso , are shown in fig .
8 , one for acn ( top ) and the other for dmso ( bottom ) .
the noteworthy features of both panels are the obvious strong orientational order near h@xmath1 for both solvent species .
however note that although there is little discernible difference between the two solvent cases , ssips exist for clusters of dmso but do not for acn .
these two pictures further illustrate the fact that the cluster shape fluctuations are strong and these fluctuations must be taken into account in any model of the cluster solvation dynamics .
furthermore , one sees that surface forces do manifest themselves in the orientational order of the solvent molecules on the cluster surface which tend to lie parallel to the surface
. however , as observed earlier , even for fairly small clusters ( @xmath121 ) there are only rather small quantitative effects on the mean potential and such surface force effects become important only at longer ion pair separations or for smaller cluster sizes .
the calculations of the ionization of hcl in acn and dmso solvents indicated some of the solvation features that make the ionization process rather different in these solvents in spite of their similar dipole moments . the differences could be ascribed to the details of the charge distributions in the molecules and their effects on the quantitative aspects of the solvation structure .
the gross features of the solvation structure were shown to be similar for both solvents . however , in spite of the apparent similarities in the local solvation structure , dmso supports solvent separated ion pairs while the tendency to form such pairs is considerably reduced or absent in acn .
the cluster environment has an important influence on the ionization process , especially for clusters with fewer than @xmath121 molecules . for ion pair separations as large as that for the ssip state
, the cluster potential of mean force closely corresponds to that of the bulk , even for clusters with as few as @xmath121 solvent molecules . for much smaller clusters ,
as might be expected , there are significant changes in the mean potential reflecting the finite size of the cluster environment . of course , as the ion pair is stretched the finite size of the cluster gives rise to much more dramatic effects on the mean potential as reflected in the rubber band "- like solvent configurations that lead to attractive forces between the ions that can not exist in the bulk .
since these forces correspond to an equilibrium cluster and involve averages over all cluster configurations with the ion pair separation fixed , they may have little relevance for ion pair dissociation processes in clusters since the time scale to establish such equilibrium structures may be long compared to typical times for dissociation .
it would be interesting to investigate the dynamics of dissociation in clusters in this context . in order to estimate the importance of the explicit incorporation of the quantum nature of the proton in our calculations
, we performed a series of simulations in an @xmath121 acn cluster where the proton was treated as a classical point charge .
the results for the computed potential of mean force are included in fig .
4 ; they clearly show the necessity to treat the proton quantum mechanically in order to accurately represent the activation free energy for ionization in these solvents .
classical mechanics underestimates the barrier for passage from the cip to ssip or free ion states . in this respect ,
the present behavior is opposite to that found in proton transfer reactions where quantum tunneling reduces the height of the effective free energy barrier.@xcite quantum dispersion tends to delocalize the proton charge leading to a less effective ionic solvation , as well as smaller solvation energies for dissociated ions in comparison to results from classical models where the proton is taken to be a localized point charge .
the simulations presented in this paper show that acid ionization in molecular clusters can mimic that in the bulk even for quite small clusters .
however , for clusters below a certain size , about @xmath121 in this study , distinctive finite - size effects influence the ionization .
the study points to some of the new chemistry that can occur as a result of environmental changes on the microscopic and mesoscopic levels .
d.l . is a recipient of a nserc foreign researcher award .
the research of r.k . was supported in part by a grant from the natural sciences and engineering research council of canada .
this research was also partially supported by the eec network contract erbchrxct930351 `` molecular dynamics and monte carlo simulations of quantum and classical systems '' .
we finally thank m. ferrario for stimulating discussions . c. cooke , c. mccallum , a.d .
pethybridge and j.e .
prue , electrochim .
acta , * 20 * , 591 ( 1975 ) .
kolthodd , s. bruchenstein and m.k .
chantooni , jr .
soc . * 83 * , 3927 ( 1961 ) . for
a discussion of this mechanism see , for example , s. glasstone , _ physical chemistry _ , ( d. van nostrand company , new york , 1959 ) .
simulations of ssip configurations for classical ions in various model solvents have been carried out .
see , for instance , o.a .
karim and j.a .
mccammon , j. am .
soc . * 108 * , 1762 ( 1986 ) ; chem .
. lett . * 132 * , 219 ( 1986 ) ; g. ciccotti , m. ferrario , j.t . hynes and r. kapral , chem . phys . *
129 * , 241 ( 1989 ) ; j. chem
* 93 * , 7137 ( 1990 ) . | acid ionization in aprotic media is studied using molecular dynamics techniques . in particular ,
models for hcl ionization in acetonitrile and dimethylsulfoxide are investigated .
the proton is treated quantum mechanically using feynman path integral methods and the remaining molecules are treated classically .
quantum effects are shown to be essential for the proper treatment of the ionization .
the potential of mean force is computed as a function of the ion pair separation and the local solvent structure is examined .
the computed dissociation constants in both solvents differ by several orders of magnitude which are in reasonable agreement with experimental results .
solvent separated ion pairs are found to exist in dimethylsulfoxide but not in acetonitrile .
dissociation mechanisms in small clusters are also investigated .
solvent separated ion pairs persist even in aggregates composed of rather few molecules , for instance , as few as thirty molecules . for smaller clusters or for large ion pair separations cluster finite - size effects come into play in a significant fashion .
2true cm |
one of the recurring questions appearing in the study of hadronic resonances is their internal structure @xcite .
the simple picture of mesons and baryons being @xmath3 and @xmath4 , respectively , has given rise to more complicated structures in many cases , with light scalar mesons widely accepted to be some kind of molecular states stemming from the interaction of pseudoscalar mesons @xcite , or the case of the two @xmath5 , widely accepted as composite states of @xmath6 and @xmath7 @xcite , among many others @xcite .
one of the pioneer works to determine whether states are composite , or more of the elementary type , is the one of weinberg , determining that the deuteron is a simple bound state of a proton and a neutron , which gets bound by an interacting potential @xcite .
the method has been used to determine that some resonances are not elementary , like the @xmath8 and the @xmath9 @xcite .
it relies basically upon determining from experiment the coupling of a state to its assumed components and then making the test of compositeness . in our language
@xmath10 is the condition to have a composite state @xcite .
the coupling can be obtained from known scattering amplitudes of the components . in most cases
this is difficult and one has only access to certain resonances through decay of heavier ones .
this is for instance the decay @xmath11 .
the @xmath12 is dynamically generated from the interaction of @xmath2 as a single channel in the chiral unitary approach @xcite .
the inclusion of higher order terms in the lagrangian barely makes any change in this resonance @xcite and then reactions producing this resonance become a good testing ground for the method that we propose . essentially , the method compares two quantities : the production of the resonance , irrelevant of its decay , and the strength of the invariant mass distribution for the production of the assumed molecular components of the resonance .
we show that the ratio of these two magnitudes , removing the phase space factors provides a useful information from which the compositeness of the resonances can be determined . in the present paper
we develop the formalism and apply it to the case of @xmath13 .
we show that the method is rather stable with respect to fair changes of some parameters and that one can determine the compositeness with some precision . experimental information on the rate for this reaction is available from ref .
@xcite . however , the second part of the information needed in the test , the @xmath14 invariant mass distribution in @xmath15 , is not yet available .
the idea exposed in this paper should provide an incentive for measuring such reaction .
in fig . [ fig : fig1 ] the @xmath16 decay at microscopical level is depicted @xcite . component . ] in order to hadronize the final @xmath17 component we must bear in mind the molecular structure of the @xmath12 in our picture .
the state is a @xmath18 . with
the prescription followed here for @xmath19-parity , such that @xmath20 , with @xmath21 a vector meson , the representation for the @xmath12 is @xmath22 let us see how this state emerges from the mechanism of fig .
[ fig : fig1]b ) . in the hadronization
we will include a @xmath23 pair with the quantum numbers of the vacuum , @xmath24 , and will obtain two pairs of @xmath25 .
one of them will correspond to a vector meson and the other one to a pseudoscalar . in terms of pseudoscalars and vectors
the @xmath25 matrix @xmath26 , @xmath27 can be written as @xmath28 which implements the standard @xmath29 mixing @xcite , and @xmath30 the matrix @xmath26 obeys @xmath31 hence , a hadronized @xmath25 component can be written in terms of matrix elements of @xmath32 and then as matrix elements of @xmath33 or @xmath34 .
it is easy to see that the combination @xmath35 has @xmath19 parity @xmath36 and @xmath37 respectively .
the @xmath38 state ( up to normalization ) , will be @xmath39 we can see that the @xmath40 term cancels because @xmath41 .
the combination in eq .
( [ eq : qq ] ) is then the same one as in eq .
( [ eq : f1 ] ) and it is also projected over @xmath42 because it comes from @xmath17 after hadronization with @xmath25 with the quantum numbers of the vacuum .
in this section we write the formalism for the @xmath16 decay , irrelevant on how the @xmath12 decays later .
diagrammatically it is represented in fig .
[ fig : fig2 ] , in the process @xmath44 . ]
the idea is that being the @xmath12 a molecule , to form it we must produce the @xmath45 components which merge into the resonance .
this process , evaluated at the bound state energy @xmath46 is given by @xmath47 where @xmath48 is the loop function of the intermediate propagator of @xmath49 and @xmath50 and @xmath51 stands for the coupling of the @xmath12 to the @xmath52 component of eq .
( [ eq : f1 ] ) with that normalization .
recall that @xmath53 is the residue at the pole . in the following we drop the index of @xmath54 .
@xmath55 in eq .
( [ eq : t ] ) factorizes the weak and the hadronization processes in the relatively narrow region of energies from the mass of the @xmath12 to about @xmath56 mev above the @xmath57 threshold .
@xmath55 is unknown , and in some works it is given in terms of form factors @xcite , which ultimately are parametrized to some data .
we assume @xmath55 constant in the range of energies that we study , which finds support in the works of @xcite .
our strategy is to cancel the factor in some ratios for which we can make predictions with no free parameters .
note also that direct resonance production without a @xmath45 intermediate state is possible .
the relevance of this contribution is discussed in sec .
the @xmath12 is bound by about @xmath59 mev with respect to the @xmath57 threshold , hence by looking at the @xmath57 production we shall not see the peak , but just the tail of the resonance .
diagrammatically , the process proceeds as in fig .
[ fig : fig3 ] . , @xmath60 production .
a ) tree level , b ) rescattering . here
the @xmath21 and @xmath61 represent vector and pseudoscalar meson , respectively . ]
note that for resonance creation shown in fig .
[ fig : fig2 ] , the intermediate @xmath62 states merges into the @xmath12 resonance . in contrast , the @xmath62 final state of eq .
( [ eq : f1 ] ) can also be produced at tree level ( fig .
[ fig : fig3 ] a ) ) , apart from the rescattering mechanism ( fig . [
fig : fig3 ] b ) ) .
analytically , we have @xmath63 where @xmath57 refers to the combination of eq .
( [ eq : f1 ] ) and @xmath64 stands for the scattering matrix of the normalized state of eq .
( [ eq : f1 ] ) . in practice , assuming one knows that the positive @xmath19-parity state is produced , an experimentalist will measure any of the four components of eq .
( [ eq : f1 ] ) , each of which has a probability @xmath65 with respect to the production of that normalized combination .
the sum of the four components would give the same result as the production of the normalized component of eq .
( [ eq : f1 ] ) that we are calculating .
the idea is to construct a ratio of the two rates of the mechanisms of figs .
[ fig : fig2 ] and [ fig : fig3 ] in order to cancel the unknown factor @xmath55 and make predictions which are tied to the nature of this resonance .
the decay rate for @xmath16 , which is known experimentally @xcite , is given in terms of @xmath66 of eq .
( [ eq : t ] ) as @xmath67 where @xmath68 is the momentum of the @xmath69 in the rest frame of the @xmath70 , calculated at the pole of the @xmath12 .
on the other hand , we obtain the invariant mass distribution for the @xmath71 decay by means of @xmath72 where @xmath73 is now the @xmath69 momentum for the @xmath74 .
the @xmath57 or @xmath60 states in eq .
( [ eq : rat ] ) have an invariant mass @xmath75 , and @xmath76 is the @xmath49 momentum in the @xmath57 rest frame . following the suggestion of ref .
@xcite , we define a dimensionless quantity , where we have also removed the phase space factors in @xmath77 , @xmath78
to illustrate the method to determine compositeness , an explicit model for the @xmath12 is studied .
we follow here the chiral unitary approach of ref .
@xcite , based on the chiral lagrangian @xcite .
this chiral lagrangian is readily obtained using the local hidden gauge approach @xcite , exchanging vector mesons between the @xmath49 and the @xmath79 and neglecting the square of the transfered four - momentum versus the square of the vector - meson mass .
higher - order terms are considered in ref . @xcite but the changes in this channel are minimal compared to the lowest order .
the @xmath80-wave potential obtained in ref .
@xcite is of the type @xmath81 , where @xmath82 are the polarization vectors of the initial and final vectors and @xmath21 is given by @xmath83\label{eq : rocap}\end{aligned}\ ] ] where @xmath84 , @xmath85 and @xmath86 , and @xmath87 should be identified with @xmath88 of eq .
( [ eq : rat ] ) . the scattering matrix , of the type @xmath89 , is given by @xmath90^{-1}v=\left[v^{-1}-g\right]^{-1}\ .\label{eq : bethe}\end{aligned}\ ] ] the potential of eq .
( [ eq : rocap ] ) is attractive and leads to a bound state .
we then use a cutoff method to regularize the @xmath91 function which becomes @xmath92 where @xmath93 , and @xmath94 stands for @xmath95 .
we then fix @xmath96 in order to get the bound state at @xmath97 mev , obtaining @xmath98 mev , which is a value of natural size .
we next calculate @xmath54 given by @xmath99 where lhpital s rule has been applied in the last step . we thus have all the elements to evaluate @xmath100 of eq .
( [ eq : rat2 ] ) and show the results in fig .
[ fig : fig4 ] . from eq .
( [ eq : rat2 ] ) as a function of @xmath75 . ]
note that @xmath101 decreases with increasing invariant mass near threshold .
should we have evaluated @xmath102 it would have an accumulation of strength around threshold , due to the existence of the resonance below threshold .
however , in @xmath101 we have divided by the phase space factors @xmath103 , and then , up to some factors , what we see in fig . [
fig : fig4 ] is essentially @xmath104 hence , this quantity has two important elements .
its shape reveals that there is a resonance below threshold .
its strength is related to the coupling of the resonance to the @xmath105 state .
the coupling contains information concerning the nature of the @xmath12 resonance , through the method which we briefly discuss below .
in a reknown paper @xcite weinberg studied @xmath106 scattering and determined that the deuteron was not an `` elementary '' particle but a `` composite '' one made from a neutron and a proton interacting through the potential responsible for the scattering properties of @xmath106 at small energies .
one could envisage an opposite extreme in which the deuteron could be a compact object of six quarks with practically no coupling to @xmath106 , which we would accept as `` elementary '' .
the coupling of the deuteron to the @xmath106 component is what describes the amount of compositeness , and the method of weinberg determines that quantitatively for lightly bound systems in @xmath87-wave .
application of the method to different cases is done in @xcite .
the method is generalized to the case of more bound states and coupled channels in @xcite , to resonances in @xcite and to other partial waves in @xcite .
different derivations and reformulations are given in a series of papers @xcite .
in the present case with one channel and one bound state we can reformulate the issue in a very simple way .
let us start from eq .
( [ eq : g2 ] ) .
we find @xmath107 according to refs .
@xcite the first term in eq .
( [ eq : com ] ) , @xmath108 , measures the compositeness of the state ( in this case as made of ( @xmath109 ) , while the second term @xmath110 is the amount of non - composite nature , which is often called elementariness @xcite , but , as discussed in ref .
@xcite , it also accounts for other composite channels not explicitly included in the basis that one has chosen to describe the state . indeed , in ref .
@xcite it is explicitly shown how , starting from two channels and energy independent potentials , one can eliminate one channel and still describe the other one by means of an effective potential , which , however , becomes energy dependent .
the second term in eq .
( [ eq : com ] ) , @xmath111 , gives in this case the probability of the channel that we have eliminated . in any case
, our aim here is to extract , from an experimental measurement of @xmath112 , the probability to have @xmath105 in the wave function of the @xmath12 , which will be given by @xmath113 . since in @xmath101
there is information about @xmath114 ( see eq .
( [ eq : coup ] ) ) , it is clear that one can get information on @xmath115 , or @xmath116 from @xmath117 .
how to do that without the need to know the explicit form of the potential @xmath21 and the value of the cutoff to regularize @xmath91 , is shown in the next section .
we will use two forms of potentials to estimate systematic uncertainties .
the potential of eq . ( [ eq : rocap ] ) , by an expansion around the pole , becomes @xmath119 up to linear terms in @xmath87 .
the condition that eq .
( [ eq : bethe ] ) has a pole at @xmath120 can be written as @xmath121 then , @xmath122 and from eq .
( [ eq : lpot ] ) we have @xmath123 eq .
( [ eq : com ] ) is now rewritten as @xmath124 where the first term is @xmath115 and the second one @xmath116 , which implies @xmath125 negative for physical solutions .
our aim now is to write @xmath126 in terms of @xmath116 and quantities which can be calculated without knowing the potential and the regulator of @xmath91 , such that from the measurement of @xmath127 one can determine @xmath116 without the need of a model to interpret it . for this
we write @xmath128 in terms of @xmath116 and calculable quantities .
the first step is to eliminate @xmath125 of eq .
( [ eq : lpot ] ) in terms of @xmath116 . for that purpose ,
recall from eq .
( [ eq : com ] ) that @xmath129 from where @xmath130 the latter equation allows to obtain @xmath125 in terms of @xmath91 and @xmath131 , but this is not of much help since @xmath91 needs an unknown regulator . yet ,
the factor that we want for @xmath132 in eq .
( [ eq : rat2 ] ) can be rewritten taking into account that @xmath133 and we can write @xmath134 this equation is most appropiate because @xmath91 is logarithmically divergent and must be regularized , but @xmath131 is convergent .
we can then use values of @xmath96 around @xmath135 gev and see the stability of the results to make a claim of weak model dependence .
we also have the term @xmath136 , which is again convergent , and @xmath137 goes to unity as @xmath138 , which means that @xmath137 is smoothly dependent on the cutoff and multiplies a term in eq .
( [ eq : rat3 ] ) which should be small compared to @xmath136 , certainly for small values of @xmath116 . in any case , we shall test the stability of the results by changing @xmath96 in a reasonable range of natural values . in order to take into account possible `` elementary '' components , often an analysis using a cdd pole @xcite
is performed @xcite .
assume that the potential is of the type @xmath139 where @xmath140 accounts for a `` bare '' pole of a possible elementary component .
the condition of a pole at @xmath46 implies @xmath141 where we have used that @xmath142 at the pole .
we can write from eqs .
( [ eq:20 ] ) , ( [ eq:21 ] ) and ( [ eq : vcdd ] ) , @xmath143 and the sum rule of eq .
( [ eq : com2 ] ) becomes @xmath144 where the first term stands for @xmath115 and the second one for z , where now @xmath145 will be positive for physical solutions . eliminating @xmath145 in terms of @xmath116 and proceeding like in the former subsection , we obtain compared to eq .
( [ eq : rat3 ] ) , eq .
( [ eq : rat4 ] ) has the extra factor @xmath146 in the last term of the denominator .
as far as this term is relatively smaller than @xmath136 and @xmath140 is relatively far from threshold , this extra term will have no much relevance and we can get about the same value of @xmath116 as with the other method from an analysis of the experimental data .
the value of @xmath140 is in principle unknown in the analysis .
one can estimate systematic uncertainties from varying its value .
sometimes , one has information about the range of possible values as we discuss in the next section . and finally it should be noted that when @xmath147 we recover the results of eq .
( [ eq : rat3 ] ) . certainly , if @xmath116 is close to zero , this second term is also very small and we expect the same result from eq .
( [ eq : rat3 ] ) and ( [ eq : rat4 ] ) . in the next section
we study the sensitivity of the results to the used method and to the cutoff @xmath96 .
in all figures the @xmath12 is fixed at @xmath148 . in fig .
[ fig : fig5 ] we show @xmath149 for different values of @xmath116 as a function of @xmath75 , using the linear potential of eq .
( [ eq : lpot ] ) and eq . ( [ eq : rat3 ] ) .
we test @xmath150 and @xmath151 mev , a wide range of values of natural size . recall that the value of @xmath96 used in the chiral unitary approach of section [ sec : chroca ] was @xmath98 mev , which is within the chosen range .
we observe that the results for @xmath152 barely depend on the value of @xmath96 , as we anticipated , in view of the fact that @xmath153 in eq .
( [ eq : rat3 ] ) only depends on @xmath131 and @xmath136 , which are both convergent . with the potential from eq .
( [ eq : lpot ] ) as a function of @xmath75 for several @xmath116 values and different values of @xmath150 and @xmath151 mev . ] with the potential from eq .
( [ eq : vcdd ] ) as a function of @xmath75 for several @xmath116 values and different values of @xmath150 and @xmath151 mev . ]
we can see that the dispersion of the results when changing the cutoff is a bit bigger for @xmath154 , but even then the band of values is narrow enough such that we can differentiate between @xmath155 and @xmath156 , and certainly between @xmath155 and @xmath157 .
note that for @xmath155 the value of @xmath149 at threshold is about @xmath158 while for @xmath157 it is about @xmath159 .
such difference is clearly visible in an experiment with present statistics .
the ratio @xmath160 is bigger for larger compositeness @xmath115 .
this is expected from its definition given in eq .
( [ eq : rat2 ] ) , i.e. , as the ratio between the resonance decay into its dynamical @xmath161 components over its decay into any final state . fig .
[ fig : fig7 ] shows @xmath162 evaluated with the cdd potential of eq .
( [ eq : vcdd ] ) .
we see again that the dispersion of the results by varying @xmath96 is small and similar to the previously discussed method . for states which are mostly molecular
, @xmath163 is far away from threshold .
we have chosen @xmath164 mev .
the choice is based on findings in the study of lattice results of the @xmath165 system by means of an auxiliary potential of the type of eq .
( [ eq : vcdd ] ) , which demanded @xmath163 to be more than @xmath166 mev above the @xmath165 threshold @xcite . the results that we obtain ,
using values of @xmath163 bigger than @xmath167 mev , barely change the results that we have shown .
smaller values of the mass excess bring changes in the upper part of @xmath75 in the plot but not very close to the threshold . by comparing figs .
[ fig : fig5 ] and [ fig : fig7 ] we observe the following features . for @xmath155 ,
both methods are identical .
for @xmath154 the results with both methods change a bit , with the differences bigger than with the change of the cutoff .
the differences become bigger as @xmath116 increases . yet , for @xmath156 and close to threshold the results vary from @xmath168 to @xmath169 , an @xmath170 % change , or @xmath171 % from the average of the two methods . for @xmath157
the differences are bigger , from @xmath172 to @xmath173 or @xmath174 % from the average . from this study and the value of @xmath175 at threshold
we find that the uncertainties of @xmath116 are of the order of @xmath0 .
so far , we have only discussed the ratio close to threshold .
however , the invariant mass dependence is also useful to determine @xmath116 .
the method of the cdd pole is more general since it has one extra parameter , @xmath140 .
in particular for larger values of @xmath140 one would encounter the potential of eq .
( [ eq : lpot ] ) and conclude that a cdd pole is unnecesary .
one could make a fit to the spectrum and determine @xmath116 and @xmath140 , from where one gets extra information about the nature of the resonance .
it is interesting to quote here what we get for @xmath116 using the chiral unitary approach of section [ sec : chroca ] . by means of the coupling
obtained there and using the cutoff of @xmath176 mev , demanded to fix the pole at @xmath177 mev , we obtain @xmath178 , very similar to the value @xmath179 in ref .
@xcite . in this case
the distribution of @xmath175 is shown in fig .
[ fig : fig4 ] and is undistinguible from what we obtain by using the linear potential and eq .
( [ eq : rat3 ] ) with this value of @xmath116 . as to the meaning of this value , stemming from the energy dependence of the potential of eq .
( [ eq : rocap ] ) , it is unclear whether it reflects missing meson - meson channels or some possible elementary component .
what matters for the present work is the accuracy of the method presented here to determine @xmath116 from the @xmath175 magnitude .
the idea to divide @xmath180 by the phase space to see the shape of the resonance has also being exploited experimentally in ref .
@xcite in the @xmath181 reaction to see the tail of the @xmath8 resonance from the @xmath182 spectrum .
see also the related theoretical paper of ref .
so far , we have assumed that the resonance formation and the @xmath57 formation proceed via the mechanism of figs .
[ fig : fig2 ] and [ fig : fig3 ] . yet ,
if we have a non-@xmath161 component we can also think of producing it directly in the weak decay .
this means that we must include the new mechanisms depicted in fig .
[ fig : fig9 ] .
this follows the line of argumentation of ref .
@xcite .
production form the `` elementary '' components . ] in fig .
[ fig : fig9 ] we have two unknowns , the coupling of @xmath70 to @xmath69 plus the `` elementary '' component , and the coupling of the resonance to this component .
we choose to parametrize this contribution as a fraction , @xmath183 , of the former one , such that the new mechanism for the coalescence will give a contribution @xmath184 as a consequence , the contribution to the @xmath57 production process of fig .
[ fig : fig9]b ) will be @xmath185 where in the last step we have used the pole approximation to the amplitude , @xmath186 .
hence , the numerator in eqs .
( [ eq : rat3 ] ) , ( [ eq : rat4 ] ) is changed to @xmath187 while the denominator is changed as @xmath188 thus the factor of eq .
( [ eq : rat3 ] ) or eq .
( [ eq : rat4 ] ) becomes now @xmath189 with @xmath190 and @xmath191 it might look that eq . ( [ eq : rat5 ] ) does not have the nice property of eq .
( [ eq : rat4 ] ) with respect to the cutoff invariance . actually , following the same steps that led to this latter equation one can readily see that the new expression in eq .
( [ eq : rat5 ] ) can be recast like eq .
( [ eq : rat4 ] ) simply replacing the @xmath135 in the numerator of the last factor by @xmath192 now @xmath193 depends on the cutoff , but we can reunify @xmath194 and then @xmath195 is a soft function of the cutoff
. furthermore , the second term in the bracket multiplying @xmath183 in eq .
( [ eq : del1 ] ) is reasonably smaller than the unity preceding it .
all this guarantees a smooth cutoff dependence of the new term .
but more important , changes induced by changes in the cutoff in eq .
( [ eq : rat5 ] ) close to threshold can be incorporated by changing @xmath183 , and since @xmath183 is unknown and will be changed to see the stability of the results , by performing this test we implicitly accommodate the cutoff dependence of the results . in fig .
[ fig : delz ] we show the results obtained by using @xmath196 and @xmath197 fractions .
note that @xmath198 for @xmath199 already introduces a @xmath200 % increase in the rate for the coalescence process .
since the idea is to apply the present method for cases where we have large molecular components , such values of @xmath183 are reasonable .
then , in order to quantify uncertainties from this new mechanism , we evaluate again @xmath160 including the new corrections .
we do the exercise using the cdd version of the potential . from eq .
( [ eq : rat5 ] ) as a function of @xmath75 for several @xmath116 and different values of the cutoff , @xmath150 and @xmath151 mev .
@xmath183 is fixed to @xmath201 . ] in fig .
[ fig : delz ] we see the results for @xmath175 for @xmath202 and @xmath157 and different values of @xmath183 ( @xmath98 is used ) the result coincides with fig . [
fig : fig7 ] . ] .
we see small changes by varying @xmath183 . for @xmath156
the changes are moderate .
they are a bit bigger for @xmath203 and even bigger for @xmath157 .
but even for @xmath157 the results at threshold from @xmath204 to @xmath205 vary only by about @xmath206% .
the reason for this stability is that the ratio with the new mechanisms alone is also of the type of @xmath207 as in eq .
( [ eq : coup ] ) , so the mechanisms are practically proportional and the ratio is maintained . to observe the uncertainties in the cutoff , in fig .
[ fig : delz2 ] we show the ratio from eq .
( [ eq : rat5 ] ) for several @xmath116 and different values of the cutoff , @xmath150 and @xmath151 mev .
@xmath183 is fixed to @xmath201 .
we can see that the curves barely move as commented above .
altogether , and summing different sources of uncertainties in quadratures , the conclusions that we drew before hold also now and one can obtain with this method the value of @xmath116 with uncertainties of about @xmath208 .
let us note that through the derivations done we have always assumed that we have the @xmath38 state of @xmath2 in the final state . actually , the @xmath209 combination is also possible .
the test conducted demands that we isolate the positive @xmath19-parity in the final state .
this is in principle possible experimentally as shown in refs .
certainly it would help if the process is dominated by the @xmath12 .
nevertheless , the study carried here is quite general and can be applied to multiple cases where one suspects that one resonance has large components of a composite system .
we propose a new method to determine the amount of compositeness of resonances mostly made from two hadron components , and bound with respect to these components .
the method relies on the comparison of two independent but related quantities . on the one hand ,
one measures the production rate of the resonance in some reaction , independently of the decay channel of the resonance . on the other hand ,
the mass distribution of the two components close to threshold is measured , which appears above the resonance energy since the resonance is supposed to be a bound state of these components .
the resonance tail appears as an enhancement in the invariant mass distribution close to threshold , easily distinguishable from a pure phase space distribution .
the method consists of taking the ratio of this mass distribution over the rate for the resonance production in the same reaction , also removing the phase space factors .
we show that the strength is related to the coupling of the resonance to the two hadron component , and the shape to the square of the resonant amplitude .
the ratio decreases as a function of the invariant mass at threshold .
then , a method is used to relate the measured ratio to the compositeness of the resonance .
we show that it is possible to determine the compositeness , 1-z , from that ratio .
systematic uncertainties are estimated to conclude that it is possible to measure values of z with about 0.1 of uncertainty .
we used a particular case for the numerical evaluations in the decay of @xmath11 which has a large composite component of @xmath2 . yet , the method is general and can be applied to any case where one has hints , or theoretical backing , for the molecular nature of a resonance . in view of this , we encourage the simultaneous measurement of these quantities , that for the moment appear in different reactions , or are measured by different groups , and usually are given in terms of counts and not absolute measurements , as is demanded by the method proposed .
this research has been supported by the spanish ministerio de economa y competitividad ( mineco ) , european feder funds under the contracts fis2014 - 51948-c2 - 2-p and sev-2014 - 0398 , by generalitat valenciana under contract prometeoii/2014/0068 . also , this work is supported by the national science foundation ( career grant no
. 1452055 , pif grant no .
1415459 ) , and by gwu ( startup grant ) .
a. martnez torres , e. oset , s. prelovsek and a. ramos , jhep * 1505 * , 153 ( 2015 ) l. s. geng , x. l. ren , y. zhou , h. x. chen and e. oset , phys .
d * 92 * , no . 1 , 014029 ( 2015 ) p.
del amo sanchez _ et al .
_ [ babar collaboration ] , phys .
d * 83 * , 052001 ( 2011 ) j. m. dias , f. s. navarra , m. nielsen and e. oset , arxiv:1601.04635 [ hep - ph ] .
z. f. sun , m. bayar , p. fernandez - soler and e. oset , phys .
d * 93 * , no .
5 , 054028 ( 2016 ) | we develop a method to measure the amount of compositeness of a resonance , mostly made as a bound state of two hadrons , by simultaneously measuring the rate of production of the resonance and the mass distribution of the two hadrons close to threshold . by using different methods of analysis
we conclude that the method allows one to extract the value of 1-z with about @xmath0 of uncertainty .
the method is applied to the case of the @xmath1 decay , by looking at the resonance production and the mass distribution of @xmath2 . |
the parity - violating asymmetries of slac e158 and sld experiments have been interpreted as the asymmetries between opposite helicities for the standard model @xcite using the chiral projection , approximated from the spin projection operator .
helicity @xmath0 is defined by the momentum of a fermion and its spin orientation ; if the spin orientation is in the same direction as its momentum , it is called right - handed helicity(@xmath1 ) . and
helicity can be observed to be opposite depending on the reference frame and is especially indefinite in the rest frame where momentum is zero .
for example , a right - handed helicity massive fermion @xmath2 in one reference frame can be observed at the same time as @xmath3 with left - handed helicity by observers in other reference frames .
chirality ( l , r ) is defined to indicate either of the two - component objects in a massive fermion field , and no physical measurement is available for the chirality of massive fermions and only when fermions are massless chirality becomes helicity .
after a review of spin polarization , we investigate the validity of the chiral projection as an approximation of the spin polarization operator in the practical calculation of matrix elements @xcite .
spin polarization is defined by the direction of the spin relative to a given coordinate system , whereas helicity is relative to the momentum direction of the particle .
the spin polarization vector @xmath4 can be derived from the spin projection operator @xmath5 for an electron with @xmath6 being the spin direction at rest : @xmath7 where the spin polarization 4-vector @xmath4 is @xmath8 satisfying that @xmath9 where it is normalized to one particle per unit volume in the rest frame by the normalization factor @xmath10 @xcite .
for example , the spin projection operator for an electron with the spin direction @xmath11 and momentum @xmath12 is : @xmath13 and the spin polarization vector is : @xmath14 in general , the spin projection operators of a fermion with @xmath15 can be given by : @xmath16 note that the location of the non - zero matrix elements is determined by the spin direction from the product of the two - component spinors @xmath17 .
the approximation of the spin projection operator @xmath5 should be performed after the full evaluation of the matrix element , taking into account the normalization factor .
for example , the trace of the matrix element calculation @xmath18 = 1 $ ] vanishes for the massless approximation ( @xmath19 ) ignoring the normalization factor @xmath20 : @xmath21 \rightarrow { 1 \over 2 m } \left ( \begin{array}{cccc } 0 & 0 & e - p_{z } & 0 \\ 0 & 0 & 0 & 0 \\ e+p_{z } & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 \end{array } \right ) \rightarrow 0.\end{aligned}\ ] ] the common practice of spin polarization approximation is inaccurate compared with the accurate evaluation of the full spin projection operator . for the relativistic limit of @xmath22 , the spin projection operators for longitudinally polarized electrons with @xmath6 parallel to @xmath23
can be reduced as : @xmath24 whereas the explicit expression of the spin polarization @xmath25 is now given by : @xmath26 and although the approximation holds true for some specific calculations such as @xmath18 = 1 $ ] it is inconsistent with the accurate evaluation of the matrix element in general . the omission of the normalization factor is also misleading because it neglects the electron mass term in the relativistic limit @xcite : @xmath27 whereas the contribution of the @xmath28 term to the matrix element should be the same regardless of the energy @xmath29 once the proper normalization is considered , since the mass factor in @xmath30 is normalized to be one by @xmath31 .
the simplification of the matrix element calculation in the relativistic limit can be given by : @xmath32 whereas the explicit expression of the spin polarization @xmath25 is now given by : @xmath33 this is inconsistent with the accurate evaluation of the general matrix element calculation , since the spin direction represented by a nonzero matrix element is inconsistent for massive electrons .
the matrix element for a right - handed chirality relativistic electron to scatter to a left - handed chirality electron with interaction @xmath34 is proportional to : @xmath35 whereas from the exact representation of massive fermions in terms of helicity , not of chirality , the corresponding matrix element is in general nonzero : @xmath36 } \nonumber \\ & & { } = \mathrm{tr}\left[\overline{u}_{h-}(p')\gamma^{\mu}u(p ) \overline{u}(p)\left({1 - \gamma^{5 } { { s}\!\!\!\slash } \over 2 } \right)\gamma^{\nu}u(p')_{h- } \right ] .
\nonumber\end{aligned}\ ] ] note that @xmath37 is supposed to act on @xmath38 , not on @xmath39 alone , and the chiral projection @xmath40 on @xmath39 is inconsistent with being a proper approximation of @xmath37 for a massive fermion .
since the chiral projection as an approximation of spin projection operators could differ from the exact calculation of spin projection operators , the matrix element should be carefully evaluated comparing with the full representation of the spin projection operators .
the parity - violating asymmetry for polarized electrons scattering on an unpolarized target is given by : @xmath41 where @xmath42 denotes the differential cross section for an incoming electron of helicity @xmath43 on an unpolarized target .
for the parity - violating asymmetry in slac e158 and sld @xcite , left- and right - handed helicity massive fermions are considered as distinguishable particles with the weak interaction structure of @xmath44 , as suggested in the standard model @xcite , and the asymmetry is accounted for as parity violation between opposite helicities . for the electron scattering of @xmath45 where @xmath46 denote the four momenta and helicities of incoming and outgoing electrons , respectively , the interaction lagrangian for the electromagnetic and weak interactions with the z boson
is given by : @xmath47z^{\mu } \nonumber \\ & = & -g_{\gamma}\overline{e}\gamma_{\mu}ea^{\mu } + g_{z}\overline{e}\gamma_{\mu}(v - a\gamma_{5})ez^{\mu } , \label{weaklageq}\end{aligned}\ ] ] where @xmath48 , @xmath49^{-1}}$ ] and @xmath50 , @xmath51 .
note that @xmath40 in eq .
( [ weaklageq ] ) are not chiral projections but chiral interaction structures , and electrons @xmath52 are later to be classified as left- and right - handed helicity electrons @xmath53 using the spin polarization projection operator @xmath54 @xcite . the total tree - level amplitude for @xmath55 scattering via @xmath56 and @xmath57 exchange in the center - of - mass ( cm ) frame is then given by : @xmath58 where @xmath59 with the cm scattering angle @xmath60 and the cm energy @xmath61^{1/2}$ ] .
for the explicit spin polarization calculation with the operator @xmath54 for inward and outward relative spin polarization corresponding to @xmath62 , the nonzero parity - violation asymmetry arises from such terms as : @xmath63'_{\lambda_{1 } } + [ \mathcal{m}_{\gamma}^{d}\mathcal{m}_{z}^{c*}]'_{\lambda_{1 } } + [ \mathcal{m}_{\gamma}^{c}\mathcal{m}_{z}^{d*}]'_{\lambda_{1 } } + [ \mathcal{m}_{\gamma}^{c}\mathcal{m}_{z}^{c*}]'_{\lambda_{1}}$ ] , @xmath64'_{\lambda_{1 } } % { 2 g_{\gamma}^{2 } g_{z}^{2 } \over ysm_{z}^{2 } } } \hspace{3.3 in } \nonumber \\
\lefteqn{= { g_{\gamma}^{2 } g_{z}^{2 } \over ysm_{z}^{2 } } \mathrm{tr}[\overline{u}(p_{3})\gamma^{\mu}(1 \mp \gamma^{5}{{s_{1}}\!\!\!\!\!\slash}\,)u(p_{1 } ) \overline{u}(p_{1})\gamma^{\nu}(v - a \gamma^{5})u(p_{3})]}\hspace{3.1 in } \nonumber \\ \lefteqn { \mathrm{tr}[\overline{u}(p_{4})\gamma_{\mu}u(p_{2 } ) \overline{u}(p_{2 } ) \gamma_{\nu}(v - a\gamma^{5})u(p_{4})]}\hspace{2.7 in } \nonumber \\ \lefteqn{= [ \mathcal{m}_{\gamma}^{d}\mathcal{m}_{z}^{d * } ] \mp 32 { g_{\gamma}^{4 } \over y}\beta ( -e_{1}p_{2z}+e_{2}p_{1z } ) v a } \hspace{3.1in}\end{aligned}\ ] ] @xmath65'_{\lambda_{1 } } } \hspace{3.3 in } \nonumber \\
\lefteqn{= { g_{\gamma}^{2 } g_{z}^{2 } \over ysm_{z}^{2 } } \mathrm{tr}[\overline{u}(p_{3})\gamma^{\mu}(1 \mp \gamma^{5}{{s_{1}}\!\!\!\!\!\slash}\,)u(p_{1 } ) \overline{u}(p_{1})\gamma^{\nu}(v - a \gamma^{5})u(p_{4})}\hspace{3.1 in } \nonumber \\ \lefteqn { \overline{u}(p_{4})\gamma_{\mu}u(p_{2 } ) \overline{u}(p_{2 } ) \gamma_{\nu}(v - a\gamma^{5})u(p_{3})]}\hspace{2.45 in } \nonumber \\ \lefteqn{= [ \mathcal{m}_{\gamma}^{d}\mathcal{m}_{z}^{c * } ] \mp 32 { g_{\gamma}^{4 } \over y}\beta ( -e_{1}p_{2z}+e_{2}p_{1z } ) v a } \hspace{3.1in}\end{aligned}\ ] ] where the incoming and target electrons are given by @xmath66 and @xmath67 . the parity - violating asymmetry @xmath68 is given by : @xmath69 where @xmath70 is lorentz invariant holding for any arbitrary frame when @xmath71 is parallel to @xmath72 and the massless approximation of @xmath68 in the cm frame @xmath73 and @xmath74 is consistent with the approximated parity - violating asymmetry of the slac e158 denoted as @xmath75 @xcite . in sld ,
the polarized differential cross section of the @xmath76 process with longitudinally polarized electrons and unpolarized positrons is @xmath77_{h\pm } = { n_{c } \alpha^{2 } \over 8 s \sin^{4 } \theta_{w } \cos^{4 } \theta_{w } [ ( s - m^{2}_{z})^{2 } + s^{2}\gamma^{2}_{z}/m^{2}_{z } ] } }
\hspace{3.3 in } \nonumber \\ \lefteqn{\times [ 2(v_{e}^2+a_{e}^2)(v_{f}^2+a_{f}^2)(e_{1}e_{2}e_{4}e_{3 } + p_{1z}p_{2z}p_{4z}p_{3z})}\hspace{3.in } \nonumber \\ \lefteqn{\mp2v_{e}a_{e}(v_{f}^2+a_{f}^2)(2e_{1}p_{2z}p_{4z}p_{3z } + 2p_{1z}e_{2}e_{4}e_{3}}\hspace{2.9 in } \nonumber \\ -(e_{1}e_{2}+p_{1z}p_{2z})(p_{4z}e_{3}+e_{4}p_{3z } ) ) \nonumber \\ \lefteqn{+4v_{e}a_{e}v_{f}a_{f}(p_{1z}e_{2}-e_{1}p_{2z } ) ( e_{4}p_{3z}-p_{4z}e_{3})}\hspace{2.9 in } \nonumber \\ \lefteqn{\mp2(v_{e}^2+a_{e}^2)v_{f}a_{f}(e_{1}e_{2}-p_{1z}p_{2z } ) ( e_{4}p_{3z}-p_{4z}e_{3})}\hspace{2.9 in } \nonumber \\ \lefteqn{-(v_{e}^2+a_{e}^2)(v_{f}^2+a_{f}^2)(e_{1}p_{2z } + p_{1z}e_{2 } ) } \hspace{2.9 in } \nonumber \\ \lefteqn{\times ( e_{4}p_{3z}+p_{4z}e_{3 } ) ] } \hspace{1.5in}\end{aligned}\ ] ] note that its massless approximation in the cm frame @xmath73 and @xmath74 is consistent with the approximated parity - violating calculation of the sld @xcite and the total cross sections are lorentz invariant .
let us investigate the exact cross section calculation under parity regarding to spin polarization and momentum of incoming and target particles .
in the exact calculations , the factor that determines the signs of parity - violating terms depending on spin polarization and momentum is given by @xmath78 .
the signs of parity - violating terms become reversed when either the spin polarization ( @xmath79 ) of the incoming particle or the relative velocity between two particles @xmath80 is reversed .
note that the signs of parity - violating terms can remain constant even when the helicity of incoming electron becomes opposite as its momentum is reversed ( @xmath81 ) under the lorentz transformations such as in fig .
[ fig : lt2particle ] , since the factor @xmath82 ) is lorentz invariant along the @xmath83 direction . for a two - particle system
, the spin polarization can be characterized by its direction relative to the velocity difference between the incoming and target particles @xmath84 , indicating whether the spin direction of the incoming particle points outward from or toward the target particle . under lorentz transformations , the differential cross sections for a right - handed helicity
incoming electron can be observed as a left - handed helicity electron , but the relative spin polarization remains inward as in fig .
[ fig : lt2particle ] .
thus , the exact evaluation of parity - violating asymmetry @xmath68 remains asymmetric in general : @xmath85 whereas the approximated calculation of parity - violating asymmetry , interpreted as the asymmetry between opposite helicities , vanishes @xmath86 .
therefore , the parity - violating asymmetry measured in slac e158 and sld should be interpreted in terms of relative spin polarization not of helicity , since the asymmetry between opposite helicities vanishes under lorentz transformations whereas the exact calculation does not .
since the approximate spin polarization @xmath40 may significantly differ from the exact spin polarization in evaluating the matrix element , here exact calculations of cross sections for parity - violating asymmetries in slac e158 and sld have been performed using the full expression of spin projection @xmath87 . and the parity - violating factor incorporating with spin polarization and momentum @xmath78 has been identified and shown that its sign depends on the spin polarization of incoming particle and the relative velocity of incoming and target particles , not on the helicity of incoming particle .
therefore , i suggest a new concept of relative spin polarization to interpret the parity - violating asymmetry as contributed by the antisymmetric nature of the weak interactions depending on whether the spin direction of the incoming electron is inward or outward relative to the target electron . a. salam , in _ elementary particle theory _
( _ nobel symposium no
. 8 _ , ed .
n. svartholm , stockholm , 1968 ) 367 .
c. y. prescott _
et al . _ ,
b * 77 * , 347 ( 1978 ) .
anthony _ et al .
_ , phys .
* 95 * ( 2005 ) 081601 . | parity - violating asymmetries in polarized electron scattering have been interpreted as the asymmetries between opposite helicities of incoming fermion based on the approximation of the spin polarization operator . here
exact calculations of cross sections for parity - violating asymmetries in slac e158 and sld have been performed using spin projection operators . and the parity - violating factor incorporating with spin polarization and momentum has been identified and shown that its sign depends on the spin polarization of incoming particle and the relative velocity of incoming and target particles .
therefore , i suggest a new concept of relative spin polarization to interpret the parity - violating asymmetry as contributed by the antisymmetric nature of the weak interactions depending on whether the spin direction of the incoming electron is inward or outward relative to the target electron . |
when the motion occurs along the @xmath16-axis , it is possible to carry out the integration over @xmath92 in equation analytically .
we define @xmath93 , \ ] ] where @xmath94 , @xmath60 is the reflection coefficient , and @xmath95 in these expressions @xmath96 and @xmath97 are respectively the polar and azimuthal spherical angles describing the dipole vector @xmath18 , while @xmath98 is the modified bessel function of the second kind and order @xmath99 .
in addition , @xmath100 $ ] and @xmath101 $ ] .
it is convenient to introduce the dimensionless variables @xmath102 , @xmath103 , @xmath104 , and @xmath105 , where @xmath106 represents a characteristic frequency of the surface ( e.g. , the surface plasmon frequency in the case of a metallic medium ) .
we also define two other auxiliary functions , @xmath107 and @xmath108 , where the latter differs from the former only by the absence of the heaviside theta function in the integrand . using these definitions , the complex atomic polarizability in equation
can be rewritten as @xmath109 where @xmath110 and @xmath111 .
these expressions and the power spectrum given in equation allow us to recast the quantum frictional force as we are now in a position to analyze the low - velocity limit , which corresponds to @xmath113 . assuming an ohmic surface , for small @xmath114 we have @xmath115 , where @xmath116 is the static reflection coefficient .
the prefactor @xmath117 is related to the material resistivity @xmath63 by @xmath118 ( for a metallic surface @xmath119 ) .
the heaviside function and the modified bessel functions in the integrand of equation limit the integration range of @xmath120 and @xmath121 , and in the limit @xmath113 , we can approximate @xmath122 and @xmath123\frac{v}{z},\frac{v}{z}\right ) \approx
\theta(w-\nu)\frac{v}{z}\eta\frac{w-\nu}{24\pi}\mathcal{a}^{0}(\varphi,\theta)\\ + \frac{v}{z}\eta\int_{{\left\vertw-\nu\right\vert}}^{\infty}\frac{dw_{1}}{2\pi}\frac{(w_{1}-{\left\vertw-\nu\right\vert})}{6\pi}w_{1}^{2}\mathcal{k}(w_{1},\varphi,\theta ) .
\label{gthetasplit}\end{gathered}\ ] ] here , we have defined @xmath124 ( where @xmath125 and @xmath126 were already defined after equation ) and , for simplicity , have dropped the arguments . in the same @xmath113 limit , we approximate the polarizability by its @xmath127 value , i.e. , @xmath128v / z , z , v / z\right ) \approx \alpha_{\rm
sp } \left[1-(\alpha_{\rm sp}r_{0}\mathcal{a}^{0})/(24\pi z^{3 } ) \right]^{-1}$ ] .
from these expressions we can already conclude that for low velocities @xmath129 .
the two terms on the right - hand side of equation are related to the two contributions to the non - equilibrium fluctuation - dissipation theorem , equation .
the first term gives the low - velocity behavior of @xmath130 , and for a metallic surface it can be written as @xmath131 where we have defined @xmath132 . upon neglecting the frequency shift in the denominator of the polarizability and averaging over all dipole orientations ,
we recover the expression given in equation .
the second term in gives the low - velocity expansion of @xmath133 .
its evaluation is more involved than the previous one , and we will only sketch the main steps . from the second term in
we can write @xmath134 where we have defined the function @xmath135 inserting this result back in and integrating over @xmath120 , we obtain an expression similar to equation but where the prefactor @xmath136 is replaced by @xmath137 , with @xmath138^{2}$ ] . proceeding as for the previous expression
we obtain equation . 1 . max - born - institut , 12489 berlin , germany 2 .
department of applied physics , yale university , new haven , connecticut 06511 , usa 3 .
institute of physics and astronomy , university of potsdam , karl - liebknecht - str .
24/25 , 14476 potsdam , germany 4 .
humboldt - universitt zu berlin , institut fr physik , ag theoretische optik & photonik , 12489 berlin , germany 5 . theoretical division , ms b213 , los alamos national laboratory , los alamos , new mexico 87545 , usa | recent progress in manipulating atomic and condensed matter systems has instigated a surge of interest in non - equilibrium physics , including many - body dynamics of trapped ultracold atoms and ions , near - field radiative heat transfer , and quantum friction .
under most circumstances the complexity of such non - equilibrium systems requires a number of approximations to make theoretical descriptions tractable .
in particular , it is often assumed that spatially separated components of a system thermalize with their immediate surroundings , although the global state of the system is out of equilibrium .
this powerful assumption reduces the complexity of non - equilibrium systems to the local application of well - founded equilibrium concepts .
while this technique appears to be consistent for the description of some phenomena , we show that it fails for quantum friction by underestimating by approximately @xmath0 the magnitude of the drag force .
our results show that the correlations among components of driven , but steady - state , quantum systems invalidate the assumption of local thermal equilibrium , calling for a critical reexamination of this approach for describing the physics of non - equilibrium systems . in recent years
the physics of non - equilibrium systems has attracted a lot of attention from different disciplines , such as stochastic thermodynamics and many - body quantum dynamics @xcite . in particular , there has been a renewed interest in non - equilibrium dispersion forces .
better known for equilibrium phenomena such as the van der waals / casimir - polder force @xcite and the casimir effect @xcite , these interactions play an important role in several fields of physics , including atomic @xcite and statistical physics @xcite , gravitation @xcite and cosmology @xcite .
non - equilibrium physics enters in the description of these phenomena when , for example , temperature gradients or mechanical motion become relevant elements of the system . from the theoretical standpoint
, one must often rely on approximations in order to predict the non - equilibrium physics of a specific system .
one of the most ubiquitous approaches relies on the local thermal equilibrium ( lte ) approximation , which consists in treating the individual components of a system as if they were in local thermal equilibrium with their immediate surroundings .
the main advantage of such a technique is that common equilibrium tools , such as the fluctuation - dissipation theorem ( fdt ) @xcite , can be applied locally , and then these local results are combined to describe the non - equilibrium dynamics of the full system .
the usual justification for the lte approximation is that the correlation length of the fields that mediate the interactions is often rather short ( the dynamics in sufficiently well - separated locations are incoherent and can be treated as being independent @xcite ) , and the sub - systems locally relax to equilibrium on a fast time - scale .
the lte approximation has been used in several non - equilibrium contexts , such as near - field radiative heat transfer @xcite , casimir forces between bodies at different temperatures @xcite , and quantum friction @xcite . in these previous cases , however , a quantitative assessment of the lte approximation is missing . in this work ,
we show that this common approach actually fails to provide reliable predictions for quantum friction .
let us consider an atom moving in vacuum with nonrelativistic velocity @xmath1 at a distance @xmath2 above and parallel to a flat surface placed at @xmath3 ( see fig .
1 ) .
the atom couples to the electromagnetic field via its dipole moment @xmath4 . in previous work
@xcite it was shown that the zero - temperature frictional force ( quantum friction ) acting on the atom is given by @xmath5 .
\label{friction2t}\end{gathered}\ ] ] here , @xmath6 is the non - equilibrium velocity - dependent dipole power spectrum tensor ( related to the spectral distribution of energy in the dipole ) , and @xmath7 is the fourier transform ( in time and along the @xmath8 plane ) of the green tensor describing the electromagnetic response of the surface . in the following the subscript @xmath9 ( @xmath10 ) means that the imaginary ( real ) part has to be ( component - wise ) considered .
the standard approach used in the literature to compute the frictional force has been to resort to the lte approximation .
it is assumed that the particle and the surface surrounded by its electromagnetic field are _ locally _ at thermal equilibrium at @xmath11 in their respective rest frames , and that the fluctuation - dissipation theorem separately applies to each subsystem @xcite . in this case
one assumes that @xmath6 is related to the imaginary part of the particle s polarizability tensor @xmath12 via the zero - temperature fdt , @xmath13 ( the function @xmath14 is the heaviside function ) . upon implementing the lte
the resulting frictional force at low velocities takes the form @xcite @xmath15 , \label{lowvelocity}\ ] ] where the primes denote frequency derivatives and we assumed that the motion is along the @xmath16-direction .
a detailed quantitative evaluation of equation requires the low - frequency behavior of the polarizability , which is often calculated within second - order perturbation theory @xcite .
although the use of the lte approximation can be justified within a second - order perturbative approach in the dipole strength for particles with large intrinsic dissipation @xcite , it becomes less rigorous for atoms where dissipation is induced by the interaction with the electromagnetic field . for systems where dissipation is caused by radiative
damping , quantum friction requires a higher - than - second - order perturbative calculation , and hence the local thermal equilibrium approximation fails because the description of such systems necessarily encompasses the correlations between the atom and the surface ( see fig .
[ ltefig ] ) .
this is the key insight of this paper . in order to test the validity of the lte approximation in quantum friction
, we are going to compute the dipole power spectrum @xmath6 , evaluate the resulting drag force , and compare it to the lte result .
this entails the computation of the non - equilibrium steady - state ( ness ) of the joint atom+field+matter system .
this difficult problem becomes manageable by modeling the internal atomic dynamics as a harmonic oscillator @xcite , for which it is possible to obtain an exact , non - perturbative form for the dipole power spectrum thanks to the quadratic nature of the full system hamiltonian @xcite .
we work in the heisenberg picture to calculate the dipole correlator in the steady state and derive the power spectrum .
( see the main text ) .
, width=321 ] in the non - relativistic approximation the equation of motion of a dipole moving along a prescribed trajectory @xmath17 and with a fixed direction @xmath18 is given by @xmath19 where @xmath20 is the oscillator s frequency and @xmath21 is the electric field .
we assume that the oscillator has no intrinsic dissipation - all dissipative dynamics arises from the coupling to the electromagnetic field .
the electric field at the instantaneous position of the atom is given by @xmath22 where @xmath23 is the field that is generated by the quantum fluctuating currents in the medium . in the stationary @xmath24 limit
, we use that @xmath25 and @xmath26 . upon inserting into the equation of motion for the dipole
, we obtain the stationary solution in fourier space as @xmath27 where we have defined the velocity - dependent polarizability @xmath28 in this expression , @xmath29 is the radiative damping while @xmath30 is related to a frequency shift @xcite , and they are given by @xmath31 @xmath32 the dipole correlation function , derived from equation , defines the power spectrum @xmath33 , where the average is taken over the initial factorized state of the system , @xmath34 . here , @xmath35 is the atom s initial density matrix and @xmath36 represents the state of the coupled field plus matter subsystem .
both the atom and the field+matter are assumed to be initially in their respective ground states .
because of equation , we can compute the dipole - dipole correlation in terms of the field - field correlator . since @xmath23 is the field generated solely by the surface we can use the fdt .
this gives @xmath37 where we have defined @xmath38/(2 { \mathrm{i}})$ ] . by combining all the above equations and using that @xmath39 , we obtain @xmath40 since the matrix @xmath41 is a symmetric tensor , we replaced @xmath42 by @xmath43 , where only the symmetric part of @xmath44 contributes to the tensor product @xcite . by noting that the polarizability and the green tensor are related via @xmath45
the dynamic power spectrum can be expressed as @xmath46 where @xmath47 \\ \times\underline{\alpha}(\omega;\mathbf{v})\cdot \underline{g}_{i}(\mathbf{k},z_{a},\omega+\mathbf{k}\cdot\mathbf{v})\cdot\underline{\alpha}^{*}(\omega;\mathbf{v}).\end{gathered}\ ] ] equation constitutes the _ generalized non - equilibrium fdt _ for the moving harmonic oscillator .
it shows that , when the system is in a ness , an extra term @xmath48 is added to the standard fdt , equation .
the expression in is similar to classical non - equilibrium generalizations of the fdt ( see , for example , refs.@xcite ) , where the additional term is related to entropy production .
however , these works often include assumptions ( e.g. markovianity ) which are incompatible with the description of quantum friction @xcite . upon inserting equation into ,
we obtain two distinct contributions to the quantum frictional force , @xmath49 which respectively arise from the first and second terms on the right hand side of .
as we will show below , the low - velocity expansion of @xmath50 corresponds to equation @xcite , while @xmath51 is entirely due to the non - equilibrium dynamics of our system .
we now compute the low - velocity expansion of the force in equation . as before
, we assume that the motion occurs along the @xmath16-direction , so that @xmath52 ( here @xmath53 is the unit vector along the @xmath16 direction ) .
the total green tensor in equation can be decomposed as the sum of the vacuum @xmath54 and the scattered contribution @xmath55 .
because of lorentz invariance , the vacuum contribution @xmath56 does not contribute to the frictional force @xcite .
for simplicity , we consider the near - field limit for @xmath55 , whose symmetric part has an imaginary part given by @xcite @xmath57 where @xmath58 , @xmath59 is the vacuum permittivity , and @xmath60 is the quasi - static approximation of the transverse magnetic reflection coefficient for the planar surface . using the previous expression one can show that that in the low - velocity limit the first term in the right hand side of equation gives ( see supplemental material ) @xmath61 where @xmath62 is the static isotropic atomic polarizability , @xmath63 is the material resistivity , and @xmath64 is a geometrical factor coming from the average over all dipole orientations .
the above expression reduces to equation when we use the polarizability given in . for the non - equilibrium correction term in
, we obtain similarly ( see supplemental material ) @xmath65 where @xmath66 . adding the low - velocity expansions of @xmath67 and @xmath68
, the full quantum frictional force becomes @xmath69 which differs by almost a factor of two from the approximate lte result in equation .
this is the main result of our paper and demonstrates that the non - equilibrium contribution to the frictional force is certainly _ not _ negligible . in fig .
[ frictionfull ] we depict the quantum frictional force equation as a function of velocity . for simplicity
the dipole is oriented along a direction for which @xmath70 and moving above a metallic surface described by the drude model @xmath71^{-1}$ ] , where @xmath72 is the plasma frequency and @xmath73 is the metal s relaxation rate ( in this case the material resistivity is given by @xmath74 ) .
for small velocities the friction is well described by the asymptotic expression equation ( black dotted line ) . in this region the integrals in equation are dominated by the low frequency behavior of @xmath6 and @xmath7 , resulting in the power - law dependency on velocity and separation ( see supplemental material ) .
the relative difference between the exact and lte results is more than @xmath75 in this region ( see inset of fig .
[ frictionfull ] ) . at high velocities ( @xmath76 for the parameters in fig .
[ frictionfull ] ) a kink is visible at the crossing between the previous asymptotic expressions and @xmath77 ( black dashed curve in fig . [ frictionfull ] ) , where @xmath78 is the surface plasmon frequency .
equation is the result of a second - order perturbative expansion of , and can be explained by a resonant process involving the atom - surface interaction @xcite . as shown in the figure , the expression in describes well the behavior of the quantum frictional force immediately after the kink .
since the impact of radiative damping is negligible in this second - order expansion , in the region right after the kink the atom and the surface can be considered uncorrelated to a good approximation , and the lte description is sufficient to characterize the quantum frictional process .
this is clearly seen in the sharp decrease of the relative difference between the exact and lte results ( see inset ) .
a further increase of the velocity leads again to a deviation from the lte approximation due to the strengthening of the non - equilibrium - induced atom - surface correlations . above a metallic surface described by the drude model .
the dipole is oriented along the direction @xmath79 .
the oscillator has a resonance frequency @xmath80 and moves at a distance @xmath81 above and parallel to the surface with dissipation rate @xmath82 . at low velocities
the lte approximation underestimates the frictional force by approximately @xmath0 ( see inset ) . as the velocity increases ( @xmath76 for the parameters above ) ,
the oscillator s radiative damping becomes less relevant and the force is accurately described by the asymptotic expression given in equation ( dashed black line ) , which corresponds to @xmath83 .
a further increase in the velocity enhances the non - equilibrium contribution to the force and a deviation from the lte description occurs again ( see inset ) .
the normalization is @xmath84 . for a @xmath85 atom ( @xmath86 @xcite ) and a plasma frequency @xmath87 ev
, we have @xmath88 .
, width=321 ] due to its small value , an experimental detection of quantum friction is challenging and designing setups that increase the strength of the interaction is certainly desirable .
specifically , equation can be rewritten as @xmath89 where @xmath90 is the leading - order ( i.e. , low - frequency and small - velocity ) expansion of the function @xmath91 defined in equation .
this demonstrates that , at low velocities , quantum friction is proportional to the square of the induced decay rate .
this feature suggests possible pathways to increase the strength of the quantum frictional force .
for instance , material properties or geometric configurations , such as hyperbolic nanostructures @xcite , which are known for producing large purcell factors , are potentially favorable for enhancing the quantum frictional force . in conclusion
, we have shown that the local thermal equilibrium approximation fails in quantum friction .
we demonstrated this point with an exact solution to a model of a harmonic oscillator moving parallel to a surface , in which the lte approach underestimates the quantum friction force by approximately @xmath75 .
motion - induced atom - surface correlations are ultimately responsible to the breakdown of the local equilibrium assumption .
it is worth emphasizing that , despite its extensive application and even if quite reasonable in most circumstances , the lte approximation relies more on phenomenological considerations than on quantitative estimations .
our results in quantum friction call for a critical assessment of the range of applicability of local thermal equilibrium in other non - equilibrium dispersion interactions .
such an analysis could potentially provide new insights and unravel important features of these and other non - equilibrium systems .
_ acknowledgments .
_ we acknowledge support by the lanl ldrd program , and by the deutsche forschungsgemeinschaft ( dfg ) through project b10 within the collaborative research center ( crc ) 951 hybrid inorganic / organic systems for opto - electronics ( hios ) .
fi further acknowledges financial support from the european union marie curie people program through the career integration grant no .
pcig14- ga-2013 - 631571 .
ch and fi acknowledge support from the dfg through the dip program ( grant fo 703/2 - 1 and schm 1049/7 - 1 ) 10 t. speck and u. seifert , europhys .
lett . * 74 * , 391 ( 2006 ) .
u. seifert , eur .
phys .
j. b * 64 * , 423 ( 2008 ) .
m. rigol , v. dunjko , and m. olshanii , nature * 452 * , 854 ( 2008 ) .
h. b. g. casimir and d. polder , phys . rev . * 73 * , 360 ( 1948 ) .
h. b. g. casimir , proc .
k. ned .
akad . wet . *
51 * , 793 ( 1948 ) .
j. fortgh and c. zimmermann , rev .
mod .
phys . * 79 * , 235 ( 2007 ) .
m. e. fisher and p .-
g .
de gennes , cr acad .
sci .
ser .
b * 287 * , 207 ( 1978 ) .
a. gambassi , journal of physics : conference series * 161 * , 012037 ( 2009 ) .
r. onofrio , new j. phys .
* 8 * , 237 ( 2006 ) .
r. j. adler , b. casey , and o. c. jacob , am . j. phys .
* 63 * , 620 ( 1995 ) .
h. b. callen and t. a. welton , phys . rev . * 83 * , 34 ( 1951 ) .
d. polder and m. van hove , phys .
rev .
b * 4 * , 3303 ( 1971 ) .
w. eckhardt , phys .
rev .
a * 29 * , 1991 ( 1984 ) .
i. a. dorofeyev , j. phys .
a : math . gen . * 31 * , 4369 ( 1998 ) .
j. m. obrecht , r. j. wild , m. antezza , l. p. pitaevskii , s. stringari , and e. a. cornell , phys .
rev .
lett . * 98 * , 063201 ( 2007 ) .
j. b. pendry , j. phys . : condes .
matter * 9 * , 10301 ( 1997 ) .
a. i. volokitin and b. n. j. persson , rev .
mod .
phys . * 79 * , 1291 ( 2007 ) .
j. s. hye and i. brevik , europhys . lett . * 91 * , 60003 ( 2010 ) . m. f. maghrebi , r. l. jaffe , and m. kardar , phys .
rev .
lett . * 108 * , 230403 ( 2012 ) . m. f. maghrebi , r. golestanian , and m. kardar , phys . rev .
a * 88 * , 042509 ( 2013 ) .
j. s. hye and i. brevik , eur .
phys .
j. d * 68 * , 1 ( 2014 ) .
f. intravaia , r. o. behunin , and d. a. r. dalvit , phys .
rev . a * 89 * , 050101(r ) ( 2014 ) .
p. w. milonni and r. w. boyd , phys .
rev .
a * 69 * , 023814 ( 2004 ) .
s. y. buhmann , l. knoell , d .-
g .
welsch , and h. t. dung , phys .
rev .
a * 70 * , 052117 ( 2004 ) .
g. lach , m. dekieviet , and u. d. jentschura , phys .
rev .
lett .
* 108 * , 043005 ( 2012 ) .
d. jentschura , u. , g. lach , m. de kieviet , and k. pachucki , phys .
rev .
lett . * 114 * , 043001 ( 2015 ) .
f. intravaia , r. o. behunin , c. henkel , k. busch , and d. a. r. dalvit , eprint : arxiv:1603.05165 .
a. einstein and l. hopf , ann .
physik * 338 * , 1105 ( 1910 ) .
j. klatt , r. bennett , and s. y. buhmann , eprint : arxiv:1601.02765 .
g. s. agarwal , z. physik * 252 * , 25 ( 1972 ) .
r. chetrite , g. falkovich , and k. gawedzki , j. stat .
mech . *
2008 * , p08005 ( 2008 ) .
m. baiesi , c. maes , and b. wynants , phys .
rev .
lett . * 103 * , 010602 ( 2009 ) .
p. talkner , m. campisi , and p. hnggi , j. stat . mech . * 2009 * , p02025 ( 2009 ) .
u. seifert and t. speck , europhys .
lett . * 89 * , 10007 ( 2010 ) .
g. dedkov and a. kyasov , phys . solid state * 45 * , 1815 ( 2003 ) . a. a. kyasov and g. v. dedkov , nuclear instrum
. methods b * 195 * , 247 ( 2002 ) .
a. i. volokitin and b. n. j. persson , phys .
rev .
b * 78 * , 155437 ( 2008 ) .
g. pieplow and c. henkel , new j. phys .
* 15 * , 023027 ( 2013 )
. j. m. wylie and j. e. sipe , phys .
rev .
a * 30 * , 1185 ( 1984 ) .
d. a. steck , technical report , oregon center for optics and department of physics , university of oregon ( http://steck.us/alkalidata ) .
f. intravaia and k. busch , phys .
rev . a * 91 * , 053836 ( 2015 ) . |
recent galaxy redshift surveys ( e.g. , 2dfgrs , * ? ? ?
* ) have allowed us to map the universe on the largest scale ever and study systematically the detailed properties of the cosmic large - scale structures .
it is confirmed by these surveys that the galaxies in the universe are not distributed evenly but develop pattern of network connected by filaments and sheets which in turn are separated by immense voids .
voids are defined observationally as large regions with very low galaxy number density and shown to occupy approximately @xmath0 of the cosmic volume @xcite .
the striking vastness of the observed void volumes inspired a flurry of research on the origin and properties of cosmic voids @xcite .
it is now generally believed that voids are originated from the local minima of the primordial density fluctuations and expand faster than the rest of the universe .
the very fact that voids are extremely underdense and expand faster has led many authors to assume somewhat naively that the shapes of voids should be spherical .
in fact previous analytical and numerical works on voids are based on this assumption ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
very recently , however , @xcite investigated systematically the shapes of voids by applying the excursion set approach to the density field in n - body simulations and demonstrated that the intrinsic shapes of voids are far from spherical symmetry .
they claimed that voids are most vulnerable to external shears due to their low - density , so that their shapes are disturbed to be nonspherical in spite of their faster expansion .
although @xcite focused on the shapes and morphology of voids , we note here that if the tidal shear plays a significant role in the case of voids , then it should cause not only the shapes of voids to be nonspherical but also voids to acquire the spin angular momentum .
our goal here is to construct an analytic model for the spin angular momentum of voids originated from the tidal shear effect and test the analytic predictions against the data from the recent millennium run simulation .
the organization of this paper is as follows . in @xmath1 , the analytic model for the void spin angular momentum and
the void spin correlation properties are presented . in @xmath2
, the millennium run galaxy catalog is analyzed to determine void spin statistics , and the numerical results are compared with the analytic predictions . in @xmath3 , the achievements and caveats of our works are discussed , and a final conclusion is drawn .
let us consider a lagrangian region in the linear smoothed density field . according to the linear tidal torque theory @xcite , the spin angular momentum of matter which make up this region , @xmath4 , is proportional to the tensor product as @xmath5 where @xmath6 is the inertia tensor of the region , and @xmath7 is the local tidal tensor at the center of the mass of the region defined as the second derivative of the perturbation potential : @xmath8
. strictly speaking , equation ( [ eqn : ang ] ) gives a valid first order approximation to the spin angular momentum only when the lagrangian regions correspond to bound halos .
however , to quantify the effect of tidal field on voids , we extend the validity of the linear tidal torque theory to the case of underdense proto - void regions and define the _ spin angular momentum of a void _ by equation ( [ eqn : ang ] ) .
since a void is an unbound system , its spin angular momentum has a distinct meaning from that of a bound halo although defined in a similar way .
basically , the spin angular momentum of a void is introduced to quantify how a proto - void region reacts to the effect of the local tidal force .
the local tidal field will generate non - radial motion of the material in the proto - void region , leading the distribution of matter in the void to deviate from spherical symmetry .
the magnitude of the void spin angular momentum quantifies the degree of this deviation , and its direction indicates the direction along which the maximum deviation occurs .
it was shown by n - body simulations that in the overdense proto - halo regions the principal axes of @xmath9 are in fact strongly correlated with that of @xmath10 @xcite .
very recently , @xcite has shown analytically that the degree of the misalignment between the principal axes of @xmath9 and @xmath10 decreases monotonically with the linear density of the region .
in other words , the more underdense a region is , the more misaligned the principal axes of @xmath9 and @xmath10 are . given these numerical and analytical findings , we propose a new hypothesis that a void forms from the initial underdense region where the principal axes of @xmath9 and @xmath10 are maximally misaligned with each other . this hypothesis combined with equation ( [ eqn : ang ] ) leads us to expect that the alignment between the void spin axes and the principal axes of the local tidal tensor should be strongest for the proto - void regions .
@xcite showed that the alignment between the spin axis of a lagrangian region and the intermediate principal axes of the local tidal tensor can be best quantified by calculating the expectation value of the unit spin given the tidal tensor , and they derived an approximate formula for it .
their original formula is characterized by a correlation parameter , @xmath11 , in the range of @xmath12 $ ] , which was introduced to represent the degree of the alignment , which is nothing but @xmath13 times @xmath11 . ] between the spin axes and the principal axes of the tidal tensors .
the case of maximum alignment corresponds to @xmath14 while no alignment case to @xmath15 . for the case of halo spins , they found @xmath16 empirically . in accordance of our hypothesis ,
@xmath14 for void spins .
now , with the value of @xmath11 set at unity , the formula of @xcite is rewritten for the case of void spins as @xmath17 where @xmath18 and @xmath19 with @xmath20 .
equation ( [ eqn : spincorr ] ) indicates that the spin axes of voids are spatially correlated due to the spatial correlation of the tidal field .
the alignments between the spin axes of neighbor void spins can be quantified by the two - point spin - spin correlation function which was introduced by @xcite for the case of dark halos : @xmath21 here , the constant @xmath22 represents the value of @xmath23 for the case of no alignment . for the case of a power - law power spectrum , the void spin - spin correlation function , @xmath24 ,
can be approximated @xcite as @xmath25 here @xmath26 is the correlation function of the linear density field smoothed on a typical lagrangian scale of voids , defined as @xmath27 where @xmath28 is the dimensionless power spectrum and @xmath29 is the top - hat spherical filter of scale radius , @xmath30 . since voids and galaxy clusters can be thought of the counterparts which are supposed to form at the local minima and maxima of the initial density field , respectively , we assume the typical lagrangian scale of voids are same as that of galaxy clusters , setting the value of @xmath31 at @xmath32mpc . since the concordance @xmath33cdm cosmology is well represented by a power - law power spectrum @xmath34 on the scale of @xmath32mpc , one can regard equation ( [ eqn : eta ] ) as a valid approximation .
it is worth mentioning here that the misalignment between the principal axes of @xmath10 and @xmath9 ( i.e. , the value of the correlation parameter @xmath14 ) is characteristic for those underdense regions in the initial density field that will eventually become voids .
although the majority of the initial underdense regions will evolve into voids , some underdense regions could merge into overdense region to collapse rather than to become voids @xcite .
@xcite showed both analytically and numerically that for the case of dark halos the separation vectors between the neighbor halos , @xmath35 , are aligned with the major axes of the tidal tensors . to quantify the degree of this alignment , @xcite proposed the following formula : @xmath36 where @xmath37 and @xmath38 is a correlation parameter in the range of @xmath12 $ ] .
the negative sign in front of the parameter @xmath38 indicates that @xmath39 is aligned with the _ major _ principal axis of @xmath40 unlike @xmath41 .
the value of @xmath38 was determined empirically to be approximately @xmath42 for the case of halos @xcite .
similarly , we expect that the separation vectors between the neighbor voids are aligned with the _ minor _ axes of the tidal tensors
. then , equation ( [ eqn : dencorr ] ) should be also true for the case of voids .
we assume further that @xmath38 has the same value of @xmath42 .
now that the spin axes of voids are aligned with the intermediate axes of the local tidal tensors ( eq.[[eqn : spincorr ] ] ) while the separation vectors to the nearest voids are aligned with the minor axes of the local tidal tensors , one can expect that the spin axes of voids should be anti - aligned with the separation vectors to the nearest voids .
the expected spin - direction anti - alignment can be quantified by the following correlation function @xcite : @xmath43 it was found by @xcite that for the case of small - scale galactic halos equation ( [ eqn : dencorrft ] ) can be approximated as @xmath44 where @xmath45 is the typical lagrangian size of a galactic halo and @xmath46 .
this approximation was made under the assumption that the tidal field does not vary significantly within the lagrangian distance @xmath47 .
for the cases of large - scale voids , however , this assumption may not be valid since within the void separation the tidal field is likely to vary significantly .
taking into account the spatial variance of the tidal field , we find that for the case of voids the spin - direction correlation function is better approximated as @xmath48
to test the analytic predictions made in @xmath1 , we use the millennium run galaxy catalog carried out by the virgo consortium @xcite .
the catalog consists of total @xmath49 galaxies at redshift @xmath50 in a periodic box of linear size @xmath51mpc , with information on various galaxy properties such as position , velocity , mass , magnitude , and star formation rate .
voids are identified in the catalog by the void - finding algorithm described in ( * ? ? ?
* hereafter hv02 ) which was in fact first suggested by @xcite .
to apply the hv02 algorithm to the millennium run catalog , we first determine the values of the key parameters , @xmath52 and @xmath53 .
the two parameters , @xmath52 and @xmath53 , represent the distance within which a field galaxy is required not to have any three neighbors and a threshold for the minimum linear size of a void , respectively .
for the millennium run catalog , the key parameter values we obtain are @xmath54mpc and @xmath55mpc , respectively . for the detailed description of the procedures to identify voids from the millennium run catalog , see our companion paper ( d. park & j. lee 2006 in preparation ) .
the total @xmath56 voids are found from the catalog .
the volume @xmath57 of each void is measured by means of the monte - carlo integration method described in hv02 . the effective radius of each void is then determined as @xmath58 .
the mean value of @xmath59 is obtained to be @xmath60mpc .
the density of each void , @xmath61 , is measured by counting the number of void galaxies : @xmath62 where @xmath63 is the number density of void galaxies , and @xmath64 is the mean number density of the total catalog galaxies .
figure [ fig : den ] plots the probability distribution of @xmath61 .
as can be seen , the voids found in the catalog are extremely underdense with the mean density @xmath65 .
we first choose only those voids which contain more than @xmath66 void galaxies .
total number of such voids is found to be @xmath67 .
the specific spin angular momentum , @xmath68 , of each selected void is determined as @xmath69 where @xmath70 and @xmath71 are the mass of the @xmath72-th galaxy and the total mass of all galaxies in each void , respectively .
the position and the velocity of the @xmath72-th void galaxy in the center of mass frame are represented by @xmath73 and @xmath74 , respectively .
we rescale this specific spin angular momentum to be dimensionless as @xmath75 with @xmath76 , which is analogous to the definition of the halo spin parameter @xcite .
figure [ fig : void ] illustrates a few examples of the voids identified from the catalog with the directions of the void spin axes marked as short solid line on the voids .
figure [ fig : log ] plots the distribution of @xmath77 for the voids from the catalog as histograms , which shows that @xmath78 has a log - normal shape .
we fit the following log - normal formula to the histogram with adjusting the values of @xmath79 and @xmath80 : @xmath81 the best - fit values of @xmath79 and @xmath80 are found to be @xmath82 and @xmath83 , respectively .
as can be seen , the value of @xmath77 spreads over a large range , revealing the strong shear effect on voids .
we measure the cosine of the relative angle , @xmath84 , between the spin axes of each void pair in the catalog , and determine the average of @xmath85 as a function of the separation distance between the two voids .
for this , we also use only those voids which have more than @xmath66 galaxies . the two - point spin - spin correlation function ( eq.[[eqn : eta ] ] ) is then determined simply by subtracting one third from the average , @xmath86 . figure [ fig : spincorr ] plots the numerical result as solid squares with errors .
the errors are calculated as the standard deviation of @xmath85 for the case of no correlation .
the horizontal dotted line corresponds to the case of no alignment .
as can be seen , there is a clear signal of alignment up to the distance of @xmath87mpc , which is almost an order of magnitude stronger compared with the case of halos . the analytic prediction ( eq.[[eqn : spincorrft ] ] ) is also plotted as solid line for comparison . as can be seen ,
the analytic prediction is in excellent agreement with the numerical result .
it is worth emphasizing that the analytic prediction is not a fitting model but derived from first physical principles under the assumption that voids form in the initial regions where the tidal and the inertia tensors are maximally uncorrelated with each other .
the good agreement between the numerical result and the analytical prediction supports this idea . in a similar manner
, we also determine the void spin - direction correlation function ( eq.[[eqn : omega ] ] ) .
figure [ fig : dencorr ] plots the numerical result as solid squares with errors .
as can be seen , there is a clear signal of anti - alignment between the void spin axes and the direction toward the nearest neighbor voids , which is also an order of magnitude stronger compared with the case of halos .
the analytic prediction ( eq.[[eqn : dencorrft ] ] ) is also plotted as solid line for comparison . we find a good agreement between the analytical and the numerical results for the spin - direction correlation , too .
in the light of the work of @xcite who claimed that the tidal shear must play a strong role in case of voids , we have proposed a scenario that voids are originated from the regions in the initial density field whose inertia tensors are completely uncorrelated with the local tidal tensors . since the alignment between the principal axes of the inertia and tidal tensors plays a role of reducing the tidal shear effect according to the linear tidal torque theory , the hypothesis implies that voids should have very high spin angular momentum .
we define the concept of the spin angular momentum for an unbound void to quantify the effect of the tidal field on the distribution of matter that make up voids .
our analytic model based on the linear tidal torque theory have predicted that there is a strong spatial alignment between the spin axes of neighbor voids and a strong anti - alignment between the spin axes of voids and the directions to the nearest voids .
our predictions have been tested against the millennium run galaxy catalog where @xmath67 large voids with more than @xmath66 galaxies are identified .
we have measured numerically the distribution and correlation functions of void spins , and found that the predictions of our analytic model are in excellent agreement with the numerical results .
the success of our work is , however , subject to a couple of caveats .
the first caveat comes from our implicit assumption that the matter voids are same as the galaxy voids . in our analytic model
, we consider the matter voids . while in the numerical analysis we deal with the galaxy voids .
although the two types of voids are not necessarily same , they should be at least closely correlated with each other .
the second caveat lies in the fact that there is no standard way to identify voids .
unlike the case of halo - finding , the use of different algorithms could produce different results for some properties of voids .
especially , the shapes and volumes of voids can differ significantly by the underlying criterion of the void - finding algorithm .
if one use an algorithm which regards a void as a spherical region of low density , then all voids found by that algorithm would be spherical . in spite of this ambiguity in finding voids
, we expect that our result on the correlation properties of the void spin angular momentum would not change significantly by the different choice of the void finding algorithm , since the correlation properties are basically determined not by the shapes of voids but by the galaxies in voids .
another issue we would like to discuss here is a possibility of testing numerically our key assumption that a void forms in the initial underdense region where the misalignment between the principal axes of the inertia and the tidal tensors becomes maximum .
although @xcite has already shown analytically that our assumption is true , it would be highly desirable to test it against numerical data from the millennium run simulation .
for this , it would be necessary to find the lagrangian positions of all the particles that make up each void and to determine the inertia momentum tensor of each lagrangian proto - void region in the principal axis frame of the initial tidal field .
as the initial density field of the millennium run simulation is not available to the public yet , the numerical test of our assumption is postponed .
we hope to report the test result elsewhere in the future .
we thank the anonymous referee for helpful suggestions .
the millennium run simulation used in this paper was carried out by the virgo supercomputing consortium at the computing center of the max - planck society in garching .
the semi - analytic galaxy catalogue is publicly available at http://www.mpa-garching.mpg.de/galform/agnpaper .
this work is supported by the research grant no .
r01 - 2005 - 000 - 10610 - 0 from the basic research program of the korea science and engineering foundation .
100 benson , a. j. , hoyle , f. , torres , f. , vogeley , m. s. 2003 , , 340 , 160 bullock , j. s. , dekel , a. , kolatt , t. s. , kravtsov , a. v. , klypin , a. a. , porciani , c. & primack , j. r. 2001 , , 555 , 240 colberg j. m. , sheth r. k. , diaferio a. , gao l. , & yoshida n. 2005 , , 360 , 216 colless et al .
2001 , , 328 , 1039 davis , m. , efstathious , g. , frenk , c. s. , & white , s. d. m. , 1985 , , 292 , 371 doroshkevich , a. g. 1970 , astrofizika , 6 , 581 dubinski , j. , da costa , l. n. , goldwirth , d. s. , lecar , m. , & piran , t. 1993 , , 410 , 458 el - ad , h. , & piran , t. 1997 , , 491 , 421 friedmann , y. , & piran , t. 2001 , , 548 , 1 gottlber , s. , lokas , e. l. , klypin , a. , & hoffman , y. 2003 , , 344 , 715 hoyle , f. , rojas , r. , vogeley , m. s. , & brinkmann , j. 2005 , , 620 , 618 hoyle , f. , & vogeley , m. s. 2002 , , 566 , 641 hoyle , f. , & vogeley , m. s. 2004 , , 607 , 751 icke v. 1984 , , 206 , 1 lee , j. & pen , u. l. 2000 , , 532 , l5 lee , j. & pen , u. l. 2001 , , 555 , 106 lee , j. 2006 , , 644 , l5 little b. , & weinberg d. h. 1994 , mnras , 267 , 605 mathis , h. , & white , s. d. m. 2003 , , 337 , 1193 patiri , s. g. , betancort - rijo , j. e. , & prada , f. 2004 , preprint ( astro - ph/0407513 ) peebles , p. j. e. 2001 , , 557 , 495 pen , u.l . , lee , j. , & seljak , u. 2000 , , 543 , l107 plionis , m. , & basilakos , s. 2002 , , 330 , 399 porciani , c. , dekel , a. , & hoffman , y. 2002 , , 332 , 339 rojas , r. r. , vogeley , m. s. , hoyle , f. , & brinkmann , j. 2004 , , 617 , 50 rojas , r. r. , vogeley , m. s. , hoyle , f. , & brinkmann , j. 2005 , , 624 , 571 sahni , v. , & shandarin , s. 1996 , mnras , 282 , 641 schmidt , j. d. , ryden b. s. , & melott a. l. 2001 , , 546 , 609 shandarin , s. , feldman , h. a. , heitmann , k. , & habib , s. 2006 , , 367 , 1629 sheth , r. k. , & van de weygaert , r. 2004 , , 350 , 517 springel , v. et al .
2005 , , 435 , 629 strauss , et al .
2002 , , 124 , 1810 white , s. d. m. 1984 , , 286 , 38 van de weygaert , r. , & van kampen , e. 1993 , , 263 , 481 | we present a theoretical study of void spins and their correlation properties .
the concept of the spin angular momentum for an unbound void is introduced to quantify the effect of the tidal field on the distribution of matter that make up the void .
both the analytical and numerical approaches are used for our study .
analytically , we adopt the linear tidal torque model to evaluate the void spin - spin and spin - density correlations , assuming that a void forms in the initial region where the inertia momentum and the tidal shear tensors are maximally uncorrelated with each other .
numerically , we use the millennium run galaxy catalog to find voids and calculate their spin statistics .
the numerical results turn out to be in excellent agreement with the analytic predictions , both of which consistently show that there are strong spatial alignments between the spin axes of neighbor voids and strong anti - alignments between the void spin axes and the directions to the nearest voids .
we expect that our work will provide a deeper insight into the origin and properties of voids and the large scale structure . |
the evolutionary prisoner s dilemma game ( pdg ) @xcite and the snowdrift game ( sg ) @xcite have become standard paradigms for studying the possible emergence of cooperative phenomena in a competitive setting .
physicists find such emergent phenomena fascinating , as similar cooperative effects are also found in interacting systems in physics that can be described by some minimal models , e.g. models of interacting spin systems .
these games are also essential in the understanding of coexistence of ( and competition between ) egoistic and altruistic behavior that appear in many complex systems in biology , sociology and economics
. the basic pdg @xcite consists of two players deciding simultaneously whether to cooperate ( c ) or to defect ( d ) .
if one plays c and the other plays d , the cooperator pays a cost of @xmath0 while the defector receives the highest payoff @xmath1 ( @xmath2 ) .
if both play c , each player receives a payoff of @xmath3 .
if both play d , the payoff is @xmath4 .
thus , the pdg is characterized by the ordering of the four payoffs @xmath5 , with @xmath6 . in a single round of the game , it is obvious that defection is a better action in a fully connected ( well - fixed ) population , regardless of the opponents decisions .
modifications on the basic pdg are , therefore , proposed in order to induce cooperations and to explain the wide - spread cooperative behavior observed in the real world .
these modifications include , for example , the iterated pdg @xcite , spatially extended pdg @xcite and games with a third strategy @xcite .
the snowdrift game ( sg ) , which is equivalent to the hawk - dove or chicken game @xcite , is a model somewhat favorable for cooperation .
it is best introduced using the following scenario @xcite .
consider two drivers hurrying home in opposite directions on a road blocked by a snowdrift .
each driver has two possible actions to shovel the snowdrift ( cooperate ( c ) ) or not to do anything ( not - to - cooperate or defect " ( d ) ) .
if they cooperate , they could be back home earlier and each will get a reward of @xmath7 . shovelling is a laborious job with a total cost of @xmath8 .
thus , each driver gets a net reward of @xmath9 .
if both drivers take action d , they both get stuck , and each gets a reward of @xmath4 .
if only one driver takes action c and shovels the snowdrift , then both drivers can also go home .
the driver taking action d ( not to shovel ) gets home without doing anything and hence gets a payoff @xmath10 , while the driver taking action c gets a sucker " payoff of @xmath11 .
the sg refers to the case when @xmath12 , leading to the ranking of the payoffs @xmath13 .
this ordering of the payoffs _ defines _ the sg .
therefore , both the pdg and sg are defined by a payoff matrix of the form @xmath14 and they differ only in the ordering of @xmath15 and @xmath16 .
it is this difference that makes cooperators persist more easily in the sg than in the pdg . in a well - mixed population ,
cooperators and detectors coexist . due to the difficulty in measuring payoffs and the ordering of the payoffs accurately in real world situations where game theory is applicable @xcite , the sd has been taken to be a possible alternative to the pdg in studying emerging cooperative phenomena @xcite .
the present work will focus on two aspects of current interest . in many circumstances ,
the connections in a competing population are better modelled by some networks providing limited interactions than a fully - connected network .
previous studies showed that different spatial structures might lead to different behaviors @xcite .
for example , it has been demonstrated that spatial structures would promote cooperation in the pdg @xcite , but would suppress cooperation in the sg @xcite .
there are other variations on the sg that resulted in improved cooperation @xcite . here
, we explore the effects of an underlying network on the evolutionary sg in a population in which there exists an additional type of players .
the latter is related to the fact that real - world systems usually consist of people who would adopt a strategy other than just c and d. for example , there may be people who do not like to participate in the competition and would rather take a small but fixed payoff .
_ studied the effects of the presence of such persons , called loners @xcite , in a generalization of the pdg called the public goods game(pgg ) .
motivated by these works of hauert _ et al . _
@xcite , we study the effects of risk averse loners in the evolutionary sg . in our model , evolution or adaptation is built in by allowing players to replace his character or strategy by that of a better - performing connected neighbor .
we focus on both the steady state and the dynamics , and study how an underlying network structure affects the emergence of cooperation .
it is found that in a fully - connected network , the c - players and d - players _ can not _ coexist with the loners . in a square lattice , however , cooperators are easier to survive . depending on the payoffs , there are situations in which c - players , d - players and loners can coexist . in sec . [ sec : model ] , the evolutionary sg with loners in a population with connections is presented . in sec .
[ sec : simulation results and discussions ] , we present detailed numerical results in fully - connected networks and in square lattices , and discuss the physics of the observed features .
the effects of noise are also discussed .
we summarize our results in sec . [
sec : conclusion ] .
we consider an evolutionary snowdrift game in which the competitions between players are characterized by the payoff matrix @xmath17 here , each player takes on one of three possible characters or strategies : to cooperate ( c ) , to defect ( d ) , or to act as a loner ( l ) .
the matrix element gives the payoff to a player using a strategy listed in the left hand column when the opponent uses a strategy in the top row . in the basic sg , it is useful to assign @xmath18 so that the payoffs can be characterized by a single parameter @xmath19 representing the cost - to - reward ratio @xcite . in terms of @xmath20 , we have @xmath21 , @xmath18 , @xmath22 , and @xmath4 . a competition involving
a loner leads to a payoff @xmath23 for both players . here
, we explore the range of @xmath24 .
spatial networking effects and evolutions are incorporated into the sg as follows . at the beginning of the game ,
the players are arranged onto the nodes of a network and the character @xmath25 of each player is assigned randomly among the choices of c , d , and l. our discussion will be mainly on fully - connected graphs and regular lattices . in a fully - connected network ,
every player is connected to all other players . in a square lattice ,
a player is linked only to his four nearest neighbors .
numerical studies are carried out using monte carlo simulations as reported in the work of szab _ et al . _
@xcite ( see also refs .
the evolution of the character of the players is governed by the following dynamics . at any time during the game , each player competes with all the players that he is linked to and hence has a payoff . a randomly chosen player @xmath26 reassesses his own strategy by comparing his payoff @xmath27 with the payoff @xmath28 of a randomly chosen connected neighbor @xmath29 . with probability @xmath30 = \frac{1}{1 + \exp\left([p(i ) - p(j)]/k\right)},\ ] ] the player @xmath26 adopts the strategy of player @xmath29 .
otherwise , the strategy of player @xmath26 remains unchanged . here
@xmath31 is a noise parameter @xcite that determines the likelihood that player @xmath26 replaces his strategy when he meets someone with a higher payoff . for @xmath32 ,
a player @xmath26 is almost certain to replace ( not to replace ) his strategy when he meets someone with a better ( worse ) payoff . for large @xmath31
, a player has a probability of @xmath33 to replace his strategy , regardless of whether @xmath28 is better or worse than @xmath27 . in a fully connected network ,
a player s character may be replaced by any player in the system . in a square lattice ,
a player s character can only be replaced by one of his four connected neighbors .
as the game evolves , the fractions of players with the three characters also evolve .
these fractions are referred to as frequencies .
depending on the parameters @xmath34 and @xmath23 , the cooperator frequency @xmath35 , defector frequency @xmath36 , and loner frequency @xmath37 take on different values in the long time limit .
we performed detailed numerical studies on our model . the number of players in the system is taken to be @xmath38 . in getting the fraction of players of different characters in the long time limit
, we typically average over the results of @xmath39 monte - carlo time steps per site ( mcs ) , after allowing @xmath40 mcs for the system to reach the long time limit .
averages are also taken over 10 initial configurations of the same set of parameters .
figure 1 shows the results for fully connected networks .
a value of @xmath41 is taken .
the cooperator frequency @xmath35 , defector frequency @xmath36 , and loner frequency @xmath37 are obtained as a function of the cost - to - benefit ratio @xmath34 for three different values of the loner s payoff @xmath42 , @xmath43 , and @xmath44 . in the absence of loners @xcite , @xmath45 and @xmath46 in a fully connected network . from figure 1 , the loners extinct for a range of values of @xmath47 in which the behavior is identical to the basic sg . for @xmath48 ,
the loners invade the whole population and both cooperators and defectors disappear .
this is similar to the results in the pdg @xcite and in the pgg @xcite . in a fully connected network ,
the three characters _ can not _ coexist .
this is in sharp contrast to the rock - scissors - paper game @xcite on a fully connected network in which the three strategies coexist .
we obtained @xmath49 numerically .
the result is shown in figure 1(d ) as a curve in the @xmath34-@xmath23 parameter space .
it is found that @xmath49 follows the functional form @xmath50 , which will be explained later .
the curve @xmath49 represents a phase boundary that separates the @xmath34-@xmath23 space into two regions .
the region below ( above ) the curve corresponds to a phase in which cooperators and defectors ( only loners ) coexist ( exist ) .
we also studied the temporal evolution in both phases , i.e. , for @xmath47 and @xmath48 .
taking @xmath51 , for example , @xmath52 .
figure 2 shows @xmath53 , @xmath54 and @xmath55 in the first @xmath39 mcs .
the initial frequencies are @xmath56 for all three characters . for values of @xmath34 deep into either phase ( see fig .
2 ) , the transient behavior dies off rapidly and the extinct character typically vanishes after @xmath57 mcs . in the phase where c and d coexist ,
@xmath35 and @xmath36 oscillate slightly with time in the long time limit , due to the dynamical nature of the game .
it is noted that for @xmath58 , the strategies compete for a long while and the transient behavior lasts for a long time .
this slowing down behavior is typical of that near a transition .
the behavior of @xmath59 follows from the rule of character evolution . in a fully - connected network ,
all c - players have the _ same _ payoff @xmath60 and all d - players have the _ same _ payoff @xmath61 .
these payoffs depend on @xmath35 , @xmath36 , and @xmath37 at each time step .
the payoff for a loner is @xmath62 at all time , for a system with @xmath63 . for small @xmath31
, @xmath37 decays exponentially with time if @xmath60 and @xmath61 are both greater than @xmath62 .
in addition , the phase with only non - vanishing @xmath35 and @xmath36 is achieved by having @xmath64 . for this phase in the long time limit , @xmath65 and @xmath66 . together with @xmath67 ( since @xmath68 in the phase under consideration ) , the condition @xmath64 implies @xmath69 and @xmath46 .
these results are identical to the basic sg ( without loners ) in a fully connected network .
the validity of this solution requires @xmath70 ( and hence @xmath71 ) , which is equivalent to @xmath72 .
this is exactly the phase boundary shown in figure 1(d ) .
the behavior of the game in a square lattice is expected to be quite different , due to the restriction that a player can only compete with his connected neighbors .
we carried out simulations on @xmath73 square lattices with periodic boundary conditions .
figure 3(a)-(c ) shows @xmath74 , @xmath75 and @xmath76 for three different values of the loner payoff @xmath23 .
the results for the spatial sg ( without loners ) on a square lattice @xcite is also shown ( solid lines in figure 3(a ) and 3(b ) ) for comparison .
a value @xmath41 is used .
several features should be noted . for @xmath77 ,
the loners eventually vanish with @xmath35 and @xmath36 take on the mean values in the spatial sg without loners .
this behavior is similar to that in fully connected networks . for @xmath78 , however , the behavior is different from that in fully connected networks . here ,
c , d , and l characters coexist . above @xmath79 , @xmath36 drops with @xmath34 to a finite value , leaving rooms for @xmath37 to increase with @xmath34 .
the cooperator frequency remains finite above @xmath79 .
therefore , the cooperator frequency or the cooperative level in the system as a whole is significantly _ improved _ by the presence of loners . for @xmath80 , increasing the payoff @xmath23 of loners leads to a higher cooperator frequency and lower defector frequency . reading out @xmath79 for different values of @xmath23 , we get the phase boundary as shown in figure 3(d ) that separates a region characterized by the coexistence of three characters and a region in which only c and d coexist .
the results indicate that , due to the restriction imposed by the spatial geometry that a player can only interact with his four nearest neighbors , it takes a certain non - vanishing value of @xmath34 for loners to survive even in the limit of @xmath81 .
the behavior is therefore different from that in a fully connected network for which the boundary is given by @xmath50 .
note that there exists a region of small values of @xmath34 in which the steady state consists of a uniform population of c strategy ( see fig .
3(a ) and fig .
3(d ) ) . for small @xmath23 ,
loners are easier to survive , when compared with the fully connected case . putting these results together , the phase diagram ( see fig .
3(d ) ) for a square lattice , therefore , shows three different phases .
the most striking effect of the spatial structure is that cooperators now exist in every phase .
interestingly , we found that the phase boundary @xmath82 in figure 3(d ) can be described quantitatively as follows .
we _ assume _ that the survival of loners is related to the cooperator frequency .
in particular , loner survival requires the cooperator frequency to drop below a certain level @xmath83 and that this value is the same in a square lattice as in a fully connected network .
that is to say , we assume that loners could survive , for a given value of @xmath23 and @xmath31 , only when @xmath84 . numerical results also indicate that when all loners extinct , @xmath35 and @xmath36 follow the results in a spatial sg without loners .
this is shown as the solid line in figure 3(a ) . therefore ,
for a given value of @xmath23 , we can simply read out the value of @xmath34 such that @xmath85 from the results in spatial sg in a square lattice . for different values of @xmath23 , this procedure results in the dashed line shown in figure 3(d ) which describes the phase boundary quite well .
figure 4 shows the temporal dependence of @xmath35 , @xmath36 , and @xmath37 in a square lattice for two values of @xmath34 at @xmath51 . for @xmath86 ( fig .
4(a ) ) , which corresponds to a case in which only cooperators and defectors coexist , the number of loners decay rapidly in time , typically within 100 mcs . after the transient behavior
, the cooperator and defector frequencies only oscillate slightly about their mean values .
this behavior is similar to that in the c and d coexistence phase in figure 1(d ) for fully - connected networks . for @xmath87 ( fig .
4(b ) ) , which corresponds to a case with the three characters coexist , the long time behavior of @xmath35 , @xmath36 and @xmath37 is oscillatory .
similar behavior has been found in the rock - scissors - paper game @xcite and in the voluntary pdg @xcite . due to the dynamical nature of character evolution
, there are continuous replacements of one character by another and this oscillatory behavior is expected .
the major difference between a square lattice and a fully - connected network is that in a fully - connected network , each player competes with all other players . as a result , there are only three payoffs in the system
one for each type of player , at each time step .
the loners , for example , have a constant payoff of @xmath62 , while the cooperators and defectors have payoffs that depend on @xmath53 and @xmath54 .
once @xmath62 is higher than the payoffs of cooperators and defectors , the number of loners grows until they take over the whole population . in a square lattice
, however , each player has a payoff that depends on his character _ and _ the detail of his neighborhood , i.e. , the characters of his four connected neighbors .
this implies that the c - players and d - players in a square lattice may have different payoffs depending on the characters of his connected neighbors .
the loners have a constant payoff of @xmath88 .
the non - uniform payoffs among c - players and d - players in a lattice allow some c and d players to coexist with the loners , by evolving to spatial local configurations that favor their survivals . since the adaptive rule is related to the payoff of each character , it will be interesting to compare the payoffs in a spatial sg without and with loners .
figure 5(a ) shows the mean payoffs of cooperators and defectors as a function of @xmath34 in a spatial sg in a square lattice _ without _ loners .
the averaged payoff over all players is also shown . for small @xmath34 ,
there is a phase with all c players and the payoff is 4 for each of the c players . for large @xmath34
, there is a phase with all d players and the payoff is zero . for intermediate @xmath34 where c and d players coexist ,
the mean payoff drops gradually with @xmath34 . in a spatial sg _ with _
loners ( fig .
5(b ) ) , it is observed that the mean payoffs basically follow that in figure 5(a ) in the phase where loners are completely replaced .
when loners can survive , the presence of these loners increases the payoffs of both the remaining cooperators and defectors .
the loners themselves have a payoff of @xmath89 in a 2d square lattice .
the cooperators payoff is enhanced once loners survive and the increase follows the same form as the increase in the loner frequency with @xmath34 ( compare the circles in fig .
5(b ) with the squares in fig .
3(c ) in the range of @xmath34 when loners survive ) .
when loners survive , the payoff averaged over all players is significantly _ enhanced _ due to their presence .
this is similar to what was found in the voluntary pdg @xcite .
all the results reported so far are for the case of @xmath41 .
this corresponds to a case where the player is highly likely to replace his character when he meets a better - performing player . in figure 6
, we show the effects of the noise parameter for a fixed @xmath42 . as @xmath31 increases
, the step - like structure in @xmath35 as a function of @xmath34 becomes less obvious and @xmath35 is gradually suppressed in the @xmath90 limit .
the most important effect of a 2d square lattice is that each player is restricted to interact with his four neighbors .
take a player of character @xmath25 , he will only encounter a finite number of configurations for which he is competing in .
for example , his four neighbors may consist of 4 c - players ; 3 c - players and 1 d - player or 1-loner , etc .
each of these configurations corresponds to a @xmath27 . in a square lattice , therefore , there will be a finite number of payoffs for a c - player , depending on the characters of the neighbors .
similarly , there are finite number of payoffs for a d - player .
the loners always get a payoff of @xmath88 .
for @xmath91 , the adaptive mechanism is strictly governed by the ordering of these payoffs .
the distribution of players in a square lattice will then evolve in time according to how the payoffs are ordered . in the long time
limit , only a few favorable local configurations will survive and the number of players in each of these favorable configurations is high . as one increases @xmath34 slightly , the ordering of the finite number of payoffs may not change .
therefore , @xmath35 will not change with @xmath34 until we reach certain values of @xmath34 that the ordering of the payoffs is changed .
this gives rise to the more sudden changes in @xmath35 as observed at some values of @xmath34 and it is the reason for having step - like features in @xmath35 and @xmath36 for small values of @xmath31 .
as the noise parameter @xmath31 increases , the adaptive mechanism is less dependent on the exact ordering of the payoffs .
therefore , the changes in @xmath35 with @xmath34 becomes more gradual as @xmath31 increases .
interestingly , less obvious step - like structures in @xmath35 are also observed in the spatial sg without loners in 2d lattices with a larger coordination number @xcite .
this is also related to the picture we just described .
a lattice with more neighbors will give a higher number of neighborhood configurations and hence more values of the payoffs .
more configurations also imply the number of players encountering a certain configuration is smaller .
thus , the number of players involved in a change in the ordering of the payoffs as @xmath34 changes is smaller .
this has the effect of making the drop in @xmath35 gradual .
therefore , increasing @xmath31 for a given fixed coordination number is similar in effect as increasing the coordination number for fixed @xmath31 .
we studied the effects of the presence of loners in a snowdrift game with loners in fully - connected networks and in square lattices . in a fully - connected network , either cooperators live with defectors or loners take over the whole population .
the condition for loners to take over is found to be @xmath92 .
this result can be understood by following the payoffs of each strategy . in a fully - connected network ,
the strategies payoffs are particularly simple in that they depend only on the strategy frequencies at the time under consideration , with each type of player having the same payoff .
in a square lattice , the spatial sg with loners behave quite differently .
it is found that the cooperators can survive in the fully parameter space covering @xmath93 and @xmath24 .
depending on the values of these parameters , there are three possible phases : a uniform c - player population , c - players and d - players coexist , and coexistence of the three characters .
the underlying lattice thus makes the survival of cooperators easier .
the presence of loners is also found to promote the presence of cooperators .
there average payoff among all players is also found to be enhanced in the presence of loners .
we discussed the influence of a square lattice in terms of the payoffs of the players . in a square lattice ,
spatial restriction is imposed on the players in that a player can only interact with the four nearest neighbors .
this leads to a payoff that does not only depend on the character but also depend on the local environment in which the player is competing in .
the players in the local environment , in turns , are also competing in their own local environment .
this will lead to clustering or aggregation of players in the square lattice into configurations that the payoffs favored .
the dependence of the frequencies on @xmath34 in a square lattice then reflects the change in preferred configurations as @xmath34 is changed .
we also studied the effects of the noise parameter in the adaptive mechanism .
it is found that as the noise parameter increases , the change of the frequencies with @xmath34 becomes more gradual .
this is related to the importance of the ordering of the many payoffs in the adaptive mechanism .
as the noise parameter increases , the exact ordering of the payoffs becomes less important and the change in frequencies becomes more gradual . in closing , we note that it will be interesting to further investigate the effects of loners in the snowdrift game in networks of other structures . among them
are the re - wiring of regular lattices into a small - world network or a random network and the scale - free networks @xcite .
this work was supported in part by the national natural science foundation of china under grant nos .
70471081 , 70371069 , and 10325520 , and by the scientific research foundation for the returned overseas chinese scholars , state education ministry of china .
one of us ( p.m.h . ) acknowledges the support from the research grants council of the hong kong sar government under grant no .
cuhk-401005 and from a direct research grant at cuhk . | the effects of an additional strategy or character called loner in the snowdrift game are studied in a well - mixed population or fully - connected network and in a square lattice .
the snowdrift game , which is a possible alternative to the prisoner s dilemma game in studying cooperative phenomena in competing populations , consists of two types of strategies , c ( cooperators ) and d ( defectors ) . in a fully - connected network , it is found that either c lives with d or the loners take over the whole population . in a square lattice ,
three possible situations are found : a uniform c - population , c lives with d , and the coexistence of all three characters .
the presence of loners is found to enhance cooperation in a square lattice by enhancing the payoff of cooperators .
the results are discussed in terms of the effects in restricting a player to compete only with his nearest neighbors in a square lattice , as opposed to competing with all players in a fully - connected network . |
psr j0205 + 6449 in supernova remnant ( snr ) 3c58 is a young energetic pulsar detected at x , @xmath1-ray , and radio wavelengths @xcite .
it has a spin period @xmath765 ms and the third highest spin down energy flux , @xmath8 erg s@xmath9 kpc@xmath10 ( where @xmath11 is the pulsar distance ) , after the crab and vela pulsars .
3c58 was thought to be young , associated with supernova sn 1181 , @xcite , and consequently should share many of the characteristics of the crab nebula , including the presence of a young pulsar .
yet its pulsar , psr j0205 + 6449 , defied detection for over twenty years .
it was only about ten years ago that it was discovered as an x - ray pulsar by _
( murray et al . 2002 ) , while its identification as a radio pulsar came soon after ( camilo et al .
was also detected in the hard x - rays ( kuiper et al .
2010 ) by the _ high energy x - ray timing experiment _
( _ hexte _ ) aboard the rossi x - ray timing explorer ( _ rxte _ ) and was also one of the first to be identified as a @xmath1-ray pulsar by _ fermi _ ( abdo et al .
psr j0205 + 6449 is clearly located within a pulsar wind nebula ( pwn ) , detected both in the x - rays ( slane et al . 2002 ; 2004 ) ,
optical , and near - infrared ( slane et al . 2008 ) . from the x - ray determined hydrogen column density ( marelli et al . 2011
; marelli 2012 ) , and using the relation , we can estimate an @xmath12 - 2.5 , similar to the crab pulsar .
the distance to , however , is possibly larger than the crab .
hi radio observations gives a distance to 3c58 of 3.2 kpc , lower but still consistent with the value of @xmath13 kpc obtained from the pulsar dispersion measure of @xmath14 @xmath15 pc @xcite and the electron distribution model of @xcite .
if the crab and 3c58 snrs were of the same age , we might also expect similarities between their pulsars .
however , psr j0205 + 6449 was found to be considerably weaker than the crab pulsar ; its x - ray emission is 1000 times lower than the crab and its radio emission is 120 times lower . in @xmath1-rays , psr j0205 + 6449
s luminosity is about 10% of the crab pulsar s but its efficiency , for an assumed 3.2 kpc distance , is about 0.20.3% compared to the crab s
0.1% @xcite consistent with psr j0205 + 6449 being an older pulsar .
some of this discrepancy can be attributed to the mounting evidence that the characteristic age of the pulsar ( given by @xmath16 @xmath17 5400 years ) is near to its true age @xcite , with the age estimates for the 3c58 snr ranging between 3000 and 5100 years , thus breaking the association between supernova sn1181 and /3c58 .
recently , however , a re - analysis of the existing hi radio observations ( kothes 2010 ) suggests a distance as small as 2 kpc for 3c 58 and an age as low as @xmath0 1000 years , reopening the debate on its association with sn1181 . the multi - wavelength properties , age , and energetics of combine to make it a likely candidate for optical emission studies .
if we assume that the pulsar s optical luminosity scales with the light cylinder magnetic field @xmath18 @xcite , then we estimate that it should have a visual magnitude in the range 2325 , depending on interstellar absorption and effects of beaming geometry .
the first deep observations of 3c58 showed evidence of an optical nebulosity at the same location as the x - ray counterpart to psr j0205 + 6449 but could not resolve the pulsar optical counterpart .
the motivation behind this work was to identify a candidate optical counterpart to psr j0205 + 6449 using observations taken under the 2009 international time programme at the la palma observatory ( pi : a. shearer ) , combined with observations recently obtained with the 10.4 m gran telescopio canarias ( gtc ) as well as archival gemini and _ hubble space telescope _
( _ hst _ ) images . table 1 reports the summary of the observations with the different facilities .
we first observed the field with the 2.5 m uk isaac newton telescope ( int ) at the la palma observatory ( roque de los muchachos , canary islands , spain ) on 2009 august 3 .
we used the wide field camera ( wfc ) , a mosaic of 4 thinned eev ccds , with an unbinned pixel size of 033 and with a field of view of @xmath19 , including the @xmath0 1gap between the chips .
we only observed through the harris r - band filter ( @xmath20 ; @xmath21 ) for a total integration time of 7320 seconds , with an average airmass of 1.35 and a seeing of 13 .
observations were performed in grey time , with the moon at an angular distance of @xmath22 but mostly below the horizon at the moment of the observations .
the integration was split in shorter dithered exposures to remove cosmic ray hits and compensate for the gaps between the ccd chips .
on 2009 august 20 and 21 , we re - observed the field with the 3.5 m italian telescopio nazionale galileo ( tng ) , also at the la palma observatory .
we used the dolores ( device optimized for the low resolution ) camera , a single chip e2v ccd with a field of - view of @xmath23 and a pixel size of 0252 .
we observed through the standard johnson v ( @xmath24 ; @xmath25 ) and r ( @xmath26 ; @xmath27 ) filters for total integration times of 6060 and 4830 seconds , respectively . in both cases ,
the average airmass was around 1.5 and the seeing @xmath28 .
observations were performed in dark time .
for both the int and tng images , we applied standard data reduction ( bias subtraction and flat - fielding ) using the tools in the iraf packages mscred and ccdred .
the dithered exposures were then aligned , stacked , and filtered for cosmic - rays so as to produce mosaic images .
we obtained additional observations of the field with the gtc at the la palma observatory on 2011 september 1 and 2011 november 20 as part of the spanish time programme ( pi : n. rea ) .
we observed with the optical system for imaging and low resolution integrated spectroscopy ( osiris ) .
the instrument is equipped with a two - chip e2v ccd detector with a nominal field of view of @xmath29 , which is actually decreased to @xmath30 due to the vignetting of one of the two chips .
the unbinned pixel size of the ccd is 0125 .
we took a sequence of dithered exposures in the sdss r ( @xmath31 ; @xmath32 ) and i ( @xmath33 ; @xmath34 ) bands on the first and second night , respectively , with exposure time of 140 seconds to minimise the saturation of bright stars in the field and correct for the fringing .
the pulsar was positioned at the nominal aim point in chip 2 .
observations were performed with an average airmass of 1.25 for both the r and i bands .
seeing conditions were @xmath35 and @xmath3610 for the first and second night , respectively .
in both nights , observations were performed in dark time and under clear conditions . for the i-band observations ,
no valid sky - flats were taken for the night of november 20 , therefore we used closest in time sky - flats taken on november 17 .
as done for the int and tng data , we reduced the data using standard tools in the iraf package ccdred .
we then stacked and averaged the single dithered exposures using the task drizzle that also performs the cosmic - ray filtering .
.observation log for the psr j0205 + 6449 observations [ cols="<,^,^,>,^ " , ] [ comp1 ] we computed the ratio between the x - ray and optical fluxes of from the available _ chandra _ and gemini measurements
. the spectral fit of marelli ( 2012 ) , based on a pl plus bb model , gives an unabsorbed x - ray flux @xmath37 erg cm@xmath10 s@xmath9 in the 0.310 kev energy range for the pl component only , whereas the total x - ray flux is @xmath39 erg cm@xmath10 s@xmath9 .
hereafter , we assume the total x - ray flux as a reference . on the basis of the gemini g detection , and using the same extinction correction as above
, we computed the unabsorbed flux of the pulsar in the g band would be @xmath40 erg cm@xmath10 s@xmath9 . by assuming a distance of 3.2 kpc , the unabsorbed fluxes give g-band and x - ray luminosities of @xmath41 erg s@xmath9 and @xmath42 erg s@xmath9 .
this gives an optical to x - ray luminosity ratio @xmath43 .
similarly , we computed the unabsorbed optical to@xmath1-ray luminosity ratio . as a reference
, we assumed the @xmath1-ray flux above 100 mev measured by _
fermi _ , @xmath44 erg cm@xmath10 s@xmath9 ( abdo et al .
for the assumed distance of 3.2 kpc the @xmath1-ray luminosity of is @xmath45 erg s@xmath9 .
this gives a luminosity ratio @xmath46 .
the computed optical , x - ray , and @xmath1-ray luminosities of , together with their ratios , are listed in table 3 , where they are compared with the corresponding values measured for the other @xmath1-ray pulsars identified in the optical . as a reference , for all pulsars we assumed the same pulsar distances as used in abdo et al .
( 2010 ) , the unabsorbed x - ray fluxes in the 0.310 kev energy range listed in marelli ( 2012 ) , and the @xmath1-ray fluxes above 100 mev listed in abdo et al .
we computed the optical luminosities according to the observed v - band magnitudes and extinction @xmath47 ( mignani 2011 ) .
as seen , would be the @xmath1-ray pulsar with the second highest optical luminosity after the crab pulsar , as expected from its low spin - down age ( @xmath48 5400 year ) .
the optical to x - ray and optical
to@xmath1-ray luminosity ratios of would be consistent with those measured for other young pulsars ( @xmath49 kyr ) , with the crab pulsar being that with the highest ratios .
however , the luminosity ratios for would be larger than those of the slightly older vela pulsar ( @xmath50 kyr ) , owing to the fact that the latter is about two orders of magnitude fainter in the optical .
we also computed the ratio between the derived optical luminosity @xmath51 erg s@xmath9 of and its rotational energy loss @xmath52 erg s@xmath9 . assuming a 3.2 kpc distance
, we obtained @xmath53 .
again , in table [ comp1 ] this value is compared with those of the other @xmath1-ray pulsars identified in the optical domain .
interestingly enough , the optical efficiency for , defined as @xmath54 , would be lower than for the other young pulsars ( crab and psr b1509@xmath658 ) but higher than the vela pulsar , confirming a trend for a decrease in the optical emission efficiency of young pulsars as a function of the spin - down age .
the existence of such a trend has been already proposed ( e.g. zharikov et al . 2006 ) , but , so far , the assumption only relied on the computed lower emission efficiency of the vela pulsar with respect to the other young pulsars crab , psr b1509@xmath658 , and psr b0540@xmath669 , with the latter ( @xmath55 ) not listed in table 3 because it has not been yet detected in @xmath1-rays .
thus , it has been unclear whether the optical emission efficiency of vela - like pulsars was indeed intrinsically lower than the crab - like ones , or vela stood out as a peculiar case .
the probable optical identification of the @xmath05400 year - old , which is ideally half way between the two classes , now represents an important piece of evidence in favour of this interpretation .
the upper limits on the optical luminosities of vela - like @xmath1-ray pulsars , such as psr b1706@xmath644 , psr j1357@xmath66429 , psr j1028@xmath65819 , psr j0007 + 7303 ( mignani et al . 1999 ; 2011 ; 2012b ; 2013 ) and psr j1048@xmath65832 ( razzano et al . 2013 ) , are consistent with this trend . of course
, the optical identification of some of these pulsars will eventually give the long sought proof .
such a trend can be interpreted as the result of the secular decrease of the pulsar non - thermal optical luminosity , an effect predicted by pacini & salvati ( 1983 ) as a result of the pulsar spin down .
we note that the optical emission ( non - thermal ) efficiency tends to increase again for the middle - aged pulsars ( psr b0656 + 14 and geminga ) .
we performed multi - band optical observations of with a variety of ground - based facilities , including the 8 m gemini and 10.4 m gtc telescopes , and the _ hst_. we detected a possible candidate counterpart to the pulsar , with an i-band magnitude of 25.5 ( 5.5@xmath3 detection ) , based upon its positional coincidence with the recently measured _
chandra _ and radio coordinates ( bietenholz et al .
thus , would possibly be the fourteenth pulsar with either an identified or proposed optical counterpart ( mignani 2011 ) , and the eighth of the @xmath1-ray pulsars detected by _ fermi_.
the pulsar s spectrum would be consistent with a pl with photon index @xmath4 , similar to the x - ray one ( @xmath5 ) .
the multi - wavelength sed of show that the optical fluxes would lie below and above the extrapolations in the optical domain of the x and @xmath1-ray pl , respectively , indicating a break in the optical / x - ray region , after that in the x/@xmath1-ray .
the optical luminosity , @xmath56 erg s@xmath9 , would imply an emission efficiency @xmath57 , higher than the two times older vela pulsar , confirming a trend for a decrease of the optical emission efficiency with the spin - down age in young pulsars . due to the presence of a bright pwn ,
deeper high spatial resolution observations with the _ hst _ are better suited to resolve the presence of a point source , confirm the optical identification of , and better study its spectrum . moreover , monitoring for changes in the flux of the candidate counterpart on a few month / year time scale would allow one to rule out the possibility that it is an emission knot in the pwn .
based on observations made with the gran telescopio canarias ( gtc ) , installed in the spanish observatorio del roque de los muchachos of the instituto de astrofsica de canarias , in the island of la palma . the research leading to these results has received funding from the european commission seventh framework programme ( fp7/2007 - 2013 ) under grant agreement n. 267251 .
pm is grateful for his funding from the irish research council ( irc ) .
we are grateful to the anonymous referee whose comments improved the quality of the final version of the paper .
fesen r. , kirshner r. p. , becker r. h. , 1988 , in roger r. s. , landecker t. l. , eds .
, iau colloq . 101 : supernova remnants and the interstellar medium , 3c58 s filamentary radial velocities , line intensities , and proper motions , p. 55 | is a young ( @xmath0 5400 years ) , crab - like pulsar detected in radio and at x and @xmath1-ray energies and has the third largest spin - down flux among known rotation - powered pulsars .
it also powers a bright synchrotron nebula detected in the optical and x - rays . at a distance of @xmath2 kpc and with
an extinction comparable to the crab , is an obvious target for optical observations .
we observed with several optical facilities , including 8 m class ground - based telescopes , such as the gemini and the gran telescopio canarias .
we detected a point source , at a significance of 5.5@xmath3 , of magnitude [email protected] , at the centre of the optical synchrotron nebula , coincident with the very accurate _
chandra _ and radio positions of the pulsar .
thus , we discovered a candidate optical counterpart to .
the pulsar candidate counterpart is also detected in the g ( @xmath027.4 ) band and weakly in the r ( @xmath026.2 ) band .
its optical spectrum is fit by a power law with photon index @xmath4 , proving that the optical emission if of non - thermal origin , is as expected for a young pulsar .
the optical photon index is similar to the x - ray one ( @xmath5 ) , although the optical fluxes are below the extrapolation of the x - ray power spectrum .
this would indicate the presence of a double spectral break between the x - ray and optical energy range , at variance with what is observed for the crab and vela pulsars , but similar to the large magellanic cloud pulsar psr b0540@xmath669 .
[ firstpage ] pulsars ; individual psr j0205 + 6449 |
understanding the formation of disks around young low - mass protostars is an important target for studies of star formation .
although rotationally supported protostellar disks have been found in class i sources , they have been reported in only a few class 0 sources ( tobin et al 2012 ; yen et al .
2013 ; murillo et al . 2013 ; lindberg et al .
even absence of such a disk down to @xmath5 au is claimed for the class 0 protostar ngc 1333 iras 2a ( maret et al .
2014 ; brinch et al . 2009 ) .
hence , disk formation in the class 0 stage is still controversial .
this is because disks are generally difficult to identify in class 0 sources due to the overwhelming emission from protostellar envelopes and outflows .
furthermore , the disk structure in class 0 sources is expected to be small , and high spatial resolution and high sensitivity observations are essential .
iras 153983359 is a low - mass class 0 protostar in the lupus 1 molecular cloud at a distance of @xmath6 pc ( lombardi et al .
a molecular outflow was detected by single - dish observations of co emission ( tachihara et al .
1996 ; van kempen et al .
based on the relatively large overlap between the blue and red lobes of the outflow ( co @xmath7@xmath8 ) observed with the james clerk maxwell telescope ( @xmath9 ) , the outflow was thought to have a pole - on geometry ( van kempen et al .
mardones et al .
( 1997 ) observed h@xmath1co and cs lines toward this source to search for signs of infall without success . on the other hand , kristensen et al .
( 2012 ) reported the presence of an inverse p - cygni profile of the h@xmath1o @xmath10@xmath11 at @xmath12 ghz , indicating an infalling motion of the envelope on a scale of about @xmath13 au .
recently , jrgensen et al . (
2013 ) detected a ring structure of the h@xmath14co@xmath15@xmath16 emission on @xmath17@xmath18 au scales .
they proposed that the ring structure is caused by the destruction of hco@xmath19 through reactions with h@xmath1o that is evaporated by the enhanced luminosity due to a recent accretion burst .
iras 153983359 also shows peculiar chemical features in other ways .
various carbon - chain molecules such as cch , c@xmath20h , and ch@xmath21cch are detected in the vicinity of the protostar , which is characteristic of so - called warm carbon - chain chemistry ( wccc ) sources ( sakai et al .
2008 , 2009 ; sakai & yamamoto 2013 ) .
recently , the kinematic structure of another wccc source , l1527 , was resolved in atacama large millimeter / submillimeter array ( alma ) observations , and was well reproduced by a model of an infalling rotating envelope ( sakai et al .
the radius of the centrifugal barrier , at which all the kinetic energy is converted to the rotational energy , was determined by comparing the observations with the model . from this radius ,
the protostellar mass was evaluated .
since both l1527 and iras 153983359 are wccc sources , it is interesting to investigate whether the envelope of iras 153983359 has a similar structure . to explore when and how a rotationally supported disk is formed around protostars ,
a deep insight into the structure of the envelope of class 0 protostars is useful . with these motivations
, we conducted alma observations toward iras 153983359 in several molecular lines .
observations of iras 153983359 were carried out with alma in cycle 0 operations on 2012 december 31 .
spectral lines of h@xmath1co and cch were observed with the band 7 receiver at frequencies of @xmath22@xmath23 ghz , and @xmath24 ghz .
the spectral line parameters are listed in table @xmath25 .
twenty - five antennas were used in the observations , where the baseline length ranged from @xmath26 to @xmath27 m. the field center of the observations was ( @xmath28 , @xmath29 ) = ( @xmath30 , @xmath31 ) .
the typical system temperature was @xmath32@xmath33 k. the backend correlator was tuned to a resolution of @xmath34 khz and a bandwidth of @xmath35 mhz , which corresponds to the velocity resolution of @xmath36 km s@xmath37 at @xmath38 ghz .
j1517243 was used for phase calibration every @xmath39 minutes .
the bandpass calibration was carried out on j1256057 for h@xmath1co and on j1924292 for cch , whereas the absolute flux density scale was derived from mars and titan , respectively .
the data calibration was performed in the antenna - based manner and uncertainties are less than @xmath40 .
images were obtained by using the clean algorithm .
the continuum image was prepared by averaging line - free channels and the line maps were obtained after subtracting the continuum directly from the visibilities .
the primary beam ( half - power beam width ) is @xmath41 .
the total on - source time was @xmath42 minutes for h@xmath1co and @xmath43 minutes for cch .
the synthesized - beam size is @xmath44 ( p.a .
@xmath45 ) for the continuum image and @xmath46 ( p.a .
@xmath47 ) for the h@xmath1co image .
the rms noise levels for the continuum and h@xmath1co emission are @xmath48 and @xmath49 , respectively .
the continuum peak is : ( @xmath50 .
the cch data were combined with those taken in another observing program ( 2011.0.00628.s ; pi : jes jrgensen ) with @xmath51 antennas . the signal - to - noise ratio ( s / n ) was improved from @xmath52 to @xmath53 with this addition .
the synthesized beam of the combined cch image is @xmath54 ( p.a .
@xmath55 ) , and the rms noise levels is @xmath56 .
we also detected the ch@xmath21oh line as already published in a separate publication ( jrgensen et al .
we did not observe any other significant line features in the observed frequency range .
figure @xmath57(a ) shows the moment 0 ( integrated intensity ) map of h@xmath1co ( @xmath58@xmath59 ) .
the most prominent feature is a well - collimated outflow extending symmetrically from the protostar along a northeast southwest axis .
the outflow has a very straight - wall structure , as in the case of the hh46 outflow ( arce et al .
the northeastern lobe of the outflow is redshifted , while the southwestern lobe is blueshifted , as shown in the moment 1 ( velocity field ) map of figure @xmath57(b ) . in these observations ,
the emission extending over scales of @xmath60 or larger is not reliable due to the lack of short baselines .
although the outflow size looks compact ( about @xmath61 for each lobe ) , the emission of the outer part may be resolved out or may be weak due to insufficient excitation conditions .
the apparent width of the outflow is @xmath62 at a distance of @xmath61 from the protostar .
figure @xmath57(c ) shows the moment 0 map of a high - excitation line of h@xmath1co ( @xmath63@xmath64 ) .
the distribution of h@xmath1co ( @xmath63@xmath64 ) is essentially similar to that of h@xmath1co ( @xmath58@xmath59 ) , although the s / n ratio of the former is rather poor . in these maps , a bright knot ( clump a in figure @xmath57(a ) )
can be seen in the redshifted lobe , which could be a shocked region caused by an impact of the outflow with dense clumps in a surrounding cloud .
in addition to the outflow feature , a centrally concentrated component with a single - peaked distribution can also be recognized , as shown in a zoom of the central part of figure @xmath57(a ) ( figure @xmath57(d ) ) .
a blowed - up of the moment 1 map is also shown in figure @xmath57(e ) .
the approximate extent of the central component is estimated to be about @xmath65 in diameter based on the intensity distribution along the line perpendicular to the outflow ( figure @xmath66 ) , which corresponds to @xmath67 au .
figure @xmath68 shows the spectral line profiles of h@xmath1co and cch toward the protostar position averaged over the synthesized beam .
the line width of h@xmath1co is as narrow as @xmath69 even toward the protostar position . using the radex program ( van der tak et al .
2007 ) to fit the intensity of the two temperature sensitive lines of para - h@xmath1co ( @xmath70@xmath71 , @xmath63@xmath64 ) , the column density of h@xmath1co and the kinetic temperature toward the protostar are estimated to be @xmath72 and @xmath73@xmath74 on the assumption that h@xmath1 density is @xmath75@xmath76 and the line width is @xmath77 .
in addition , the ortho / para ratio is estimated to be @xmath78 with the intensity of an ortho - h@xmath1co line ( @xmath58@xmath59 ) .
the optical depths for these lines are @xmath79 ( @xmath70@xmath71 ) , @xmath80 ( @xmath63@xmath64 ) , @xmath81 ( @xmath58@xmath59 ) , respectively , therefore the lines are not opaque .
figure @xmath82(a ) shows the moment 0 map of cch ( @xmath83=43 , @xmath84=7/25/2 , @xmath85=43 and 32 ) , while figure @xmath82(b ) is a zoom of its central part .
the outflow cavity is prominent and the centrally concentrated component can also be seen .
the cch distribution is more extended along the southeast northwest axis than the h@xmath1co distribution .
it seems to have a slight dip toward the protostar position as shown in the intensity profile along the line perpendicular to the outflow axis ( figure @xmath66 ) .
this feature of cch is consistent with that reported by jrgensen et al .
although the spectrum of cch at the protostar position is complicated because of the two hyperfine components ( figure @xmath68 ) , the line width of each hyperfine component of cch is @xmath69 or less , as in the case of h@xmath1co .
we first analyze the outflow feature observed in the h@xmath1co ( @xmath58@xmath59 ) line .
figure @xmath86(a ) shows the position velocity ( pv ) diagram of the h@xmath1co ( @xmath58@xmath59 ) line emission along the outflow axis through the protostar position shown in figure @xmath57(a ) .
the outflow has a redshifted component extending to the northeast from the protostar and a blueshifted component extending to the southwest from the protostar .
since the redshifted and blueshifted lobes of the outflow show little overlap with each other around the protostar , the outflow axis seems to be close to the plane of the sky , indicating that the disk / envelope geometry is almost edge - on . the highest velocity at a certain distance to the protostar linearly increases as a function of the distance , as often observed for outflow cavities ( e.g. , lee et al . 2000 ; arce et al .
the intense knot in the redshifted component corresponds to clump a in figure @xmath57(a ) .
in addition to the high - velocity component , another velocity component can be seen around the systemic velocity ( @xmath87 km s@xmath37 ; sakai et al .
this component is slightly blueshifted on the northeastern side of the protostar and redshifted on the southwestern side of the protostar , which is the reverse case compared to the high - velocity component .
figure @xmath86(b ) is the pv diagram of h@xmath1co ( @xmath58@xmath59 ) along a line perpendicular to the outflow axis in the redshifted lobe indicated in figure @xmath57(a ) .
the pv diagram shows an elliptic feature with a knot - like distribution at the higher - velocity range .
this knot - like structure corresponds to clump a in figure @xmath57(a ) .
the gas in the cavity wall seems to be expanding .
we employ the standard model of an outflow cavity from lee et al .
( 2000 ) to analyze the observed geometrical and kinematical structures of the outflow , where the outflow cavity is assumed to have a parabolic shape and its velocity is proportional to the distance to the protostar : @xmath88 where @xmath89 denotes the distance to the protostar along the outflow axis , and @xmath90 the radial size of the cavity perpendicular to @xmath89 .
@xmath91 and @xmath92 are both normalization constants , and are set to be @xmath93 .
@xmath94 and @xmath95 are free parameters .
the best results are obtained with an inclination angle of @xmath3 , as shown by the blue lines in figures @xmath86(a ) and ( b ) . when the inclination angle is higher than @xmath96 or less than @xmath97
, the model does not reproduce the observations well with any values of @xmath94 and @xmath95 .
hence , the inclination angle is determined to be @xmath98 , where the quoted error is the estimated limit of error based on the above analysis .
the derived parameters are : @xmath99 arcsec@xmath37 and @xmath100 km s@xmath37 for an inclination angle of @xmath3 . as mentioned in section 1 , van kempen et al .
( 2009 ) reported an inclination angle of @xmath101 .
this discrepancy seems to originate from the limited spatial resolution of their data .
another possibility is that the outflow direction on small scales is different from that on larger scales ( e.g. , sakai et al .
2012 ; kristensen et al . 2013 ; yildiz et al . 2012 ; mizuno et al .
on the other hand , recent atacama pathfinder experiment observations of co @xmath84=65 indicate an inclination angle of @xmath3 ( u.a . , yildiz et al .
submitted to a&a ) , which is consistent with our result .
figure @xmath102 shows the pv diagram of cch ( @xmath103 ) along the outflow axis through the protostar position .
the pv diagram is complicated , because the two hyperfine components are blended with a separation of only @xmath104 . in comparison with the h@xmath1co case ,
the low - velocity component close to the systemic velocity is relatively bright in comparison with the high - velocity component .
this means that the cch emission from the outflow part primarily traces the compressed ambient gas around the outflow cavity rather than the entrained outflow gas .
cch may be formed by gas - phase reactions in dense photodissociation region layers ( e.g. jansen et al .
1995a , 1995b ; sternberg & dalgarno 1995 ) . as shown in figures @xmath57(d ) and @xmath82(b ) ,
there is a centrally concentrated component in the h@xmath1co and cch emission .
the distributions of c@xmath105s , c@xmath106o , and ch@xmath21oh also have such a component ( jrgensen et al .
2013 ) . here
, we investigate the kinematics of this component .
a zoom of the moment 1 map of h@xmath1co ( figure @xmath57(e ) ) is dominated by the overwhelming outflow motion . however
, a slightly skewed feature around the continuum peak is marginally recognized . in order to reveal the motion near the protostar more carefully ,
we prepared pv diagrams . figures @xmath107(a ) and ( b ) show the pv diagrams of h@xmath1co ( @xmath58@xmath59 ) along the two lines centered at the protostar position shown in figure @xmath57(f ) .
figure @xmath107(a ) is along the axis perpendicular to the outflow axis , which is shown by a broken arrow labeled as @xmath108 " in figure @xmath57(f ) .
figure @xmath107(b ) is along the outflow axis , which is shown by a broken arrow labeled as @xmath109 " in figure @xmath57(f ) . in figure @xmath107(a ) , we can see a marginal trend that the intensity peaks in the redshifted and the blueshifted velocity ranges are in the southeastern and northwestern sides of the protostar position , respectively , although a rotation signature is not obvious in the centrally concentrated component . on the other hand ,
two emission peaks are seen in figure @xmath107(b ) , one of which is redshifted in the northeastern side of the protostar position and the other is blueshifted in the southwestern side .
this systematic velocity gradient in figure @xmath107(b ) is the same case of that in the outflow , but the modest velocity gradient in the vicinity of the protostar is difficult to attribute to the outflow , according to our outflow model .
considering the outflow direction and the inclination angle , the outflow - envelope structure of this source is expected to be like a schematic illustration shown in the upper panel of figure @xmath110 .
therefore , the velocity gradient could be a signature of an infalling motion in the envelope rather than outflow .
an infalling / rotating signature is not clearly seen in the pv diagrams for cch ( @xmath103 ) ( figures @xmath107(c ) and ( d ) ) , in contrast with the well - studied low - mass class 0 protostar l1527 ( sakai et al .
2014a , 2014b ) at a distance of @xmath111 pc ( torres et al .
2007 ) . in l1527 , whose protostellar mass and inclination angle are @xmath112 and @xmath113 , respectively , the velocity shift from the systemic velocity observed for the cch line is @xmath114 km s@xmath37 at a radius of @xmath115 au from the protostar ( sakai et al .
on the other hand , the velocity shifts from the systemic velocity are as small as @xmath116 km s@xmath37 and @xmath117 km s@xmath37 for h@xmath1co and cch , respectively , in iras 153983359 ( figure @xmath107 ) , despite the low inclination angle of this source ( @xmath3 ) .
although iras 153983359 is similar to l1527 in its large - scale ( @xmath118 a few @xmath119 au ) chemical composition ( sakai et al .
2009 ) , the infalling / rotating motion is not very clear . in this observation , we resolve structure down to @xmath120 au ( @xmath0 ) around the protostar .
nevertheless the observed line width is narrow .
this means that the doppler shift due to infalling / rotating motions around the protostar should be small . in principle , the small velocity shift could be explained if the dense gas were not associated with the protostar , that is , if it were mostly present in the outflow
. however , this possibility seems unlikely because the outflow motion can not well explain the velocity gradient in figure @xmath107(b ) for h@xmath1co , as mentioned above .
hence , the small velocity shift likely implies a low protostellar mass .
the upper limit of the central mass is roughly estimated from the maximum velocity shifts from the systemic velocity . under the energy conservation law
, the central mass @xmath121 can be represented in terms of the infalling velocity ( @xmath122 ) and the rotation velocity ( @xmath123 ) at the distance @xmath124 to the protostar as : @xmath125 by use of this relation , the central mass can roughly be estimated to be smaller than @xmath4 with a maximum velocity of less than @xmath116 km s@xmath37 at @xmath0 ( @xmath120 au ) from the protostar .
here , we assume as a robust case that the infalling velocity and the rotating velocity are both the observed maximum value of @xmath126 in figure @xmath107(a ) . in the case of no rotation ( free fall ) ,
the upper limit to the central mass is estimated to be about @xmath127 . in the case of l1527 , which has similar chemical features to iras 153983359 , the toy model of the infalling rotating envelope shown in figure
@xmath110 has been applied to analyze the observational results for c - c@xmath21h@xmath1 and cch ( sakai et al .
2014a , 2014b ) .
this model assumes that the gas motion follows the particle motion , which is governed by the gravity of the central mass with conservation of energy and angular momentum .
a power - law distribution of the density is employed , and the intensity is assumed to be proportional to the column density along the line of sight . in spite of such a simple model , the basic features of the pv diagrams observed in l1527 for c - c@xmath21h@xmath1 and cch are well reproduced .
therefore , it is worth applying the same model to the velocity structure observed for h@xmath1co and cch in iras 153983359 . in l1527 , cch resides only in infalling rotating the envelope .
hence , it is preferable to examine the toy model with the cch line .
however , the pv diagrams of cch in iras 153983359 are complex due to the hyperfine structure , and their s / ns are insufficient for critical comparison .
in contrast , the emission of h@xmath1co ( @xmath58@xmath59 ) is bright enough in iras 153983359 .
although h@xmath1co is present in the inner disk - like structure as well as the infalling rotating envelope in l1527 , the latter contribution is significant ( sakai et al .
hence , we compared the model results with the h@xmath1co ( @xmath58@xmath59 ) data of iras 153983359 .
details of the calculation with the model is described in appendix .
figure @xmath128 shows the pv diagrams of h@xmath1co ( @xmath58@xmath59 ) along the envelope axis .
the blue contours represent the model results . unlike in the l1527 case
, we can not determine the radius of the centrifugal barrier from the pv diagram because of absence of an obvious rotation signature .
nevertheless , we can roughly estimate its upper limit from the cch distribution .
if cch is present only in the infalling rotating envelope as in the case of l1527 and the radius of the centrifugal barrier is larger than the synthesized beam ( @xmath0 ; @xmath120 au ) , a hole in the cch distribution toward the protostar position should be resolved .
although an intensity dip toward the center is marginally seen in the cch distribution ( figure @xmath66 ) , it is not well resolved in the present observation .
hence , we set the upper limit of the centrifugal radius to @xmath129 au .
we simulated the pv diagram for various sets of the protostellar mass and the radius of the centrifugal barrier under this constraint , as shown in figure @xmath128 .
the models with a central mass of @xmath130 and a radius of the centrifugal barrier of @xmath131 au are chosen as the best simulations among the @xmath132 models shown in figure @xmath128 by eye .
the mass is consistent with the upper limit estimated in section 3.3 .
thus , the low protostellar mass can be confirmed with this simulation .
we also simulated the pv diagrams along the different directions through the protostar position shown in figure @xmath82 ( the six broken arrows labeled as @xmath108 " , @xmath96 " , @xmath133 " , @xmath109 " , @xmath134 " , @xmath135 " ) for certain values of the central mass ( @xmath130 ) and the centrifugal barrier ( @xmath136 au ) as an example ( figure @xmath137 ) . the trends of a velocity gradient around the protostar are reproduced by the model .
the low mass of the protostar iras 153983359 ( @xmath138 ) is essentially derived from the narrow line width in the inner envelope ( @xmath139 au ) which has an almost edge - on configuration .
the above mass estimate may suffer from the uncertainty of the inclination angle of the envelope / disk .
if the inclination angle is larger , the mass evaluated by the model becomes larger .
however , the upper limit of the protostellar mass does not change significantly , even if an inclination angle of @xmath96 is employed . by use of the dynamical timescale of the extended outflow ( @xmath116@xmath140 yr for each lobe ; u.a .
, yildiz et al . submitted to a&a ) and the upper limit of the central mass ( @xmath138 ) obtained in this study , the average accretion rate is calculated to be less than @xmath141 yr@xmath37 .
the dynamical timescale of the outflow can be regarded as the lower limit , because the older part of the outflow may not be detected .
hence , the above accretion rate is regarded as the upper limit .
nevertheless , this estimate is roughly consistent with the typical accretion rate for low - mass protostars of ( @xmath142@xmath143 ) @xmath144 yr@xmath37 ( e.g. , hartmann et al . 1997 ) , and may be higher than that for another wccc source l1527 ( @xmath145 ; ohashi et al .
1997 ) . by use of the mass and the dynamical timescale of the outflow reported by van kempen ( 2009 )
, the mass outflow rate is calculated to be @xmath146 and @xmath147 for the red and blue lobes , respectively .
these values correspond to @xmath148 and @xmath149 , when an inclination angle of @xmath3 is employed .
hence , the mass accretion rate and the mass outflow rate are roughly comparable to each other . on the other hand , the mass accretion rate can be estimated by use of the equation ( palla & stahler 1991 ) : @xmath150 where the @xmath151 is the luminosity and the @xmath152 the radius of the protostar . by using the obtained upper limit to @xmath121
( @xmath4 ) , the mass accretion rate @xmath153 is estimated to be larger than @xmath154 with @xmath155 ( jrgensen et al .
2013 ) and @xmath156 ( e.g. , palla 1999 ; baraffe & chabrier 2010 ) . this lower limit is consistent with the abovementioned upper limit ( @xmath157 ) . in spite of the episodic accretion suggested by jrgensen et al .
( 2013 ) , the average mass accretion rate is not very different from the canonical value ( @xmath142@xmath145 ) . on the other hand , the upper limit of the protostar / envelope mass ratio is evaluated to be @xmath158 from the upper limit of the protostellar mass of @xmath4 and an envelope mass of @xmath159 ( kristensen et al .
the ratio is @xmath160 , if the mass of @xmath130 is employed .
hence , the ratio seems smaller than that in l1527 ( @xmath161 ; tobin et al . 2012 ) .
this implies that the protostar is in the infant stage and is still growing .
since the specific angular momentum brought into the inner region is smaller in the earlier stage of the protostellar evolution , it seems likely that the rotation signature is not so evident as in the l1527 case .
the h@xmath1co distribution has a centrally concentrated component with a single - peaked distribution , whereas the cch distribution is more flattened ( figure @xmath66 ) .
therefore , the h@xmath1co line traces the inner region in comparison with the cch line .
this trend is also seen in the observations of l1527 ( sakai et al .
it should be noted that figure @xmath137 shows a faint high - velocity component of h@xmath1co ( @xmath162 km s@xmath37 or @xmath163 km s@xmath37 ) toward the protostar position .
if this component really represents the contribution of a rotationally supported disk , it suggests that a disk structure is formed already in the early stages of the low - mass star formation .
if the upper limit of the central mass of @xmath4 is employed , h@xmath1co may be present even at @xmath164 au , according to its maximum velocity shift ( @xmath165 km s@xmath37 ) .
we here stress that extensive studies of the disk structure in class 0 sources have now become possible with alma and that a chemical approach will be of help to such studies .
our result reveals that the disk / envelope structure can be observed with various molecular lines .
hence , the chemical evolution can be investigated even in the disk - forming stage .
this paper makes use of the following alma data set ads / [email protected] .
alma is a partnership of the eso ( representing its member states ) , the nsf ( usa ) and nins ( japan ) , together with the nrc ( canada ) and the nsc and asiaa ( taiwan ) , in cooperation with the republic of chile . the joint alma observatory is operated by the eso , the aui / nrao and the naoj .
the authors are grateful to the alma staff for their excellent support . n.s . and s.y .
acknowledge financial support from grant - in - aid from the ministry of education , culture , sports , science , and technologies of japan ( 21224002 , 25400223 , and 25108005 ) .
thanks the advanced leading graduate course for photon science ( alps ) for financial support .
in our envelope model , the three - dimensional space is sectioned into meshes centered around the protostar , as shown in figure @xmath167 , and each mesh has data of the abundance of the gas and the velocity field .
the intensities of the emission from the meshes are calculated under the assumptions as follows . *
the motion of the gas is approximated by the particle motion , where the gas pressure is neglected .
the particle motion is governed only by the gravity of the central mass with conservation of its energy and angular momentum . *
the envelope has a flared shape with a flared angle of @xmath3 .
its outer radius is fixed to be @xmath168 au , which is the radius shown in figure @xmath107(a ) . *
although the envelope is not spherical , the density of the gas in the envelope is assumed to be proportional to @xmath169 outside the centrifugal barrier for simplicity ( e.g. , shu 1977 ; ohashi et al .
1997 ; harvey et al .
this corresponds to the density profile of an infalling cloud .
however , this choice is rather arbitrary in this study , because our main interest in this model is on the velocity field of the gas around the protostar .
furthermore , we assume that the gas is absent inward of the centrifugal barrier .
namely , no rotationally supported disk structure is considered .
this can be justified by a lack of the clear keplerian rotation feature in the pv diagram . *
the molecular emission is simply assumed to be optically thin .
therefore , the intensity is just proportional to the number of molecules along the line of sight .
no radiative transfer effects are considered . *
the spectral line has a gaussian profile with a line width of @xmath170 km s@xmath37 , and the emission is observed with a gaussian beam with a fwhm of @xmath171 . *
the number of the meshes is fixed to be @xmath172 and each mesh is a cube with a size of @xmath173 ( @xmath174 au ) .
the velocities of the gas are discretized with a step of @xmath160 km s@xmath37 .
the model parameters are the inclination angle @xmath175 , the central mass @xmath121 , and the radius of the centrifugal barrier @xmath176 .
the radius of the centrifugal barrier is represented as @xmath177 where @xmath178 is the gravitational constant .
it is the radius at which all the kinetic energy is converted to the rotational energy , and is determined by the relative strength of the gravitational force and the centrifugal force .
hence , specification of @xmath176 and @xmath121 means specification of the specific angular momentum @xmath179 .
the infalling and rotating velocities of particles at the distance @xmath124 to the protostar are represented as follows : @xmath180 arce , h.g . ,
mardones , d. , corder , s.a . , et al .
2013 , apj , 774 , 39 baraffe , i. , & chabrier , g. , 2010 , a&a , 521 , a44 brinch , c. , jrgensen , j.k . , & hogerheijde , m.r . , 2009 , a&a , 502 , 199 hartmann , l. , cassen , p. , & kenyon , s.j .
, 1997 , apj , 475 , 770 harvey , d.w.a . , wilner , d.j . , & myers , p.c . , 2003 ,
apj , 583 , 809 jansen , d.j . , spaans , m. , hogerheijde , m.r . , & van dishoeck , e.f . , 1995b , a&a , 303 , 541 jansen , d.j . , van dishoeck , e.f . , black , j.h . ,
spaans , m. , & sosin , c. , 1995a , a&a , 302 , 223 jrgensen , j.k . , visser , r. , sakai , n. , et al .
2013 , apjl , 779 , l22 kristensen , l.e . ,
klaassen , p.d . ,
mottram , j.c .
, schmalzl , m. , & hogerheijde m.r . , 2013 , a&a , 549 , l6 kristensen , l.e . , van dishoeck , e.f . ,
bergin , e.a .
, et al . 2012 , a&a , 542 , a8 lee , c.f . ,
mundy , l.g .
, reipurth , b. , ostriker , e.c . , & stone , j.m . , 2000 , apj , 542 , 925 lindberg , j.e . ,
jrgensen , j.k . , brinch , c. , et al .
2014 , a&a , 566 , a74 lombardi , m. , lada , c.j . , & alves , j. , 2008 , a&a , 480 , 785 mardones , d. , myers , p.c .
, tafalla , m. , et al .
1997 , apj , 489 , 719 maret , s. , belloche , a. , maury , a.j . ,
2014 , a&a , 563 , l1 mizuno , a. , fukui , y. , iwata , t. , & nozawa , s. , 1990 , apj , 356 , 184 mller , h.s.p .
, schlder , f. , stutzki , j. , & winnewisser , g. , 2005 , jmost , 742 , 215 murillo , n.m . ,
lai , s.p . ,
bruderer , s. , harsono , d. , & van dishoeck , e.f . , 2013 ,
a&a , 560 , a103 ohashi , n. , hayashi , m. , ho , p.t.p . , & momose , m. , 1997 , apj , 475 , 211 palla , f. , 1999 , in the origin of stars and planetary systems , ed .
lada & , n.d .
kylafis ( dordrecht : kluwer ) , 375 palla , f. , & stahlar , s.w . , 1991 , apj , 375 , 288 sakai , n. , ceccarelli , c. , bottinelli , s. , sakai , t. , & yamamoto , s. , 2012 , apj , 754 , 70 sakai , n. , oya , y. , sakai , t. , et al .
2014a , apjl , 791 , l38 sakai , n. , sakai , t. , hirota , t. , burton , m. , & yamamoto , s. , 2009 , apj , 697 , 769 sakai , n. , sakai , t. , hirota , t. , & yamamoto , s. , 2008 , apj , 672 , 371 sakai , n. , sakai , t. , hirota , t. , et al .
2014b , natur , 507 , 78 sakai , n. , & yamamoto , s. , 2013 , chrv .
, 113 , 8981 shu , f.h . , 1977 , apj , 214 , 488 sternberg , a. , & dalgarno , a. , 1995 , apjs , 99 , 565 tachihara , k. , dobashi , k. , mizuno , a. , ogawa , h. , & fukui , y. , 1996 , pasj , 48 , 489 tobin , j.j , hartmann , l. , chiang , h.f .
, et al . 2012 , natur , 492 , 83 torres , r.m . ,
loinard , l. , mioduszewski , a.j .
, & rodr guez , l.f .
, 2007 , apj , 671 , 1813 van der tak , f.f.s . , black , j.h . ,
schier , f.l . ,
jansen , d.j . , &
van dishoeck , e.f . , 2007 , a&a , 468 , 627 van kempen , t.a . , van dishoeck , e.f . ,
hogerheijde , m.r . , & gsten , r. , 2009 , a&a , 508 , 259 yen , h.w . ,
takakuwa , s. , ohashi , n. , & ho , p.t.p .
, 2013 , apj , 772 , 22 yildiz , u.a . ,
kristensen , l.e .
, van dishoeck , e.f . ,
2012 , a&a , 542 , a86 yildiz , u.a . , kristensen , l.e .
, van dishoeck , e.f . ,
et al . 2014 , submitted to a&a lccc transition & frequency & @xmath181 & s@xmath182 + & ( ghz ) & ( k ) & ( d@xmath183 ) + + @xmath70@xmath71 & 362.7360480 & 42 & 27.168 + @xmath58@xmath59 & 351.7686450 & 49 & 26.096 + @xmath63@xmath64 & 363.9458940 & 75 & 22.834 + + @xmath83=43 , @xmath84=7/25/2 , @xmath85=32 & 349.4006712 & 34 & 1.6942 + @xmath83=43 , @xmath84=7/25/2 , @xmath85=43 & 349.3992756 & 34 & 2.2712 + | sub - arcsecond ( @xmath0 ) images of h@xmath1co and cch line emission have been obtained in the @xmath2 mm band toward the low - mass protostar iras 153983359 in the lupus 1 cloud as one of the cycle 0 projects of the atacama large millimeter / submillimeter array .
we have detected a compact component concentrated in the vicinity of the protostar and a well - collimated outflow cavity extending along the northeast southwest axis .
the inclination angle of the outflow is found to be about @xmath3 , or almost edge - on , based on the kinematic structure of the outflow cavity .
this is in contrast to previous suggestions of a more pole - on geometry .
the centrally concentrated component is interpreted by use of a model of the infalling rotating envelope with the estimated inclination angle , and the mass of the protostar is estimated to be less than @xmath4 .
higher spatial resolution data are needed to infer the presence of a rotationally supported disk for this source , hinted at by a weak high - velocity h@xmath1co emission associated with the protostar . |
wetting of chemically and topographically heterogeneous surfaces gives rise to a rich phenomenology and a corresponding wealth of theoretical challenges.@xcite a remarkable example is that a careful combination of surface roughness and chemistry yields highly liquid - repellent and self - cleaning surfaces under given environmental conditions : this class of properties is often referred to as _
superhydrophobicity_. superhydrophobicity is related to the trapping of gaseous pockets ( air and/or vapor ) inside surface roughness.@xcite we will loosely refer to this scenario as the cassie state .
the superhydrophobic cassie state also favors the emergence of liquid slippage under flow conditions.@xcite however , superhydrophobicity breaks down as soon as the surface roughness becomes wet in the so - called wenzel state . as a consequence of the very different properties of the cassie and wenzel states ,
there is a growing interest in designing surfaces that are capable of stabilizing the superhydrophobic cassie state . in order for such engineering effort to be effective , a thorough knowledge of the cassie - wenzel transition
is required . with this objective ,
we analyze here the mechanism of the cassie - wenzel transition with the string method applied to molecular dynamics simulations .
the phase transition between the cassie and the wenzel states is , in most cases , characterized by large free - energy barriers.@xcite the superhydrophobic cassie state on the same surface can be stable , metastable , or unstable depending on the environmental conditions .
therefore it is not correct to speak about _
superhydrophobic surfaces _ , but rather about _
superhydrophobic states_.@xcite the problem of designing surfaces with superhydrophobic properties is therefore one of maximizing the range of temperatures , pressures , and characteristics of the liquids / vapor phases in contact for which the cassie state is stable . in the conditions where the cassie state is not thermodynamically stable , it is nonetheless possible to exploit metastabilities to obtain long - living superhydrophobicity : this is the case , _
e.g. _ , of omniphobic surfaces that present a metastable cassie state even with `` wetting '' liquids thanks to a special reentrant geometry.@xcite indeed , free - energy barriers must be much larger than the thermal energy @xmath0 in order for the metastable cassie state to survive for times that are significant for experiments and applications .
knowledge of how the cassie - wenzel transition starts and evolves the transition path or wetting mechanism may yield new insights for the design of engineered surfaces . according to the transition state theory
, the rate of the cassie - wenzel transition depends exponentially on the free - energy barrier between these two states .
thus , designing surfaces with the desired cassie - wenzel free - energy barrier is an effective tool for controlling the rate of the process .
for instance it has recently been suggested@xcite that the cassie state is generally metastable underwater .
this statement , which is based on experiments on a small number of surfaces , could be made more precise if the way in which free - energy barriers depend on the geometry of the surface roughness was known .
the crucial point for applications is not whether the desired state is metastable or stable but whether it will last longer than the experiment / application .
the first step in engineering surfaces is , therefore , the characterization of the wetting mechanism and of its dependence on the topography and chemistry of the surface , as well as on the thermodynamic conditions .
previous works typically considered the mechanism of the cassie - wenzel transition on drops , see _ e.g. _ ref . , but there is a growing interest in submerged surfaces where the external pressure plays a key role in the stability of the superhydrophobic state .
we focus here on the latter case by studying a model system that is simple enough to allow comparison of different approaches and yet shows a surprisingly rich phenomenology ( see fig .
[ fig : intro ] ) . .
free - energy profile of the cassie - wenzel transition as computed from the path above at coexistence ( _ bottom _ ) .
[ fig : intro],title="fig:",width=340 ] + .
free - energy profile of the cassie - wenzel transition as computed from the path above at coexistence ( _ bottom _ ) .
[ fig : intro],title="fig:",width=340 ] in previous works @xcite we characterized the free - energy barriers of the cassie - wenzel transition occurring in isolated hydrophobic roughness elements under different conditions of pressure and temperature .
the systems considered spanned from few nanometers ( explored _ via _ molecular dynamics simulations@xcite ) to macroscopic dimensions , for which the continuum rare events methods , or cream , was developed.@xcite over this broad range of systems , at coexistence when the cassie and wenzel state have the same free - energy , free - energy barriers are much larger than @xmath1 accounting for strong metastabilities .
for all previous approaches , the wetting path was characterized following the changes in the ( coarse - grained ) density field of the fluid , @xmath2 .
this , in turn , was considered a parametric function of the filling of the surface corrugation ( or liquid volume inside it ) , @xmath3 .
the resulting path of the transition is the sequence @xmath4 of density fields minimizing the free - energy at a given @xmath5 .
under suitable conditions , explained in sec .
[ sec : cream ] , this represent a realistic description of the wetting path .
however , when this description was applied in combination with a macroscopic , sharp - interface macroscopic model,@xcite we obtained a discontinuous wetting path ( see fig .
[ fig : intro ] ) .
the discontinuity corresponds to an instantaneous switching from a symmetric liquid / vapor meniscus to an asymmetric bubble in one of the corners of the corrugation ( morphological transition ) .
this discontinuity occurs at the `` transition state '' and results in a non - differentiability of the free energy profile at this point .
this sharp point may have two distinct origins , * an algorithmic one , related to the parameterization of the wetting path with the volume of liquid in the groove used in cream * a modeling one , that is , it could arise as a genuine feature of the sharp - interface model , which was used in combination with cream.@xcite the goal of this paper is therefore to address the question about the sharp point of the free - energy profile and , more generally , to lay on solid statistical grounds the discussion about the wetting mechanism on rough surfaces . in order to achieve this goal , we compute the _ minimum free - energy path _ ( mfep ) using the string method in collective variables.@xcite we employ atomistic simulations with the aim of making minimal assumptions on the liquid / vapor interface and the interactions with the walls .
the collective variable that characterizes the microscopic configuration implemented is the coarse - grained density field .
we compare atomistic with continuum , sharp - interface model paths and free - energy profiles .
the continuum path is obtained with the string and cream methods .
qualitatively , the atomistic and continuum wetting are consistent .
surprisingly , atomistic string free - energy profile shows a better agreement with continuum cream .
this is due to an `` error cancellation '' , with cream compensating intrinsic limitations in the sharp - interface model with respect to the atomistic case .
the second goal of this work is to elucidate the effect of the shape of the surface corrugation and its size on the mechanism of the cassie - wenzel transition and on the related free - energy barrier . anticipating our results , the concept of transition path itself may break down if the corrugations are sufficiently small .
the paper is organized as follows : in sections [ sec : md ] , [ sec : interface ] , and [ sec : cream ] the methods employed here , the _ atomistic string _ , the _ interface string _ , and the _ continuum rare events method _ ( cream ) are introduced and compared .
this first part contains the main methodological findings . in section [ sec : results ]
the atomistic and continuum results are presented and discussed , concentrating on the physics of the cassie - wenzel transition .
the last section summarized all conclusions .
the mechanism of the cassie - wenzel transition was investigated with the string method in collective variables applied to molecular dynamics simulations@xcite .
molecular dynamics simulations were performed with the lammps engine@xcite equipped for the string calculations with the plumed@xcite plugin as explained below . the isothermal / isobaric ensemble ( npt )
was used for all simulations by using the algorithm of martyna _ _ et al.__@xcite .
the standard lennard jones ( lj ) potential was used for the fluid - fluid interactions ; fluid - solid interactions were also of lj type , with the attractive term that was tuned through the factor @xmath6 : @xmath7 \text { , } \ ] ] where @xmath8 is the distance between the atoms @xmath9 and @xmath10 , while @xmath11 and @xmath12 set the scales of energy and length , respectively . in order to obtain a hydrophobic solid , we set @xmath13 , which corresponds to a contact angle of @xmath14 .
periodic boundary conditions are applied in the three directions .
the lower wall featured a rectangular groove ( or trench ) extending through the @xmath15 direction .
two kinds of grooves were considered , the first having a width of @xmath16 and square aspect ratio and the second having a width of @xmath17 and a rectangular aspect ratio , see fig .
[ fig : cv ] .
the collective variable used to describe the intrusion of liquid inside of the groove was the coarse - grained density field .
this quantity was computed from the atomic positions as the number of atoms inside the cells sketched in fig .
[ fig : cv ] . we used a mollified version of the characteristic function of the cells based on the fermi functions in order to prevent impulsive forces on atoms crossing the cell boundaries
( see sec .
[ sec : algorithm ] ) .
the cells occupied the whole @xmath15 dimension of the groove thus being effectively two - dimensional : for the square groove a total of @xmath18 cells were used , while for the rectangular groove @xmath19 , as sketched in fig .
[ fig : cv ] .
, height=151 ] the landau free - energy of the system as a function of the realization , @xmath20 , of the ( vector ) collective variable @xmath21 is defined as : @xmath22 where @xmath23 is the boltzmann constant , @xmath24 is the system temperature , @xmath25 is the probability to find the system at state @xmath20 , and @xmath26 is the @xmath27 dimensional vector of particles positions , with @xmath28 the number of particles in the system . the probability @xmath25 is expressed in the second equality of eq . as the integral over the @xmath29-dimensional configurational space of the probability density @xmath30 of the relevant ensemble ( the boltzmann factor ) times dirac deltas centered at value @xmath31 of the @xmath32 components of the collective variable .
the collective variable is assumed to depend only on the @xmath29 configurational degrees of freedom . for the general derivation of the string method in collective variables we refer the reader to the original work of maragliano _ _ et al.__@xcite briefly ,
this method allows one to identify the minimum free - energy path ( mfep ) , that is , the path of maximum likelihood .
the mfep is the continuous curve in the space of collective variables in this case the coarse - grained density field satisfying the equation @xmath33 where @xmath34 is a parameterization of the mfep , @xmath35 means `` parallel to '' , the indices @xmath9 and @xmath10 run over the @xmath32 collective variables ( which in vector notation are indicated as @xmath20 ) , @xmath36 is the free - energy defined in eq . , and @xmath37 is a metric matrix due to projection of the phase space onto the collective variable space , and defined as@xcite @xmath38 where @xmath39 and @xmath40 is the potential energy of the system . loosely speaking , when the metric matrix is unitary , eq
prescribes that the mfep joins two minima of the free - energy landscape passing through the bottom of the valleys and the saddle point connecting them ( the transition state ) .
the string method is an algorithm that allows one to identify the mfep .
the string itself is a discretization of the path connecting two metastable states , that is , two minima in the free - energy landscape .
the string is parameterized according to its relative arc - length , @xmath41 , with @xmath42 and @xmath43 beginning and end of the string .
the discrete points along the string are called _ images _ and are labeled with their position on the string @xmath44 , where @xmath45 is the index of the images .
we use here the version of the string method by e _
_ et al.__@xcite , which consists of three steps : 1 .
calculation of the free - energy gradient and of the metric matrix , see rhs of eq . , at the current position of the images ; 2 .
evolution of one timestep of the images according to the ( time - discretized ) pseudo - dynamics @xmath46 3 .
parameterization of the string to enforce equal arc - length parameterization among contiguous images . for sufficiently large @xmath47
, the algorithm guarantees that @xmath48 converges to the mfep of eq . .
@xcite steps 2 and 3 of the string algorithm above can be written also as@xcite @xmath49 where the term @xmath50 projects out the component of @xmath51 along the string .
his projector , indeed , implements the constraint @xmath52 in the dynamics of @xmath53 ( see eq . 2 of ref .
summarizing , the string method is a first order dynamics with the constraint of constant value of the parameter @xmath44 following the generalized force @xmath54 . in other words ,
the string method is a constrained ( local ) minimization of the system at a set of states at fixed @xmath44 in a space equipped with the metric @xmath55 . in order to compute the free - energy gradient and the metric matrix required to evolve the string _ via _ eq .
( point @xmath56 of the algorithm ) we used _
restrained molecular dynamics _
( rmd , see ref . ) . in practice ,
a restraining potential was added to the hamiltonian of the system in the form @xmath57 , where @xmath58 is the current value of the @xmath9-th component of the collective variable vector and @xmath31 is its target value . for sufficiently large @xmath59 , the dynamics driven by the restrained hamiltonian samples the conditional ensemble at @xmath60 , and allows one to compute the relevant quantities at a given position of the string @xcite .
once the string method has converged , it is possible to explore local approximations to the isocommittor surfaces , which are the hyperplanes orthogonal ( _ via _ the metric @xmath55 ) to the string . in particular
, we considered here the hyperplane intersecting the string at the transition state ; this plane bears the relevant information about the transition region , that is , the region where the probability to fall into the products side is the same as in the reactants @xcite .
the transition ensemble is a collection of microscopic states with the constraint that the system lies on this plane .
the _ transition state ensemble _ just discussed was computed _ via _ restrained molecular dynamics implementing the prescription @xmath61 where @xmath62 is the tangent to the string at the transition state ( ts ) .
the configurations extracted from the transition state ensemble are discussed in sec .
[ sec : results ] .
rather than describing the configuration of the system by the coarse - grained density field , @xmath63 , one can characterize the configuration of the liquid inside the cavity by the position , @xmath64 , of the liquid - vapor interface .
conceptually , one can obtain the corresponding landau free - energy , @xmath65 $ ] , by eq .
( [ eq : landau ] ) using the collective variable , @xmath66 .
this strategy requires a definition of the interface position , @xmath67 , in terms of the microscopic degrees of freedom , @xmath68 , which can be achieved via the coarse - grained density field . in the following ,
however , we do not compute the landau free - energy from the particle simulations but , instead , we use the mean - field , sharp - interface approximation to estimate @xmath69 .
this grand potential takes the form @xcite @xmath70 where @xmath71 , @xmath72 , and @xmath73 denote the total volume and the volumes occupied by the liquid and the vapor , respectively .
@xmath74 , @xmath75 , and @xmath76 characterize the liquid - vapor interface tension , and the surface tension of the confining solid with the liquid and the vapor .
@xmath77 , @xmath78 , and @xmath79 are the contact areas of the three coexisting phases . additionally , like in capillary - wave hamiltonians or solid - on - solid models , we assume that the interface position , @xmath80 , is a single - valued function , i.e. , configurations with overhangs and bubbles are ignored . using a completely filled cavity as reference state , @xmath81 with @xmath67 being
the height of the cavity and @xmath45 its widths , we obtain for the excess grand potential @xmath82 @xmath83 with the abbreviations @xmath84 , @xmath85 , and young s equation @xmath86 .
next , we express the vapor volume , @xmath73 , the interface area , @xmath77 , and the solid - vapor contact area , @xmath79 , through the interface position , @xmath87 @xmath88 the expressions will be appropriate , if the liquid - vapor interface is above the bottom of the cavity , i.e. , @xmath89 . if the liquid - vapor interface touches the bottom of the cavity , however , the liquid will be in contact with the solid , @xmath75 , rather than two separate liquid - vapor and vapor - solid interfaces .
we account for this effect by the correction term @xmath90 & = & \left(\gamma_{sl}-[\gamma_{lv}+\gamma_{sv}]\right)a_{sl}^{\rm bottom}\label{eq : omega_bottom } \\
\frac{\delta \omega[h_{i}]}{v_{tot}}&= & \frac{1}{lh}\int_{0}^{l } { \rm d}x\;g(h_{i}(x))\end{aligned}\ ] ] with the interface potential @xmath91 .
this introduces an additional length scale , @xmath92 , which characterizes the interaction range between the liquid - vapor interface and the bottom of the cavity . in the absence of long - range forces , this length scale is set by the width of the liquid - vapor interface , and we use a value that is smaller than all other length scales of our macroscopic model . in the numerical calculations we employ the following shape of the interface potential @xmath93 which smoothly interpolates between @xmath94 and @xmath95 .
thus , the excess grand potential is given by the following functional of the interface position , @xmath96 @xmath97}{v_{tot}\delta p_{\max } } & = & \delta \tilde p \frac{1}{lh } \int_{0}^{l } { \rm d}x\;h_{i}(x ) - \frac{1}{2h\cos \theta_{y } } \int_{0}^{l } { \rm d}x\ ; \sqrt{1+\left(\frac{{\rm d}h_{i}}{{\rm d}x}\right)^{2 } } \nonumber \\ & & - \frac{h_{i}(0)+h_{i}(l)+l}{2h } + \frac{1+\cos \theta_{y}}{2h\cos\theta_{y } } \int_{0}^{l } { \rm d}x\;f(h_{i}(x))\end{aligned}\ ] ] rescaling the spatial coordinate by the width of the cavity , @xmath98 , and the position of the interface by the height of the cavity , @xmath99 , we arrive at the final expression @xmath100}{v_{tot}\delta p_{\max } } & = & \delta \tilde p \int_{0}^{1 } { \rm d}\tilde x\;\tilde h(x ) - \frac{1}{2\cos \theta_{y } } \int_{0}^{1 } { \rm d}\tilde x\ ; \sqrt{\alpha^{2}+\left(\frac{{\rm d}\tilde h}{{\rm d}\tilde x}\right)^{2 } } \nonumber \\ & & - \frac{\tilde h(0)+\tilde h(1)+\alpha}{2 } + \frac{\alpha(1+\cos \theta_{y})}{2\cos\theta_{y } } \int_{0}^{1 } { \rm d}\tilde x\;f(\tilde h(\tilde x))\end{aligned}\ ] ] with @xmath101 , and @xmath92 is also measured in units of @xmath67 .
in the numerical calculations , the function @xmath102 approximated by @xmath32 values , @xmath103 for @xmath104 .
thus , the excess grand potential becomes a function of the @xmath32 rescaled interface positions ,
@xmath105 @xmath106 using this explicit expression , we compute the chemical potential @xmath107 @xmath108 and a similar expression holds for @xmath109 .
we use these expressions to calculate the minimum free - energy path that is a string of interface positions @xmath110 defined by the condition that the chemical potential perpendicular to the path vanishes . unlike eq .
( [ eq : mfep ] ) , we do not include the jacobian , @xmath111 , of the transformation from the microscopic degrees of freedom , @xmath68 , of the particle model to the collective variables , @xmath112 because we did not explicitly specify the transformation between the microscopic and collective variables .
we note , however , that ( i ) the free - energy maximum of the string without @xmath111 corresponds to the saddle point of the free - energy landscape and ( ii ) that the string corresponds to a continuous evolution of the liquid - vapor interface and , in particular , mimics a locally conserved dynamics of the coarse - grained density .
, @xmath113 , @xmath114 , @xmath115 , and @xmath116 ( in green ) .
the profile is compared to the atomistic and cream results at the same conditions . on the abscissa the parametric variable of the cream method @xmath117
is reported ; for the interface string @xmath117 is computed from the path in the top panel , while for the atomistic string @xmath118 , where @xmath119 is the total number of particles inside the coarse - graining cells .
the saddle point corresponds to a configuration where the liquid - vapor interface touches the bottom of the capillary ( between image 3 and 4 of the path above ) .
( c ) behavior of the saddle point at different @xmath120 .
[ fig : istring],scaledwidth=60.0% ] here we discuss the relation between the wetting path identified by the continuum rare events method ( cream ) previously introduced in giacomello _ _ et al.__@xcite and the interface `` string '' .
we start by introducing the committor function and isocommittor surfaces .
the committor is the probability that a trajectory passing by a state , an atomistic or a macroscopic configuration , will reach next the product state , say wenzel , rather than the reactant state .
an isocommittor surface is the ( hyper)surface in the state space of constant committor function value .
thus , for example , the isocommittor @xmath121 is the surface of states having @xmath122 % of probability to reach first the products and @xmath122 % to reach first the reactants .
this surface is , by definition , the transition state .
consider an isocommittor surface , @xmath123 which intersects the mfep at a point @xmath124 .
let us denote by @xmath125 the normal to @xmath123 at @xmath124 , and by @xmath126 the tangent to the mfep at the same point . in ref .
it is shown that : @xmath127 if we compare eq . with eq
, we conclude that the mfep is the path connecting constrained minima of the free - energy on the isocommittor surfaces @xmath128 foliating the space .
similar to the string method , cream aims at identifying a wetting path .
however , at variance with the string method , in cream this path is not parametrized with its arc - length but with an observable of relevance for the problem at hand .
this very general framework can be applied to different models , ranging from the micro- to the macroscale ; for the macroscopic scale , in the sharp - interface model that was originally used to describe the cassie - wenzel transition , the wetting path is parametrized with the volume of liquid in the groove , @xmath129 .
like in the case of the string method , in cream the free - energy is minimized subject to the constraint that the variable parametrizing the path is constant . in other words
, the path obtained by cream connects the constrained minima of the free - energy on the constant volume surfaces @xmath130 . comparing this formulation of the cream path with the formulation of the string path in terms of constrained minimization of the free - energy in the isocommittor surface ,
we conclude that the two paths coincide if and only if @xmath131 , at least locally to the path . in fig .
[ fig : istring]b we compare the cream and interface string free energy profiles along the respective paths .
we remark that the profiles coincide for most of the path , departing from each other only in a relatively small region around the transition state .
this is not surprising because , at variance with the string , cream does not impose the continuity of the path .
thus , in the case of the cassie - wentzel transition , which takes place through a morphological transition ( see ref . and sec .
[ sec : results ] ) , cream does not map the continuous path , along which the symmetric meniscus configuration goes into the gas - bubble - in - a - corner one ( see fig .
[ fig : creamvsstringpaths ] ) . in fig .
[ fig : istring]b we report also the free - energy profile obtained from atomistic simulations .
the agreement seems to be better between atomistic and cream results than with interface string .
the reason for this is discussed more in detail in the results section ; here we remark that this better agreement is due to `` errors cancellation '' , with the underestimation of the barrier in cream compensating for an overestimation intrinsic to the sharp - interface models . while the other represents the complementary degrees of freedom .
the dashed line in panel ( a ) is the hyperplane @xmath132 , where @xmath133 is a generic vector .
this is the plane on which @xmath134 is zero .
the dashed line in panel ( b ) , instead , represents the hypersurface @xmath135 , i.e. one of the infinite hypersurfaces on which the minimum of the free energy is sought in the cream method .
the string path follows the valley of the reactants and then moves smoothly to the valley of the products .
when the reactants and products valleys become parallel , the cream path moves abruptly from one to the other , following the minimum of the free - energy corresponding to a given level of progress of the reaction .
this is shown on the panel ( b ) : the yellow line denotes the two branches of the cream path , and the gray double arrow highlights its discontinuity . [
fig : creamvsstringpaths],scaledwidth=80.0% ] summarizing , the paths obtained from cream and the string method are not identical but give the same qualitative description of the process .
while the string method gives a detailed and continuous description of the most likely wetting path all along the process , cream represents the segment around the transition state as a sharp morphological transition ( fig .
[ fig : creamvsstringpaths ] ) .
indeed , cream and the string can be used as complementary tools .
cream allows to efficiently compute all the possible `` reactive '' channels .
the string method can then be used to further refine the cream paths . when the system is relatively simple , like the case of wetting of a square groove , @xcite it is possible to obtain the analytical solution of the cream equations .
thanks to cream it was possible to derive an extended version of the laplace equation , which relates the liquid / gas meniscus curvature to the surface tension . this relation , that was introduced for the first time in ref .
, is valid along most of the wetting path , apart in the region connecting the symmetric meniscus and gas - bubble - in - a - corner morphologies .
finally , cream has a high parallel efficiency , even higher than the string method as it does not require any exchange of data among images , and can be run on non connected heterogeneous computers .
in this section , we present the `` physical '' results obtained _ via _ the atomistic string and the sharp - interface calculations ( string and cream ) introduced in the previous sections .
we focus on the transition path for the cassie - wenzel transition and on the related free - energy profiles .
the length scales covered range from few particle diameters @xmath136 of the smallest atomistic system simulated to macroscopic scales , which are described in terms of sharp - interface models .
[ [ the - mechanism - of - the - cassie - wenzel - transition ] ] the mechanism of the cassie - wenzel transition + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + ( wenzel state ) .
the matrix elements are normalized with the maximum value . on the @xmath137 and @xmath15 axes
is reported the cell number .
values lower than one are observed on the diagonal where the density is lower than the bulk liquid one ( at the wall corners , see fig .
[ fig : path ] ) .
[ fig : mm],scaledwidth=60.0% ] we computed the transition path of the cassie - wenzel transition on two geometries , a square and a rectangular groove ( fig .
[ fig : cv ] ) .
a total of @xmath138 images was used to discretize the string : the images and graphs that follow are labeled with the image number .
the pressure of the npt simulations was chosen to be close to the coexistence between the cassie and the wenzel states .
the strings were initialized from configurations extracted from rmd simulations with a single collective variable , number of particles in the groove , analogous , apart for the ensemble ( here npt ) , to those presented and discussed in ref . .
we ensure that all initial images in the string have the same symmetry , that is , all menisci lie in the same corner , the left one .
a general result of the atomistic string calculations is the form of the metric matrix along the string .
a representative one is reported in fig .
[ fig : mm ] , showing that the most significant elements are those on the main diagonal .
this result supports the assumption of a unitary metric matrix as is usually done in macroscopic , sharp - interface models . however , there are other very small but non - zero elements related to the surrounding coarse - graining cells as reflected by the multi - diagonal character of the metric matrix .
these fine details could be encompassed in macroscopic models once the metric matrix is known from atomistic simulations .
the detailed analysis of the effect of the metric matrix is deferred to a future study . ]
the square groove measures around @xmath139 .
the thermodynamic conditions were @xmath140 and @xmath141 in lj units , where @xmath142 is the global pressure observable computed for all atoms . around @xmath143 steps of evolution of the string ( for each of which the mean forces were computed _ via _ rmd simulations , see sec .
[ sec : algorithm ] ) were required to ensure convergence . in fig .
[ fig : path]a we show the transition path for the square groove , _ i.e. _ , the sequence of average density fields forming the string at convergence .
the meniscus is initially flat close to the cassie state .
as the transition proceeds , the meniscus descends into the groove with constant curvature ( images @xmath144-@xmath145 ) until close to the bottom a liquid finger is formed on the right side of the groove ( images @xmath146-@xmath147 ) .
eventually the liquid wets one corner of the groove forming a circular bubble that gradually shrinks ( images @xmath148-@xmath149 ) until it is completely absorbed and the wenzel state is reached ( images @xmath143-@xmath138 ) .
the initial and final parts of the path ( the initial pinning , the symmetric meniscus , and the final bubble in the corner ) are in fair agreement with previous restrained molecular dynamics simulations @xcite and the macroscopic cream results @xcite ( see fig . [
fig : intro ] ) . however , close to the transition state the liquid - vapor interface forms a finger thus departing from the constant curvature menisci prescribed by cream .
this discrepancy is explained by the interface string path which , close to the transition state , exhibits a point of the meniscus with high curvature ( similar to the atomistic finger ) that eventually touches the bottom wall creating a small and a large bubble ( see fig . [
fig : istring]a ) .
the fine details of the process , however , are not easy to compare , because the diffuse nature of the atomistic interface tends to smear out sharp points and small vapor domains ; to this must be added that computational constraints limit the number of images and coarse - graining cells in the atomistic string . the rectangular groove measures around @xmath150 and is therefore twice as wide as the square one .
the thermodynamic conditions of the npt simulations were @xmath140 and @xmath151 for this case .
more than @xmath143 steps of string evolution were required for convergence .
the mfep for the rectangular groove is shown in fig .
[ fig : path]b .
it is seen that before the cassie minimum the meniscus curvature is allowed to vary while the triple line is pinned at the top corners of the groove ( images @xmath95-@xmath152 ) , as is expected from the macroscopic gibbs criterion @xcite .
this is an evidence that pinning happens also at the nanoscale , even though in a particle description of the system the continuum concept of `` geometrical singularity '' ( corners at the top of the groove , in the present case ) has no meaning .
however , the string resolution ( number of images ) does not allow us to quantify the range of contact angles for which pinning occurs .
the intrusion into the groove happens when a sufficiently large meniscus curvature is reached , around image @xmath152 .
the meniscus bends towards one corner at images @xmath153-@xmath154 , earlier in the progress of the transition than in the case of the square groove .
this observation can be made more quantitative by considering the length of the string up to the transition state and normalizing it with the total length of the string concerned with the activated event , that is , @xmath155 , where @xmath9 is the image number ; for the rectangular groove @xmath156 , while for the square one @xmath157 .
the contact line touches the bottom wall ( on the right hand side ) and recedes `` rapidly '' towards the opposite corner while the contact line at the vertical wall on the left does not move . as a consequence , during the shrinking process the bubble at the corner starts with a slightly flattened shape and then tends to a circular one .
this mechanism seems a generic one for the contact with the wall , since the interface string path for the square groove also shows that the left contact line behaves as if it were `` pinned '' at the vertical wall although there is no defect while the right contact line slides on the bottom wall ( see fig .
[ fig : istring]a ) . in @xmath1 units .
the square groove is at @xmath141 and the rectangular groove at @xmath151 .
the image count for the rectangular groove is shifted by @xmath158 in order to make the two wenzel states coincide .
the reference free - energy is taken to be that relative to the wenzel state , @xmath159 .
[ fig : profiles],scaledwidth=60.0% ] the free - energy profiles related to the mfeps just presented are shown in fig .
[ fig : profiles ] .
we remind that the profiles are at slightly different pressures . in both cases ,
the free - energy barriers connected to the cassie - wenzel transition are large as compared to the thermal energy @xmath0 , a fact that corroborates the presence of strong metastabilities on nano - rough hydrophobic surfaces .
the free - energy barrier for the square groove is @xmath160 @xmath0 , while for the rectangular one is @xmath161
@xmath0 , around @xmath162 larger , suggesting that both the size and the aspect ratio of the groove have an effect on the kinetics of the cassie - wenzel transition .
[ [ sec : ts_ensemble ] ] validity of the concept of transition path + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + in fig .
[ fig : ts_ensemble ] we show atomistic configurations extracted from the transition state ensemble of the square groove ; these microstates were computed _ via _ rmd as detailed in sec .
[ sec : ts ] .
it is apparent that the microscopic configurations correspond to several macroscopic states : bubble on the right corner , on the left one , and in the center .
for the larger rectangular groove , instead , the transition state ensemble is connected with a well defined configuration featuring a bubble in the left corner , similar to that shown in fig .
[ fig : path]b ( see also the related movie for movies of the transition paths and the transition state ensemble . ] ) .
we ascribe this behavior to the flat free - energy landscape along the hyperplane orthogonal to the transition state .
when the barrier separating the symmetry - related wetting paths ( bubble - on - the - left and bubble - on - the - right corners ) is @xmath163 the system can easily jump from one to the other . in this case , the wetting path identified by the string ( and cream ) method has little statistical significance .
in other words , the transition tubes @xcite around the two specular paths , i.e. , the region of state space in which most of the transition trajectories pass through , overlap .
thus , we can not describe the wetting trajectories as `` fluctuations '' around the mfep . in these conditions , other methods , such as the finite temperature string @xcite ,
would be needed to capture information about the transition .
these results suggest that the macroscopic models of capillary systems have a lower length scale below which they break down : the free - energy landscape becomes too flat to identify a single _
macroscopic state_. in the language of transition path theory , for sufficiently small grooves the transition tube becomes large and the single transition path found in the zero temperature limit has no statistical significance .
, scaledwidth=60.0% ] rescaled by @xmath164 .
the energetic costs of forming liquid - vapor or solid - vapor interfaces and of wetting the bottom wall are defined in eqs . and , respectively .
the results are computed via the interface string with @xmath165 , @xmath113 , @xmath114 , @xmath115 , and @xmath116 .
[ fig : energy_balance],scaledwidth=60.0% ] the interface string calculations were performed for the square groove in the region of the transition state ( the maximum of the profile ) , which is critical both for evaluating the free - energy barriers and for testing the cream results . as shown by the free - energy profiles in fig .
[ fig : istring]b at low filling levels ( @xmath166 ) and high ones ( @xmath167 ) the interface string coincides with cream . around the transition state , instead , the interface string significantly departs from cream .
the transition state itself coincides with a cusp in the free - energy profile , while in cream it is a non - differentiable point ( see fig .
[ fig : intro ] ) .
ironically , the jump discontinuity in the derivative at the transition state found with cream induced us to further investigate the phenomenon , which is actually more severe in the full - fledged string which shows an infinite discontinuity of the derivative ( in the sharp - interface limit @xmath168 ) .
the interface string however guarantees the continuity of the transition path , which is consistent with a continuous dynamics .
the cusp arises because in touching the bottom wall the radius of curvature of the meniscus has to change sign from positive ( symmetric meniscus ) to negative ( asymmetric bubble ) .
this is realized with a single point of zero radius of curvature ( and infinite curvature ) developing on a side of the liquid - vapor interface just before the contact with the bottom wall ( see fig . [
fig : istring]a ) .
a different point of view is that the area of the liquid - vapor interface significantly increases during the formation of the liquid `` finger '' while @xmath129 remains almost constant , thus giving rise to the sharp increase of the free - energy at the transition state as clearly shown by the energy balance in fig .
[ fig : energy_balance ] .
the thermodynamic force @xmath169 corresponding to the finger becomes infinite at the transition state .
this divergence of @xmath169 is integrable , thus giving rise to the cusp in the free - energy profile of fig .
[ fig : istring]b . in all cases analyzed atomistic , interface string , and cream the transition state is connected with the contact of the liquid domain with the bottom wall of the groove ( compare the paths in figs . [
fig : path ] , [ fig : istring]a , and [ fig : intro ] with the free free - energy profiles in figs .
[ fig : istring]b ) : while the meniscus descends symmetrically in the groove , the free - energy grows steadily because of the substitution of vapor - solid interface with liquid - solid one which is unfavorable for hydrophobic materials ( see fig .
[ fig : energy_balance ] ) ; when contact of the meniscus with the bottom wall eventually occurs , new liquid - solid interface replaces the liquid - vapor and solid - vapor interfaces at the bottom , resulting in an overall reduction of the free - energy ( the negative term @xmath170 in fig .
[ fig : energy_balance ] ) . in the atomistic string
, the transition state is found around one - half of the string for the rectangular groove while it is towards the end of it for the square one ( see fig .
[ fig : profiles ] ) .
the location of the transition thus depends on the aspect ratio of the groove ; this fact is easily explained by the balance of the energy contributions above : for a taller groove the ( growing ) branch of the free - energy connected with the meniscus sliding on the vertical walls of the groove is longer , while the descending branch due to the shrinking bubble is steeper .
the atomistic free - energy profiles reported in fig . [ fig : profiles ] are smooth at the transition state , differently from those obtained _ via _ the interface string and _ via _ cream ( see fig . [
fig : istring]b ) . in the atomistic case ,
indeed , thermal fluctuations tend to smear out the extreme events seen in the sharp - interface models .
in particular , the formation of a point in the meniscus with very high curvature is impossible in an atomistic picture . in order to account for this effect , we repeat the interface string calculations at increasing values of @xmath92 , which correspond to wider liquid - vapor interfaces , see fig . [
fig : istring]c .
the case @xmath171 roughly corresponds to the atomistic one , where the ratio of the interface thickness and the groove width is also @xmath172 .
figure [ fig : istring]c demonstrates that in the case of diffuse interfaces the transition state is smooth and the height of the free - energy barrier tends to decrease .
the atomistic string method in the density field collective variable was applied for the first time to the cassie - wenzel transition to determine rigorously the mechanism of the transition .
the results of this work are both methodological and physical . from the methodological point of view
, we demonstrated the relationship between approximate macroscopic methods@xcite and the full - fledged interface string .
the former methods are algorithmically simple and computationally convenient but fail where more parallel valleys are present in the free - energy landscape .
the string simulations also offered physical insight into the mechanism of the cassie - wenzel transition from the nanoscale to the macroscale .
a morphological transition was observed during the rare event with the meniscus changing from a symmetric to an asymmetric configuration .
the contact of the meniscus with the bottom wall determines the position and shape of the transition state ; at the nanoscale the transition state is smooth , while in macroscopic models it shows a cusp - like behavior .
the free - energy barriers are large compared to @xmath0 even in nanoscale grooves ; the depth of the groove and its aspect ratio are the critical parameters to determine the kinetics of the cassie - wenzel transition .
it was also shown that for very small grooves ( width @xmath173 ) the concept of transition path breaks down and it is not possible to identify a unique sequence of macroscopic configurations that describe the cassie - wenzel transition .
the research leading to these results has received funding from the european research council under the european union s seventh framework programme ( fp7/2007 - 2013)/erc grant agreement n. [ 339446 ] .
s.m . acknowledges financial support from the miur - firb grant no .
thanks the sfb 803 ( b03 ) for financial support .
we acknowledge prace for awarding us access to resource fermi based in italy at casalecchio di reno .
27ifxundefined [ 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 * * , ( ) link:\doibase 10.1007/s10404 - 014 - 1376 - 0 [ * * , ( ) ] @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) in @noop _
_ , vol .
( , ) pp . @noop * * , ( ) | the string method is a general and flexible strategy to compute the most probable transition path for an activated process ( rare event ) . we apply here the atomistic string method in the density field to the cassie - wenzel transition , a central problem in the field of superhydrophobicity .
we discuss in detail the mechanism of wetting of a submerged hydrophobic cavity of nanometer size and its dependence on the geometry of the cavity .
furthermore , we discuss the algorithmic analogies between the string method and cream [ giacomello _ et al .
_ , phys .
rev .
lett . * 109 * , 226102 ( 2012 ) ] , a method inspired by the string that allows for a faster and simpler computation of the mechanism and of the free - energy profiles of the wetting process .
this approach is general and can be employed in mesoscale and macroscopic calculations . |
the formalism of cosmological perturbations about a bianchi i universe with a scalar field was only developed recently in refs .
@xcite and in ref .
@xcite for the subcase of axisymmetric spacetimes . in such anisotropic inflationary models ,
the shear is always dominating at early time so that the universe behaves as a kasner spacetime .
it was realized that this anisotropic early era has an important signature since gravity waves are amplified during this era @xcite .
indeed , this stage is usually short and this instability is transient .
however this has led to question the stability of a pure kasner universe @xcite .
the kasner spacetimes @xcite are vacuum solutions of the einstein field equations .
they describe universes which are spatially homogeneous and euclidean but with an anisotropic expansion .
they play an important role in cosmology since they are a key structure in the discussion of the dynamics of spatially homogeneous spacetimes close to the singularity .
belinsky , khalatnikov and lifshitz @xcite and misner @xcite investigated the nature of the cosmological singularity by means of a bianchi ix model , whose temporal behaviour toward the singularity was shown to be described by a sequence of anisotropic kasner era .
this has led to the mixmaster picture @xcite and the idea of the cosmic billiard @xcite that can offer a description of the geometry of the universe prior to inflation .
the stability of this picture was already adressed in ref .
@xcite and later numerically in refs .
@xcite . only recently was it revisited @xcite in light of the recent developments concerning bianchi i universe perturbations @xcite in the particular case of a kasner universe with an axial symmetry , but no general study of the stability to linear perturbations of a kasner spacetime with arbitrary exponents has been performed yet .
this is the goal of this article .
we limit our analysis to the study of the stability of the kasner spacetime to linear perturbations . indeed , this can exhibit some instabilities but does not allow to demonstrate the general stability when no instability is exhibited at linear order in the perturbations .
for instance , the stability of the minkowski spacetime to linear perturbation is an easy exercise while the general proof of stability is extremely challenging @xcite .
also some spacetimes may be unstable to a particular class of perturbations while stable in other cases .
for instance , the einstein static universe was shown to be unstable with respect to spatially homogeneous and isotropic perturbations @xcite while it was then understood that the issue was not that simple when harrison @xcite demonstrated that all physical inhomogeneous modes are oscillatory if the matter content was a radiation fluid , while gibbons @xcite demonstrated the stability against conformal metric perturbations when the matter content was a fluid with sound speed larger that @xmath0 .
finally @xcite , it was shown that the einstein static spacetime is neutrally stable against small inhomogeneous vector and tensor perturbations and neutrally stable against adiabatic scalar density inhomogeneities as long as the condition exhibited by gibbons holds , and unstable otherwise .
our analysis is somehow simplified by the fact that the kasner spacetimes are empty so that we need no assumption on the matter content or on the type of perturbations considered .
once the evolution of the linear perturbation is determined , in a gauge invariant formalism to ensure that no gauge mode can spoil the conclusion , we compute the invariants of the metric and compare them to those of the background . again , since the kasner spacetime is empty , only the square of the weyl tensor has to be considered .
this ensures that the conclusion is not affected by the choice of gauge or coordinates system . in this article
, we consider the perturbation theory around generic kasner spacetimes , as described in [ sec1 ] , in order to discuss their stability .
first , in [ sec2 ] , starting from the general formalism we developed in refs .
@xcite , we show that for a vacuum spacetime only two perturbations are propagating , as expected .
we then exhibit some asymptotic analytical solutions for the behaviour of the perturbations in
[ sec3 ] which allow to discuss the stability at late and early times in the generic case . in
[ secaxial ] , we consider the axially symmetric case considered in ref .
@xcite , mostly in order to discuss the agreement with this previous analysis .
we consider a kasner spacetime , that is a bianchi i universe with metric usually written in terms of the cosmic time @xmath1 as @xmath2 for the vacuum , where @xmath3 is the volume averaged scale factor and @xmath4 the metric on the constant time hypersurfaces .
the vacuum einstein equations imply that @xmath5 , from which we deduce that the averaged hubble parameter is @xmath6 it is easily integrated to deduce that @xmath7 and we can always set @xmath8 .
the kasner metric can then be shown to be of the form @xmath9
^ 2,\ ] ] with the coefficients
@xmath10 satisfying the constraints @xmath11 that are fulfilled if @xmath12 @xmath13 varies in an interval of length @xmath14 and we always have @xmath15 if @xmath16 $ ] but it is enough to let it vary in @xmath17\pi/6,\pi/2[$ ] , which always ensures that @xmath18 ( see fig . [ f1 ] ) .
the two limiting cases @xmath19 or @xmath20 for which @xmath21 and @xmath22 respectively , correspond to space - times with an extra axial symmetry so that @xmath23 or @xmath24 . with @xmath13 ( @xmath25 in solid line , @xmath26 in dashed line and @xmath27 in dotted line ) . we can choose @xmath28 $ ] so that @xmath18.,width=321 ] it is thus clear that the spatial metric is explicitely given by @xmath29 \quad\hbox{with } \quad q_i = p_i-\frac{1}{3}\ ] ] so that @xmath30 the @xmath31 are explicitely given by @xmath32 we also define the useful parameterisation @xmath33 such that @xmath34 $ ] .
we define the conformal time @xmath35 from @xmath36 , and the metric ( [ metricb1 ] ) can then be recast as @xmath37.\ ] ] from the previous analysis , we have @xmath38 so that we set @xmath39 and also @xmath40 we easily deduce that @xmath41 , \quad \gamma^{ij } = { \rm diag}\left[\left(\frac{\eta}{\eta_0}\right)^{-3q_i}\right].\ ] ] we define the comoving hubble parameter by @xmath42 , where a prime refers to a derivative with respect to the conformal time so that @xmath43 the shear tensor is defined as @xmath44 and is traceless ( @xmath45 ) so that @xmath46\ ] ] and thus @xmath47 $ ] so that @xmath48 and the einstein equations imply that the shear satisfies @xmath49 we also define the shear in cosmic time by @xmath50 the kasner spacetime is a vacuum solution of einstein equations and this implies that @xmath51 it is thus only characterized by its weyl tensor , both at the background and perturbation level .
the weyl tensor has just two types of non - vanishing components which are the @xmath52 and the @xmath53 .
the former are related to the components of the type @xmath54 ( thanks to the traceless conditions ) and to the electric part of the weyl tensor .
the latter are related to its magnetic part , and it vanishes for the background kasner spacetime .
the only non - vanishing components of the background weyl tensor are thus expressed as @xmath55}^{j } + 2\hat \sigma^i_{[k}\hat \sigma_{l]}^j\nonumber\\ & & + 2h\left(\gamma^i_{[k}\hat \sigma_{l]}^j + \hat \sigma^i_{[k}\gamma_{l]}^j \right),\end{aligned}\ ] ] where @xmath56 $ ] refers to the antisymmetrisation of indices such that for an antisymmetric tensor @xmath57}$ ] ( and similarly later on , @xmath58 refers to the symmetrisation of indices such that for a symmetric tensor @xmath59 ) .
these weyl components are explicitely given by @xmath60}^j,\end{aligned}\ ] ] and it is easy to notice explicitely that these two types of components are not independent and are related thanks to eq .
( [ magicrelation123 ] ) .
we thus obtain that @xmath61 is given by @xmath62 where in the last equalities , @xmath63 can take any of the values @xmath64 . the function @xmath65 is given by [ using eq .
( [ eqrelteta ] ) ] @xmath66 the weyl tensor and his algebraic properties are used to classify spacetimes by their petrov type . defining the tracefree symmetric rank-2 tensor @xmath67 in terms of the electric ( @xmath68 ) and magnetic ( @xmath69 ) parts of the weyl tensor ,
provides a way to classify the weyl tensors @xcite . for a kasner spacetime ,
since the magnetic part of the weyl tensor vanishes we obtain @xmath70 the eigenvalues @xmath71 of @xmath72 define its petrov type .
they are all real and different so that a generic kasner spacetime ( @xmath73 ) has petrov type i while the two axially symmetric solutions with degenerate eigenvalues are of type o when @xmath20 and of type d when @xmath19 @xcite .
the kasner spacetime being a particular case of bianchi universe , the study of the evolution of the perturbations derives simply from our previous formalism @xcite that we specialize to the vacuum , using that @xmath74 so that @xmath75 and the property ( [ propkasner ] ) . following the formalism developed in refs .
@xcite , we consider the general metric of an almost bianchi i spacetime , @xmath76.\end{aligned}\ ] ] @xmath77 and @xmath78 can be further decomposed into scalar , vector and tensor components as @xmath79 with @xmath80 we then construct the gauge invariant quantities and define the conformal newtonian gauge by the conditions @xmath81 so that @xmath82 we also introduce the extremely useful variable @xcite @xmath83 the tensor variable @xmath84 is readily gauge invariant . at this stage , we are left with @xmath85 degrees of freedom : two scalars ( @xmath86 and @xmath87 ) , two vectors ( @xmath88 ) and two tensors ( @xmath89 ) .
contrary to the perturbation theory around a spatially homogeneous and isotropic spacetime @xcite , the three types of perturbations do not decouple .
as we shall now see , the perturbation equations will exhibit four constraints so that only two dynamical degrees of freedom related to the gravity waves remain , as in minkowski or de sitter spacetimes .
the perturbation equations are conveniently written in fourier space and we decompose any quantity in fourier modes as follows .
using the cartesian comoving coordinates system @xmath90 on the constant time hypersurfaces , we decompose any scalar function as @xmath91 the comoving wave co - vectors @xmath92 are constant , @xmath93 .
we now define @xmath94 which is now time - dependent and explicitely given by @xmath95 so that @xmath96 for any mode @xmath92 , a basis @xmath97 of the subspace perpendicular to @xmath92 can be constructed from the natural cartesian basis as ( see appendix a of ref .
@xcite for the details of the construction ) @xmath98 with @xmath99 @xmath100 the three euler angles @xmath101 depend on time and are explicitely given by @xmath102 so that @xmath103 and @xmath104 the previous relations determine @xmath105 and @xmath106 .
the requirement that @xmath97 remains orthogonal to @xmath92 imposes that @xmath107 which can be rewritten as @xmath108 which determines @xmath109 up to an integration constant .
the vector and tensor modes can then be decomposed respectively as @xmath110 and @xmath111 where the polarisation tensors have been defined as @xmath112 the perturbation equations in fourier space involve the components of the decomposition of the shear on the basis @xmath113 .
since it is a symmetric trace - free tensor , it can be decomposed as @xmath114 this decomposition involves 5 independent components of the shear in a basis adapted to the wavenumber @xmath92 .
we must stress however that @xmath115 must not be interpreted as the fourier components of the shear , even if they explicitely depend on @xmath92 .
this dependence arises from the local anisotropy of space .
a similar decompositon for @xmath116 defines the coefficients @xmath117 and we obtain from einstein equation in vacuum that it must satisfy the constraint @xmath118 which is independent of @xmath92 .
the equation of evolution of the shear ( [ ein - tt ] ) implies that @xmath119 where the matrix @xmath120 is defined by @xmath121 the equation ( [ decshear ] ) can be inverted to get @xmath122 the last equation implies , using eq .
( [ defespilonij ] ) , that @xmath123\label{s4}\\ \stcross & = & 2\sqrt{2}\sum_i q_i\omega_1^i\omega_2^i.\label{s5}\end{aligned}\ ] ] we use the equations derived in refs .
@xcite when applied to the particular case of a vacuum solution ( so that @xmath124 ) .
the two scalar perturbations are explicitely obtained from the tensor modes as @xmath125 we then deduce that @xmath87 is given by @xmath126 while the second bardeen potential is then given by @xmath127 the vector mode is then given by @xmath128 e_\lambda.\ ] ] this shows that the scalar and vector modes are obtained algebraically from the tensor modes , which are the only degrees of freedom that propagate . using the shorthand notation @xmath129 for the opposite polarisation of @xmath130 , i.e. meaning that if @xmath131 , then @xmath132 , and vice - versa , and introducing @xmath133 we have @xmath134 the general solution of this linear equation can always be obtained in terms of four transfer functions as @xmath135 where @xmath136 and @xmath137 are the initial conditions .
the notations @xmath138 and @xmath139 refer respectively to a decaying and a growing mode .
when the two polarisations are decoupled ( which is the case asymptotically at early and late times , as we whall see below ) then we only have two transfer functions to consider @xmath140 and @xmath141 since the transfer of power from one polarisation to the other is negligible .
similarly we define transfer functions for @xmath87 , @xmath142 and @xmath143 by @xmath144 @xmath145 @xmath146 in conclusion , for a given model , one can determine @xmath117 for each mode @xmath92 and then solve eq .
( [ eqformu ] ) for the gravitational waves .
one can then deduce @xmath86 , @xmath87 and @xmath143 algebraically .
as expected , only two degrees of freedom can propagate .
the existence of the vector and tensor modes arises from the fact that isotropy is violated so that the svt modes do not decouple .
the kasner case , being a vacuum solution , is thus simpler than the bianchi i case studied in refs .
in full generality , the stability analysis requires first to solve eq .
( [ eqformu ] ) to determine the four transfer functions defined in eq .
( [ e338 ] ) .
this allows to evaluate the effect of the perturbations on the square of the weyl tensor , which is by construction independent of the choices of gauge and coordinates system .
we thus introduce @xmath147 where @xmath148 is the background value of @xmath149 given in eq .
( [ valc ] ) such that at the background level @xmath150 , and at the perturbed level up to second order we have when @xmath151 @xmath152 the perturbation of @xmath153 at first order in the perturbations , @xmath154 , can be decomposed in fourier modes as @xmath155 in order to assess the behaviour of @xmath149 at early and late time , we assume that at an arbitrary initial time @xmath156 the initial conditions for the two modes defined in eq .
( [ e338 ] ) are such that ( 1 ) they are not correlated , @xmath157 and that ( 2 ) they have the same initial power spectrum , @xmath158 @xmath159 the initial power spectrum is an unknown function of the comoving wave - vector @xmath92 .
it is clear that @xmath160 at all time .
this is indeed not the case for the perturbation of @xmath153 at second order in perturbations , @xmath161 , since it is quadratic .
the previous definitions allow to compute that at lowest order @xmath162 with @xmath163 and we define @xmath164 the function @xmath165 can in principle be expressed in terms of the transfer functions that appear in eq .
( [ e338 ] ) and in terms of the coefficients @xmath166 ... from a numerical point of view , the situation is thus clear but very time consuming .
it is also unnecessary to solve the evolution equations in their full generality since we are interested in the asymptotic behaviour of @xmath167 when @xmath168 or @xmath169 . in these two regimes ,
the wave numbers tend to focus along a principal axis since [ see eq .
( [ deff2b ] ) ] @xmath170 and @xmath171 we shall thus discuss the dynamics of the perturbations when their mode is aligned along a principal axis ( [ 4b ] ) and their contribution to @xmath165 ( [ 4c ] ) .
these two steps can be performed completely analytically so that we can then discuss the general behaviour of @xmath167 at late time ( [ 4d ] ) and early time ( [ 4e ] ) .
we assume that only one of the components of the fourier space , indexed by @xmath63 , satisfies @xmath172 .
it is then clear that @xmath173 so that @xmath174 then , since @xmath175 , we conclude from eq .
( [ background_sigma ] ) that @xmath176 because of eq .
( [ background_sigma2 ] ) and thus @xmath177 this implies , from eq .
( [ eqphia ] ) , that @xmath178 it follows from eq .
( [ background_sigma3 ] ) that @xmath179 , and it can be checked that @xmath180 where we have defined @xmath181 which is explicitely given by @xmath182 the scalar perturbations are then given algebraically by @xmath183 that derives from eq .
( [ e : temp2 ] ) from which we deduce that eqs .
( [ magic_equation]-[e : scal2b ] ) reduce to @xmath184 , with @xmath13 ( @xmath185 in solid line , @xmath186 in dashed line and @xmath187 in dotted line ) .
we recall that @xmath188 whatever @xmath13.,width=321 ] the only equations to solve are eqs .
( [ eqformu ] ) for the gravity .
they decouple for each polarisation and lead to the system @xmath189\mu_+ = 0,\label{evomualigned1}\\ & & \mu_\times''+ \left[k_i^2\left(\frac{\eta}{\eta_0}\right)^{-2 q_i } + \frac{1 - 9\delta_i^2}{4\eta^2 } \right ] \mu_\times=0.\label{evomualigned2}\end{aligned}\ ] ] these two equations are compatible with the general equations ( [ mu2axial]-[omega2axial2 ] ) obtained for @xmath19 and , in this particular case , with those derived in ref .
the general solutions of eqs .
( [ evomualigned1]-[evomualigned2 ] ) can both be obtained as [ see eqs .
( [ trucjp1]-[trucjp2 ] ) ] the linear combination ( [ e338 ] ) with @xmath190 where @xmath191,\\ { \cal n}^\lambda_{i}(k_i,\eta)&= & n_{\nu_i^\lambda } \left[\frac{k_i\eta_0}{1-q_i}\left(\frac{\eta}{\eta_0}\right)^{1-q_i } \right].\end{aligned}\ ] ] here , @xmath192 and @xmath193 are the bessel and newmann functions of index @xmath194 and @xmath195 is depicted on fig . [ f2 ] ( note that @xmath195 is always non - negative while @xmath196 can be negative ) .
it is then clear that @xmath197 so that the two bardeen potentials are given by @xmath198 if follows that for these particular modes , the solution of the dynamics is obtained completely analytically .
figures [ fcp1 in ] give an example of the behaviour of the tensor and scalar modes .
the late time behaviour of the perturbations can be easily understood using the large argument expansions ( [ besseljinf ] ) and ( [ besselninf ] ) of the bessel functions , since @xmath199 \nonumber\\ t^{\downarrow}_{\lambda}&\simeq&\sqrt{\frac{2}{\pi\ell_1}}\left(\frac{\eta}{\eta_0}\right)^{-\frac{1-q_1}{2 } } \sin\left[\ell_1\left(\frac{\eta}{\eta_0}\right)^{1-q_1 } \!\!\!\!\!\!-\nu_1\frac{\pi}{2 } -\frac{\pi}{4 } \right ] \nonumber\end{aligned}\ ] ] with @xmath200 it follows that the late time behaviour is in general of the form @xmath201,\ ] ] where @xmath202 and @xmath203 are two constants that can easily be obtained in terms of @xmath204 and @xmath205 .
@xmath142 is then proportional to @xmath206 [ see eq .
( [ xki ] ) ] and @xmath207 , which completely determines the solution [ see eqs .
( [ bb1]-[bb2 ] ) ] .
asymptotically , the dominant term of the wavevector is aligned along the direction expanding the fastest , i.e @xmath208 .
the two polarisations behave differently because @xmath209 . using the small argument expansions ( [ besselj0 ] ) and ( [ besseln0b ] ) of the bessel function
, we obtain that , since @xmath210 , when @xmath211 @xmath212^{\nu_3^\times } \nonumber\\ t^\downarrow_\times & \sim & \frac{\gamma(\nu_3^\times)}{\pi } \left[\frac{\ell_3}{2}\left(\frac{\eta}{\eta_0}\right)^{1-q_3 } \right]^{-\nu_3^\times } \label{e430 } \end{aligned}\ ] ] with @xmath213 these expressions hold as long as @xmath214 , that is for @xmath215 , a particular case that is discussed in appendix [ secpi2 ] .
the transfer functions for the @xmath216 polarisation behave as @xmath217\\ t^\downarrow_+ & \sim & \frac{2}{\pi}\left[\ln\left(\frac{\ell_3}{2 } \left(\frac{\eta}{\eta_0}\right)^{1-q_3}\right ) + \gamma_e\right ] .
\label{e432 } \end{aligned}\ ] ] for these modes , the expressions of @xmath218 and @xmath165 turn out to simplify greatly and can be expressed in terms of the dimensionless parameter @xmath219 first , @xmath218 can be expressed in terms of the two transfer functions and their first derivatives as @xmath220 the two coefficients @xmath221 can be computed and are given by @xmath222,\\ f_+^{(1 ) } & = & \sqrt{2 } \delta_i [ -9 ( 1 + 2 q_i ) + ( 7 + 2 q_i ) x^2].\end{aligned}\ ] ] interestingly , the @xmath223 polarisation does not contribute . in the above
expression @xmath224 stands either for @xmath225 or @xmath226 depending on the solution ( decaying , growing , or any linear combination of them ) that is considered . when @xmath227 , @xmath228 vanishes identically and , again , since @xmath229 we shall not consider this linear part any further
. the second order @xmath230 can be expanded as @xmath231 with the same conventions as for @xmath228 . the eight coefficients @xmath232 are explicitely given by @xmath233 @xmath234\\ f_{\times}^{(0,1)}&=&f_{\times}^{(1,0)}\nonumber\\ & = & -4(1-q_i)[(1 + 2 q_i)^2-x^2]\end{aligned}\ ] ] and @xmath235\nonumber\\ f_{\times}^{(1,1)}&=&2[(1 + 2 q_i)^2 + 4x^2].\end{aligned}\ ] ] at late time ( @xmath236 ) , only the growing mode dominates and we can safely neglect the decaying mode since it would only lead to a redefinition of the phase @xmath237 ; see eq .
( [ eqasymptoteinfty ] ) . as can be concluded from the evolution ( [ klarge ] ) of the wave number , the parameter @xmath238 defined in eq .
( [ defx ] ) behaves as @xmath239 which is always a growing function of @xmath35 ( remember @xmath240 and thus @xmath241 ; see fig .
we thus need to take the limit when @xmath242 of the transfer functions @xmath243 and of their derivatives . from eq .
( [ eqasymptoteinfty ] ) , we conclude that @xmath244.\ ] ] since @xmath245 it implies that , at leading order in @xmath238 , @xmath246.\end{aligned}\ ] ] using the expansion ( [ expchi2 ] ) , we conclude that @xmath247}{(1 + 2 q_1)^2(1-q_1 ) } \nonumber\\ & \to&\frac{8 x^3}{3 \pi ( 1 + 2 q_1)^2}\,.\label{e444}\end{aligned}\ ] ] this expression is valid for any kasner spacetime but for the particular case @xmath20 that is discussed in appendix [ secpi2 ] ( since in that case @xmath248 ) .
it follows that @xmath249 which is unbounded at late time . as can be concluded from the evolution ( [ ksmall ] ) of the wave number , the parameter @xmath238 defined in eq .
( [ defx ] ) behaves , when @xmath250 , as @xmath251 which is always a decreasing function of @xmath35 ( except if @xmath20 , see appendix [ secpi2 ] for that particular case ) .
it follows that we need to evaluate the limit @xmath252 of the transfer functions . at early time , one can however not disregard the decaying mode since it diverges faster when @xmath211 .
the expression ( [ expchi2 ] ) behaves as @xmath253 in order to determine the behaviour of the transfer functions , we consider two cases .
first , if we consider only the growing mode [ i.e. @xmath254 and @xmath255 in eq .
( [ e338 ] ) ] then we deduce from the expressions ( [ e430 ] ) and ( [ e432 ] ) of the transfer functions that @xmath256 at leading order , the contributions of the two polarisations to @xmath230 are thus @xmath257 eqs .
( [ e430 ] ) and ( [ e432 ] ) tells that @xmath258 and @xmath259 so that @xmath260 ( since @xmath261 ) and @xmath262 thus , the contribution of the growing mode leads to a bounded contribution to @xmath230 . on the other hand ,
if we consider the decaying mode [ i.e. @xmath263 and @xmath264 in eq .
( [ e338 ] ) ] , i.e. the most diverging mode when @xmath265 , then @xmath266 we conclude that at leading order the contributions of the two polarisations to @xmath230 are @xmath267 since , from eqs .
( [ e430 ] ) and ( [ e432 ] ) , @xmath268 with @xmath269 for all cases , the @xmath223 contribution diverges faster when @xmath250 and we conclude that @xmath270 during the cosmic evolution , for any modes , the components of the shear evolve from an early value corresponding to the situation of the mode being aligned to the third spatial axis ( index @xmath271 ) to a situation corresponding to the mode being aligned to the first spatial axis ( index @xmath272 ) .
this dynamics arises from the fact that the euler angles are not constant over time and that @xmath273 evolves from 0 to @xmath274 and @xmath275 from @xmath274 to 0 ( except in the axisymmetric cases @xmath276 for which the value of @xmath275 does not matter ) .
this implies that @xmath277 varies from @xmath278 to @xmath279 , @xmath280 and @xmath281 from 0 to 0 but with a transient deviation from 0 and @xmath196 from @xmath282 to @xmath283 , as depicted on fig .
[ f3 ] .
it follows that the asymptotic behaviour of the perturbation of the weyl tensor can be discussed by looking at these two particular regimes .
figure [ fcp1axial ] shows how a numerical solution interpolates between these two regimes . using this procedure , analog to the one developed in ref .
@xcite , we conclude that the kasner spacetime is unstable at late time since @xmath284 decreases with time slower than @xmath148 , which means that the weyl tensor of the perturbation always dominate the weyl tensor of the homogeneous space at late time .
indeed , for a given mode the late time asymptotic regime is reached when @xmath285\,.\ ] ] at any given time @xmath35 , we denote by @xmath286 the set of modes such that @xmath287 so that eq .
( [ e47 ] ) implies that @xmath288 since @xmath289 is positive .
given that @xmath290 is converging toward @xmath291 which is diverging at late times , we deduce that @xmath290 is diverging at late times .
this is compatible with the analysis of ref .
@xcite for the case of an axisymmetric kasner spacetime , as well as with earlier findings @xcite on the amplification of gravity waves in anisotropic inflation during the shear dominated era , which can be described by a kasner era of finite duration . at early time
the conclusion depends on the choice of the modes that are considered .
we have shown that if decaying modes are present then the spacetime is unstable when @xmath292 , following the same argument as above but with the modes such that @xmath293,\ ] ] whereas it is stable if one imposes to have only growing modes since then @xmath230 remains bounded . as pointed out in @xcite ( see appendix b ) , this result is compatible with the analysis by ref .
@xcite where it was pointed out that a condition on the perturbations should be imposed for the spacetime to be stable when @xmath292 . as explained in ref .
@xcite , this condition precisely kills the decaying modes @xmath294 for both perturbations .
in order to compare our analysis to the existing literature , we consider the particular case of an axially symmetric case with @xmath19 . this particular case was studied in ref .
@xcite and corresponds to the axisymmetric case @xmath295 so that @xmath296 in that case , and in that case only , we perform a permutation of the axis ( @xmath297 , @xmath298 and @xmath299 ) with respect to the parameterisation of angles given in [ secmodedecomposiiton ] so that the special direction is the first axis ( labelled by the index @xmath272 ) . after deriving the gravity wave equation ( [ 5a ] ) and showing that the two polarisations are decoupled at all times ( and not only asymptotically as for a generic case ) , and expressing the vector and scalar perturbation ( [ 5b ] )
, we investigate the late ( [ 5c ] ) and early ( [ 5d ] ) time behaviours ( and the interpolation of the exact numerical integration between these two asympotic regimes in [ 5e ] ) .
we set @xmath300 , so that eq .
( [ deff2b ] ) takes the form @xmath301 and we shall assume that @xmath302 , since otherwise we are back to [ 4b ] . because of the rotational invariance , eq .
( [ euler2 ] ) implies that the euler angle @xmath275 is constant @xmath303 so that eq .
( [ euler3 ] ) implies that @xmath304 .
we can thus choose @xmath305 the euler angle @xmath273 is obtained from eq .
( [ euler1 ] ) @xmath306 it follows that eqs .
( [ euler4]-[euler5 ] ) simplify to @xmath307 the components of the shear then take very simple expressions .
( [ s1 ] ) reduces to @xmath308 while eqs .
( [ s2]-[s3 ] ) give @xmath309 and eqs .
( [ s4]-[s5 ] ) lead to @xmath310 these expressions satisfy indeed eq .
( [ sumsurs ] ) and are depicted on fig .
we recover the behaviour ( [ e411 ] ) for @xmath277 , ( [ svagen ] ) for @xmath280 and ( [ e414 ] ) for @xmath311 .
( solid ) , @xmath312 ( dashed ) and @xmath196 ( dotted ) for @xmath313 ( heavy ) and @xmath314 ( light ) or @xmath315 ( light ) for a kasner universe with @xmath19.,width=321 ] since @xmath316 , the two polarisations decouple and their equations of evolution ( [ eqformu ] ) simplify to @xmath317 with @xmath318 \label{omega2axial1}\\ \omega^2_{+ } & = & k^2 + \frac{1}{4\eta^2 } \left[1 + 2\svun^2\left(\frac{\sqrt{2}\stplus}{2-\spar}-1\right)^2 \right]\label{omega2axial2}\end{aligned}\ ] ] where we have used that @xmath319 and @xmath320 .
we emphasize that our expressions agree with those of ref .
@xcite ( see for instance eq . ( 40 ) of that reference ) .
the expressions of the scalar perturbations are therefore @xmath321 that derives from eq .
( [ e : temp2 ] ) from which we deduce that eqs .
( [ magic_equation]-[e : scal2b ] ) reduce to @xmath322\nonumber\end{aligned}\ ] ] and @xmath323 the absolute value of the coefficient in eq .
( [ axial - x ] ) evolves from @xmath324 to @xmath325 and @xmath143 vanishes both at early and late times . at late times
, the wave - number behaves as @xmath326 so that @xmath327 .
for the two polarisations , the growing and decaying modes entering the solution ( [ e338 ] ) behave as @xmath328=j_0\left(\frac{x}{2}\right)\\ t^\downarrow_\lambda & \sim&n_0\left[\frac{k_1\eta_0}{2 } \left(\frac{\eta}{\eta_0}\right)^{2 } \right]=n_0\left(\frac{x}{2}\right ) .
\end{aligned}\ ] ] the two modes combine to modify the phase so that the general solution is @xmath329.\end{aligned}\ ] ] the behaviour of the scalar and vector perturbations are therefore easily obtained since @xmath330 , i.e. @xmath331 so that @xmath332 while @xmath333 and @xmath334 .
it follows that @xmath335 and @xmath336 these expressions can then be inserted in eq .
( [ expchi2 ] ) that is valid for @xmath19 to conclude that at late time , the polarisations contribute in the same way to @xmath230 as @xmath337 which is equivalent to our prior result ( [ e444 ] ) .
the result for the @xmath223 polarisation matches the results of ref .
@xcite up to normalisation factors of the transfer functions , but it appears to be different from the conclusions on the @xmath216 polarisation .
asymptotically , the fourier mode is aligned with the preferred direction of the spacetime , and it turns out to be difficult to compare our result with ref .
@xcite given that most expressions of their formalism are singular for a mode along that special direction .
we are confident that our result is correct in that limit since from eq .
( [ limxilate ] ) we find that @xmath338 , which is the perturbation of the square of the weyl , when expressed in terms of the transfer function takes the form @xmath339.\ ] ] this is formally the same expression as for a pure minkowski space - time [ see ( eq . [ eb5 ] ) ] , which is expected when one takes the small scale limit . at early times
, the wave - number behaves as @xmath340 so that @xmath341 , which goes to zero . as in [ 4e ] , we can not neglect the decaying mode .
we can use the expression of @xmath230 obtained in [ 4e ] since it is well - behaved when @xmath342 . if we first consider the growing mode [ i.e. @xmath254 and @xmath255 in eq .
( [ e338 ] ) ] then we deduce that @xmath343= j_0(2x)\\ t_\times^\uparrow & \sim&j_3\left[2 k_\perp\eta_0 \left(\frac{\eta}{\eta_0}\right)^{\frac{1}{2 } } \right ] = j_3(2x ) , \end{aligned}\ ] ] from which it follows that @xmath344 asymptotically , we have from eq .
( [ besselj0 ] ) that @xmath345 since @xmath346 @xmath347 while @xmath348 and @xmath349 and it follows that @xmath350 and @xmath351 with these expressions , we conclude that at leading order , the contributions of the two polarisations to @xmath230 are @xmath352 which both remain bounded . on the other hand , if we consider the decaying mode [ i.e. @xmath263 and @xmath264 in eq .
( [ e338 ] ) ] , i.e. the most diverging mode when @xmath265 , then @xmath353= n_0(2x)\\ t_\times^\downarrow & \sim & n_3\left[2 k_\perp\eta_0 \left(\frac{\eta}{\eta_0}\right)^{\frac{1}{2 } } \right ] = n_3(2x ) , \end{aligned}\ ] ] the asymptotic behaviours of which are @xmath354 it allows to conclude that the contributions of the two polarisations to @xmath230 are @xmath355 as expected both terms diverge when @xmath211 .
the transition between the two asympotic regimes is illustrated on fig .
[ f3 ] for the components of the shear .
while the transition is quite sharp , it is shifted depending on the value of @xmath356 .
then , we can integrate numerically eq .
( [ mu2axial ] ) for different modes .
this exact numerical solution is compared in fig .
[ fcp1axial ] to the two asymptotic forms , which demonstrates that the transition is sharp even for the evolution of the modes . in fig .
[ fcp2axial ] , the full numerical solution of @xmath357 is compared to the two asymptotic behaviours described in [ 5c ] and [ 5d ] .
this shows explicitely the validity of the asymptotic expansions .
we have performed the stability study of a generic kasner spacetime with respect to linear perturbations using the formalism developed in refs . @xcite .
since kasner spacetimes are solutions of the vacuum einstein equations , one expects only two degrees of freedom to propagate , which we explicitely demonstrate .
they can be identified with the two polarisations of gravity waves .
the extra scalar and vector components of the metric are then obtained algebraically from constraint equations .
contrary to a minkowski or de sitter spacetime , they do not strictly vanish because the spatial sections are not isotropic anymore . we have shown that for any fourier mode , the behaviour of the gravity waves interpolates between two asymptotic regimes , at early and late times , in which the mode can be considered almost aligned with a principal axis of the spatial metric .
interestingly , in this limit the gravity wave equation can be integrated analytically .
this allows to set a lower bound on the square of the weyl tensor generated by the perturbations and compare it to the one of the background .
we have concluded that at late time the kasner spacetimes were unstable with respect to linear perturbations but in the particular case of @xmath20 , which corresponds to an axisymmetric configuration , product of a two - dimensional milne spacetime and a two - dimensional euclidean spacetime , which actually maps to one quarter of the minkowski spacetime . at early time , the conclusion depends on the modes included in the analysis .
if one includes decaying modes , the kasner spacetimes are again unstable while stability to linear perturbations is recovered only if the growing modes are excited .
this latter result was already known since ref .
@xcite and our analysis confirms the one of ref .
@xcite that was limited to the particular case of an axisymmetric configuration with @xmath19 .
this result is also compatible with the amplification of gravity waves during the shear dominated era of an anisotropic inflationary phase described by a bianchi i spacetime @xcite .
this is amplification of a remnant of the instability described in this article for a finite kasner era .
this result has some importance concerning the dynamics of the universe close to the singularity . on the one hand ,
the succession of kasner era in the bkl formalism and the rotation of the kasner axes from one era to the other was studied by assuming homogeneous behaviour , i.e. assuming the inhomogeneity scale was larger than the mean hubble patch @xcite .
the instability at early time ( i.e. going toward the singularity ) can alter the conclusions on the approach of the singularity , even though each kasner era has a finite duration .
in particular , as soon as the effect of the gravity wave is large enough , the backreaction would have to be taken into account , and it is not clear that the evolution toward the singularity follows a series of bkl oscillations .
this requires further investigations that go beyond perturbation theory , see e.g. ref
. @xcite . on the other hand ,
the instability at late time could also be relevant in models where the pre - inflationary era is described by an anisotropic era .
the gravity waves modes generated during this era can potentially be observed @xcite if the number of e - folds during inflation is not too large and their presence can modify the onset of inflation .
besides the speculative phenomenology of the early universe , our result sheds some light on the peculiarity of the kasner spacetimes among the class of homogeneous vacuum solutions of einstein equations .
cp is supported by the stfc ( uk ) grant st / h002774/1 , and thanks the royal astronomical society for financial support and the university of cape town for hospitality during part of this work was undertaken .
jpu thanks the pncg for financial support for this project and george ellis for discussions .
it is a pleasure to thank m. peloso and e. gmrkolu for the many comments they gave us when comparing our results .
we also thank francis bernardeau , dick bond and andrei linde for their kind advices .
this work is dedicated to the memory of our friend and colleague lev kofman with whom this project was initiated during his last trip to paris in june 2009 .
there exist several ways to parameterize , the coefficients @xmath358 of the kasner metric . because of the two constraints ( [ eqconstraintspi ] ) , it is in fact a 1-parameter family of triplets that is completely specified by only one number .
three parameterisations have been widely used . * _ single exponant parameterisation . _ as we have seen in [ sec2-a ] , it is sufficient to give one of the kasner exponants to specify the two others .
for instance , fixing the smallest one , @xmath359 , gives that @xmath360 so that @xmath361 , from which we deduce that @xmath362,\ ] ] @xmath363 .\ ] ] the two constraints ( [ eqconstraintspi ] ) allow to deduce that the @xmath10 satisfy the following useful relations .
first , @xmath364 so that @xmath365 also conveniently rewritten as @xmath366 from @xmath367 , one deduces that @xmath368 this tells us that @xmath369 which implies in particular @xmath370 , with similar relations obtained after permutation of the indices @xmath371 . in terms of the coefficient @xmath31 , it leads to the relations @xmath372 in terms of the @xmath373 these relations take the simple form @xmath374 this shows that all the properties of the kasner spacetime , but also the evolution of its perturbations when a mode is aligned with an eigendirection of the shear , are characterized by the choice of a single kasner exponent corresponding to that particular direction . *
_ angular parameterisation . _ as we have seen in [ sec2-a ] , the constraints ( [ eqconstraintspi ] ) imply that the triplet can be specified by a choice of an angle @xmath375 $ ] and its expression is given by eq .
( [ paravarpi ] ) and we have seen that @xmath376 * _ bkl parameterisation . _ refs . @xcite use the parameterisation @xmath377 with @xmath378 . it can be obtained @xcite by solving the two constraints ( [ metricb1 ] ) once having set @xmath379 and the condition @xmath18 imposes that @xmath380 , with @xmath381 corresponding to @xmath19 and @xmath382 to @xmath20 .
the particular case @xmath20 requires special attention .
it corresponds to a spacetime with metric @xmath383 which appears to be the product of a two - dimensional milne space with a two - dimensional euclidean space .
it is the taub representation @xcite of the minkowski metric . with the change of coordinates defined by @xmath384 this metric
is rewritten as @xmath385 i.e. has a minkowski metric but only covers the patch @xmath386 , the upper cone ( @xmath387 ) corresponding to an expanding universe while the lower one ( @xmath388 ) describes a contracting spacetime .
this explains why @xmath389 [ eq . ( [ valc ] ) ] . the choice of the time slicing plays an important role when studying the perturbations
. in minkowskian coordinates the spacetime is explicitely homogeneous and isotropic so that the tensor modes decouple from the other types of perturbations and evolve according to @xmath390 their amplitude is thus constant and the perturbation of the square of the weyl tensor @xmath391 remains constant . under its kasner form , this no more obvious and one needs to check the consistency with the result in minkowskian coordinates @xmath392\ ] ] where the two modes have constant amplitude ( so that the distinction between growing and decaying is no more relevant ) .
we follow the general formalism of this article to study the perturbation in the particular case @xmath393 so that @xmath394 it implies that @xmath395 obviously , this case can be deduced from the case @xmath19 by @xmath396 so that the euler angles are again @xmath397 it follows that the expressions for the shear components are also given by eqs .
( [ s1axial]-[s3axial ] ) but with all signs changed , ie .
@xmath398 etc .
the evolution of the gravity waves is dictated by eq .
( [ mu2axial ] ) with @xmath399 @xmath400 and @xmath401 the only difference concerns the euler angle @xmath273 which is now given by @xmath402 in the limit @xmath403 , @xmath404 so that @xmath405 .
this implies that @xmath406 @xmath407 , @xmath408 and @xmath409 so that we conclude @xmath410 the solution of the gravity wave equation is then explicitely given by @xmath411,\end{aligned}\ ] ] for the two polarisations , and where the growing and decaying modes are combined together , leading to a redefinition of the phase . since @xmath412
, we can not use the variable @xmath153 anymore and thus work directly with @xmath284 . in the limit @xmath236
, we obtain @xmath413\nonumber\\ & \to&\frac{24 x^3 } { \pi \eta^4 s^4}\propto\eta^{-\frac{3}{2}}.\end{aligned}\ ] ] it follows that @xmath414 decays at late time so that the spacetime is stable against the perturbations . note
that this is the same behaviour as for the case @xmath215 , but the main difference arises from the fact that @xmath412 since in the general case @xmath284 decreases with time but slower than @xmath148 so that the effect of the perturbations dominate over the background quantity at late time .
the stability of the taub metric ( [ eqtaubmetric ] ) was studied in ref .
@xcite with the same conclusions as our analysis .
the differential equation @xmath415 u = 0\ ] ] has the general solution @xmath416,\ ] ] where @xmath417 is a linear combination of a bessel function of the first kind ( @xmath418 ) and of the second kind ( @xmath419 or newmann function ) .
gibbons , `` the entropy and stability of the universe '' , nucl .
b * 292 * , 784 ( 1987 ) ; g.w .
gibbons , `` sobolevs inequality , jensens theorem and the mass and entropy of the universe '' , nucl .
b * 310 * , 636 ( 1988 ) . | this article investigates the stability of a generic kasner spacetime to linear perturbations , both at late and early times .
it demonstrates that the perturbation of the weyl tensor diverges at late time in all cases but in the particular one in which the kasner spacetime is the product of a two - dimensional milne spacetime and a two - dimensional euclidean space . at early times
, the perturbation of the weyl tensor also diverges unless one imposes a condition on the perturbations so as to avoid the most divergent modes to be excited . |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.